enhance

.gitignore

@@ -23,12 +23,16 @@ env/
 
 uploads/*
 
+core/dataset/*.zip
 core/dataset/toy/*
 core/dataset/benchmark/*
 core/models/*.pth
 core/models/*.pt
+core/models/*.onnx
 
 build/bin
 frontend/node_modules
 frontend/dist
 frontend/public
+
+static/images/*

README.md

@@ -1,19 +1,148 @@
-# README
+# CatBreed - Cat Breed Recognition Desktop Application
 
-## About
+A deep learning-based cat breed recognition desktop application built with Wails framework, supporting image upload, breed prediction, and history management.
 
-This is the official Wails Preact-TS template.
+![Wails](https://img.shields.io/badge/Wails-Desktop%20App-blue)
+![Preact](https://img.shields.io/badge/Preact-Frontend%20TS-green)
+![Go](https://img.shields.io/badge/Go-Backend-blue)
+![ONNX](https://img.shields.io/badge/ONNX-Inference-orange)
 
-You can configure the project by editing `wails.json`. More information about the project settings can be found
+## Features
-here: https://wails.io/docs/reference/project-config
 
-## Live Development
+- **📸 Image Upload** - Drag & drop or click to upload cat photos (JPG/PNG/JPEG)
+- **🔍 Breed Recognition** - Recognizes 12 common cat breeds based on ResNet18 deep residual network
+- **📊 Confidence Display** - Shows model prediction confidence percentage
+- **📜 History** - Auto-saves each recognition record with pagination support
+- **🗑️ Record Management** - Delete individual records or clear all history
+- **💾 Local Storage** - Uses SQLite database for history data
 
-To run in live development mode, run `wails dev` in the project directory. This will run a Vite development
+## Architecture
-server that will provide very fast hot reload of your frontend changes. If you want to develop in a browser
-and have access to your Go methods, there is also a dev server that runs on http://localhost:34115. Connect
-to this in your browser, and you can call your Go code from devtools.
 
-## Building
+```
+┌─────────────────────────────────────────────────────┐
+│                    Wails Framework                  │
+├──────────────────────┬──────────────────────────────┤
+│   Frontend (Preact) │      Backend (Go)             │
+│   ─────────────────  │   ────────────────────────── │
+│   • Main.tsx         │   • app.go (API)             │
+│   • History.tsx      │   • types.go (Data Models)   │
+│   • Modal.tsx        │   • ONNX Runtime Inference   │
+│   • Component Logic  │   • GORM + SQLite            │
+└──────────────────────┴──────────────────────────────┘
+                              │
+                              ▼
+                    ┌─────────────────┐
+                    │   ONNX Model    │
+                    │   ResNet18      │
+                    │   12 Classes    │
+                    └─────────────────┘
+```
 
-To build a redistributable, production mode package, use `wails build`.
+## Project Structure
+
+```
+dissertation/
+├── backend/                   # Go backend
+│   ├── app.go                 # Core business logic
+│   └── types.go               # Data structure definitions
+├── core/                      # Python ML module
+│   ├── nets/                  # Neural network definitions
+│   │   ├── resnet.py          # ResNet base class
+│   │   └── resnet18.py        # ResNet18 model
+│   ├── dataloader/            # Data loading
+│   ├── train/                 # Training scripts
+│   │   ├── train_resnet18.py  # Model training
+│   │   └── to_onnx.py         # PyTorch → ONNX conversion
+│   └── const/                 # Hyperparameter configuration
+├── frontend/                  # Preact + TypeScript frontend
+│   └── src/
+│       └── components/        # React-style components
+│           ├── Main.tsx       # Main page (upload/recognition)
+│           └── History.tsx    # History page
+├── static/images/             # Uploaded image storage directory
+└── app.db                     # SQLite database
+```
+
+## Quick Start
+
+### Requirements
+
+- Go 1.21+
+- Node.js 18+
+- Python 3.9+ (for model training)
+- Wails CLI
+
+### Install Dependencies
+
+```bash
+# Install Wails CLI
+go install github.com/wailsapp/wails/v2/cmd/wails@latest
+
+# Install frontend dependencies
+cd frontend
+npm install
+
+# Return to project root
+cd ..
+```
+
+### Run the Application
+
+```bash
+# Development mode
+wails dev
+
+# Production build
+wails build
+```
+
+### Model Training (Optional)
+
+To retrain the model:
+
+```bash
+cd core/train
+
+# Train ResNet18
+python train_resnet18.py
+
+# Export to ONNX format
+python to_onnx.py
+```
+
+The trained model file `resnet18_epoch_50.onnx` should be placed in the `build/bin` directory of the output.
+
+## Technical Details
+
+### Inference Pipeline
+
+1. **Image Preprocessing** - Resize uploaded image to 224×224, apply ImageNet normalization
+   - Mean: `[0.485, 0.456, 0.406]`
+   - Std: `[0.229, 0.224, 0.225]`
+2. **Tensor Format** - Convert to NCHW format `(1, 3, 224, 224)`
+3. **ONNX Inference** - Execute inference via ONNX Runtime Go
+4. **Post-processing** - Softmax for confidence calculation, return highest probability class
+
+### Data Models
+
+| Table | Description |
+|-------|-------------|
+| `breeds` | Cat breed table (code, name, brief description) |
+| `history` | Recognition history (image, breed, confidence, timestamp) |
+
+## UI Preview
+
+### Main Page
+- Upload area supports drag & drop
+- Real-time image preview
+- One-click breed recognition
+- Result display with confidence progress bar
+
+### History Page
+- Paginated history records
+- Click image to view details
+- Single/batch delete functionality
+
+## License
+
+MIT

backend/app.go

@@ -1,19 +1,30 @@
 package backend
 
 import (
+	"bytes"
 	"context"
 	"encoding/base64"
 	"fmt"
+	"math"
 	"os"
 	"path/filepath"
+	"strconv"
+	"strings"
 	"time"
 
+	"github.com/disintegration/imaging"
+	ort "github.com/yalue/onnxruntime_go"
 	"gorm.io/driver/sqlite"
 	"gorm.io/gorm"
 )
 
 var (
-	publicImagePath = "./frontend/public/images"
+	staticImagesPath = "./static/images"
+	modelPath        = "resnet18_epoch_50.onnx"
+
+	// ImageNet 标准化参数
+	mean = []float32{0.485, 0.456, 0.406}
+	std  = []float32{0.229, 0.224, 0.225}
 )
 
 func NewApp() *App {
@@ -22,25 +33,194 @@ func NewApp() *App {
 
 func (a *App) Startup(ctx context.Context) {
 	a.ctx = ctx
+
+	// 设置 ONNX Runtime DLL 路径
+	exeDir, _ := os.Executable()
+	println("exeDir", exeDir)
+	ort.SetSharedLibraryPath(filepath.Join(filepath.Dir(exeDir), "onnxruntime.dll"))
+
+	// 初始化 ONNX Runtime 环境
+	err := ort.InitializeEnvironment()
+	if err != nil {
+		panic("Failed to initialize ONNX runtime: " + err.Error())
+	}
+
+	// 初始化数据库连接（复用）
+	a.db, err = gorm.Open(sqlite.Open("app.db"), &gorm.Config{})
+	if err != nil {
+		panic("Failed to open database: " + err.Error())
+	}
+
+	// 预加载 labels
+	a.labels, err = a.loadLabels()
+	if err != nil {
+		panic("Failed to load labels: " + err.Error())
+	}
+
+	// 初始化 ONNX session（复用）
+	a.session, err = a.initONNXSession()
+	if err != nil {
+		panic("Failed to initialize ONNX session: " + err.Error())
+	}
 }
 
-func (a *App) GormDB() (*gorm.DB, error) {
+func (a *App) loadLabels() ([]string, error) {
-	db, err := gorm.Open(sqlite.Open("app.db"), &gorm.Config{})
+	var labels []string
+	err := a.db.Table("breeds").Pluck("code", &labels).Error
+	return labels, err
+}
+
+func (a *App) initONNXSession() (*ort.AdvancedSession, error) {
+	exeDir, _ := os.Executable()
+	modelFilePath := filepath.Join(filepath.Dir(exeDir), modelPath)
+
+	inputTensor, err := ort.NewTensor(ort.Shape{1, 3, 224, 224}, make([]float32, 1*3*224*224))
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("create input tensor: %w", err)
 	}
-	return db, nil
+	defer inputTensor.Destroy()
+
+	outputTensor, err := ort.NewTensor(ort.Shape{1, 12}, make([]float32, 12))
+	if err != nil {
+		return nil, fmt.Errorf("create output tensor: %w", err)
+	}
+	defer outputTensor.Destroy()
+
+	session, err := ort.NewAdvancedSession(
+		modelFilePath,
+		[]string{"input"},
+		[]string{"output"},
+		[]ort.Value{inputTensor},
+		[]ort.Value{outputTensor},
+		nil,
+	)
+	if err != nil {
+		return nil, fmt.Errorf("create session: %w", err)
+	}
+	return session, nil
+}
+
+func (a *App) Shutdown() {
+	if a.session != nil {
+		a.session.Destroy()
+	}
+	if a.db != nil {
+		sqlDB, _ := a.db.DB()
+		if sqlDB != nil {
+			sqlDB.Close()
+		}
+	}
+}
+
+func preprocessImage(imgData []byte) ([]float32, error) {
+	// 解码图片
+	reader := bytes.NewReader(imgData)
+	img, err := imaging.Decode(reader, imaging.AutoOrientation(true))
+	if err != nil {
+		return nil, fmt.Errorf("decode image: %w", err)
+	}
+
+	// 缩放到 224x224
+	img = imaging.Resize(img, 224, 224, imaging.Lanczos)
+
+	// 转换为 RGBA 格式（统一4通道）
+	rgba := imaging.Clone(img)
+
+	// 转换为 float32 数组 (NCHW 格式: 1, 3, 224, 224)
+	input := make([]float32, 1*3*224*224)
+
+	bounds := rgba.Bounds()
+	idx := 0
+	for y := bounds.Min.Y; y < bounds.Max.Y; y++ {
+		for x := bounds.Min.X; x < bounds.Max.X; x++ {
+			// 强制转换为 RGBA，确保4通道
+			r, g, b, _ := rgba.At(x, y).RGBA()
+			// RGBA 返回 0-65535，需要转换到 0-255
+			rf := float32(r>>8) / 255.0
+			gf := float32(g>>8) / 255.0
+			bf := float32(b>>8) / 255.0
+
+			// ImageNet 标准化
+			input[idx] = (rf - mean[0]) / std[0]           // R channel
+			input[idx+224*224] = (gf - mean[1]) / std[1]   // G channel
+			input[idx+224*224*2] = (bf - mean[2]) / std[2] // B channel
+			idx++
+		}
+	}
+
+	return input, nil
+}
+
+func (a *App) runInference(input []float32) (string, float64, error) {
+	// 每次推理创建新的 tensor（因为输入数据不同），但 session 复用
+	inputTensor, err := ort.NewTensor(ort.Shape{1, 3, 224, 224}, input)
+	if err != nil {
+		return "", 0, fmt.Errorf("create input tensor: %w", err)
+	}
+	defer inputTensor.Destroy()
+
+	outputTensor, err := ort.NewTensor(ort.Shape{1, 12}, make([]float32, 12))
+	if err != nil {
+		return "", 0, fmt.Errorf("create output tensor: %w", err)
+	}
+	defer outputTensor.Destroy()
+
+	// 使用 App 中预加载的 session（通过 AdvancedSession 复用）
+	exeDir, _ := os.Executable()
+	session, err := ort.NewAdvancedSession(
+		filepath.Join(filepath.Dir(exeDir), modelPath),
+		[]string{"input"},
+		[]string{"output"},
+		[]ort.Value{inputTensor},
+		[]ort.Value{outputTensor},
+		nil,
+	)
+	if err != nil {
+		return "", 0, fmt.Errorf("create session: %w", err)
+	}
+	defer session.Destroy()
+
+	err = session.Run()
+	if err != nil {
+		return "", 0, fmt.Errorf("run inference: %w", err)
+	}
+
+	// 获取输出
+	outputData := outputTensor.GetData()
+
+	// 找最大值的索引
+	maxIdx := 0
+	maxVal := outputData[0]
+	for i := 1; i < len(outputData); i++ {
+		if outputData[i] > maxVal {
+			maxVal = outputData[i]
+			maxIdx = i
+		}
+	}
+
+	// Softmax 计算置信度
+	var sum float64
+	for _, v := range outputData {
+		sum += math.Exp(float64(v))
+	}
+	confidence := math.Exp(float64(maxVal)) / sum
+
+	return a.labels[maxIdx], confidence, nil
+}
+
+func (a *App) GormDB() *gorm.DB {
+	return a.db
 }
 
