Title: Research on robot vision target tracking detection method based on image segmentation

Authors: Bin Liu

Addresses: Hunan College of Information, Chang Sha, 410000, China

Abstract: Aiming at the problems of large tracking error and low tracking and detection accuracy in robot vision target tracking and detection, a robot vision target tracking and detection method based on image segmentation is proposed. First of all, the different thresholds of the visual target image are determined, and the target image is pre-processed. Then, an image segmentation algorithm is introduced to segment the target image according to the colour distance between the super-pixel foreground and background of the target image. Finally, the Gaussian filter is used to smooth the segmented target image, and the optical flow method and Weber local description factor are combined to construct the target tracking detection method to achieve the final research. The experimental results show that the target tracking and detection error of the proposed method is less than 1.0%, the detection accuracy is up to 99%, and the tracking and detection time is only 0.69 s. This method effectively improves the target tracking and detection effect.

Keywords: image segmentation; robot: visual target; tracking detection.

DOI: 10.1504/IJCISTUDIES.2023.132494

International Journal of Computational Intelligence Studies, 2023 Vol.12 No.1/2, pp.155 - 169

Received: 28 Nov 2022
Accepted: 12 Feb 2023
Published online: 24 Jul 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article