基于自适应伽马校正的异常驾驶行为检测方法

doi:10.19678/j.issn.1000-3428.0065832

摘要/Abstract

摘要：

针对低照度条件下重型卡车司机异常驾驶行为检测方法存在检测准确率低、检测速度慢等问题，结合图像自适应增强方法和轮廓定位检测思想，提出一种基于自适应伽马校正的异常驾驶行为检测方法。对传入视频图像进行自适应伽马校正，通过抑制噪声、改善暗部和提升信息熵来提高识别准确率。基于图像灰度值和信息熵对双阈值伽马函数进行自适应调节，从而获得更丰富的边缘信息和色彩信息。利用K-近邻背景建模法将驾驶员前景图像分离以确定检测区域，通过边缘检测进行驾驶员头部和手部轮廓识别，获得关键定位点间的欧氏距离，并进行异常驾驶行为判断。在此基础上，结合异常行为次数和时间阈值，解决环境干扰和图像噪声的问题。实验结果表明，与单尺度Retinex、多尺度Retinex以及自适应直方图均衡方法相比，所提方法在提高检测准确率的同时有效提高了检测速度，能在不同环境下有效检测异常驾驶行为。

关键词: 异常驾驶检测, 关键点定位, 伽马校正, 边缘检测, K-近邻背景建模

Abstract:

To address the problems of low detection accuracy and slow detection speed of heavy truck driver abnormal driving behavior detection methods under low illumination conditions, an abnormal driving behaviors detection method based on adaptive Gamma correction is proposed by combining image adaptive enhancement methods and contour localization detection ideas.The adaptive Gamma correction is applied to the incoming video image to realize noise suppression, darkness improvement, and information entropy enhancement to improve the accuracy of recognition. The double threshold Gamma function is adaptively adjusted based on the image grey value and information entropy to obtain richer edge information and color information.A K-Nearest Neighbor(KNN) background subtraction model is constructed to separate the driver foreground image to determine the detection area, while the driver head and hand contours are identified by edge detection to obtain the Euclidean distance between key localization points for abnormal driving behavior judgement. Based on this, the number of abnormal behaviors and time thresholds are fused to solve the problem of environmental interference and image noise. The experimental results show that when compared with the Single-Scale Retinex(SSR), Multi-Scale Retinex(MSR) and the adaptive histogram equalization method, the proposed method not only improves detection accuracy but also effectively improves detection speed. Furthermore, the proposed method can effectively detect abnormal driving behavior in different environments.

Key words: abnormal driving detection, key points location, Gamma correction, edge detection, K-Nearest Neighbor(KNN) background modeling

艾青松, 张皓喆, 严俊伟. 基于自适应伽马校正的异常驾驶行为检测方法[J]. 计算机工程, 2023, 49(9): 279-286.

Qingsong AI, Haozhe ZHANG, Junwei YAN. Abnormal Driving Behavior Detection Method Based on Adaptive Gamma Correction[J]. Computer Engineering, 2023, 49(9): 279-286.

http://www.ecice06.com/CN/Y2023/V49/I9/279

图/表 10

图1 不安全驾驶行为示例

Fig.1 Examples of unsafe driving behavior

图2 基于自适应伽马校正的异常驾驶行为检测方法框架

Fig.2 Framework of abnormal driving behavior detection method based on adaptive Gamma correction