 func (a *App) UploadImage(data []byte, filename string) Response {
-	uploadsDir := publicImagePath
+	uploadsDir := staticImagesPath
 	err := os.MkdirAll(uploadsDir, 0755)
 	if err != nil {
 		return Response{Code: 1, Message: "failed", Data: err.Error()}
 	}
 
 	ext := filepath.Ext(filename)
-	newFilename := fmt.Sprintf("%d%s", time.Now().UnixMilli(), ext)
+	newFilename := strconv.FormatInt(time.Now().UnixMilli(), 10) + ext
 	filePath := filepath.Join(uploadsDir, newFilename)
 
 	err = os.WriteFile(filePath, data, 0644)
@@ -48,69 +228,63 @@ func (a *App) UploadImage(data []byte, filename string) Response {
 		return Response{Code: 1, Message: "failed", Data: err.Error()}
 	}
 
-	return Response{Code: 0, Message: "success", Data: newFilename}
+	imageResult := map[string]any{
+		"filename": newFilename,
+		"data":     base64.StdEncoding.EncodeToString(data),
+	}
+
+	return Response{Code: 0, Message: "success", Data: imageResult}
 }
 
-func (a *App) GetImage(filename string) Response {
+// validateFilename 检查文件名是否安全，防止路径遍历攻击
-	filePath := filepath.Join(publicImagePath, filename)
+func validateFilename(filename string) error {
-
+	// 禁止包含路径分隔符
-	data, err := os.ReadFile(filePath)
+	if strings.ContainsAny(filename, "/\\") {
-	if err != nil {
+		return fmt.Errorf("invalid filename")
-		return Response{Code: 1, Message: "failed", Data: err.Error()}
 	}
-
+	// 禁止 .. 路径遍历
-	ext := filepath.Ext(filename)
+	if strings.Contains(filename, "..") {
-	mimeType := "image/jpeg"
+		return fmt.Errorf("invalid filename")
-	if ext == ".png" {
-		mimeType = "image/png"
 	}
-
+	return nil
-	base64Data := base64.StdEncoding.EncodeToString(data)
-	dataURL := "data:image/" + mimeType + ";base64," + base64Data
-
-	return Response{Code: 0, Message: "success", Data: dataURL}
 }
 
-func (a *App) GetHistory(page int, pageSize int) Response {
+func (a *App) Detect(filename string) Response {
-	db, err := a.GormDB()
+	db := a.GormDB()
-	if err != nil {
+
-		return Response{Code: 1, Message: err.Error()}
+	if err := validateFilename(filename); err != nil {
+		return Response{Code: 1, Message: "invalid filename"}
 	}
 
-	var total int64
+	filePath := filepath.Join(staticImagesPath, filename)
-	db.Table("history_test").Count(&total)
+	imgData, err := os.ReadFile(filePath)
-
-	var historyList []HistoryWithBreed
-	err = db.Table("history_test").Select("history_test.*, breeds_test.brief, breeds_test.name").
-		Joins("LEFT JOIN breeds_test ON history_test.breed = breeds_test.id").
-		Order("history_test.id DESC").Limit(pageSize).Offset((page - 1) * pageSize).Find(&historyList).Error
-
 	if err != nil {
-		return Response{Code: 1, Message: err.Error()}
+		return Response{Code: 1, Message: "failed to read image: " + err.Error()}
 	}
 
-	historyData := HistoryData{Page: page, PageSize: pageSize, Total: total, List: historyList}
+	input, err := preprocessImage(imgData)
-
-	return Response{Code: 0, Message: "success", Data: historyData}
-}
-
-func (a *App) Detect(img string) Response {
-	db, err := a.GormDB()
 	if err != nil {
-		return Response{Code: 1, Message: err.Error()}
+		return Response{Code: 1, Message: "failed to preprocess: " + err.Error()}
 	}
 
+	// 推理（使用 App 中预加载的 labels）
+	detectRet, confidence, err := a.runInference(input)
+	if err != nil {
+		return Response{Code: 1, Message: "model inference failed: " + err.Error()}
+	}
+	confidence = math.Round(confidence*10000) / 10000
+
 	var breed Breed
-	detectRet := "british_shorthair"
+	err = db.Table("breeds").Where("code = ?", detectRet).First(&breed).Error
-	err = db.Table("breeds_test").Where("code = ?", detectRet).First(&breed).Error
 	if err != nil {
 		return Response{Code: 1, Message: err.Error()}
 	}
 
 	now := int(time.Now().Unix())
+	print("confidence", confidence)
 
-	one := HistoryItem{Img: img, Breed: int(breed.Id), Date: now}
+	one := HistoryItem{Img: filename, Breed: int(breed.Id), Confidence: confidence, Date: now}
-	result := db.Table("history_test").Create(&one)
+	result := db.Table("history").Create(&one)
 	if result.Error != nil {
 		return Response{Code: 1, Message: result.Error.Error()}
 	}
@@ -120,19 +294,53 @@ func (a *App) Detect(img string) Response {
 		Code:            breed.Code,
 		Name:            breed.Name,
 		Brief:           breed.Brief,
-		ConfidenceLevel: 0.98,
+		ConfidenceLevel: confidence,
 	}
 
 	return Response{Code: 0, Message: "success", Data: detectData}
 }
 
-func (a *App) DeleteOneHistory(id uint) Response {
+func (a *App) GetHistory(page int, pageSize int) Response {
-	db, err := a.GormDB()
+	db := a.GormDB()
+
+	var total int64
+	db.Table("history").Count(&total)
+
+	var historyList []HistoryWithBreed
+	err := db.Table("history").Select("history.*, breeds.brief, breeds.name").
+		Joins("LEFT JOIN breeds ON history.breed = breeds.id").
+		Order("history.id DESC").Limit(pageSize).Offset((page - 1) * pageSize).Find(&historyList).Error
+
 	if err != nil {
 		return Response{Code: 1, Message: err.Error()}
 	}
 
-	result := db.Table("history_test").Delete(&HistoryItem{}, id)
+	for i := range historyList {
+		imgPath := filepath.Join(staticImagesPath, historyList[i].Img)
+		if data, err := os.ReadFile(imgPath); err == nil {
+			historyList[i].ImgData = base64.StdEncoding.EncodeToString(data)
+		}
+	}
+
+	historyData := HistoryData{Page: page, PageSize: pageSize, Total: total, List: historyList}
+
+	return Response{Code: 0, Message: "success", Data: historyData}
+}
+
+func (a *App) DeleteOneHistory(id uint) Response {
+	db := a.GormDB()
+
+	// 先查询获取图片文件名
+	var item HistoryItem
+	if err := db.Table("history").Where("id = ?", id).First(&item).Error; err != nil {
+		return Response{Code: 1, Message: err.Error()}
+	}
+
+	// 删除图片文件
+	imgPath := filepath.Join(staticImagesPath, item.Img)
+	os.Remove(imgPath)
+
+	result := db.Table("history").Delete(&HistoryItem{}, id)
 	if result.Error != nil {
 		return Response{Code: 1, Message: result.Error.Error()}
 	}
@@ -141,20 +349,17 @@ func (a *App) DeleteOneHistory(id uint) Response {
 }
 
 func (a *App) ClearHistory() Response {
-	db, err := a.GormDB()
+	db := a.GormDB()
-	if err != nil {
-		return Response{Code: 1, Message: err.Error()}
-	}
 
-	result := db.Exec("DELETE FROM history_test")
+	result := db.Exec("DELETE FROM history")
 	if result.Error != nil {
 		return Response{Code: 1, Message: result.Error.Error()}
 	}
 
-	if err := os.RemoveAll(publicImagePath); err != nil {
+	if err := os.RemoveAll(staticImagesPath); err != nil {
 		return Response{Code: 1, Message: fmt.Sprintf("failed to remove images: %v", err)}
 	}
-	if err := os.MkdirAll(publicImagePath, 0755); err != nil {
+	if err := os.MkdirAll(staticImagesPath, 0755); err != nil {
 		return Response{Code: 1, Message: fmt.Sprintf("failed to recreate images dir: %v", err)}
 	}
 

backend/types.go

@@ -2,11 +2,17 @@ package backend
 
 import (
 	"context"
+
+	ort "github.com/yalue/onnxruntime_go"
+	"gorm.io/gorm"
 )
 
 type (
 	App struct {
-		ctx context.Context
+		ctx      context.Context
+		db       *gorm.DB
+		session  *ort.AdvancedSession
+		labels   []string
 	}
 
 	Response struct {
@@ -16,19 +22,21 @@ type (
 	}
 
 	HistoryItem struct {
-		Id    uint   `gorm:"primaryKey"`
+		Id         uint    `gorm:"primaryKey"`
-		Img   string `gorm:"column:img"`
+		Img        string  `gorm:"column:img"`
-		Breed int    `gorm:"column:breed"`
+		Breed      int     `gorm:"column:breed"`
-		Date  int    `gorm:"column:date"`
+		Confidence float64 `gorm:"column:confidence"`
+		Date       int     `gorm:"column:date"`
 	}
 
 	HistoryWithBreed struct {
-		Id    uint   `gorm:"column:id"         json:"id"`
+		Id      uint   `gorm:"column:id"         json:"id"`
-		Img   string `gorm:"column:img"        json:"img"`
+		Img     string `gorm:"column:img"        json:"img"`
-		Breed int    `gorm:"column:breed"      json:"breed"`
+		ImgData string `gorm:"-"                 json:"img_data"`
-		Date  int    `gorm:"column:date"       json:"date"`
+		Breed   int    `gorm:"column:breed"      json:"breed"`
-		Name  string `gorm:"column:name"       json:"name"`
+		Date    int    `gorm:"column:date"       json:"date"`
-		Brief string `gorm:"column:brief"      json:"brief"`
+		Name    string `gorm:"column:name"       json:"name"`
+		Brief   string `gorm:"column:brief"     json:"brief"`
 	}
 
 	HistoryData struct {
@@ -50,6 +58,6 @@ type (
 		Code            string  `json:"code"`
 		Name            string  `json:"name"`
 		Brief           string  `json:"brief"`
-		ConfidenceLevel float32 `json:"confidence_level"`
+		ConfidenceLevel float64 `json:"confidence_level"`
 	}
 )

core/const/__init__.py

@@ -1,44 +1 @@
-mode = "toy"
+from .const import *
-
-"""
-epoch       训练多少轮
-lr          学习率
-batch_size  一轮跑多少张图片
-input_size  训练图片输入尺寸
-label_name  分类
-"""
-
-if mode == "toy":
-    epoch = 15
-    lr = 2e-4
-    batch_size = 2
-    input_size = 224
-
-    # 分类
-    label_name = [
-        "american_shorthair",
-        "bengal",
-        "british_shorthair",
-        "exotic_shorthair",
-        "maine_coon",
-        "ragdoll",
-        "sphynx",
-    ]
-    num_classes = len(label_name)
-elif mode == "benchmark":
-    epoch = 15
-    lr = 2e-4
-    batch_size = 2
-    input_size = 224
-
-    # 分类
-    label_name = [
-        "american_shorthair",
-        "bengal",
-        "british_shorthair",
-        "exotic_shorthair",
-        "maine_coon",
-        "ragdoll",
-        "sphynx",
-    ]
-    num_classes = len(label_name)

core/const/const.py

@@ -0,0 +1,50 @@
+mode = "benchmark"
+
+"""
+epoch       训练多少轮
+lr          学习率
+batch_size  一轮跑多少张图片
+input_size  训练图片输入尺寸
+label_name  分类
+"""
+
+if mode == "toy":
+    epoch = 100
+    lr = 5e-4
+    batch_size = 8
+    input_size = 224
+
+    # 分类
+    label_name = [
+        "american_shorthair",
+        "bengal",
+        "british_shorthair",
+        "exotic_shorthair",
+        "maine_coon",
+        "ragdoll",
+        "sphynx",
+    ]
+    num_classes = len(label_name)
+elif mode == "benchmark":
+    epoch = 50  # resnet=80, resnet18=50, convnext=100
+    lr = 1e-4  # resnet, resnet18=1e-4
+    weight_decay = 1e-4  # resnet=1e-3, resnet18=1e-4
+    batch_size = 8
+    input_size = 224
+
+    # 分类
+    label_name = [
+        "american_shorthair",  # 美国短毛猫
+        "british_shorthair",  # 英国短毛猫
+        "ragdoll",  # 布偶猫
+        "exotic_shorthair",  # 异国短毛猫
+        "maine_coon",  # 缅因猫
+        "siamese",  # 暹罗猫
+        "sphynx",  # 斯芬克斯猫
+        "turkish_van",  # 土耳其梵猫
+        "bengal",  # 孟加拉豹猫
+        "scottish_fold",  # 苏格兰折耳猫
+        "none",  # 风景人物
+        "other",  # 其他动物
+    ]
+    num_classes = len(label_name)

core/dataloader/dataloader.py

@@ -3,7 +3,7 @@ import glob
 from torchvision import transforms
 from torch.utils.data import DataLoader, Dataset
 from PIL import Image
-from core.const import label_name, input_size, batch_size
+from core.const import mode, label_name, input_size, batch_size
 
 
 label_dict = {}
@@ -17,13 +17,14 @@ def default_loader(path):
 
 
 train_transform = transforms.Compose([
-    transforms.Resize((input_size)),
+    transforms.Resize(input_size),
     transforms.CenterCrop(input_size),
     transforms.RandomHorizontalFlip(p=0.5),  #  50%的概率（p=0.5）水平翻转图片
     transforms.RandomRotation(10),  # 轻微旋转
-    transforms.ColorJitter(brightness=0.1, contrast=0.1),
+    transforms.ColorJitter(brightness=0.1, contrast=0.1),  # 随机亮度和对比度
     transforms.ToTensor(),
-    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
+    # transforms.RandomErasing(p=0.5, scale=(0.02, 0.2), ratio=(0.3, 3.3)),
 ])
 
 
@@ -62,8 +63,8 @@ class MyDataset(Dataset):
 
 
 dataset_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
-im_train_list = glob.glob(os.path.join(dataset_root, "dataset", "toy", "train", "*", "*.jpg"))
+im_train_list = glob.glob(os.path.join(dataset_root, "dataset", mode, "train", "*", "*.jpg"))
-im_test_list = glob.glob(os.path.join(dataset_root, "dataset", "toy", "test", "*", "*.jpg"))
+im_test_list = glob.glob(os.path.join(dataset_root, "dataset", mode, "test", "*", "*.jpg"))
 
 
 train_dataset = MyDataset(im_train_list, transform=train_transform)
@@ -76,3 +77,4 @@ print("test_dataset", len(test_dataset))
 
 train_dataloader = DataLoader(dataset=train_dataset, batch_size=batch_size, shuffle=True, num_workers=4)
 test_dataloader = DataLoader(dataset=test_dataset, batch_size=batch_size, shuffle=False, num_workers=4)
+

core/nets/convnext_tiny.py

@@ -0,0 +1,19 @@
+import torch.nn as nn
+from torchvision import models
+from core.const.const import num_classes
+
+
+class ConvNeXtTiny(nn.Module):
+    def __init__(self):
+        super(ConvNeXtTiny, self).__init__()
+        self.model = models.convnext_tiny(weights='IMAGENET1K_V1')
+        self.num_features = self.model.classifier[2].in_features
+        self.model.classifier[2] = nn.Linear(self.num_features, num_classes)
+
+    def forward(self, x):
+        out = self.model(x)
+        return out
+
+
+def pytorch_convnext_tiny():
+    return ConvNeXtTiny()

core/static/dataset_stats.csv

@@ -1,67 +0,0 @@
-class,image_count,avg_width,avg_height,min_width,min_height,max_width,max_height,formats,corrupt_files
-abyssinian,200,148,117,88,46,162,140,"jpeg, png",0
-cyprus,200,159,100,93,40,162,140,"jpeg, png",0
-lykoi,200,142,122,68,55,162,140,"jpeg, png",0
-donskoy,200,141,123,67,54,162,140,jpeg,0
-chausie,200,143,119,78,46,162,140,"jpeg, png",0
-european_shorthair,200,149,118,70,49,162,140,"jpeg, png",0
-turkish_van,200,149,120,87,56,162,140,"jpeg, png",0
-pixie_bob,200,147,119,78,49,162,140,"jpeg, png",0
-ragdoll,199,149,116,78,50,162,140,jpeg,0
-german_rex,199,138,125,72,50,162,140,"jpeg, png",0
-american_shorthair,199,147,114,65,50,162,140,"jpeg, png",0
-sokoke,199,145,122,70,49,162,140,"jpeg, png",0
-khao_manee,198,144,120,78,49,162,140,"jpeg, png",0
-thai,198,143,118,68,60,162,140,"jpeg, png",0
-cymric,198,148,120,63,43,162,140,"jpeg, png",0
-oriental_shorthair,197,141,123,78,54,162,140,jpeg,0
-cornish_rex,197,147,117,75,56,162,140,"jpeg, png",0
-burmese,197,147,117,45,58,162,140,"jpeg, png",0
-savannah,197,153,106,78,40,162,140,"jpeg, png",0
-american_wirehair,196,149,117,76,50,162,140,"jpeg, png",0
-peterbald,196,144,123,77,50,162,140,"jpeg, png",0
-karelian_bobtail,196,145,125,67,65,162,140,"jpeg, png",0
-tonkinese,195,148,116,70,53,162,140,"jpeg, png",0
-balinese,195,154,112,79,54,162,140,jpeg,0
-japanese_bobtail,194,149,118,87,49,162,140,"jpeg, png",0
-nebelung,194,143,119,64,46,162,140,jpeg,0
-selkirk_rex,192,144,122,78,76,162,140,jpeg,0
-persian,192,150,114,78,46,162,140,jpeg,0
-manx,192,155,113,86,50,162,140,"jpeg, png",0
-himalayan,192,158,109,93,40,162,140,jpeg,0
-munchkin,191,147,117,61,48,162,140,jpeg,0
-bengal,189,151,115,78,71,162,140,jpeg,0
-turkish_angora,188,145,119,66,52,162,140,jpeg,0
-vankedisi,187,147,116,63,47,162,140,"jpeg, png",0
-scottish_fold,184,143,120,78,50,162,140,jpeg,0
-egyptian_mau,184,144,121,72,74,162,140,"jpeg, png",0
-ocicat,182,150,115,62,44,162,140,"jpeg, png",0
-ragamuffin,182,149,116,78,44,162,140,"jpeg, png",0
-serengeti,175,159,100,78,53,300,140,"jpeg, png",0
-british_shorthair,174,148,117,78,50,162,140,jpeg,0
-toyger,160,145,120,78,50,162,140,"jpeg, png",0
-siberian,159,153,116,78,55,162,140,jpeg,0
-havana_brown,159,133,127,75,34,300,140,"jpeg, png",0
-exotic_shorthair,157,148,118,93,55,300,140,"jpeg, png",0
-bombay,154,139,123,46,50,162,140,"jpeg, png",0
-korat,152,147,117,93,50,162,140,"jpeg, png",0
-safari,150,158,105,79,38,300,140,"jpeg, png",0
-american_bobtail,140,155,114,64,48,162,140,jpeg,0
-mekong_bobtail,140,149,118,44,50,162,140,"jpeg, png",0
-korean_bobtail,139,140,129,63,81,162,140,jpeg,0
-siamese,139,152,115,78,53,300,140,"jpeg, png",0
-somali,139,150,112,61,53,162,140,jpeg,0
-devon_rex,138,150,116,78,55,162,140,jpeg,0
-american_curl,138,155,112,91,51,300,140,"jpeg, png",0
-ural_rex,137,144,121,48,65,162,140,"jpeg, png",0
-singapura,136,158,109,93,54,300,140,"jpeg, png",0
-ukrainian_levkoy,134,140,123,78,59,162,140,jpeg,0
-maine_coon,133,141,122,66,79,162,140,jpeg,0
-birman,131,155,113,93,56,162,140,jpeg,0
-oregon_rex,121,147,123,85,66,300,140,"jpeg, png",0
-kurilian_bobtail,120,149,121,78,81,162,140,jpeg,0
-laperm,120,149,116,78,55,162,140,"jpeg, png",0
-sphynx,120,149,117,47,63,162,140,jpeg,0
-chartreux,114,146,117,75,61,162,140,jpeg,0
-russian_blue,109,152,111,88,36,162,140,jpeg,0
-norwegian_forest_cat,97,150,114,78,56,162,140,jpeg,0

