LeNet-5 简介

在神经网络的和深度学习领域，Yann LeCun 在 1998 年在 IEEE 上发表了 (Gradient-based learning applied to document recognition，1998)， 文中首次提出了 卷积-池化-全连接 的神经网络结构， 由 LeCun 提出的七层网络命名为 LeNet-5，因而也为他赢得了 CNN 之父的美誉。

LeNet-5 网络结构

作为标准的卷积网络结构，LeNet-5 对后世的影响深远，以至于在 16 年后，谷歌提出 Inception 网络时也将其命名为 GoogLeNet， 以致敬 Yann LeCun 对卷积神经网络发展的贡献。然而 LeNet-5 提出后的十几年里，由于神经网络的可解释性问题和计算资源的限制， 神经网络的发展一直处于低谷。

LeNet-5 共有 5 层(输入输出层不计入层数，池化层与卷积层算 1 层). 每层都有一定的训练参数，其中三个卷积层的训练参数较多，每层都有多个滤波器(特征图)， 每个滤波器都对上一层的输出提取不同的像素特征：

• Layer input：输入
• Layer 1：卷积-池化
• Layer 2：卷积-池化
• Layer 3：卷积(全连接)
• Layer 4：全连接
• Layer 5：全连接
• Layer output：输出

原始论文中的网络结构：

下面是对网络结构的详解：

As the name indicates，LeNet5 has 5 layers with two convolutional and three fully connected layers. Let’s start with the input. LeNet5 accepts as input a greyscale image of 32x32， indicating that the architecture is not suitable for RGB images (multiple channels). So the input image should contain just one channel. After this，we start with our convolutional layers.

The first convolutional layer has a filter size of 5x5 with 6 such filters. This will reduce the width and height of the image while increasing the depth (number of channels). The output would be 28x28x6. After this，pooling is applied to decrease the feature map by half， i.e，14x14x6. Same filter size (5x5) with 16 filters is now applied to the output followed by a pooling layer. This reduces the output feature map to 5x5x16.

After this，a convolutional layer of size 5x5 with 120 filters is applied to flatten the feature map to 120 values. Then comes the first fully connected layer，with 84 neurons. Finally，we have the output layer which has 10 output neurons， since the MNIST data have 10 classes for each of the represented 10 numerical digits.

参考

• Gradient-Based Learning Applied to Document Recognition
• 卷积神经网络（CNN）基础及经典模型介绍