卷积神经网络-学习笔记

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/sinat_35406909/article/details/83714925

1. Edge Detection

• Vertical Edge Detection: Convolve the image with Matrix $\begin{matrix} 1 & 0 & -1 \\ 1 & 0 & -1 \\ 1 & 0 & -1 \end{matrix}$
• Horizontal Edge Detection: Convolve the image with Matrix $\begin{matrix} 1 & 1 & 1 \\ 0 & 0 & 0 \\ -1 & -1 & -1 \end{matrix}$
• Sobel Filter: $\begin{matrix} 1 & 0 & -1 \\ 2 & 0 & -2 \\ 1 & 0 & -1 \end{matrix}$
  Give more attention to center
• Scharr Filter: $\begin{matrix} 3 & 0 & -3 \\ 10 & 0 & -10 \\ 3 & 0 & -3 \end{matrix}$

卷积函数:
python : conv_forward
opencv : filter2D

2. Padding

• Pad the image to save information from borders before convolution
• Valid Convolution: Direct Convolution, Size of output is restricted without padding
  Same Convolution: Size of output is the same as input

3. Strided Convolution

• The step of convolution is not one.
• size of input is n, stride s, filter size f, padding p, size of output is $\frac {n+2p-f} {s} +1$
• cross-correlation / convolution: flip the matrix both vertically and horizontally

4. Convolution on 3-D Images

5. Convolution Layer in CNN

• Convolve with filter
• Add bias
• Apply it to function $Relu$
• Go to the next layer

6. Pooling Layer in CNN

• Max Pooling: Reduce the size of features by taking MAX from every small matrix
• Average Pooling: taking the AVERAGE value

7. Fully-Connected Layer in CNN

• can be seen as a layer in neural network
• neighboring layers are fully-connected