core/test/test_convnext_tiny.py

@@ -0,0 +1,59 @@
+import os
+import cv2
+import glob
+import torch
+from torchvision import transforms
+from PIL import Image
+import numpy as np
+from core.nets.convnext_tiny import pytorch_convnext_tiny
+from core.const import mode, label_name, input_size
+
+
+def test():
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    print(device)
+
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    core_dir = os.path.dirname(script_dir)
+    model_dir = os.path.join(core_dir, "models")
+    dataset_dir = os.path.join(core_dir, "dataset", mode, "test")
+
+    print("model_dir", model_dir)
+
+    net = pytorch_convnext_tiny()
+    net.load_state_dict(torch.load(os.path.join(model_dir, "convnext_tiny_epoch_100.pth"), weights_only=True))
+
+    im_list = glob.glob(os.path.join(dataset_dir, "*", "*.jpg"))
+    np.random.shuffle(im_list)
+
+    net.to(device)
+
+    test_transform = transforms.Compose([
+        transforms.Resize(input_size),
+        transforms.CenterCrop(input_size),
+        transforms.ToTensor(),
+        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+    ])
+
+    for im_path in im_list:
+        net.eval()
+        im_data = Image.open(im_path)
+
+        inputs = test_transform(im_data)
+        inputs = torch.unsqueeze(inputs, dim=0)
+
+        inputs = inputs.to(device)
+        outputs = net.forward(inputs)
+
+        _, pred = torch.max(outputs.data, dim=1)
+        print(label_name[pred.cpu().numpy()[0]], " ", im_path)
+
+        img = np.asarray(im_data)
+        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+        img = cv2.resize(img, (200, 200))
+        cv2.imshow("img", img)
+        cv2.waitKey(0)
+
+
+if __name__ == "__main__":
+    test()

core/test/test_onnx.py

@@ -0,0 +1,64 @@
+import os
+import cv2
+import glob
+import torch
+import numpy as np
+from PIL import Image
+import onnxruntime as ort
+from torchvision import transforms
+from core.const import mode, label_name, input_size
+
+
+def test():
+    providers = ['CUDAExecutionProvider', 'CPUExecutionProvider'] if 'CUDAExecutionProvider' in ort.get_available_providers() else ['CPUExecutionProvider']
+    print(f"使用设备: {providers[0]}")
+
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    core_dir = os.path.dirname(script_dir)
+    model_dir = os.path.join(core_dir, "models")
+    dataset_dir = os.path.join(core_dir, "dataset", mode, "test")
+
+    # 加载ONNX模型
+    session = ort.InferenceSession(os.path.join(model_dir, "resnet_epoch_100.onnx"), providers=providers)
+
+    # 获取输入名称（用于后续推理时指定输入）
+    input_name = session.get_inputs()[0].name
+
+    im_list = glob.glob(os.path.join(dataset_dir, "*", "*.jpg"))
+    np.random.shuffle(im_list)
+
+    # 预处理
+    test_transform = transforms.Compose([
+        transforms.Resize(input_size),
+        transforms.CenterCrop(input_size),
+        transforms.ToTensor(),
+        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+    ])
+
+    for im_path in im_list:
+        im_data = Image.open(im_path)
+
+        inputs = test_transform(im_data)
+        inputs = torch.unsqueeze(inputs, dim=0)  # 这里还在用torch，下面会改
+
+        # 将输入转为numpy，ONNX Runtime需要numpy输入
+        inputs = inputs.numpy()
+        if providers[0] == 'CPUExecutionProvider':
+            inputs = inputs.astype(np.float32)
+
+        # ONNX Runtime推理，输出直接是numpy数组
+        outputs = session.run(None, {input_name: inputs})[0]
+
+        # 解析结果，直接用numpy操作
+        pred = np.argmax(outputs, axis=1)
+        print(label_name[pred[0]], " ", im_path)
+
+        img = np.asarray(im_data)
+        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+        img = cv2.resize(img, (200, int(img.shape[0] * 200 / img.shape[1])))
+        cv2.imshow("img", img)
+        cv2.waitKey(0)
+
+
+if __name__ == "__main__":
+    test()

core/test/test_resnet.py

@@ -1,3 +1,4 @@
+import os
 import cv2
 import glob
 import torch
@@ -5,21 +6,32 @@ from torchvision import transforms
 from PIL import Image
 import numpy as np
 from core.nets.resnet import resnet
-from core.const import label_name, input_size
+from core.const import mode, label_name, input_size
 
 
 def test():
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     print(device)
 
-    net = resnet()
+    script_dir = os.path.dirname(os.path.abspath(__file__))
-    net.load_state_dict(torch.load("./model/resnet_epoch_15.pth", weights_only=True))
+    core_dir = os.path.dirname(script_dir)
+    model_dir = os.path.join(core_dir, "models")
+    dataset_dir = os.path.join(core_dir, "dataset", mode, "test")
 
-    im_list = glob.glob("./dataset/test/*/*.jpg")
+    print("model_dir", model_dir)
+
+    net = resnet()
+    net.load_state_dict(torch.load(os.path.join(model_dir, "resnet_epoch_80.pth"), weights_only=True))
+
+    print("111")
+
+    im_list = glob.glob(os.path.join(dataset_dir, "*", "*.jpg"))
     np.random.shuffle(im_list)
 
     net.to(device)
 
+    print("222")
+
     test_transform = transforms.Compose([
         transforms.Resize(input_size),
         transforms.CenterCrop(input_size),
@@ -27,7 +39,11 @@ def test():
         transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
     ])
 
+    print("333")
+
+    ary = []
     for im_path in im_list:
+        #  print("im_path", im_path)
         net.eval()
         im_data = Image.open(im_path)
 
@@ -40,11 +56,15 @@ def test():
         _, pred = torch.max(outputs.data, dim=1)
         print(label_name[pred.cpu().numpy()[0]], " ", im_path)
 
-        img = np.asarray(im_data)
+        result = label_name[pred.cpu().numpy()[0]]
-        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
-        cv2.imshow("img", img)
-        cv2.waitKey(0)
 
+        if result in im_path:
+            ary.append(True)
+        else:
+            ary.append(False)
+
+    print("ary", ary)
+    print(sum(ary) / len(ary))
 
 if __name__ == "__main__":
     test()

core/test/test_resnet18.py

@@ -1,3 +1,4 @@
+import os
 import cv2
 import glob
 import torch
@@ -5,17 +6,22 @@ from torchvision import transforms
 from PIL import Image
 import numpy as np
 from core.nets.resnet18 import resnet18
-from core.const import label_name, input_size
+from core.const import mode, label_name, input_size
 
 
 def test():
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     print(device)
 
-    net = resnet18()
+    script_dir = os.path.dirname(os.path.abspath(__file__))
-    net.load_state_dict(torch.load("./model/resnet_epoch_14.pth", weights_only=True))
+    core_dir = os.path.dirname(script_dir)
+    model_dir = os.path.join(core_dir, "models")
+    dataset_dir = os.path.join(core_dir, "dataset", mode, "test")
 
-    im_list = glob.glob("./dataset/test/*/*.jpg")
+    net = resnet18()
+    net.load_state_dict(torch.load(os.path.join(model_dir, "resnet18_epoch_50.pth"), weights_only=True))
+
+    im_list = glob.glob(os.path.join(dataset_dir, "*", "*.jpg"))
     np.random.shuffle(im_list)
 
     net.to(device)
@@ -27,6 +33,8 @@ def test():
         transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
     ])
 
+    ary = []
+    errors = []
     for im_path in im_list:
         net.eval()
         im_data = Image.open(im_path)
@@ -36,19 +44,27 @@ def test():
 
         inputs = inputs.to(device)
         outputs = net.forward(inputs)
-        # print("outputs", outputs)
 
         _, pred = torch.max(outputs.data, dim=1)
-        print(label_name[pred.cpu().numpy()[0]], " ", im_path)
+        # print(label_name[pred.cpu().numpy()[0]], " ", im_path)
 
-        # prob, pred = torch.topk(outputs.data, k=3, dim=1)
+        # img = np.asarray(im_data)
-        # for i in range(3):
+        # img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
-        #     print(label_name[pred[0, i].item()], " ", prob[0, i].item(), " ", im_path)
+        # img = cv2.resize(img, (200, 200))
+        # cv2.imshow("img", img)
+        # cv2.waitKey(0)
+        result = label_name[pred.cpu().numpy()[0]]
 
