Abstract
This paper proposed an adaptive multi-object pedestrian recognition algorithm based on SLIC. First, we used SLIC to superpixel the pre-segmentation processing on the image. Then, the hash distance is added as the superpixel point aggregation parameter based on the traditional superpixel measurement unit of LAB color space distance and position distance. Finally, we identified the clustering subject by using the extreme learning machine neural network. The proposed method can adaptively determine the number of superpixels to achieve high recognition performance. This method simply needs to preset the number of pre-segments, which can reduce the number of detection targets, improve the segmentation efficiency, and shorten the image identification time.
Similar content being viewed by others
References
Wu, H., Lu, J., Liu, C., et al.: An improved density space clustering for inspection image segmentation algorithm. J. Huazhong Univ. Sci. Technol. 43, 473–476 (2015). (in Chinese)
Hu, Z., Guo, M.: Fast segmentation of improved grabcut color image based on SLIC. Comput. Eng. Appl. 52(2), 186–190 (2016)
Zhang, L., Li, C., Tang, J., et al.: Online object segmentation via fusing appearance and motion features. J. Image Graph. 20(10), 1358–1365 (2015). (in Chinese)
Liu, D., Liu, W., Fei, B.: Foregrond discrimination in local model-matching tracking. J. Image Graph. 21(5), 616–627 (2016). (in Chinese)
Shu, S., Yang, M.: Classification method of hyperspectral image based on watershed segmentation and sparse representation. Comput. Sci. 43(2), 89–94 (2016)
Bo, P., Yuan, Y., Wang, K.: Interacrive multi-phase image segmentation based on superpixels. J. Image Graph. 20(6), 764–771 (2015). (in Chinese)
Ling, C., Chen, H., Yang, X., et al.: Fundus image hard superpixels exudates detection based on SLIC and DB-SCAN clustering. J. China Univ. 36(4), 399–405 (2015). (in Chinese)
Hu, J., Zhao, Y., Cao, J.: Board wood surface defect image segmentation based on super pixel. J. Northeast Forest. Univ. 43(10), 97–102 (2015). (in Chinese)
Li, X., Shin, B., Liu, Y., et al.: Multi-target recognition method based on improved YOLOv2 model. Laser Optoelectron. Progress. 57(10), 113–122 (2020)
Shu, Y., Jia, Q., Zhao, C., et al.: Target recognition method based on multi-scale analysisand neural network. Radar ECM 40(02), 31–34 (2020)
Wu, T., Xia, J., Huang, Y.: Target recognition method of SAR images based on cascade decision fusion of SVM and SRC. J. Henan Polytech. Univ. 39(04), 118–124 (2020)
Wang, G., Xu, Z., Lu, W., et al.: Target detection and recognition based on convolutional neural network. Comput. Digit. Eng. 48(02), 338–343 (2020)
Chen, Z., Ye, D., Zhu, C., et al.: Object recognition method based on improved YOLOv3. Comput. Syst. Appl. 29(01), 49–58 (2020)
Zhang, C., Feng, C., Gao, T.: Target recognition and grabbing based on machine vision. Agric. Equip. Veh. Eng. 57(12), 93–96 (2019)
Chai, Y., Xu, J.: Target recognition and positioning system based on machine vision. Comput. Eng. Des. 40(12), 3557–3562 (2019)
Long, L., Wen, X., Lin, Y.: Target recognition method based on GA-BP neural network. Transd. Microsyst. Technol. 38(10), 47–50 (2019)
Lin, K., Zhang, Y., Li, H.: Research on HOG feature extraction algorithm weighted by information entropy. Comput. Eng. Appl. 56(06), 147–152 (2020)
Zhang, X.: Research on several new super pixel algorithms. Control Eng. 5, 902–907 (2015)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Yu, T., Wang, C., Liu, X. et al. Adaptive superpixel-based multi-object pedestrian recognition. Machine Vision and Applications 32, 16 (2021). https://doi.org/10.1007/s00138-020-01133-x
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00138-020-01133-x