Robot Stereo Vision Guidance System Based on Attention Mechanism

Abstract:

Add attention mechanism into traditional robot stereo vision system, thus got the possible workpiece position quickly by saliency image, highly accelerate the computing process. First, to get the camera intrinsic matrix and extrinsic matrix, camera stereo calibration needed be done. Then use those parameter matrixes to rectify the newly captured images, disparity map can be got based on the OpenCV library, meanwhile, saliency image was computed by Itti algorithm. Workpiece spatial pose to left camera coordinates can be got with triangulation measurement principal. After a series of coordinates transformation workpiece spatial pose to world coordinates can be got. With the robot inverse solution function, the robot joint rotation angle can be got thus driver the robot to work. At last, experiment results show the effectiveness of this method.

• Periodical: Applied Mechanics and Materials (Volumes 385-386)
• Pages: 708-711
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.385-386.708
• Online since: August 2013
• Authors: Long Yang, Nan Feng Xiao
• Keywords: Attention Mechanism, Coordinates Transformation, Machine Vision, Robot Guidance

References

[1] Dr. Velappa Ganapath: Stereo Vision based Robot Controller[C]/ 2008 IEEE International Conference on System, Man and Cybernetics(SMC 2008). Singapore, 2008: 1849-1854.

DOI: 10.1109/icsmc.2008.4811558

[2] LAN Qian. Complicated Scene Target Detection Method Based on the Significant Characteristics Describe [D]. WuHan: Huazhong University of Science and Technology, (2007).

[3] Itti, Koch. A Model of Saliency-Based Visual Attention for Rapid Science Analysis [J]: IEEE Transaction on Pattern analysis and Machine Intelligenec, 1998, 20(11): 1254-1259.

DOI: 10.1109/34.730558

[4] Yaqi Hu, Fang Meng. Analysis and Realization of Saliency Map Based on Visual Attention Mechanism [C]/The 5th International Congress on Image and Signal Processing. ChongQin, 2012: 415-419.

DOI: 10.1109/cisp.2012.6470028

[5] Azra Fetić, Davor Jurić i Dinko Osmanković. The procedure of a camera calibration using Camera Calibration Toolbox for MATLAB [C]/MIPRO 2012. Croatia, 2012: 1752-1757.

[6] Slavomir Matuska, Robert Hudec, Miroslav Benco. The Comparison of CPU Time Consumption for Image Processing Algorithm in Matlab and OpenCV[C]/Proceedings on 9th International Conference-2012. ELEKTRO, 2012: 75-78.

DOI: 10.1109/elektro.2012.6225575

[7] Gary Bradski. Learning OpenCV [M]. Beijing: Tsinghua University Press, (2009).

[8] YIN Li. Computer Vision 3D Reconstruction Method Research Based on the OpenCV [D]. Hefei: Anhui University, (2011).

[9] JIANG Wen-hui. Multi-Robot Coordination Control Method Research Based on Internet [D]. Guangzhou: South China University of Technology, (2012).