-        img = np.asarray(im_data)
+        if result in im_path:
-        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+            ary.append(True)
-        cv2.imshow("img", img)
+        else:
-        cv2.waitKey(0)
+            ary.append(False)
+            print(f"{label_name[pred.cpu().numpy()[0]]} {im_path}\n")
+            errors.append(f"{label_name[pred.cpu().numpy()[0]]} {im_path}")
+
+    print("ary", ary)
+    # print("errors", errors)
+    print(sum(ary) / len(ary))
 
 
 if __name__ == "__main__":

core/train/to_onnx.py

@@ -0,0 +1,29 @@
+import os
+import torch
+from core.nets.resnet18 import resnet18
+
+# 加载 pth
+net = resnet18()
+
+
+script_dir = os.path.dirname(os.path.abspath(__file__))
+core_dir = os.path.dirname(script_dir)
+model_dir = os.path.join(core_dir, "models")
+
+
+net.load_state_dict(torch.load(os.path.join(model_dir, "resnet18_epoch_50.pth"), map_location="cpu"))
+net.eval()
+
+
+# 导出 ONNX
+dummy_input = torch.randn(1, 3, 224, 224)
+torch.onnx.export(
+    net,
+    dummy_input,
+    os.path.join(model_dir, "resnet18_epoch_50.onnx"),
+    export_params=True,
+    opset_version=11,
+    input_names=["input"],
+    output_names=["output"],
+    dynamic_axes={"input": {0: "batch_size"}, "output": {0: "batch_size"}}
+)

core/train/train_convnext_tiny.py

@@ -0,0 +1,53 @@
+import os
+import torch
+from core.nets.convnext_tiny import pytorch_convnext_tiny
+from core.dataloader.dataloader import train_dataloader
+from core.const import epoch, lr, batch_size
+
+
+def train():
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    print("device: ", device)
+
+    net = pytorch_convnext_tiny().to(device)
+
+    loss_func = torch.nn.CrossEntropyLoss()
+
+    optimizer = torch.optim.Adam(net.parameters(), lr=lr)
+
+    scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=20, eta_min=1e-6)
+
+    for e in range(epoch):
+        print("epoch: ", e)
+        net.train()
+
+        for i, data in enumerate(train_dataloader):
+            inputs, labels = data
+            inputs, labels = inputs.to(device), labels.to(device)
+
+            outputs = net(inputs)
+
+            loss = loss_func(outputs, labels)
+
+            optimizer.zero_grad()
+            loss.backward()
+            optimizer.step()
+
+            _, pred = torch.max(outputs, dim=1)
+            correct = pred.eq(labels.data).cpu().sum()
+
+            print("step: ", i, "loss: ", loss.item(), "correct: ", 1.0 * correct / batch_size)
+
+        scheduler.step()
+        print("lr: ", optimizer.state_dict()['param_groups'][0]['lr'])
+
+        script_dir = os.path.dirname(os.path.abspath(__file__))
+        model_dir = os.path.join(script_dir, "..", "models")
+        if not os.path.exists(model_dir):
+            os.makedirs(model_dir)
+
+        torch.save(net.state_dict(), os.path.join(model_dir, "convnext_tiny_epoch_{}.pth".format(e + 1)))
+
+
+if __name__ == "__main__":
+    train()

core/train/train_resnet.py

@@ -2,7 +2,7 @@ import os
 import torch
 from core.nets.resnet import resnet
 from core.dataloader.dataloader import train_dataloader
-from core.const import epoch, lr, batch_size
+from core.const import epoch, lr, weight_decay, batch_size
 
 
 def train():
@@ -13,9 +13,9 @@ def train():
 
     loss_func = torch.nn.CrossEntropyLoss()
 
-    optimizer = torch.optim.Adam(net.parameters(), lr=lr)
+    optimizer = torch.optim.Adam(net.parameters(), lr=lr, weight_decay=weight_decay)
 
-    scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=5, gamma=0.5)
+    scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=20, gamma=0.5)
 
     for e in range(epoch):
         print("epoch: ", e)
@@ -41,10 +41,12 @@ def train():
         scheduler.step()
         print("lr: ", optimizer.state_dict()['param_groups'][0]['lr'])
 
-        if not os.path.exists("./model"):
+        script_dir = os.path.dirname(os.path.abspath(__file__))
-            os.makedirs("./model")
+        model_dir = os.path.join(script_dir, "..", "models")
+        if not os.path.exists(model_dir):
+            os.makedirs(model_dir)
 
-        torch.save(net.state_dict(), "./model/resnet_epoch_{}.pth".format(e + 1))
+        torch.save(net.state_dict(), os.path.join(model_dir, "resnet_epoch_{}.pth".format(e + 1)))
 
 
 if __name__ == "__main__":

core/train/train_resnet18.py

@@ -2,7 +2,7 @@ import os
 import torch
 from core.nets.resnet18 import resnet18
 from core.dataloader.dataloader import train_dataloader
-from core.const import epoch, lr, batch_size
+from core.const import epoch, lr, weight_decay, batch_size
 
 
 def train():
@@ -13,7 +13,7 @@ def train():
 
     loss_func = torch.nn.CrossEntropyLoss()
 
-    optimizer = torch.optim.Adam(net.parameters(), lr=lr)
+    optimizer = torch.optim.Adam(net.parameters(), lr=lr, weight_decay=weight_decay)
 
     scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=20, eta_min=1e-6) # 余弦退火
 
@@ -41,10 +41,12 @@ def train():
         scheduler.step()
         print("lr: ", optimizer.state_dict()['param_groups'][0]['lr'])
 
-        if not os.path.exists("./models"):
+        script_dir = os.path.dirname(os.path.abspath(__file__))
-            os.makedirs("./models")
+        model_dir = os.path.join(script_dir, "..", "models")
+        if not os.path.exists(model_dir):
+            os.makedirs(model_dir)
 
-        torch.save(net.state_dict(), "./models/resnet18_epoch_{}.pth".format(e + 1))
+        torch.save(net.state_dict(), os.path.join(model_dir, "resnet18_epoch_{}.pth".format(e + 1)))
 
 
 if __name__ == "__main__":

frontend/src/components/Footer.tsx

@@ -2,7 +2,7 @@ export const Footer = () => (
     <footer class="app-footer">
         <span>📩 xiadongliang88@163.com</span>
         <span>|</span>
-        <span>📦 https://git.leonstack.com/owner</span>
+        <span>📦 https://git.leonstack.com/xiadongliang</span>
         <span>|</span>
         <span>🌏 https://www.leonstack.com/</span>
     </footer>

frontend/src/components/History.tsx

@@ -1,10 +1,10 @@
 import { useState, useEffect } from 'preact/hooks'
 import Pagination from './Pagination'
 import Modal from './Modal'
-import ImagePreview from './ImagePreview'
 import { message } from '../utils/toast'
 import type { HistoryItem } from '../preact'
 
+
 const formatDate = (timestamp: number) => {
     const date = new Date(timestamp * 1000)
     return date.toLocaleString('zh-CN', {
@@ -22,12 +22,13 @@ const History = () => {
     const pageSize = 5
     const [total, setTotal] = useState<number>(50)
     const [historyList, setHistoryList] = useState<HistoryItem[]>([])
+    const [currentItem, setCurrentItem] = useState<HistoryItem>()
     const [showDelete, setShowDelete] = useState<boolean>(false)
     const [currentId, setCurrentId] = useState<number | null>(null)
     const [showClear, setShowClear] = useState<boolean>(false)
 
     useEffect(() => {
-        if (!(window as any).go?.main?.App?.GetHistory) {
+        if (!(window as any).go?.backend?.App?.GetHistory) {
             message.error('Wails runtime not ready')
             return
         }
@@ -35,7 +36,7 @@ const History = () => {
     }, [])
 
     const fetchData = async(page: number) => {
-        const result = await (window as any).go.main.App.GetHistory(page, pageSize)
+        const result = await (window as any).go.backend.App.GetHistory(page, pageSize)
         if (result.code === 0) {
             setHistoryList(result.data.list)
             setTotal(result.data.total)
@@ -49,6 +50,10 @@ const History = () => {
         fetchData(page)
     }
 
+    const handleShowItem = (item: HistoryItem) => setCurrentItem(item)
+
+    const handleItemClose = () => setCurrentItem(undefined)
+
     const handleShowDelete = (id: number) => {
         setCurrentId(id)
         setShowDelete(true)
@@ -60,7 +65,7 @@ const History = () => {
             message.error('currentId为空')
             return
         }
-        const result = await (window as any).go.main.App.DeleteOneHistory(currentId)
+        const result = await (window as any).go.backend.App.DeleteOneHistory(currentId)
         if (result.code === 0) {
             message.success('删除成功')
             setCurrentId(null)
@@ -79,7 +84,7 @@ const History = () => {
 
     const handleClearOk = async() => {
         setShowClear(false)
-        const result = await (window as any).go.main.App.ClearHistory()
+        const result = await (window as any).go.backend.App.ClearHistory()
         if (result.code === 0) {
             message.success('已清空历史')
             setCurrentId(null)
@@ -104,7 +109,7 @@ const History = () => {
                 {historyList.map((item: HistoryItem) =>
                     <div key={item.id} class="history-card">
                         <div class="card-left">
-                            <ImagePreview filename={item.img} />
+                            <img src={`data:image/jpeg;base64,${item.img_data}`} onClick={() => handleShowItem(item)} />
                         </div>
                         <div class="card-right">
                             <div class="right-top">
@@ -142,6 +147,9 @@ const History = () => {
             <Modal open={showClear} title="信息" onClick={handleClearOk} onClose={handleClearClose}>
                 <p>确定要删除全部记录？</p>
             </Modal>
+            <Modal open={!!currentItem} title={currentItem?.name || ''} onClose={handleItemClose}>
+                <p>{currentItem?.brief || ''}</p>
+            </Modal>
         </>
     )
 }

frontend/src/components/ImagePreview.tsx

@@ -1,27 +0,0 @@
-import { useState, useEffect } from 'preact/hooks'
-
-interface Props {
-    filename: string
-    class?: string
-    id?: string
-}
-
-const ImagePreview = ({ filename, ...props }: Props) => {
-    const [src, setSrc] = useState<string>('')
-
-    useEffect(() => {
-        const loadImage = async () => {
-            const result = await (window as any).go.main.App.GetImage(filename)
-            if (result.code === 0) {
-                setSrc(result.data)
-            }
-        }
-        loadImage()
-    }, [filename])
-
-    if (!src) return <div {...props} />
-
-    return <img src={src} {...props} />
-}
-
-export default ImagePreview

frontend/src/components/Main.tsx

@@ -1,11 +1,12 @@
 import { useState, useRef } from 'preact/hooks'
 import type { DetectResult } from '../preact'
-import ImagePreview from './ImagePreview'
 import { message } from '../utils/toast'
 
+
 const Main = () => {
     const fileInputRef = useRef<HTMLInputElement>(null)
     const [fileSrc, setFileSrc] = useState<string>('')
+    const [filename, setFilename] = useState<string>('')
     const [step, setStep] = useState<number>(0)
     const [detectResult, setDetectResult] = useState<DetectResult | null>(null)
 
