基于TensorFlow2.0（keras）+ flask 实现手写数字识别web应用 demo

声明 （2019年12月16日晚23点）

识别率达到95%以上。已经是个成熟的demo了

地址

http://112.126.101.188:3335/

技术选型

TensorFlow2.0，使用 keras Api ，构造 cnn 模型训练，flask 提供服务，前端使用了 canvas 绘图。

github

https://github.com/ybsdegit/Keras_flask_mnist

欢迎大家 star，fork

• 安装依赖

pip install -r requirements.txt

运行

• 启动服务

python app.py

本地启动访问地址为：http://localhost:3335/

• 训练
  源码中也包含训练好的模型 model.h5,测试集成功率99.9，也可以自行训练。

python model/train.py

效果

• flask app

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 2019/9/15 14:27
# @Author  : Paulson
# @File    : app.py
# @Software: PyCharm
# @define  : function

from flask import Flask, render_template, request
import numpy as np
import tensorflow.keras as keras
import re
import base64
from tensorflow.keras.preprocessing.image import img_to_array, load_img

app = Flask(__name__)

model_file = './model/model.h5'
global model
model = keras.models.load_model(model_file)

@app.route('/')
def index():
    return render_template("index.html")

@app.route('/predict/', methods=['Get', 'POST'])
def preditc():
    parseImage(request.get_data())
    img = img_to_array(load_img('output.png', target_size=(28, 28), color_mode="grayscale")) / 255.
    img = np.expand_dims(img, axis=0)
    response = model.predict_classes(img)[0]
    return str(response)

def parseImage(imgData):
    imgStr = re.search(b'base64,(.*)', imgData).group(1)
    with open('./output.png', 'wb') as output:
        output.write(base64.decodebytes(imgStr))


if __name__ == '__main__':
    app.run(host="0.0.0.0", port=3335)

• 训练

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 2019/9/15 14:14
# @Author  : Paulson
# @File    : train.py
# @Software: PyCharm
# @define  : function
import tensorflow as tf
import tensorflow.keras as keras
from tensorflow.keras import Sequential
from tensorflow.keras.layers import Flatten, Dense, Conv2D, MaxPooling2D, Dropout, MaxPool2D
from tensorflow.keras.datasets import mnist
from tensorflow.keras import backend as K
import json


batch_size = 128
num_classes = 10
epochs = 10

# input image dimensions
img_rows, img_cols = 28, 28

# the data, shuffled and split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()

if K.image_data_format() == 'channels_first':
    x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
    x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
    input_shape = (1, img_rows, img_cols)
else:
    x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
    x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
    input_shape = (img_rows, img_cols, 1)
    
x_train = x_train.astype('float32') / 255
x_test = x_test.astype('float32') / 255

print('x_train shape:', x_train.shape)
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')

# convert class vectors to binary class matrices
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
print(input_shape)

# 构建网络
model = Sequential()
# 第一个卷积层，32个卷积核，大小５x5，卷积模式SAME,激活函数relu,输入张量的大小
model.add(Conv2D(filters=32, kernel_size=(5, 5), padding='Same', activation='relu',
                 input_shape=(28, 28, 1)))
model.add(Conv2D(filters=32, kernel_size=(5, 5), padding='Same', activation='relu'))
# 池化层,池化核大小２x2
model.add(MaxPool2D(pool_size=(2, 2)))
# 随机丢弃四分之一的网络连接，防止过拟合
model.add(Dropout(0.25))
model.add(Conv2D(filters=64, kernel_size=(3, 3), padding='Same', activation='relu'))
model.add(Conv2D(filters=64, kernel_size=(3, 3), padding='Same', activation='relu'))
model.add(MaxPool2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Dropout(0.25))
# 全连接层,展开操作，
model.add(Flatten())
# 添加隐藏层神经元的数量和激活函数
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.25))
# 输出层
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])


model.fit(x_train, y_train, batch_size=batch_size,epochs=epochs,verbose=1,validation_data=(x_test, y_test))
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])


with open('model.json', 'w') as outfile:
    json.dump(model.to_json(), outfile)
    
model_file = 'model.h5'
model.save(model_file)