基于CNN的分类算法

import numpy as np import tensorflow as tf import os from PIL import Image import base64 import matplotlib.pyplot as plt batchSize = 30 num_epochs = 200 import pandas as pd # # #dataset = np.load('trainData.npz') # x_test = dataset['arr_0'] # # x = np.array(x) # # x = x.reshape([-1, 400, 400, 1]) # y_test = dataset['arr_1'] # division = int(len(x_test) * 0.5) # x_test = x_test[division:] # y_test = y_test[division:] def tfRecordRead(fileNameQue, heigh, width, channels, n_class): reader = tf.TFRecordReader() # 创建一个队列来维护输入文件列表 # 从文件中读出一个Example _, serialized_example = reader.read(fileNameQue) # 用FixedLenFeature将读入的Example解析成tensor features = tf.parse_single_example( serialized_example, features={ 'image': tf.FixedLenFeature([], tf.string), 'label': tf.FixedLenFeature([], tf.int64) }) # 将字符串解析成图像对应的像素数组 image = tf.decode_raw(features['image'], tf.float32) # image = tf.decode_raw(features["image"], tf.uint8) image = tf.reshape(image, [heigh, width, channels]) # image = tf.cast(image, tf.float32) * (1 / 255.0) labels = tf.cast(features['label'], tf.int64) labels = tf.one_hot(labels, n_class, 1, 0) return image, labels def tfRecordBatchRead(filename, heigh, width, channels, n_class, batchSize): fileNameQue = tf.train.string_input_producer([filename], shuffle=True, num_epochs=num_epochs) image, labels = tfRecordRead(fileNameQue, heigh, width, channels, n_class) # fetch图像和label min_after_dequeue = 100 capacity = min_after_dequeue + 3 * batchSize # 预取图像和label并随机打乱，组成batch，此时tensor rank发生了变化，多了一个batch大小的维度 imageBatch, labelBatch = tf.train.shuffle_batch([image, labels], batch_size=batchSize, capacity=capacity, min_after_dequeue=min_after_dequeue, num_threads=2) return imageBatch, labelBatch filename = r'D:\ZWJProject\tensorflow\CNN\record\Imageoutput.tfrecords' # filename = 'Imageoutput.tfrecords' heigh, width, channels, n_class = 100, 100, 3, 3 print(heigh, width, channels, n_class) imageBatch, labelBatch = tfRecordBatchRead(filename, heigh, width, channels, n_class, batchSize) init_op = tf.group(tf.global_variables_initializer(), tf.local_variables_initializer()) # -----------------构建网络---------------------- ''' 本程序cnn网络模型，共有7层，前三层为卷积层，后三层为全连接层，前三层中，每层包含卷积、激活、池化层 ''' # 占位符，设置输入参数的大小和格式 x = tf.placeholder(tf.float32, [None, heigh, width, channels]) y = tf.placeholder(tf.float32, [None, n_class]) def inference(input_tensor, train, regularizer, channels=3, n_class=3): # -----------------------第一层---------------------------- with tf.variable_scope('layer1-conv1'): # 初始化权重conv1_weights为可保存变量，大小为5x5,1（RGB），数量为32个 conv1_weights = tf.get_variable("weight", [5, 5, channels, 32], initializer=tf.truncated_normal_initializer(stddev=0.1)) # 初始化偏置conv1_biases，数量为32个 conv1_biases = tf.get_variable("bias", [32], initializer=tf.constant_initializer(0.0)) # 卷积计算，tf.nn.conv2d为tensorflow自带2维卷积函数，input_tensor为输入数据，conv1_weights为权重，strides=[1, 1, 1, 1]表示左右上下滑动步长为1，padding='SAME'表示输入和输出大小一样，即补0 conv1 = tf.nn.conv2d(input_tensor, conv1_weights, strides=[1, 1, 1, 1], padding='SAME') # 激励计算，调用tensorflow的relu函数 relu1 = tf.nn.relu(tf.nn.bias_add(conv1, conv1_biases)) with tf.name_scope("layer2-pool1"): # 池化计算，调用tensorflow的max_pool函数，strides=[1,2,2,1]，表示池化边界，2个对一个生成，padding="VALID"表示不操作。 pool1 = tf.nn.max_pool(relu1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding="VALID") # -----------------------第二层---------------------------- with tf.variable_scope("layer3-conv2"): # 同上，不过参数的有变化，根据卷积计算和通道数量的变化，设置对应的参数 conv2_weights = tf.get_variable("weight", [5, 5, 32, 64], initializer=tf.truncated_normal_initializer(stddev=0.1)) conv2_biases = tf.get_variable("bias", [64], initializer=tf.constant_initializer(0.0)) conv2 = tf.nn.conv2d(pool1, conv2_weights, strides=[1, 1, 1, 1], padding='SAME') relu2 = tf.nn.relu(tf.nn.bias_add(conv2, conv2_biases)) with tf.name_scope("layer4-pool2"): pool2 = tf.nn.max_pool(relu2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID') # -----------------------第三层---------------------------- # 同上，不过参数的有变化，根据卷积计算和通道数量的变化，设置对应的参数 with tf.variable_scope("layer5-conv3"): conv3_weights = tf.get_variable("weight", [3, 3, 64, 128], initializer=tf.truncated_normal_initializer(stddev=0.1)) conv3_biases = tf.get_variable("bias", [128], initializer=tf.constant_initializer(0.0)) conv3 = tf.nn.conv2d(pool2, conv3_weights, strides=[1, 1, 1, 1], padding='SAME') relu3 = tf.nn.relu(tf.nn.bias_add(conv3, conv3_biases)) with tf.name_scope("layer6-pool3"): pool3 = tf.nn.max_pool(relu3, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID') # -----------------------第四层---------------------------- # 同上，不过参数的有变化，根据卷积计算和通道数量的变化，设置对应的参数 with tf.variable_scope("layer7-conv4"): conv4_weights = tf.get_variable("weight", [3, 3, 128, 128], initializer=tf.truncated_normal_initializer(stddev=0.1)) conv4_biases = tf.get_variable("bias", [128], initializer=tf.constant_initializer(0.0)) conv4 = tf.nn.conv2d(pool3, conv4_weights, strides=[1, 1, 1, 1], padding='SAME') relu4 = tf.nn.relu(tf.nn.bias_add(conv4, conv4_biases)) with tf.name_scope("layer8-pool4"): pool4 = tf.nn.max_pool(relu4, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID') nodes = 6 * 6 * 128 reshaped = tf.reshape(pool4, [-1, nodes]) # -----------------------第五层---------------------------- with tf.variable_scope('layer9-fc1'): # 初始化全连接层的参数，隐含节点为1024个 fc1_weights = tf.get_variable("weight", [nodes, 1024], initializer=tf.truncated_normal_initializer(stddev=0.1)) if regularizer != None: tf.add_to_collection('losses', regularizer(fc1_weights)) fc1_biases = tf.get_variable("bias", [1024], initializer=tf.constant_initializer(0.1)) # 使用relu函数作为激活函数 fc1 = tf.nn.relu(tf.matmul(reshaped, fc1_weights) + fc1_biases) # 采用dropout层，减少过拟合和欠拟合的程度，保存模型最好的预测效率 if train: fc1 = tf.nn.dropout(fc1, 0.5) # -----------------------第六层---------------------------- with tf.variable_scope('layer10-fc2'): # 同上，不过参数的有变化，根据卷积计算和通道数量的变化，设置对应的参数 fc2_weights = tf.get_variable("weight", [1024, 512], initializer=tf.truncated_normal_initializer(stddev=0.1))