计算机应用 ›› 2019, Vol. 39 ›› Issue (9): 2505-2510.DOI: 10.11772/j.issn.1001-9081.2019030488

• 人工智能 • 上一篇    下一篇

结合语义边界信息的道路环境语义分割方法

宋小娜1,2, 芮挺1, 王新晴1   

  1. 1. 中国人民解放军陆军工程大学 野战工程学院, 南京 210018;
    2. 华北水利水电大学 机械学院, 郑州 450045
  • 收稿日期:2019-03-25 修回日期:2019-05-07 出版日期:2019-09-10 发布日期:2019-05-17
  • 通讯作者: 芮挺
  • 作者简介:宋小娜(1982-),女,河南南阳人,讲师,博士研究生,主要研究方向:图像处理、模式识别、深度学习;芮挺(1972-),男,江苏南京人,教授,博士,主要研究方向:人工智能、模式识别;王新晴(1963-),男,江苏泰州人,教授,博士,主要研究方向:信号处理、智能算法、无人化智能车辆。
  • 基金资助:

    国家重点研发计划项目(2016YFC0802904);国家自然科学基金资助项目(61472444,61671470);江苏省自然科学基金资助项目(BK20161470)。

Semantic segmentation method of road environment combined semantic boundary information

SONG Xiaona<sup>1,2</sup>, RUI Ting<sup>1</sup>, WANG Xinqing<sup>1</sup>   

  1. 1. College of Field Engineering, Army Engineering University of People's Liberation Army, Nanjing Jiangsu 210018, China;
    2. College of Mechanical Engineering, North China University of Water Resources and Electric Power, Zhengzhou Henan 450045, China
  • Received:2019-03-25 Revised:2019-05-07 Online:2019-09-10 Published:2019-05-17
  • Supported by:

    This work is partially supported by the National Key Research and Development Program of China (2016YFC0802904), the National Natural Science Foundation of China (61472444, 61671470), the Natural Science Foundation of Jiangsu Province (BK20161470).

摘要:

语义分割是实现道路语义环境解释的重要方法,深度学习语义分割由于卷积、池化及反卷积的作用使分割边界模糊、不连续以及小目标漏分错分,影响了分割效果,降低了分割精度。针对上述问题,提出了一种结合语义边界信息的新的语义分割方法,首先在语义分割深度模型中构建了一个语义边界检测子网,利用网络中的特征共享层将语义边界检测子网络学习到的语义边界信息传递给语义分割网络;然后结合语义边界检测任务和语义分割任务定义了新的模型代价函数,同时完成语义边界检测和语义分割两个任务,提升语义分割网络对物体边界的描述能力,提高语义分割质量。最后在Cityscapes数据集上进行一系列实验证明,结合语义边界信息的语义分割方法在准确率上比已有的语义分割网络SegNet提升了2.9%,比ENet提升了1.3%。所提方法可以改善语义分割中出现的分割不连续、物体边界不清晰、小目标错分漏分、分割精度不高等问题。

关键词: 语义分割, 全卷积神经网络, 道路环境感知, 边缘检测, 无人驾驶车辆

Abstract:

Semantic segmentation is an important method to interpret the road semantic environment. The convolution, pooling and deconvolution in semantic segmentation of deep learning result in blur and discontinuous segmentation boundary, missing and wrong segmentation of small objects. These influence the outcome of segmentation and reduce the accuracy of segmentation. To deal with the problems above, a new semantic segmentation method combined semantic boundary information was proposed. Firstly, a subnet of semantic boundary detection was built in the deep model of semantic segmentation, and the feature sharing layers in the network were used to transfer the semantic boundary information learned in the semantic boundary detection subnet to the semantic segmentation network. Then, a new cost function of the model was defined according to the tasks of semantic boundary detection and semantic segmentation. The model was able to accomplish two tasks simultaneously and improve the descriptive ability of object boundary and the quality of semantic segmentation. Finally, the method was verified on the Cityscapes dataset. The experimental results demonstrate that the accuracy of the method proposed is improved by 2.9% compared to SegNet and is improved by 1.3% compared to ENet. It can overcome the problems in semantic segmentation such as discontinous segmentation, blur boundary of object, missing and wrong segmentation of small objects and low accuracy of segmentation.

Key words: semantic segmentation, Fully Convolutional Network (FCN), road environment perception, boundary detection, unmanned vehicle

中图分类号: