Pedestrian Recognition Algorithm Based on Information Fusion of Visual and Millimeter Wave Radar

被引：0

作者：

Xu W. ^{[1
]}

Zhou P. ^{[1
]}

Zhang F. ^{[1
]}

Huang L. ^{[1
]}

机构：

[1] Automative Engineering Institute, Guangzhou Automobile Group Co. Ltd., Guangzhou

来源：

| 1600年 / Science Press卷 / 45期

关键词：

Information fusion; Machine vision; Millimeter wave radar; Pedestrian recognition;

D O I：

10.11908/j.issn.0253-374x.2017.s1.007

中图分类号：

学科分类号：

摘要：

In this paper, an algorithm was proposed for recognizing pedestrians using visual and millimeter-wave radar data fusion in active collision avoidance systems. First, the HOG feature was extracted from the image based on monocular vision, and the pedestrian information was obtained by using the support vector machine (SVM) classification method. At the same time, target was obtained by fast target detection by using Tolerance-Fast Beat Frequency Matching, and the pedestrian information based on vision was transferred to the millimeter-wave radar. Then, pedestrian information based on vision and target information based on the millimeter-wave radar was compared and target identified was matched as pedestrians. Finally, the vision-based pedestrian feature information was fused with the pedestrian characteristic information detected by the millimeter-wave radar, and the new characteristic information of the pedestrian target was obtained. The correct recognition rate of pedestrians was verified by collecting the video and radar data of the road environment. Experimental results demonstrate that based on the algorithm proposed in this paper, the correct recognition rate is higher under the premise of obtaining more accurate pedestrian information. © 2017, Editorial Department of Journal of Tongji University. All right reserved.

引用

页码：37 / 42and91

页数：4254

共 12 条

[1]

Song X., Feng G., Yang J., The current state and trends of automotive active collision avoidance system, Automotive Engineering, 30, 4, (2008)

[2]

Huang S.N., Yang J.K., Eklund F., Analysis of car-pedestrian impact scenarios for evaluation of a pedestrian sensor system based on accident data from Sweden, Expert Symposium on Accident Research, (2006)

[3]

Llorca D.F., Milanes V., Alonso I.P., Et al., Autonomous pedestrian collision avoidance using a fuzzy steering controller, IEEE Transactions on Intelligent Transportation System, 12, 2, (2011)

[4]

Milanes V., Llorcab D.F., Villagraa J., Et al., Vision-based active safety system for automatic stopping, Expert Systems with Applications, 39, 12, (2012)

[5]

Dalal N., Triggs B., Histograms of oriented gradients for human detection, IEEE Computer Society Conference on Computer Vision & Pattern Recognition, pp. 886-893, (2005)

[6]

Sun W., Xue M., Sun T., Et al., The optimized pedestrian tracking-learning-detection algorithm based on SVM, Journal of Hunan University(Natural Sciences), 43, 10, (2016)

[7]

Cao W., Wang Y., Wen Y., Research on obstacle recognition based on wavelet moments and SVM for deicing robot on high voltage transmissi on line, Journal of Hunan University(Natural Sciences), 39, 9, (2012)

[8]

Yu H., Liu J., A vision-based method to estimate passenger flow in bus, Journal of Image and Graphics, 13, 4, (2008)

[9]

Kannala J., Brandt S.S., A generic camera model and calibration method for conventional, wide-angle, and fish-eye lenses, Pattern Analysis and Machine, 28, 8, (2006)

[10]

Jung H.G., Lee Y.H., Yoon P.J., Et al., Radial distortion refinement by inverse mapping-based extrapolation, Pattern Recognition Conference, pp. 675-678, (2006)

← 1 2 →