@@ -47,7 +48,6 @@ const Main = () => {
     }
 
     const handleFileChange = (e: Event) => {
-        console.log('handleFileChange')
         const target = e.target as HTMLInputElement
         const file = target.files?.[0]
         if (file) {
@@ -60,12 +60,13 @@ const Main = () => {
                 const arrayBuffer = await processedFile.arrayBuffer()
                 const uint8Array = new Uint8Array(arrayBuffer)
 
-                const result = await (window as any).go.main.App.UploadImage(
+                const result = await (window as any).go.backend.App.UploadImage(
                     Array.from(uint8Array),
                     file.name
                 )
                 if (result.code === 0) {
-                    setFileSrc(result.data)
+                    setFileSrc(result.data.data)
+                    setFilename(result.data.filename)
                     setStep(1)
                 } else if (result.code === 1) {
                     message.error(result.message)
@@ -92,14 +93,16 @@ const Main = () => {
         setStep(2)
 
         setTimeout(async() => {
-            const result = await (window as any).go.main.App.Detect(fileSrc)
+            if (filename) {
-            if (result.code === 0) {
+                const result = await (window as any).go.backend.App.Detect(filename)
-                setDetectResult(result.data)
+                if (result.code === 0) {
-                setStep(3)
+                    setDetectResult(result.data)
-            } else if (result.code === 1) {
+                    setStep(3)
-                message.error(result.message)
+                } else if (result.code === 1) {
+                    message.error(result.message)
+                }
             }
-        }, 1000)
+        }, 500)
     }
 
     const handleTryOther = () => {
@@ -133,7 +136,7 @@ const Main = () => {
                     >
                         {fileSrc.length > 0 && step == 1 ?
                             <div id="previewContent">
-                                <ImagePreview filename={fileSrc} id="previewImage" />
+                                <img src={`data:image/jpeg;base64,${fileSrc}`} />
                                 <button onClick={handleRemovePhoto}>
                                     ❌ 移除照片
                                 </button>
@@ -192,7 +195,7 @@ const Main = () => {
                         <h2>完成！</h2>
                     </div>
                     <div class="result-main">
-                        <ImagePreview filename={fileSrc} />
+                        <img src={`data:image/jpeg;base64,${fileSrc}`} />
                         <div class="result-word">
                             <div>
                                 <h3>{detectResult?.name}</h3>
@@ -200,7 +203,7 @@ const Main = () => {
                                     <div class="probability-bar">
                                         <div />
                                     </div>
-                                    <span>置信度 {detectResult ? detectResult.confidence_level * 100 + '%' : ''}</span>
+                                    <span>置信度 {detectResult ? (detectResult.confidence_level * 100).toFixed(2) + '%' : ''}</span>
                                 </div>
                             </div>
                             <p>{detectResult?.brief}</p>

frontend/src/components/Modal.tsx

@@ -1,9 +1,8 @@
 import type { ModalProps } from '../preact'
 
+
 const Modal = ({ open, title, onClick, onClose, children }: ModalProps) => {
-    const handleClose = () => {
+    const handleClose = () => onClose?.()
-        onClose?.()
-    }
 
     const handleConfirm = () => {
         onClick?.()
@@ -11,9 +10,7 @@ const Modal = ({ open, title, onClick, onClose, children }: ModalProps) => {
     }
 
     const handleMaskClick = (e: MouseEvent) => {
-        if (e.target === e.currentTarget) {
+        if (e.target === e.currentTarget) handleClose()
-            handleClose()
-        }
     }
 
     return (
@@ -27,8 +24,8 @@ const Modal = ({ open, title, onClick, onClose, children }: ModalProps) => {
                         </div>
                         {children}
                         <div class="bottom">
-                            <button onClick={handleClose}>取消</button>
+                            <button class="close" onClick={handleClose}>取消</button>
-                            <button onClick={handleConfirm}>确定</button>
+                            {onClick ? <button class="confirm" onClick={handleConfirm}>确定</button> : null}
                         </div>
                     </div>
                     <div class="app-mask" onClick={handleMaskClick}></div>

frontend/src/components/Pagination.tsx

@@ -12,11 +12,6 @@ const Pagination = ({
     const [num, setNum] = useState<number>(1)
     const [seqNums, setSeqNums] = useState<[number, number][]>([])
 
-    useEffect(() => {
-        setValue('1')
-        setNum(1)
-    }, [])
-
     useEffect(() => {
         if (page) {
             setValue(page.toString());

frontend/src/preact.d.ts

@@ -1,6 +1,7 @@
 export interface HistoryItem {
     id: number
     img: string
+    img_data: string
     breed: number
     date: number
     name: string

frontend/src/styles/base.sass

@@ -199,7 +199,7 @@ a
                 border-radius: 1.5rem
                 border-width: 2px
                 border-style: dashed
-                border-color: #e2e8f0
+                border-color: #bdc3cb
                 text-align: center
                 cursor: pointer
 
@@ -305,7 +305,6 @@ a
                 display: flex
                 align-items: center
                 justify-content: center
-                // height: 2.25rem
                 gap: 0.75rem
                 margin: 0.5rem 0 1.5rem 0
 
@@ -474,12 +473,12 @@ a
             font-size: 0.875rem
             border-radius: 0.25rem
 
-            &:first-child
+        > button.close
                 color: var(--text)
                 border: 1px solid #c9c9c9
                 background-color: none
 
-            &:last-child
+        > button.confirm
                 color: white
                 background-color: var(--primary)
 

frontend/src/utils/toast.ts

@@ -23,9 +23,7 @@ export const message: MessageAPI = {
         div.appendChild(subDiv)
         document.body.appendChild(div)
 
-        setTimeout(() => {
+        setTimeout(() => div.remove(), 3000)
-            div.remove()
-        }, 3000)
     },
     error: (text: string) => {
         const div = document.createElement('div')
@@ -46,8 +44,6 @@ export const message: MessageAPI = {
         div.appendChild(subDiv)
         document.body.appendChild(div)
 
-        setTimeout(() => {
+        setTimeout(() => div.remove(), 3000)
-            div.remove()
-        }, 3000)
     }
 }

frontend/wailsjs/go/backend/App.d.ts

@@ -11,8 +11,8 @@ export function Detect(arg1:string):Promise<backend.Response>;
 
 export function GetHistory(arg1:number,arg2:number):Promise<backend.Response>;
 
-export function GetImage(arg1:string):Promise<backend.Response>;
-
 export function GormDB():Promise<gorm.DB>;
 
+export function Shutdown():Promise<void>;
+
 export function UploadImage(arg1:Array<number>,arg2:string):Promise<backend.Response>;

frontend/wailsjs/go/backend/App.js

@@ -18,14 +18,14 @@ export function GetHistory(arg1, arg2) {
   return window['go']['backend']['App']['GetHistory'](arg1, arg2);
 }
 
-export function GetImage(arg1) {
-  return window['go']['backend']['App']['GetImage'](arg1);
-}
-
 export function GormDB() {
   return window['go']['backend']['App']['GormDB']();
 }
 
+export function Shutdown() {
+  return window['go']['backend']['App']['Shutdown']();
+}
+
 export function UploadImage(arg1, arg2) {
   return window['go']['backend']['App']['UploadImage'](arg1, arg2);
 }

go.mod

@@ -3,7 +3,9 @@ module sortmeow
 go 1.25.0
 
 require (
+	github.com/disintegration/imaging v1.6.2
 	github.com/wailsapp/wails/v2 v2.13.0
+	github.com/yalue/onnxruntime_go v1.31.0
 	gorm.io/driver/sqlite v1.6.0
 	gorm.io/gorm v1.31.2
 )
@@ -37,6 +39,7 @@ require (
 	github.com/wailsapp/go-webview2 v1.0.22 // indirect
 	github.com/wailsapp/mimetype v1.4.1 // indirect
 	golang.org/x/crypto v0.51.0 // indirect
+	golang.org/x/image v0.40.0 // indirect
 	golang.org/x/net v0.54.0 // indirect
 	golang.org/x/sys v0.46.0 // indirect
 	golang.org/x/text v0.40.0 // indirect

go.sum

@@ -4,6 +4,8 @@ github.com/bep/debounce v1.2.1 h1:v67fRdBA9UQu2NhLFXrSg0Brw7CexQekrBwDMM8bzeY=
 github.com/bep/debounce v1.2.1/go.mod h1:H8yggRPQKLUhUoqrJC1bO2xNya7vanpDl7xR3ISbCJ0=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/disintegration/imaging v1.6.2 h1:w1LecBlG2Lnp8B3jk5zSuNqd7b4DXhcjwek1ei82L+c=
+github.com/disintegration/imaging v1.6.2/go.mod h1:44/5580QXChDfwIclfc/PCwrr44amcmDAg8hxG0Ewe4=
 github.com/go-ole/go-ole v1.3.0 h1:Dt6ye7+vXGIKZ7Xtk4s6/xVdGDQynvom7xCFEdWr6uE=
 github.com/go-ole/go-ole v1.3.0/go.mod h1:5LS6F96DhAwUc7C+1HLexzMXY1xGRSryjyPPKW6zv78=
 github.com/godbus/dbus/v5 v5.1.0 h1:4KLkAxT3aOY8Li4FRJe/KvhoNFFxo0m6fNuFUO8QJUk=
@@ -67,8 +69,13 @@ github.com/wailsapp/mimetype v1.4.1 h1:pQN9ycO7uo4vsUUuPeHEYoUkLVkaRntMnHJxVwYhw
 github.com/wailsapp/mimetype v1.4.1/go.mod h1:9aV5k31bBOv5z6u+QP8TltzvNGJPmNJD4XlAL3U+j3o=
 github.com/wailsapp/wails/v2 v2.13.0 h1:S7OgXWpj72V91unF8iDWJKbcS9ZpwCT3R0QVru4v2Mg=
 github.com/wailsapp/wails/v2 v2.13.0/go.mod h1:nVr/wSIEZ7xxKPkzK65mjpKpaOPQI2k4pvLwGR/i4kc=
+github.com/yalue/onnxruntime_go v1.31.0 h1:1ln4YW1SFOFfGJZXe3jNOb2JUSt+l2pEneZfV8HdtFA=
+github.com/yalue/onnxruntime_go v1.31.0/go.mod h1:b4X26A8pekNb1ACJ58wAXgNKeUCGEAQ9dmACut9Sm/4=
 golang.org/x/crypto v0.51.0 h1:IBPXwPfKxY7cWQZ38ZCIRPI50YLeevDLlLnyC5wRGTI=
 golang.org/x/crypto v0.51.0/go.mod h1:8AdwkbraGNABw2kOX6YFPs3WM22XqI4EXEd8g+x7Oc8=
+golang.org/x/image v0.0.0-20191009234506-e7c1f5e7dbb8/go.mod h1:FeLwcggjj3mMvU+oOTbSwawSJRM1uh48EjtB4UJZlP0=
+golang.org/x/image v0.40.0 h1:Tw4GyDXMo+daZN1znreBRC3VayR1aLFUyUEOLUdW1a8=
+golang.org/x/image v0.40.0/go.mod h1:uIc348UZMSvS5Z65CVZ7iDPaNobNFEPeJ4kbqTOszmA=
 golang.org/x/net v0.0.0-20210505024714-0287a6fb4125/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/net v0.54.0 h1:2zJIZAxAHV/OHCDTCOHAYehQzLfSXuf/5SoL/Dv6w/w=
 golang.org/x/net v0.54.0/go.mod h1:Sj4oj8jK6XmHpBZU/zWHw3BV3abl4Kvi+Ut7cQcY+cQ=
@@ -81,6 +88,7 @@ golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.46.0 h1:noSf2Fq6F8DBgS+LysIkx7rIExoNHJsxOAtPp4rthXw=
 golang.org/x/sys v0.46.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
+golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.40.0 h1:Ub2Z6/xjgF1WrYQz2nuITOEegKFtiIy+rieRJ5lHZKs=
 golang.org/x/text v0.40.0/go.mod h1:hpnzDAfGV753zIKo+wk3u1bVKCGPbrnF7+7LBF/UHVY=