图3 改进伽马校正的流程

Fig.3 Procedure of improving Gamma correction

图4 坐标系转换

Fig.4 Coordinate system transformation

图5 图像预处理前后的对比

Fig.5 Comparison of the graph before and after preprocessing

图6 不同参数下的伽马校正效果

Fig.6 Gamma correction effect under different parameters

图7 暗光条件下不同增强方法的性能对比

Fig.7 Performance comparison of different enhancement methods in dark conditions

参考文献 28

1	罗兰贝格. 中国商用车车联网白皮书: 中国商用车车联网现状与发展趋势[EB/OL]. [2022-08-20]. https://wenku.baidu.com/view/7eace9bb0708763231126edb6f1aff00bed570c1.html.
	ROLAND B. White paper on the internet of commercial vehicles in China: current situation and development trend of the Internet of commercial vehicles China[EB/OL]. [2022-08-20]. https://wenku.baidu.com/view/7eace9bb0708763231126edb6f1aff00bed570c1.html. (in Chinese)
2	KAYE S A, LEWIS I, FREEMAN J. Comparison of self-report and objective measures of driving behavior and road safety: a systematic review. Journal of Safety Research, 2018, 65, 141- 151. doi: 10.1016/j.jsr.2018.02.012
3	纪辰瑾. 基于卷积神经网络的驾驶员行为检测方法的研究[D]. 长春: 吉林大学, 2021.
	JI C J. Research on driver behavior detection method based on convolutional neural network[D]. Changchun: Jilin University, 2021. (in Chinese)
4	HUANG W, LIU X, LUO M Y, et al. Video-based abnormal driving behavior detection via deep learning fusions. IEEE Access, 2019, 7, 64571- 64582. doi: 10.1109/ACCESS.2019.2917213
5	HU Y C, LU M Q, LU X B. Spatial-temporal fusion convolutional neural network for simulated driving behavior recognition[C]//Proceedings of the 15th International Conference on Control, Automation, Robotics and Vision. Washington D. C., USA: IEEE Press, 2018: 1271-1277.
6	CHIOU C Y, WANG W C, LU S C, et al. Driver monitoring using sparse representation with part-based temporal face descriptors. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(1): 346- 361. doi: 10.1109/TITS.2019.2892155
7	高文. 驾驶员行为检测的MASKR-CNN模型实现[EB/OL]. [2022-08-20]. https://d.wanfangdata.com.cn/periodical/dzsj202106021.
	GAO W. Implementation of driver behavior detection based on MASKR-CNN model[EB/OL]. [2022-08-20]https://d.wanfangdata.com.cn/periodical/dzsj202106021. (in Chinese)
8	ZHANG J, WU Z C, LI F, et al. Attention-based convolutional and recurrent neural networks for driving behavior recognition using smartphone sensor data. IEEE Access, 2019, 7, 148031- 148046. doi: 10.1109/ACCESS.2019.2932434
9	KANG H B. Various approaches for driver and driving behavior monitoring: a review[C]//Proceedings of IEEE International Conference on Computer Vision Workshops. Washington D. C., USA: IEEE Press, 2014: 616-623.
10	胡习之, 黄冰瑜. 基于面部特征分析的疲劳驾驶检测方法. 科学技术与工程, 2021, 21(4): 1629- 1636.
	HU X Z, HUANG B Y. Fatigue driving detection system based on face feature analysis. Science Technology and Engineering, 2021, 21(4): 1629- 1636.
11	ELASSAD A, ELAMRANI Z, HAJAR M, et al. The application of machine learning techniques for driving behavior analysis: a conceptual framework and a systematic literature review. Engineering Applications of Artificial Intelligence, 2020, 87(10): 103312.
12	XING Y, LV C, WANG H J, et al. Driver activity recognition for intelligent vehicles: a deep learning approach. IEEE Transactions on Vehicular Technology, 2019, 68(6): 5379- 5390. doi: 10.1109/TVT.2019.2908425
13	郑伟成, 李学伟, 刘宏哲, 等. 基于深度学习的疲劳驾驶检测算法. 计算机工程, 2020, 46(7): 21- 29. URL
	ZHENG W C, LI X W, LIU H Z, et al. Fatigue driving detection algorithm based on deep learning. Computer Engineering, 2020, 46(7): 21- 29. URL
14	杨先凤, 李小兰, 贵红军. 改进的自适应伽马变换图像增强算法仿真. 计算机仿真, 2020, 37(5): 241- 245.
	YANG X F, LI X L, GUI H J. Simulation of improved adaptive gamma transform image enhancement algorithm. Computer Simulation, 2020, 37(5): 241- 245.
15	SONG R X, LI D, YU J D. Low illumination image enhancement algorithm based on DT-CWT and tone mapping. Journal of Computer-Aided Design & Computer Graphics, 2018, 30(7): 1305- 1311.
16	HANTOUSH S A, ALRIDHA A H. Dynamic weights equations for converting grayscale image to RGB image. Journal of University of Babylon for Pure and Applied Sciences, 2018, 26(8): 122- 129. doi: 10.29196/jubpas.v26i8.1677
17	YE G D, PAN C, HUANG X L, et al. A chaotic image encryption algorithm based on information entropy. International Journal of Bifurcation and Chaos, 2018, 28(1): 1850010. doi: 10.1142/S0218127418500104
18	付长斐, 叶宾, 李会军. 基于HSV颜色空间的运动目标识别[EB/OL]. [2022-08-20]. https://d.wanfangdata.com.cn/periodical/dglbljs202002013.
	FU C F, YE B, LI H J. Moving target recognition based on HSV color space[EB/OL]. [2022-08-20]. https://d.wanfangdata.com.cn/periodical/dglbljs202002013. (in Chinese)
19	CHEN G Y, DING Y F, SHEN X P. Sweet KNN: an efficient KNN on GPU through reconciliation between redundancy removal and regularity[C]//Proceedings of the 33rd International Conference on Data Engineering. Washington D. C., USA. IEEE Press, 2017: 621-632.
20	QASIM S, KHAN K N, YU M, et al. Performance evaluation of background subtraction techniques for video frames[C]//Proceedings of International Conference on Artificial Intelligence. Washington D. C., USA: IEEE Press, 2021: 102-107.
21	ZORAN, ZIVKOVIC. Efficient adaptive density estimation per image pixel for the task of background subtraction. Pattern Recognition Letters, 2006, 27(7): 773- 780. doi: 10.1016/j.patrec.2005.11.005
22	OWOTOGBE J, IBIYEMI T S, ADU B. Edge detection techniques on digital images-a review[EB/OL]. [2022-08-20]. https://www.researchgate.net/publication/338112798_ Edge_Detection_Techniques_on_Digital_Images_-_A_ Review.
23	WANG S K, CHEN L, ZHOU Z X, et al. Human fall detection in surveillance video based on PCANet. Multimedia Tools and Applications, 2016, 75(19): 11603- 11613. doi: 10.1007/s11042-015-2698-y
24	XIONG Z G, TANG Z W, CHEN X W, et al. Research on image retrieval algorithm based on combination of color and shape features. Journal of Signal Processing Systems, 2021, 93(2): 139- 146.
25	张坤, 胡斌星, 许新鹏, 等. 基于对数极坐标变换的快速相关匹配算法. 飞行力学, 2020, 38(2): 61- 65. URL
	ZHANG K, HU B X, XU X P, et al. A fast correlation matching algorithm based on log-polar transform. Flight Dynamics, 2020, 38(2): 61- 65. URL
26	关文喆. 基于机器视觉的司机异常驾驶行为研究[D]. 北京: 北京邮电大学, 2020.
	GUAN W Z. Research on driver's abnormal driving behavior based on machine vision[D]. Beijing: Beijing University of Posts and Telecommunications, 2020. (in Chinese)
27	HUANG Y J, DING H T, ZHANG Y B, et al. A motion planning and tracking framework for autonomous vehicles based on artificial potential field elaborated resistance network approach. IEEE Transactions on Industrial Electronics, 2020, 67(2): 1376- 1386. doi: 10.1109/TIE.2019.2898599
28	周作梅, 石成远. 基于Retinex的低照度图像增强算法研究. 信息与电脑, 2020, 32(20): 48- 50. URL
	ZHOU Z M, SHI C Y. Research on low illumination image enhancement algorithm based on retinex. China Computer & Communication, 2020, 32(20): 48- 50. URL

[1]	王富平, 于俊涛, 张锲石. 基于自适应方向导数滤波器的彩色边缘检测[J]. 计算机工程, 2022, 48(9): 204-212.
[2]	范明亮, 郭子涵, 柴晓楠, 商建东. 面向FT-M7002的Sobel边缘检测算法优化实现[J]. 计算机工程, 2022, 48(6): 193-199.
[3]	黄胜, 冉浩杉. 基于语义信息的精细化边缘检测方法[J]. 计算机工程, 2022, 48(3): 204-210.
[4]	王囡, 侯志强, 赵梦琦, 余旺盛, 马素刚. 结合边缘检测的语义分割算法[J]. 计算机工程, 2021, 47(7): 257-265.
[5]	郭恒亮, 柴晓楠, 韩林, 赫晓慧, 商建东. Canny边缘检测算法在飞腾平台上的实现与优化[J]. 计算机工程, 2021, 47(7): 37-43.
[6]	陈志鹏, 郑文秀, 黄琼丹. 基于Sobel滤波器的图像风格转换算法[J]. 计算机工程, 2021, 47(12): 274-277,284.
[7]	刘畅, 张剑, 林建平. 基于改进全卷积神经网络的图像单像素边缘提取[J]. 计算机工程, 2020, 46(1): 262-270.
[8]	谢晓燕, 辛晓斐, 朱筠, 王飞龙, 刘阳. 基于边缘检测的3D-HEVC深度图运动估计算法[J]. 计算机工程, 2019, 45(7): 264-267.
[9]	张文坤,汪西原,宋佳乾. 基于分数阶微分差与高斯曲率滤波的边缘检测算法[J]. 计算机工程, 2019, 45(2): 213-219.
[10]	张然,赵凤群. 基于双向扩散与冲击滤波的雾天图像增强算法[J]. 计算机工程, 2018, 44(10): 221-227.
[11]	周恺,苏娟. 基于概率提升树的虹膜分割算法[J]. 计算机工程, 2017, 43(8): 249-252,257.
[12]	魏秋明,梁军,鲍泓,王晶,李论. 异构计算平台图像边缘检测算法优化研究[J]. 计算机工程, 2017, 43(5): 240-247.
[13]	宋祺鹏,唐晶磊,辛菁. 基于生长模型的苗期大豆植株三维重建[J]. 计算机工程, 2017, 43(5): 275-280.
[14]	彭宏,赵鹏博. 边缘检测中的改进型均值滤波算法[J]. 计算机工程, 2017, 43(10): 172-178.
[15]	付辉,吴斌,李林飞,张红英. 基于暗原色先验的雾霾图像清晰度复原[J]. 计算机工程, 2016, 42(7): 232-237.

选择文件类型/文献管理软件名称

选择包含的内容