Research on DPC Clustering Multi-Objective Detection Method for Disorderly Grasping

doi:10.3778/j.issn.1002-8331.2305-0140

Abstract

Abstract: In order to detect as many graspable targets as possible from the scene, a multi-target detection algorithm based on DPC feature point clustering is proposed. The feature points of the template image and the image to be detected are first extracted using the SIFT algorithm, and clustered using the DPC algorithm to obtain the set of feature points belonging to different clustering centres. Then the feature points belonging to different clustering centres are matched with the template image feature points, combined with the RANSAC algorithm to remove false matches and count the number of correct matches, and the homography matrix from the template image to the image to be detected is calculated based on the correctly matched feature points to obtain the target detection result. Finally, the correct detection results are filtered according to the number of correct matching points for each target, and among the correct detection results, the graspable targets are filtered according to the difference between the number of matching points and the maximum number of matching points for the target. After detecting the graspable target, the scene disparity map is obtained using the stereo matching algorithm, the 3D coordinates of the target are calculated, and the target pose is calculated using the PNP algorithm based on the correspondence between the 3D coordinates and the 2D coordinates in the image. The experimental results show that the DPC clustering-based multi-target detection method can accurately detect the target object among multiple identical stacked targets and calculate the target pose separately, effectively solving the problem of multi-target detection in unordered grasping applications.

Key words: disordered grasping, template matching, density peak clustering（DPC）, stack target

摘要： 为了尽可能多地从场景中检测出可抓取目标，提出了一种基于DPC特征点聚类的多目标检测算法。使用SIFT算法提取模板图像和待检测图像的特征点，并使用DPC算法对待检测图像特征点聚类，得到属于不同聚类中心的特征点集合。将属于不同聚类中心的特征点分别与模板图像特征点进行匹配，结合RANSAC算法去除误匹配并统计正确匹配点数量，根据正确匹配的特征点计算从模板图像到待检测图像的单应矩阵从而得到目标检测结果。根据每个目标正确匹配点数量筛选正确的检测结果，并在正确的检测结果中根据目标匹配点数量和目标最多匹配点数量的差值筛选出可抓取目标。检测出可抓取目标之后，使用立体匹配算法得到场景视差图，计算目标的三维坐标，并根据三维坐标与图像中二维坐标的对应关系使用PNP算法计算目标位姿。实验结果表明，基于DPC聚类的多目标检测方法能够在多个相同堆叠目标中准确检测出目标物体并分别计算位姿，有效解决了无序抓取应用中的多目标检测问题。

关键词: 无序抓取, 模板匹配, 密度峰值聚类（DPC）, 堆叠目标

CHEN Zeyu, LI Xiangguo, CAO Dengfeng, ZHU Denglin. Research on DPC Clustering Multi-Objective Detection Method for Disorderly Grasping[J]. Computer Engineering and Applications, 2023, 59(23): 175-182.

陈泽瑜, 李向国, 曹登锋, 朱灯林. 面向无序抓取的DPC聚类多目标检测方法研究[J]. 计算机工程与应用, 2023, 59(23): 175-182.

