基于最大熵和边缘信息的非结构化道路检测

doi:10.16180/j.cnki.issn1007-7820.2020.01.005

摘要/Abstract

摘要：

针对非结构化道路因场景干扰因素多而难以准确检测的问题,文中提出了一种基于最大熵和边缘信息相结合的非结构化道路检测算法。根据熵值大的地方,图像灰度相对较均匀;熵值小的地方,图像灰度离散性较大的特性,采用二维最大熵做道路初分割。同时,结合道路边缘信息解决道路检测因光照不均、阴影和水迹等因素干扰而导致道路检测不准确的问题。实验结果表明,该算法能够准确检测出光照、阴影、水迹的路面,并且不受道路类型的影响,满足实时性要求。

关键词: 非结构化道路, 二维最大熵, 边缘检测, 形态学滤波, 道路分割, 随机一致性算法

Abstract:

Aiming at the problem that the unstructured roads are difficult to accurately detect due to the various scene interference factors, an unstructured road detection algorithm based on the combination of maximum entropy and edge information was proposed. In this paper, the two-dimensional maximum entropy was used for the initial segmentation of the road because of the fact that the gray level of the high entropy fields of image was relatively uniform and vice versa. Furthermore, the road edge information was applied to further improve the inaccurate problem of road detection caused by uneven illumination, shadow and water stain. The experimental results indicated that the algorithm could accurately detect the road surface with light, shadow and water without the effect of the road environment and meet requirements of real-time.

Key words: unstructured road, two-dimensional maximum entropy, edge detection, morphological filtering, road segmentation, random consistency algorithm

中图分类号:

TP751

王翔,张娟,方志军. 基于最大熵和边缘信息的非结构化道路检测[J]. 电子科技, 2020, 33(1): 23-28.

WANG Xiang,ZHANG Juan,FANG Zhijun. Unstructured Road Detection Based on Edge Information and Maximum Entropy Segmentation[J]. Electronic Science and Technology, 2020, 33(1): 23-28.

