用python的tensorflow包写了个手写自动识别的py脚本
前提条件
python
pip install tensorflow pillow numpy matplotlib
python
import tensorflow as tf
from tensorflow.keras import datasets, layers, models
import numpy as np
import tkinter as tk
from PIL import Image, ImageOps, ImageDraw
from tkinter import ttk
# 加载MNIST数据集
(train_images, train_labels), (test_images, test_labels) = datasets.mnist.load_data()
train_images, test_images = train_images / 255.0, test_images / 255.0
# 构建卷积神经网络模型
model = models.Sequential([
layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),
layers.MaxPooling2D((2, 2)),
layers.Conv2D(64, (3, 3), activation='relu'),
layers.MaxPooling2D((2, 2)),
layers.Conv2D(64, (3, 3), activation='relu'),
layers.Flatten(),
layers.Dense(64, activation='relu'),
layers.Dense(10, activation='softmax')
])
# 编译并训练模型
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(train_images[..., np.newaxis], train_labels, epochs=5, validation_data=(test_images[..., np.newaxis], test_labels))
# Tkinter UI 界面,手写输入并预测数字
class DigitRecognizerApp:
def __init__(self, root):
self.root = root
self.root.title("Handwritten Digit Recognition")
# 创建画布用于手写,绑定窗口大小变化时调整画布大小
self.canvas = tk.Canvas(self.root, bg='white')
self.canvas.grid(row=0, column=0, pady=2, padx=2, sticky="nsew")
self.canvas.bind("<B1-Motion>", self.paint)
self.canvas.bind("<Configure>", self.resize_canvas)
# 初始图像对象
self.image = Image.new("L", (200, 200), 255)
self.draw = ImageDraw.Draw(self.image)
# 按钮:清除画布
self.clear_button = tk.Button(self.root, text="Clear", command=self.clear_canvas)
self.clear_button.grid(row=1, column=0, pady=2, sticky="ew")
# 按钮:预测数字
self.predict_button = tk.Button(self.root, text="Predict", command=self.predict_digit)
self.predict_button.grid(row=2, column=0, pady=2, sticky="ew")
# 结果显示区
self.result_label = tk.Label(self.root, text="Prediction: None", font=('Helvetica', 16))
self.result_label.grid(row=3, column=0, pady=2, sticky="ew")
# 概率显示区 - 显示最高概率数字和所有概率
self.prob_frame = tk.Frame(self.root)
self.prob_frame.grid(row=4, column=0, pady=2, sticky="nsew")
self.highest_prob_label = tk.Label(self.prob_frame, text="Highest Probability: None", font=('Helvetica', 12))
self.highest_prob_label.pack(pady=2)
self.prob_text = tk.Text(self.prob_frame, height=10, font=('Helvetica', 12))
self.prob_text.pack(fill=tk.BOTH, expand=True)
# 调整窗口布局
self.root.grid_rowconfigure(0, weight=1)
self.root.grid_columnconfigure(0, weight=1)
self.root.grid_rowconfigure(4, weight=1) # 使概率显示区域自适应
def paint(self, event):
# 在画布上绘制手写输入
x, y = event.x, event.y
r = 8 # 手写笔的半径
self.canvas.create_oval(x-r, y-r, x+r, y+r, fill='black')
self.draw.ellipse([x-r, y-r, x+r, y+r], fill='black')
def resize_canvas(self, event):
# 调整图像大小,保持用户手写的内容
new_width, new_height = event.width, event.height
self.image = self.image.resize((new_width, new_height), Image.ANTIALIAS)
self.draw = ImageDraw.Draw(self.image)
def clear_canvas(self):
# 清除画布
self.canvas.delete("all")
self.image = Image.new("L", (self.canvas.winfo_width(), self.canvas.winfo_height()), 255)
self.draw = ImageDraw.Draw(self.image)
self.result_label.config(text="Prediction: None")
self.highest_prob_label.config(text="Highest Probability: None")
self.prob_text.delete(1.0, tk.END)
def predict_digit(self):
# 将用户手写的图像处理为模型输入格式
img = self.image.resize((28, 28)) # 将图像调整为28x28
img = ImageOps.invert(img) # 反转颜色,黑底白字
img = np.array(img).reshape(1, 28, 28, 1) / 255.0 # 标准化
# 使用模型进行预测
predictions = model.predict(img)
predicted_digit = np.argmax(predictions[0]) # 最高概率的数字
probabilities = predictions[0] # 每个数字的概率
highest_prob = probabilities[predicted_digit] # 获取最高概率
# 更新UI显示结果
self.result_label.config(text=f"Prediction: {predicted_digit}")
self.highest_prob_label.config(text=f"Highest Probability: {predicted_digit} ({highest_prob:.4f})")
# 显示所有数字的概率
self.prob_text.delete(1.0, tk.END)
for i in range(10):
self.prob_text.insert(tk.END, f"Digit {i}: {probabilities[i]:.4f}\n")
# 启动应用程序
if __name__ == "__main__":
root = tk.Tk()
app = DigitRecognizerApp(root)
root.mainloop()
还有点缺陷就是不能ui界面不能根据画面的放大缩小自动适应