基于机器视觉的驾驶员嘴唇特征点定位算法

摘要: 为了有效监控机动车驾驶员的疲劳状态, 通过边缘检测和红色像素点提取相结合的方法确定动态阈值, 对彩色人脸图像中嘴唇进行分割和提取, 再利用逼近算法对嘴唇的各个特征点进行定位, 采用机器视觉监视嘴唇运动, 对驾驶员的疲劳状态进行判定。试验结果表明: 该定位算法可有效去除嘴唇周围的干扰, 完整地将嘴唇从原图中提取出来, 并准确进行嘴唇特征点定位; 该算法对肤色深浅不同、光照不均匀、头部偏转角度不同等情况均有很好的鲁棒性。
- 交通信息工程 /
- 驾驶员疲劳监控 /
- 嘴唇特征点定位 /
- 机器视觉 /
- 边缘检测
Abstract: In order to monitor driver fatigue state effectively, its dynamic threshold was determined by using edge-detection method and red color element point extraction method, lip was segmented and extracted in colorful face images.Each feature point of lip was located based on approximation algorithm.The movement of lip was monitored, and the state of driver fatigue was determined based on machine vision.Test result indicates that the locating algorithm can dislodge the interferences around lip effectively, pick up lip shape from original image completely, and locate the feature points of lip exactly.The algorithm has good robustness for the situations of different skin colors, different lights and different rotational angles of head.
- traffic information engineering /
- driver fatigue monitor /
- locating of lip feature point /
- machine vision /
- edge detection

HTML全文

图 1 嘴唇定位流程

Figure 1. Flow of lip location

下载: 全尺寸图片幻灯片

图 2 嘴唇的定位

Figure 2. Locating of lip

下载: 全尺寸图片幻灯片

图 3 嘴唇边缘提取

Figure 3. Extraction of lip edge

下载: 全尺寸图片幻灯片

图 4 修正的嘴唇边缘提取

Figure 4. Revised extraction of lip edge

下载: 全尺寸图片幻灯片

图 5 嘴唇的提取

Figure 5. Extraction of lip

下载: 全尺寸图片幻灯片

图 6 嘴唇提取的试验结果

Figure 6. Test results of lip extraction

下载: 全尺寸图片幻灯片

图 7 嘴唇特征点定位流程

Figure 7. Locating flow of lip characteristic points

下载: 全尺寸图片幻灯片

图 8 左边嘴角定位过程

Figure 8. Locating process for left corner o fmouth

下载: 全尺寸图片幻灯片

图 9 嘴唇处于张开状态时左边嘴角定位过程

Figure 9. Locating process for left corner of mouth when mouth is open

下载: 全尺寸图片幻灯片

图 10 嘴唇特征点定位结果

Figure 10. Locating result of lip characteristic points

下载: 全尺寸图片幻灯片

参考文献(9)

[1]	周玉彬, 俞梦孙. 疲劳驾驶检测方法的研究[J]. 医疗卫生装备, 2003, 24 (6): 25-28. https://www.cnki.com.cn/Article/CJFDTOTAL-YNWS200306009.htm ZHOU Yu-bin, YU Meng-sun. Research on method of detectingdrowsy driver[J]. Chinese Medical Equipment Journal, 2003, 24 (6): 25-28. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YNWS200306009.htm
[2]	王荣本, 郭克友, 储江伟, 等. 适用驾驶员疲劳状态监测的人眼定位方法研究[J]. 公路交通科技, 2003, 20 (5): 111-114. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200305029.htm WANG Rong-ben, GUO Ke-you, CHU Jiang-wei, et al. Study on the eye location method in driver fatigue state sur-veillance[J]. Journal of Highway and TransportationResearch and Development, 2003, 20 (5): 111-114. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200305029.htm
[3]	陶霖密, 彭振云, 徐光佑. 人体的肤色特征[J]. 软件学报, 2001, 12 (7): 1032-1041. https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB200107009.htm TAO Lin-mi, PENG Zhen-yun, XU Guang-you. The fea-tures of skin color[J]. Journal of Software, 2001, 12 (7): 1032-1041. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB200107009.htm
[4]	张桂林, 张天序, 魏洛刚, 等. 基于边缘特征的运动目标提取与跟踪[J]. 华中理工大学学报, 1994, 22 (5): 42-45. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG405.008.htm ZHANG Gui-lin, ZHANG Tian-xu, WEI Luo-gang, et al. An edge-feature-based approachtothe extraction andtrackingof a moving object[J]. Journal of Huazhong University ofScience and Technology, 1994, 22 (5): 42-45. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG405.008.htm
[5]	郝海涛, 田玉敏. 彩色图象分割方法及进展[J]. 山西师范大学学报: 自然科学版, 2005, 19 (1): 64-68. https://www.cnki.com.cn/Article/CJFDTOTAL-SFDX20050100L.htm HAO Hai-tao, TI AN Yu-min. Methods and advances on colori mage segmentation[J]. Journal of Shanxi Teachers Universi-ty: Natural Science Edition, 2005, 19 (1): 64-68. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-SFDX20050100L.htm
[6]	陆继祥, 张增芳, 李陶深, 等. 基于24位彩色人脸图像嘴唇的分割和提取[J]. 计算机工程, 2003, 29 (2): 147-148. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJC200302060.htm LUJi-xiang, ZHANG Zeng-fang, LI Tao-shen, et al. Seg-mentation and extraction of human mouth based on 24 bit colori mage[J]. Computer Engineering, 2003, 29 (2): 147-148. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSJC200302060.htm
[7]	马丽红, 余英林, 张宇. 复杂背景中的人脸特征定位与提取[J]. 数据采集与处理, 2000, 15 (1): 33-38. https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ200001007.htm MA Li-hong, YU Ying-lin, ZHANF Yu. Facial featureextraction in a complex background[J]. Journal of DataAcquisition & amp; amp; Processing, 2000, 15 (1): 33-38. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ200001007.htm
[8]	WANG Yan-jiang, YUAN Bao-zong. A novel approach forhuman face detection fromcolor i mages under complex back-ground[J]. Pattern Recognition, 2001, 34 (10): 1983-1992.
[9]	GARCI A C, TZIRITAS G. Face detection using quantizedskin color regions merging and wavelet packet analysis[J]. IEEE Transactions on Multi media, 1999, 1 (3): 264-277.