First, gradient image is constructed by using direction templates convolve with image.

首先利用方向模板对图像进行卷积，求得梯度图像。

youdao

We then convolve x_image with the weight tensor, add the bias, apply the ReLU function, and finally max pool.

我们把x_image和权值向量进行卷积，加上偏置项，然后应用ReLU激活函数，最后进行max pooling。

youdao

We then convolve x_image with the weight tensor, add the bias, apply the ReLU function, and finally max pool.

我们把x_image和权值向量进行卷积，加上偏置项，然后应用ReLU激活函数，最后进行max pooling。

youdao