图/表 15

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

图15

参考文献 20

[1]	胡晓辉, 孙苗强, 苏晓许 . 伪彩色空间完全非结构化道路检测方法[J]. 中国图象图形学报, 2012,17(2):203-208.
	Hu Xiaohui, Sun Miaoqiang, Su Xxiaoxu . Completely unstructured road detection for pseudo-color space[J]. Journal of Image and Graphics, 2012,17(2):203-208.
[2]	Álvarez J M, López A M, Gevers T , et al. Combining priors, appearance, and context for road detection[J]. IEEE Transactions on Intelligent Transportation Systems, 2014,15(3):1168-1178.
[3]	王永忠, 文成林 . 基于Haar纹理的非结构化道路消失点检测[J]. 中国图象图形学报, 2013,18(4):382-391.
	Wang Yongzong, Wen Chengling . Vanishing point detection of unstructured road based on haar texture[J]. Journal of Image and Graphics, 2013,18(4):382-391.
[4]	刘媛, 宋晓琳, 张伟伟 . 基于多模型结合优化的非结构化道路检测算法[J]. 工程设计学报, 2013(2):157-162.
	Liu Yan, Song Xiaoling, Zhang Weiwei . Unstructured road-detection algorithm based on multiple models and optimization[J]. Journal of Engineering Design, 2013(2):157-162.
[5]	陈清华, 杨静宇, 陈建亭 . 基于自适应模板的非结构化道路检测[J]. 东南大学学报(自然科学版), 2007,37(6):1102-1107.
	Chen Qinghua, Yang Jingyu, Chen Jianting . Lane and road detection based on adaptive deformable template[J]. Journal of Southeast University(Natural Science Edition), 2007,37(6):1102-1107.
[6]	王海, 蔡英凤, 贾允毅 , 等. 基于深度卷积神经网络的场景自适应道路分割算法[J]. 电子与信息学报, 2017,39(2):263-269.
	Wang Hai, Cai Yingfong, Jia Yuyi , et al. Scene adaptive road segmentation algorithm based on deep convolutional neural network[J]. Journal of Electronics & Information Technology, 2007,37(6):1102-1107.
[7]	Li J, Xue M, Prokhorov D . Deep neural network for structural prediction and lane detection in traffic scene[J]. IEEE Transactions on Neural Networks & Learning Systems, 2017,28(3):690-698.
[8]	李锋, 阚建霞 . 基于Sobel算子的图像快速二维最大熵阈值分割算法[J]. 计算机科学, 2015,42(z1):209-210,220.
	Li Feng, Kan Jianxia . Fast two-dimensional maximum mntropy threshold segmentation method based on sobel operator[J]. Computer Science, 2015,42(z1):209-210,220.
[9]	叶志伟, 王明威, 靳华中 , 等. 基于混合杜鹃搜索算法的图像二维熵阈值方法[J]. 计算机仿真, 2015,32(10):287-291.
	Ye Zhiwei, Wang Mingwei, Jin Huazhong , et al. An image segmentation approach using hybrid cuckoo search algorithm and 2-d entropy[J]. Computer Simulation, 2015,32(10):287-291.
[10]	Ye Z, Hu Z, Lai X , et al. Image segmentation using thresholding and swarm intelligence[J]. Journal of Software, 2012,7(5):176-189.
[11]	付凯城, 张竹馨, 曲毅 . 基于Canny-ROA算子和Hough变换的SAR图像道路检测[J]. 现代电子技术, 2016,39(23):1-4.
	Fu Kaicheng, Zhang Zhuxin, Qu Yi . Canny-ROA operator and hough transform based road detection according to SAR image[J]. Modern Electronic Technique, 2016,39(23):1-4.
[12]	兰丽, 胡晓辉, 孙苗强 . 一种基于灰关联和Sobel算子的完全非结构化道路边缘检测方法[J]. 湖南工程学院学报(自然科学版), 2013,23(1):42-44.
	Lan Li, Hu Xiaohui, Sun Miaoqiang . Completely unstructured road edge detection based on grey relation and soble operator[J]. Journal of Hunan Institute of Engineering(Natural Science Edition), 2013,23(1):42-44.
[13]	Wang N, Zhao H L, Ju S G , et al. An improved adaptive canny edge detection algorithm[J]. Journal of Sichuan University(Natural Science Edition), 2014,51(3):479-482.
[14]	Xu H K, Qin Y Y, Chen H R . An improved algo-rithm for edge detection based on canny[J]. Infrared Technology, 2014,36(3):210-214.
[15]	张建廷, 张立民 . 新型自适应稳健双边滤波图像分割[J]. 浙江大学学报(工学版), 2016,50(9):1703-1710.
	Zhang Jianting, Zhang Limin . Image segmentation using novel adaptive robust bilateral filter[J]. Journal of Zhejiang University(Engineering Science), 2016,50(9):1703-1710.
[16]	姒绍辉, 胡伏原, 张伟 , 等. 一种改进的双边滤波图像去噪算法[J]. 计算机工程与应用, 2015(2):185-189.
	Si Saohui, Hu Fuyuan, Zhang Wei , et al. Improved bilateral filtering algorithm[J]. Computer Engineering and Applications, 2015(2):185-189.
[17]	崔仲远, 张少辉 . 多尺度Retinex和双边滤波相融合的图像增强算法[J]. 激光杂志, 2015,36(4):90-93.
	Cui Zongyuan, Zhang Shaohui . Image enhancement algorithm based on multi-scale retinex and bilateral filter[J]. Laser Journal, 2015,36(4):90-93.
[18]	Peng H, Xiao J S, Shen S M , et al. A fast algorithm based on RANSAC for vision lane detection[J]. Journal of Shanghai Jiaotong University, 2014,48(12):1721-1726.
[19]	Chao F, Shuai D, Hou L , et al. Lane marking detection technique based on improved RANSAC algorithm[J]. Automotive Engineering, 2014,36(4):503-508.
[20]	Aharon Bar Hillel, Ronen Lerner, Dan Levi , et al. Re-cent progress in road and lane detection:a survey[J]. Machine Vision and Applications, 2014(25):727-745.