Tensorflow 卷积操作示例

代码

import  tensorflow as tf
import  matplotlib.pyplot as plt

#读取jpg文件
original_data = tf.read_file("test.jpg")
#解析数据
img_data = tf.image.decode_jpeg(original_data)
img_data = tf.cast(img_data, tf.float64)
 
#卷积核，  卷积高度 * 宽度 * 通道数 *卷积核个数
filter = tf.Variable(tf.random_normal(shape=[5,5,3,3], dtype=tf.float64))

#卷积操作
img_filter_data = tf.nn.conv2d([img_data], filter, [1, 1, 1, 1], padding='SAME')
#值类型转换
img_filter_data_u64 = tf.cast(img_filter_data, tf.uint64)
with tf.Session() as sess:
    tf.global_variables_initializer().run()

    img_filter_data_u64 = sess.run(img_filter_data_u64)

    b,h,w,c =  (img_filter_data_u64.shape)
    #如果是灰度图片，需转换一下，才能在plt上显示
    if c == 1:
        img_filter_data_u64 = img_filter_data_u64.reshape(b,h,w)
    plt.imshow(img_filter_data_u64[0])
    plt.show()

效果