A Realtime Object Detection Algorithm Based on Limited Computing Resource

Aiming at the problem of realtime object detection in humanoid robot, this paper presents a method that combines the Support Vector Machine (SVM) classifier with the Histograms of Oriented Gradients (HOG) feature of local area in the image. In order to reduce time consumption, the paper uses one image segmentation and scan policy to get candidate local area in the image. To evaluate the proposed method, we have established a data set of black and white ball images from RoboCup SPL games. Experimental results have demonstrated that recognition efficiency has been improved greatly and the algorithm can be executed in realtime on the NAO robot that has only a 1.6 GHz CPU.

[1]  Mei-Chen Yeh,et al.  Fast Human Detection Using a Cascade of Histograms of Oriented Gradients , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[2]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[3]  Chu-Song Chen,et al.  Fast Human Detection Using a Novel Boosted Cascading Structure With Meta Stages , 2008, IEEE Transactions on Image Processing.

[4]  Shuicheng Yan,et al.  An HOG-LBP human detector with partial occlusion handling , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[5]  Maurice Pagnucco,et al.  RoboCup SPL 2015 Champion Team Paper , 2015, RoboCup.

[6]  Rainer Hoch,et al.  On the evaluation of document analysis components by recall, precision, and accuracy , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[7]  Xuelong Li,et al.  Efficient HOG human detection , 2011, Signal Process..

[8]  Astha Baxi,et al.  A Review on Otsu Image Segmentation Algorithm , 2013 .

[9]  Larry S. Davis,et al.  Human detection using partial least squares analysis , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[10]  Udo Frese,et al.  Grab a mug - Object detection and grasp motion planning with the Nao robot , 2012, 2012 12th IEEE-RAS International Conference on Humanoid Robots (Humanoids 2012).

[11]  Navneet Dalal,et al.  Finding People in Images and Videos , 2006 .