【读书1】【2017】MATLAB与深度学习——卷积层(4)

图6-13 当图像矩阵与滤波器不匹配时，较大的重要元素不会起到显著的作用Whenthe image matrix does not match the filter, the significant elements are notaligned

这是因为图像矩阵与滤波器不匹配，图像矩阵的重要元素在错误的方向上对齐。

This is because the image matrix does notmatch the filter; the significant elements of the image matrix are aligned inthe wrong direction.

以同样的方式处理第二个卷积滤波器，产生如图6- 14所示的特征映射。

In the same manner, processing the secondconvolution filter produces the feature map shown in Figure 6-14.

图6-14 卷积输出的大小取决于图像矩阵与卷积滤波器是否匹配The valuesdepend on whether the image matrix matches the convolution filter

与第一次卷积操作类似，该特征映射中的元素值取决于图像矩阵是否与卷积滤波器匹配。

Similarly to the first convolutionoperation, the values in the elements of this feature map depend on whether theimage matrix matches the convolution filter or not.

综上所述，卷积层实现输入图像与卷积滤波器的运算并产生特征映射。

In summary, the convolution layer operatesthe convolution filters on the input image and produces the feature maps.

在卷积层中提取的特征由经过训练的卷积滤波器确定。

The features that are extracted in theconvolution layer determined by the trained convolution filters.

因此，卷积层提取的特征取决于使用哪一种卷积滤波器。

Therefore, the features that theconvolution layer extracts vary depending on which convolution filter is used.

卷积滤波器创建的特征映射在该层产生输出之前通过激活函数进行处理。

The feature map that the convolution filtercreates is processed through the activation function before the layer yieldsthe output.

卷积层的激活函数与普通神经网络的激活函数相同。

The activation function of the convolutionlayer is identical to that of the ordinary neural network.

虽然ReLU函数在当前的大多数应用中使用，但sigmoid函数和tanh函数也经常被使用。

Although the ReLU function is used in mostof the recent applications, the sigmoid function and the tanh function areoften employed as well.

仅供参考，在数字信号处理领域中广泛应用的滑动平均滤波器是一种特殊的卷积滤波器。

Just for the reference, the moving averagefilter, which is widely used in the digital signal processing field, is aspecial type of convolution filter.

如果你熟悉数字滤波器，将它们与之联系起来，可以更好地理解卷积滤波器蕴含的思想。

If you are familiar with digital filters,relating them to this concept may allow you to better understand the ideasbehind the convolution filter.

池化层（Pooling Layer）

池化层的作用是减小图像的大小，将图像特定区域的相邻像素合成为单个代表值。

The pooling layer reduces the size of theimage, as it combines neighboring pixels of a certain area of the image into asingle representative value.

池化是许多其它图像处理方案中已经采用的典型技术。

Pooling is a typical technique that manyother image processing schemes have already been employing.

为了在池化层中进行操作，我们应该确定如何从图像中选择池化像素以及如何计算代表值。

In order to conduct the operations in thepooling layer, we should determine how to select the pooling pixels from theimage and how to set the representative value.

通常是从正方形矩阵中选择相邻的像素，而像素合成的数量因具体问题而异。

The neighboring pixels are usually selectedfrom the square matrix, and the number of pixels that are combined differs fromproblem to problem.

代表值通常被设置为所选像素的平均值或最大值。

The representative value is usually set asthe mean or maximum of the selected pixels.

池化层的操作非常简单。

The operation of the pooling layer issurprisingly simple.

——本文译自Phil Kim所著的《Matlab Deep Learning》

更多精彩文章请关注微信号：