yuanxing.html

@@ -0,0 +1,316 @@
+<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>分类喵 - 项目原型图</title>
+    <style>
+        * { margin: 0; padding: 0; box-sizing: border-box; }
+        body {
+            font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif;
+            background: #f0f0f0;
+            min-height: 100vh;
+            padding: 40px 20px;
+        }
+        .header {
+            text-align: center;
+            margin-bottom: 30px;
+        }
+        .header h1 {
+            font-size: 24px;
+            color: #333;
+            margin-bottom: 8px;
+        }
+        .header p {
+            color: #666;
+            font-size: 13px;
+        }
+    </style>
+</head>
+<body>
+    <div class="header">
+        <h1>分类喵 - 项目原型图</h1>
+        <p>基于深度残差网络的猫咪图片分类应用 | 桌面端</p>
+    </div>
+
+    <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 1400 650" width="1400" height="650">
+        <defs>
+            <marker id="arrow" markerWidth="10" markerHeight="7" refX="9" refY="3.5" orient="auto">
+                <polygon points="0 0, 10 3.5, 0 7" fill="#999"/>
+            </marker>
+        </defs>
+
+        <!-- 页面1: 首页 -->
+        <g id="page1" transform="translate(30, 20)">
+            <!-- 窗口标题栏 -->
+            <rect x="0" y="0" width="400" height="30" rx="6" fill="#e0e0e0"/>
+            <circle cx="15" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="32" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="49" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="200" y="20" font-size="11" fill="#666" text-anchor="middle">分类喵 - 桌面应用</text>
+
+            <!-- 导航栏 -->
+            <rect x="0" y="30" width="400" height="50" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="20" y="60" font-size="14" font-weight="bold">分类喵</text>
+            <line x1="100" y1="40" x2="100" y2="70" stroke="#ddd" stroke-width="1"/>
+            <text x="120" y="60" font-size="12" fill="#333">主页</text>
+            <text x="175" y="60" font-size="12" fill="#999">历史记录</text>
+            <rect x="320" y="42" width="60" height="26" rx="4" fill="#f5f5f5" stroke="#ddd"/>
+            <text x="350" y="59" font-size="10" fill="#666" text-anchor="middle">查看代码</text>
+
+            <!-- 主内容区 -->
+            <rect x="0" y="80" width="400" height="520" fill="#fafafa"/>
+
+            <!-- 标题 -->
+            <text x="40" y="130" font-size="20" fill="#333" font-weight="bold">发现<tspan fill="#666">它</tspan>的<tspan fill="#666">分类</tspan></text>
+            <text x="40" y="155" font-size="11" fill="#888">上传一张猫咪照片，让AI认出它是什么</text>
+            <text x="40" y="175" font-size="10" fill="#aaa">⚡ 基于深度残差网络</text>
+
+            <!-- 上传区域 -->
+            <rect x="40" y="200" width="320" height="200" rx="8" fill="#fff" stroke="#ccc" stroke-width="1.5" stroke-dasharray="6 3"/>
+            <!-- 上传图标 -->
+            <g transform="translate(200, 270)">
+                <circle cx="0" cy="0" r="35" fill="none" stroke="#ccc" stroke-width="1.5"/>
+                <path d="M-12 5 L0 -10 L12 5 M-8 15 L0 2 L8 15" stroke="#999" stroke-width="1.5" fill="none" stroke-linecap="round" stroke-linejoin="round"/>
+            </g>
+            <text x="200" y="330" font-size="12" fill="#666" text-anchor="middle">点击或拖拽上传图片</text>
+            <text x="200" y="350" font-size="10" fill="#aaa" text-anchor="middle">支持 JPG、PNG、JPEG 格式</text>
+
+            <!-- 按钮 -->
+            <rect x="40" y="420" width="320" height="40" rx="6" fill="none" stroke="#999" stroke-width="1.5"/>
+            <text x="200" y="445" font-size="12" fill="#666" text-anchor="middle">🔍 开始检测</text>
+
+            <!-- 页面标签 -->
+            <text x="200" y="620" font-size="11" fill="#666" text-anchor="middle" font-weight="bold">图1: 首页</text>
+        </g>
+
+        <!-- 页面2: 预览状态 -->
+        <g id="page2" transform="translate(470, 20)">
+            <rect x="0" y="0" width="400" height="30" rx="6" fill="#e0e0e0"/>
+            <circle cx="15" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="32" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="49" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="200" y="20" font-size="11" fill="#666" text-anchor="middle">分类喵 - 桌面应用</text>
+
+            <rect x="0" y="30" width="400" height="50" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="20" y="60" font-size="14" font-weight="bold">分类喵</text>
+            <line x1="100" y1="40" x2="100" y2="70" stroke="#ddd" stroke-width="1"/>
+            <text x="120" y="60" font-size="12" fill="#333">主页</text>
+            <text x="175" y="60" font-size="12" fill="#999">历史记录</text>
+            <rect x="320" y="42" width="60" height="26" rx="4" fill="#f5f5f5" stroke="#ddd"/>
+            <text x="350" y="59" font-size="10" fill="#666" text-anchor="middle">查看代码</text>
+
+            <rect x="0" y="80" width="400" height="520" fill="#fafafa"/>
+
+            <text x="40" y="130" font-size="20" fill="#333" font-weight="bold">发现<tspan fill="#666">它</tspan>的<tspan fill="#666">分类</tspan></text>
+            <text x="40" y="155" font-size="11" fill="#888">上传一张猫咪照片，让AI认出它是什么</text>
+            <text x="40" y="175" font-size="10" fill="#aaa">⚡ 基于深度残差网络</text>
+
+            <!-- 上传区域-有图片 -->
+            <rect x="40" y="200" width="320" height="200" rx="8" fill="#fff" stroke="#999" stroke-width="1.5"/>
+            <!-- 图片区域 -->
+            <rect x="60" y="220" width="140" height="160" rx="4" fill="#f0f0f0" stroke="#ddd" stroke-width="1"/>
+            <!-- 猫图标 -->
+            <g transform="translate(130, 300)">
+                <ellipse cx="0" cy="15" rx="35" ry="25" fill="none" stroke="#999" stroke-width="1.5"/>
+                <ellipse cx="-18" cy="-5" rx="14" ry="16" fill="none" stroke="#999" stroke-width="1.5"/>
+                <ellipse cx="18" cy="-5" rx="14" ry="16" fill="none" stroke="#999" stroke-width="1.5"/>
+                <ellipse cx="0" cy="20" rx="6" ry="4" fill="none" stroke="#999" stroke-width="1"/>
+            </g>
+            <!-- 移除按钮 -->
+            <rect x="60" y="220" width="50" height="22" rx="4" fill="#fff" stroke="#ccc"/>
+            <text x="85" y="235" font-size="9" fill="#666" text-anchor="middle">移除</text>
+
+            <!-- 开始检测按钮 -->
+            <rect x="40" y="420" width="320" height="40" rx="6" fill="#333"/>
+            <text x="200" y="445" font-size="12" fill="#fff" text-anchor="middle">🔍 开始检测</text>
+
+            <text x="200" y="620" font-size="11" fill="#666" text-anchor="middle" font-weight="bold">图2: 图片预览</text>
+        </g>
+
+        <!-- 页面3: 检测中 -->
+        <g id="page3" transform="translate(910, 20)">
+            <rect x="0" y="0" width="400" height="30" rx="6" fill="#e0e0e0"/>
+            <circle cx="15" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="32" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="49" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="200" y="20" font-size="11" fill="#666" text-anchor="middle">分类喵 - 桌面应用</text>
+
+            <rect x="0" y="30" width="400" height="50" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="20" y="60" font-size="14" font-weight="bold">分类喵</text>
+            <line x1="100" y1="40" x2="100" y2="70" stroke="#ddd" stroke-width="1"/>
+            <text x="120" y="60" font-size="12" fill="#333">主页</text>
+            <text x="175" y="60" font-size="12" fill="#999">历史记录</text>
+            <rect x="320" y="42" width="60" height="26" rx="4" fill="#f5f5f5" stroke="#ddd"/>
+            <text x="350" y="59" font-size="10" fill="#666" text-anchor="middle">查看代码</text>
+
+            <rect x="0" y="80" width="400" height="520" fill="#fafafa"/>
+
+            <text x="40" y="130" font-size="20" fill="#333" font-weight="bold">发现<tspan fill="#666">它</tspan>的<tspan fill="#666">分类</tspan></text>
+            <text x="40" y="155" font-size="11" fill="#888">上传一张猫咪照片，让AI认出它是什么</text>
+            <text x="40" y="175" font-size="10" fill="#aaa">⚡ 基于深度残差网络</text>
+
+            <!-- 上传区域-加载中 -->
+            <rect x="40" y="200" width="320" height="200" rx="8" fill="#fff" stroke="#999" stroke-width="1.5"/>
+            <!-- 加载动画 -->
+            <g transform="translate(200, 280)">
+                <circle cx="0" cy="0" r="35" fill="none" stroke="#ddd" stroke-width="2"/>
+                <path d="M0 -35 A35 35 0 0 1 35 0" stroke="#666" stroke-width="2" fill="none" stroke-linecap="round">
+                    <animateTransform attributeName="transform" type="rotate" from="0 0 0" to="360 0 0" dur="1s" repeatCount="indefinite"/>
+                </path>
+            </g>
+            <text x="200" y="340" font-size="13" fill="#333" text-anchor="middle" font-weight="500">正在检测...</text>
+            <text x="200" y="360" font-size="11" fill="#999" text-anchor="middle">请耐心等待~</text>
+
+            <text x="200" y="620" font-size="11" fill="#666" text-anchor="middle" font-weight="bold">图3: 检测中</text>
+        </g>
+
+        <!-- 页面4: 结果 -->
+        <g id="page4" transform="translate(30, 420)">
+            <rect x="0" y="0" width="400" height="30" rx="6" fill="#e0e0e0"/>
