Traffic danger detection by visual attention model of sparse sampling

Abstract

A method to detect traffic dangers based on visual attention model of sparse sampling was proposed. The hemispherical sparse sampling model was used to decrease the amount of calculation which increases the detection speed. Bayesian probability model and Gaussian kernel function were applied to calculate the saliency of traffic videos. The method of multiscale saliency was used and the final saliency was the average of all scales, which increased the detection rates extraordinarily. The detection results of several typical traffic dangers show that the proposed method has higher detection rates and speed, which meets the requirement of real-time detection of traffic dangers.

Keywords

traffic dangers / visual attention model / sparse sampling / Bayesian probability model / multiscale saliency

References

[1] LiuY, LiuH, NiF-l, XuW-f. New self-calibration approach to space robots based on hand-eye vision [J]. Journal of Central South University, 2011, 18(4): 1087-1096

[2] BertozziM, BroggiA. Gold: A parallel real-time stereo vision system for generic obstacle and lane detection [J]. IEEE Transactions on Image Processing, 1998, 7(1): 62-81

[3] FleetD J, BarronJ L, BeaucheminS. Performance of optical flow techniques [J]. International Journal of Computer Vision, 1994, 12(1): 43-77

[4] MihcakK M, KozintseyI, RamchandranK, MoulinP. Low-complexity image denoising based on statistical modeling of wavelet coefficients [J]. Signal Processing Letters, IEEE, 1999, 6(12): 300-303

[5] NakaI H. Robust object detection using a posteriori probability [J]. Information Processing Society Japan, 2010, 15(4): 431-441

[6] CrismanJ D, RojasJ C. Vehicle detection in color images [C]. IEEE Conference on Intelligent Transportation System, 2010BostonIEEE Press403-408

[7] GiseokK, JaeS C. Vision-based vehicle detection and intervehicle distance estimation for driver alarm system [J]. Optical Society of Japan, 2012, 19(6): 388-393

[8] KimD S, LeeK H. Segment-based region of interest generation for pedestrian detection in far-infrared images [J]. Infrared Physics & Technology, 2013, 61(11): 120-128

[9] GavrilaD M. Pedestrian detection from a moving vehicle [C]. Proceedings of European Conference on Computer Vision, 2000Dublin, IrelandSpringer37-49

[10] RupeshG, MeeraT K, SantanuC. Visual saliency guided video compression algorithm [J]. Signal Processing: Image Communication, 2013, 28(9): 1006-1022

[11] IttiL, KochC, NieburE. A model of saliency-based visual attention for rapid scene analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998, 20(11): 1254-1259

[12] ToetA. Computational versus psychophysical bottom-up image saliency: A comparative evaluation study [J]. IEEE Trans Pattern Anal Mach Intell, 2011, 33(11): 2131-2146

[13] MatildeG M, MariaL E, FranciscoJ H. Colorcopigmentation study by tristimulus colorimetry (CIELAB) in red wines obtained from Tempranillo and Graciano varieties [J]. Food Research International, 2013, 51(1): 123-131

[14] DavidF, RamirezM, OdeliaS, JuanF. A saliency-based bottom-up visual attention model for dynamic scenes analysis [J]. Biological Cybernetics, 2013, 107(2): 141-160

[15] GuoM-w, ZhaoY-z, ZhangC-b. Fast object detection based on selective visual attention [J]. Neurocomputing, 2014, 9(1): 1-14

[16] NaveedE, IrfanM, SungW B. Feature aggregation based visual attention model for video summarization [J]. Computers and Electrical Engineering, 2014, 40(3): 993-1005

[17] JiangQ-c, YanX-f. Monitoring multi-mode plantwide processes by using mutual information-based multi-block PCA, joint probability, and Bayesian inference [J]. Chemometrics and Intelligent Laboratory Systems, 2014, 136(15): 121-137

[18] SjoerdM S, FransA V, ChrisL P. Saliency in a suppressed image affects the spatial origin of perceptual alternations during binocular rivalry [J]. Vision Research, 2010, 50(19): 1913-1921

[19] HuK-t, LeouJ J, HsiaoH H. Spatiotemporal saliency detection and salient region determination for H.264 videos [J]. Journal of Visual Communication and Image Representation, 2013, 24(7): 760-772

[20] ZhangL-y, TongM H, MarksT K, ShanH-h, CottrellG W. SUN: A Bayesian framework for saliency using natural statistics [J]. Journal of Vision, 2008, 8(7): 20-32