Skip to main content

Pedestrian Detection

Pedestrian Detection

Fused Features based on Gradient, Texture, and Local phase information in chromatic domain

A new human detection descriptor based on a combination of three major types of visual information: color, shape, and texture has been developed. Shape features are extracted based on both the gradient concept and the phase congruency in LUV color space. The Center-Symmetric Local Binary Pattern (CSLBP) approach is used to capture the texture information of the image. The fusing of these complementary information yields to capture a broad range of the human appearance details that improves the detection accuracy. The proposed features are formed by computing the phase congruency of the three-color channels in addition to the gradient magnitude and CSLBP value for each pixel in the image with respect to its neighborhood. Only the maximum phase congruency values are selected from the corresponding color channels. The histogram of oriented phase and gradients, as well as the histogram of CSLBP values for the local regions of the image, are determined. These histograms are concatenated to construct the proposed descriptor, that fuses the shape and texture features, and it is named as Chromatic domain Phase features with Gradient and Texture (CPGT). Several experiments were conducted to evaluate the performance of the proposed CPGT descriptor. The experimental results show that the proposed descriptor has better detection performance and lower error rates when compared to several state of art feature extraction methodologies.

Figure 1: The general structure of human detection system

Figure 2:  Framework of the human detection system based the proposed descriptor algorithm

Figure 3: Human detection based on CPGT descriptor

Figure 4: Human detection in extreme darker environment

Figure 5: Human detection in complex lighting environment

Figure 6: Human detection in crowded environment

References
Hussin Ragb and Vijayan K. Asari, "Multi-feature fusion for robust human detection in thermal infrared imagery," SPIE Journal of Optical Engineering, doi.org/10.1117/1.OE.58.4.043101, vol. 58, no. 4, pp. 043101: 1-10, April 2019.

Hussin Ragb and Vijayan K. Asari, "Multi-hypothesis approach for efficient human detection," Journal of Imaging Science and Technology, doi.org/10.2352/J.ImagingSci.Technol.2019.63.2.020503, January 2019.

Hussin K. Ragb and Vijayan K. Asari, "Local phase features in chromatic domain for human detection," International Journal of Monitoring and Surveillance Technologies Research, doi: 10.4018/IJMSTR.2016070104, vol. 4, no. 3, pp. 52-72, 2016.

Hussin Ragb and Vijayan K. Asari, "Fused structure and texture (FST) features for improved pedestrian detection," International Journal of Computer, Electrical, Automation, Control and Information Engineering, (World Academy of Science, Engineering and Technology), vol. 10, no. 1, pp. 202-209, 2016.

Hussin Ragb, Redha Almahdi, and Vijayan K. Asari, "Aggregate channel features based on local phase, color, texture, and gradient features for people localization," 2019 IEEE National Aerospace & Electronics Conference, Dayton, Ohio, USA, 15 - 19 July 2019.

Hussin Ragb, Theus Aspiras, and Vijayan K. Asari, "Human detection in infrared imagery using gradient and texture features and super-pixel segmentation," IEEE International Conference on Technologies for Homeland Security (IEEE HST 2018), Woburn, Massachusetts, USA, 23-24 October 2018.

Hussin K. Ragb, Vijayan K. Asari, and Theus H. Aspiras, "Human detection in infrared imagery using intensity distribution, gradient and texture features," Proceedings of the SPIE Defense + Commercial Sensing: Mobile Multimedia/Image Processing, Security, and Applications 2018 (Multimedia Algorithms and Systems) , Orlando, Florida, United States, vol. 10668, pp. 1-8, 15 - 19 April 2018.

Hussin Ragb, Theus Aspiras, and Vijayan K. Asari, "Depth and super-pixel extraction for augmenting human detection," Proceedings of the IS&T International Symposium on Electronic Imaging: Imaging and Multimedia Analytics in a Web and Mobile World , Burlingame, California, USA, pp. 336: 1-6, 28 January - 01 February 2018.

Hussin Ragb and Vijayan K. Asari, "A feature fusion strategy for human detection in omnidirectional camera imagery," Proceedings of the IS&T International Symposium on Electronic Imaging: Imaging and Multimedia Analytics in a Web and Mobile World , Burlingame, California, USA, pp. 375: 1-5, 28 January - 01 February 2018.

Theus Aspiras, Vijayan K. Asari, and Hussin Ragb, "Enhanced Human Detection using Gradient and Texture Features and Super-pixel Segmentation," IEEE Workshop on Applied Imagery and Pattern Recognition: Big Data, Analytics, and Beyond - AIPR 2017, Washington Dc, USA, 10 - 12 October 2017.

Hussin Ragb and Vijayan K. Asari, "Fused shape features based on gradients and local phase in color domain," 2017 National Aerospace & Electronics Conference: Algorithms for Tracking in Aerial Surveillance, Dayton, Ohio, USA, 27 - 30 June 2017.

Hussin Ragb, Theus Aspiras, Vijayan K. Asari and Pramud Rawat, "Omnidirectional camera image analysis for law-enforcement and border control applications," IEEE International Conference on Technologies for Homeland Security (IEEE HST 2017), Waltham, Massachusetts, USA, 25 -26 April 2017.

Hussin Ragb and Vijayan K. Asari, "Chromatic domain phase features with gradient and texture for efficient human detection," Proceedings of the IS&T International Conference on Electronic Imaging (Society for Imaging Science and Technology): Imaging and Multimedia Analytics in a Web and Mobile World , DOI: https://doi.org/10.2352/ISSN.2470-1173.2017.10.IMAWM-172, San Francisco (Burlingame), California, USA, pp. 74-79, 29 January - 02 February 2017.

Hussin Ragb and Vijayan Asari, "Multi-feature fusion and PCA based approach for efficient human detection," IEEE Computer Society Workshop on Applied Imagery and Pattern Recognition - AIPR 2016: Imaging and Artificial Intelligence: Intersection and Synergy, Washington DC, USA, 18 - 20 October 2016.

Hussin Ragb and Vijayan Asari, "Color and local phase based descriptor for human detection," National Aerospace & Electronics Conference & Ohio Innovation Summit (NAECON-OIS), Dayton, Ohio, USA, pp. 68-73, 26 - 29 July 2016.

Hussin Ragb and Vijayan K. Asari, "Histogram of oriented phase (HOP): a new descriptor based on phase congruency," SPIE Conference on Commercial + Scientific Sensing and Imaging: Mobile Multimedia/Image Processing, Security, and Applications 2016, doi:10.1117/12.2225159, Baltimore, MD, USA, vol. 9869, pp. 1-10, 17 - 21 April 2016.

Hussin Ragb and Vijayan K. Asari, "Histograms of oriented phase and gradient (HOPG) descriptor for improved pedestrian detection," IS&T International Conference on Electronic Imaging: Video Surveillance and Transportation Imaging Applications 2016, San Francisco, California, USA, 14-18 February 2016.

Hussin Ragb and Vijayan K. Asari, "Fused structure and texture (FST) features for improved pedestrian detection," Proceedings of the International Conference on Computer Vision, Imaging and Computer Graphics - ICCVICG 2016, (World Academy of Science, Engineering and Technology), Zurich, Switzerland, vol. 18, no. 1, Part V, pp. 819-826, 12-13 January 2016.

Related Videos

CONTACT

Vision Lab, Dr. Vijayan Asari, Director

Kettering Lab
300 College Park
Dayton, Ohio 45469 - 0232
937-229-1779
Email
CONNECT