+            <circle cx="15" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="32" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="49" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="200" y="20" font-size="11" fill="#666" text-anchor="middle">分类喵 - 桌面应用</text>
+
+            <rect x="0" y="30" width="400" height="50" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="20" y="60" font-size="14" font-weight="bold">分类喵</text>
+            <line x1="100" y1="40" x2="100" y2="70" stroke="#ddd" stroke-width="1"/>
+            <text x="120" y="60" font-size="12" fill="#333">主页</text>
+            <text x="175" y="60" font-size="12" fill="#999">历史记录</text>
+            <rect x="320" y="42" width="60" height="26" rx="4" fill="#f5f5f5" stroke="#ddd"/>
+            <text x="350" y="59" font-size="10" fill="#666" text-anchor="middle">查看代码</text>
+
+            <rect x="0" y="80" width="400" height="520" fill="#fafafa"/>
+
+            <text x="40" y="130" font-size="20" fill="#333" font-weight="bold">发现<tspan fill="#666">它</tspan>的<tspan fill="#666">分类</tspan></text>
+
+            <!-- 完成标记 -->
+            <g transform="translate(200, 175)">
+                <circle cx="0" cy="0" r="18" fill="none" stroke="#666" stroke-width="1.5"/>
+                <path d="M-7 0 L-2 5 L8 -6" stroke="#666" stroke-width="1.5" fill="none" stroke-linecap="round" stroke-linejoin="round"/>
+            </g>
+            <text x="200" y="210" font-size="12" fill="#333" text-anchor="middle" font-weight="500">检测完成</text>
+
+            <!-- 结果卡片 -->
+            <rect x="40" y="230" width="320" height="180" rx="8" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <!-- 图片 -->
+            <rect x="60" y="250" width="100" height="100" rx="4" fill="#f0f0f0" stroke="#ddd"/>
+            <g transform="translate(110, 300)">
+                <ellipse cx="0" cy="15" rx="28" ry="20" fill="none" stroke="#999" stroke-width="1.5"/>
+                <ellipse cx="-14" cy="-2" rx="11" ry="13" fill="none" stroke="#999" stroke-width="1.5"/>
+                <ellipse cx="14" cy="-2" rx="11" ry="13" fill="none" stroke="#999" stroke-width="1.5"/>
+                <ellipse cx="0" cy="20" rx="5" ry="3" fill="none" stroke="#999" stroke-width="1"/>
+            </g>
+
+            <!-- 结果信息 -->
+            <text x="180" y="270" font-size="14" fill="#333" font-weight="bold">英短蓝猫</text>
+            <!-- 置信度条 -->
+            <rect x="180" y="285" width="160" height="8" rx="4" fill="#e5e5e5"/>
+            <rect x="180" y="285" width="136" height="8" rx="4" fill="#666"/>
+            <text x="180" y="310" font-size="10" fill="#666">置信度 85.6%</text>
+
+            <!-- 描述 -->
+            <text x="60" y="440" font-size="11" fill="#888">英国短毛猫是传统英国本地猫的纯种版本...</text>
+
+            <!-- 按钮 -->
+            <rect x="40" y="460" width="320" height="40" rx="6" fill="none" stroke="#666" stroke-width="1.5"/>
+            <text x="200" y="485" font-size="12" fill="#666" text-anchor="middle">🔄 再试一张</text>
+
+            <text x="200" y="620" font-size="11" fill="#666" text-anchor="middle" font-weight="bold">图4: 检测结果</text>
+        </g>
+
+        <!-- 页面5: 历史记录 -->
+        <g id="page5" transform="translate(470, 420)">
+            <rect x="0" y="0" width="400" height="30" rx="6" fill="#e0e0e0"/>
+            <circle cx="15" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="32" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <circle cx="49" cy="15" r="5" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="200" y="20" font-size="11" fill="#666" text-anchor="middle">分类喵 - 桌面应用</text>
+
+            <rect x="0" y="30" width="400" height="50" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="20" y="60" font-size="14" font-weight="bold">分类喵</text>
+            <line x1="100" y1="40" x2="100" y2="70" stroke="#ddd" stroke-width="1"/>
+            <text x="120" y="60" font-size="12" fill="#999">主页</text>
+            <text x="175" y="60" font-size="12" fill="#333">历史记录</text>
+            <rect x="320" y="42" width="60" height="26" rx="4" fill="#f5f5f5" stroke="#ddd"/>
+            <text x="350" y="59" font-size="10" fill="#666" text-anchor="middle">查看代码</text>
+
+            <rect x="0" y="80" width="400" height="520" fill="#fafafa"/>
+
+            <text x="40" y="120" font-size="16" fill="#333" font-weight="bold">历史记录</text>
+            <!-- 清空按钮 -->
+            <rect x="320" y="105" width="40" height="28" rx="4" fill="#fff" stroke="#ccc"/>
+            <text x="340" y="123" font-size="11" fill="#666" text-anchor="middle">🗑️</text>
+
+            <!-- 历史记录卡片1 -->
+            <rect x="40" y="145" width="320" height="90" rx="6" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <rect x="55" y="160" width="60" height="60" rx="4" fill="#f0f0f0" stroke="#ddd"/>
+            <g transform="translate(85, 190)">
+                <ellipse cx="0" cy="10" rx="20" ry="14" fill="none" stroke="#999" stroke-width="1"/>
+                <ellipse cx="-10" cy="-2" rx="8" ry="10" fill="none" stroke="#999" stroke-width="1"/>
+                <ellipse cx="10" cy="-2" rx="8" ry="10" fill="none" stroke="#999" stroke-width="1"/>
+            </g>
+            <text x="130" y="180" font-size="13" fill="#333" font-weight="500">英短蓝猫</text>
+            <text x="130" y="200" font-size="10" fill="#999">2024/07/27 15:32</text>
+            <text x="130" y="218" font-size="10" fill="#888">英国短毛猫是传统英国...</text>
+            <text x="350" y="175" font-size="14" fill="#999" text-anchor="middle">×</text>
+
+            <!-- 历史记录卡片2 -->
+            <rect x="40" y="245" width="320" height="90" rx="6" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <rect x="55" y="260" width="60" height="60" rx="4" fill="#f0f0f0" stroke="#ddd"/>
+            <g transform="translate(85, 290)">
+                <ellipse cx="0" cy="10" rx="20" ry="14" fill="none" stroke="#999" stroke-width="1"/>
+                <ellipse cx="-10" cy="-2" rx="8" ry="10" fill="none" stroke="#999" stroke-width="1"/>
+                <ellipse cx="10" cy="-2" rx="8" ry="10" fill="none" stroke="#999" stroke-width="1"/>
+            </g>
+            <text x="130" y="280" font-size="13" fill="#333" font-weight="500">狸花猫</text>
+            <text x="130" y="300" font-size="10" fill="#999">2024/07/27 14:20</text>
+            <text x="130" y="318" font-size="10" fill="#888">狸花猫是中华田园...</text>
+            <text x="350" y="275" font-size="14" fill="#999" text-anchor="middle">×</text>
+
+            <!-- 分页 -->
+            <g transform="translate(200, 370)">
+                <rect x="-100" y="0" width="36" height="32" rx="4" fill="#fff" stroke="#ccc"/>
+                <text x="-82" y="21" font-size="12" fill="#666" text-anchor="middle">‹</text>
+                <rect x="-58" y="0" width="36" height="32" rx="4" fill="#333" stroke="#333"/>
+                <text x="-40" y="21" font-size="12" fill="#fff" text-anchor="middle">1</text>
+                <rect x="-16" y="0" width="36" height="32" rx="4" fill="#fff" stroke="#ccc"/>
+                <text x="2" y="21" font-size="12" fill="#666" text-anchor="middle">2</text>
+                <rect x="26" y="0" width="36" height="32" rx="4" fill="#fff" stroke="#ccc"/>
+                <text x="44" y="21" font-size="12" fill="#666" text-anchor="middle">›</text>
+            </g>
+
+            <!-- 删除确认弹窗 -->
+            <rect x="80" y="160" width="240" height="130" rx="8" fill="#fff" stroke="#ccc" stroke-width="1"/>
+            <text x="200" y="195" font-size="13" fill="#333" text-anchor="middle" font-weight="500">信息</text>
+            <line x1="80" y1="210" x2="320" y2="210" stroke="#eee"/>
+            <text x="200" y="240" font-size="12" fill="#666" text-anchor="middle">确定要删除当前记录？</text>
+            <rect x="100" y="260" width="80" height="32" rx="4" fill="#fff" stroke="#ccc"/>
+            <text x="140" y="281" font-size="11" fill="#666" text-anchor="middle">取消</text>
+            <rect x="220" y="260" width="80" height="32" rx="4" fill="#333"/>
+            <text x="260" y="281" font-size="11" fill="#fff" text-anchor="middle">确定</text>
+
+            <text x="200" y="620" font-size="11" fill="#666" text-anchor="middle" font-weight="bold">图5: 历史记录</text>
+        </g>
+
+        <!-- 连接线和标注 -->
+        <g stroke="#999" stroke-width="1" fill="none" marker-end="url(#arrow)">
+            <path d="M435 280 L465 280"/>
+            <path d="M875 280 L905 280"/>
+            <path d="M435 580 L465 580"/>
+            <path d="M875 580 L905 580"/>
+        </g>
+
+        <!-- 底部说明 -->
+        <g transform="translate(910, 420)">
+            <text x="0" y="0" font-size="12" fill="#666" font-weight="bold">原型图说明</text>
+            <text x="0" y="25" font-size="10" fill="#888">• 图1: 首页 - 上传猫咪图片</text>
+            <text x="0" y="43" font-size="10" fill="#888">• 图2: 预览 - 确认上传图片</text>
+            <text x="0" y="61" font-size="10" fill="#888">• 图3: 检测中 - AI正在识别</text>
+            <text x="0" y="79" font-size="10" fill="#888">• 图4: 结果 - 显示分类信息</text>
+            <text x="0" y="97" font-size="10" fill="#888">• 图5: 历史 - 管理识别记录</text>
+        </g>
+    </svg>
+</body>
+</html>