Abstract:

Semantic segmentation is a kind of pixel level segmentation and classification of different objects in images, it is important for image processing and is widely used in different fields. Deep convolutional neural networks has achieved remarkable results in machine vision in recent years. Aiming at the task of semantic segmentation of dense prediction, this paper proposes an improved method based on VGGNet in which the shallow information is fused with the deeper feature map, and feature extraction and fusion are carried out by using paralleling convolution with different sampling rates. This method is more effective in extracting features and context information from different layers, then improves the accuracy of semantic segmentation. This paper also improves the accuracy of semantic segmentation by optimizing the edge of the images with a fully connected conditional random fields. This paper reaches 71.3% mIOU in the semantic segmentation task of test set of PASCAL VOC 2012, and the result is superior to the previous main classical methods based on VGGNet.

Key words: semantic segmentation, convolutional neural networks, machine vision, dense prediction, fully connected conditional random fields

摘要：

语义分割是对图像中的不同目标进行像素级的分割和分类，是图像处理领域中的一项重要研究，应用十分广泛。深度卷积神经网络在近几年的机器视觉研究中取得了显著成效。针对密集预测的语义分割任务，提出了一种基于VGGNet网络的方法。该方法在深层特征图像中融合了浅层信息，且采用并行的不同采样率的空洞卷积进行特征提取与融合，更有效地提取不同层的特征和上下文信息，从而提高语义分割精度。采用全连接条件随机场优化图像边界，进一步提高语义分割的精度。该方法在PASCAL VOC 2012语义分割任务测试集中取得了71.3% mIOU的结果，优于之前基于VGGNet的主要经典方法。

关键词: 语义分割, 卷积神经网络, 机器视觉, 密集预测, 全连接条件随机场