U-Net论文阅读（逐句翻译+精读）

U-Net最初是一个用于二维图像分割的卷积神经网络，分别赢得了ISBI 2015细胞追踪挑战赛和龋齿检测挑战赛的冠军. U-net是基于全卷积网络拓展和修改而来，网络由两部分组成：一个收缩路径(contracting path)来获取上下文信息以及一个对称的扩张路径(expanding path）用以精确定位。下面就来精读一下这篇论文吧~

1. Abstract

There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window
convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Moreover, the network is fast. Segmentation
of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net

为了更有效的利用标注数据，我们使用了数据增强的方法(data augmentation)。

我们的网络由两部分组成：一个收缩路径(contracting path)来获取上下文信息以及一个对称的扩张路径(expanding path)用以精确定位。

这种网络可以从很少的图像中进行端到端的训练。
这个网络非常的快

2. Introduction

In the last two years, deep convolutional networks have outperformed the state of the art in many visual recognition tasks.While convolutional networks have already existed for a long time, their success was limited due to the size of the available training sets and the size of the considered networks. The breakthrough by Krizhevsky et al. was due to supervised training of a large network with 8 layers and millions of parameters on the ImageNet dataset with 1 million training images. Since then, even larger and deeper networks have been trained.

The typical use of convolutional networks is on classification tasks, where the output to an image is a single class label. However, in many visual tasks, especially in biomedical image processing, the desired output should include localization, i.e., a class label is supposed to be assigned to each pixel. Moreover, thousands of training images are usually beyond reach in biomedical tasks.
Hence, Ciresan et al. trained a network in a sliding-window setup to predict the class label of each pixel by providing a local region (patch) around that pixel as input. First, this network can localize. Secondly, the training data in terms
of patches is much larger than the number of training images. The resulting network won the EM segmentation challenge at ISBI 2012 by a large margin.

Obviously, the strategy in Ciresan et al. has two drawbacks. First, it is quite slow because the network must be run separately for each patch（对每个点都要截取一块图进行训练）, and there is a lot of redundancy due to overlapping patches. Secondly, there is a trade-off between localization accuracy and the use of context. Larger patches require more max-pooling layers that reduce the localization accuracy（如果截取的那块图过大，会损失掉局部信息）, while small patches allow the network to see only little context.

FCN的思想是：

In this paper, we build upon a more elegant architecture, the so-called “fully convolutional network”. We modify and extend this architecture such that it works with very few training images and yields more precise segmentations(见图1). The main idea in fully convolutional network is to supplement a usual contracting network by successive layers（"U"型的右侧）, where pooling operators are replaced by upsampling operators. Hence, these layers increase the resolution of the output. In order to localize, high resolution features from the contracting path are combined with the upsampled output（skip-connection）. A successive convolution layer can then learn to assemble a more precise output based on this information.

作者的改进：

One important modification in our architecture is that in the upsampling part we have also a large number of feature channels, which allow the network to propagate context information to higher resolution layers. As a consequence,
the expansive path is more or less symmetric to the contracting path, and yields a u-shaped architecture. The network does not have any fully connected layers and only uses the valid part of each convolution, i.e., the segmentation map only contains the pixels, for which the full context is available in the input image. This strategy allows the seamless segmentation of arbitrarily large images by an overlap-tile strategy (see Figure 2). To predict the pixels in the border region of the image, the missing context is extrapolated by mirroring the input image. This tiling strategy is important to apply the network to large images, since otherwise the resolution would be limited by the GPU memory.

与FCN不同的是：

我们的网络在上采样部分依然有大量的特征通道，这使得网络可以将空间上下文信息向更高的分辨率层传播。结果是，上采样路径基本对称于下采样路径，并呈现出一个U型。

网络不存在任何全连接层，并且，只使用每个卷积的valid部分，例如，分割图只包含这样一些像素点，这些像素点的完整上下文都出现在输入图像中。这种策略允许使用Overlap-tile策略无缝地分割任意大小的图像(参见下图)。

为了预测图像边界区域的像素点，我们采用镜像图像的方式补全缺失的环境像素。这个tiling方法在使用网络分割大图像时是非常有用的，因为如果不这么做，GPU显存会限制图像分辨率。

Overlap-tile策略可以无缝分割任意大小的图像(这里分割的神经元结构在EM堆叠)。黄色区域是预测的分割，需要蓝色区域内的图像数据作为输入。通过镜像的方式外推缺少的输入数据。

3. Network Architecture

The network architecture is illustrated in Figure 1. It consists of a contracting path (left side) and an expansive path (right side). The contracting path follows the typical architecture of a convolutional network. It consists of the repeated application of two 3x3 convolutions (unpadded convolutions), each followed by a rectified linear unit (ReLU) and a 2x2 max pooling operation with stride 2 for downsampling. At each downsampling step we double the number of feature channels.
Every step in the expansive path consists of an upsampling of the feature map followed by a 2x2 convolution (“up-convolution”) that halves the number of feature channels, a concatenation with the correspondingly cropped
feature map from the contracting path, and two 3x3 convolutions, each followed by a ReLU. The cropping is necessary due to the loss of border pixels in every convolution. At the final layer a 1x1 convolution is used to map each 64-component feature vector to the desired number of classes. In total the network has 23 convolutional layers. To allow a seamless tiling of the output segmentation map (see Figure 2), it is important to select the input tile size such that all 2x2 max-pooling operations are applied to a layer with an even x- and y-size.

contracting path是典型的卷积网络架构：

架构中含有着一种重复结构，每次重复中都有2个 3*3 卷积层（无padding）、非线性ReLU层和一个 2*2 max pooling层（stride为2）。每一次下采样后我们都把特征通道的数量加倍。

expansive path也使用了一种相同的排列模式：

每一步都首先使用反卷积(up-convolution)，每次使用反卷积都将特征通道数量减半，特征图大小加倍。反卷积过后，将反卷积的结果与contracting path中对应步骤的特征图拼接起来。
contracting path中的特征图尺寸稍大，将其修剪过后进行拼接。对拼接后的map再进行2次3*3的卷积。

最后一层的卷积核大小为1*1，将64通道的特征图转化为特定类别数量（分类数量，二分类为2）的结果.