References

[1] XU Y H，ZHOU B，GAN Y H，et al.Recognition of disordered targets for robot based on 3D visual clustering and matching[C]//2020 Chinese Automation Congress（CAC），2020：1407-1412.
[2] WAN G Y，LI F D，LIU B Y，et al.A novel robotic 6DOF pose measurement strategy for large-size casts based on stereo vision[J].Assembly Automation，2022，42（4）：458-473.
[3] 王高，陈晓鸿，柳宁，等.一种基于视角选择经验增强算法的机器人抓取策略[J].华南理工大学学报（自然科学版），2022，50（9）：126-137.
WANG G，CHEN X H，LIU N，et al.A robot grasping policy based on viewpoint selection experience enhancement algorithm[J].Journal of South China University of Technology（Natural Science Edition），2022，50（9）：126-137.
[4] 肖贤鹏，胡莉，张静，等.基于多尺度特征融合的抓取位姿估计[J].计算机工程与应用，2022，58（10）：172-177.
XIAO X P，HU L，ZHANG J，et al.Grasp pose estimation based on multi-scale feature fusion[J].Computer Engineering and Applications，2022，58（10）：172-177.
[5] 魏元，杨华，程正东，等.红外小目标检测的滞后阈值分割法[J].激光与红外，2020，50（1）：111-116.
WEI Y，YANG H，CHENG Z D，et al.Hysteresis threshold segmentation in infrared small target detection[J].Laser & Infrared，2020，50（1）：111-116.
[6] XIAO Y Z，TIAN Z Q，YU J C，et al.A review of object detection based on deep learning[J].Multimedia Tools and Applications，2020，79：23729-23791.
[7] 王凯，余振军，何显辉，等.改进的多尺度形状模板匹配算法[J].激光杂志，2022，43（4）：82-87.
WANG K，YU Z J，HE X H，et al.Improved multi-scale shape template matching algorithm[J].Laser Journal，2022，43（4）：82-87.
[8] 赵振民，彭国华，符立梅.基于形状模板的快速高精度可靠图像匹配[J].计算机应用，2010，30（2）：441-444.
ZHAO Z M，PENG G H，FU L M.Fast high-precision reliable image matching algorithm based on shape[J].Journal of Computer Applications，2010，30（2）：441-444.
[9] 张嘉文，史金光，徐东辉，等.基于电视末制导技术的舰船目标分割与匹配方法在复杂背景下的应用[J].南京理工大学学报，2021，45（5）：529-536.
ZHANG J W，SHI J G，XU D H，et al.Ship segmentation and recognition technology of TV-guided projectiles in complex background[J].Journal of Nanjing University of Science and Technology，2021，45（5）：529-536.
[10] 李忠海，李申，崔建国，等.基于快速SIFT特征提取的模板匹配算法[J].计算机工程，2011，37（24）：222-224.
LI Z H，LI S，CUI J G，et al.Template matching algorithm based on fast SIFT feature extraction[J].Computer Engineering，2011，37（24）：222-224.
[11] LOWE D G.Object recognition from local scale-invariant features[C]//Proceedings of the Seventh IEEE International Conference on Computer Vision，1999：1150-1157.
[12] BAY H，ESS A，TUYTELAARS T，et al.Speeded-up robust features（SURF）[J].Computer Vision and Image Understanding，2008，110（3）：346-359.
[13] RUBLEE E，RABAUD V，KONOLIGE K，et al.ORB：an efficient alternative to SIFT or SURF[C]//2011 International Conference on Computer Vision，2011：2564-2571.
[14] 王鑫鹏，王晓强，林浩，等.深度学习典型目标检测算法的改进综述[J].计算机工程与应用，2022，58（6）：42-57.
WANG X P，WANG X Q，LIN H，et al.Review on improvement of typical object detection algorithms in deep learning[J].Computer Engineering and Applications，2022，58（6）：42-57.
[15] 惠记庄，罗丽，杨永奎，等.基于SURF-BRISK的目标识别匹配与定位方法研究[J].长安大学学报（自然科学版），2016，36（3）：93-101.
HUI J Z，LUO L，YANG Y K，et al.Research on location algorithm and target recognition matching based on SURF-BRISK[J].Journal of Chang’an University（Natural Science Edition），2016，36（3）：93-101.
[16] LI Z Z，CHEN J B，RAO J J，et al.Fast robust feature-based template matching[C]//AIP Conference Proceedings，2019.
[17] WEI A H，CHEN B Y.Robotic object recognition and grasping with a natural background[J].International Journal of Advanced Robotic Systems，2020，17（2）：172988142092110.
[18] 陶杰，吴尧才，朱熙豪，等.基于3D视觉的水泵生产线零部件无序抓取研究[J].机电工程，2022，39（5）：604-611.
TAO J，WU Y C，ZHU X H，et al.Disorderly grasping parts of pump production line based on 3D vision[J].Journal of Mechanical & Electrical Engineering，2022，39（5）：604-611.
[19] ZHAO G Y，MA H B，JIN Y.A method for robust object recognition and pose estimation of rigid body based on point cloud[C]//15th International Conference on Intelligent Robotics and Applications，Harbin，China，August 1-3，2022.Cham：Springer International Publishing，2022：468-480.
[20] JAGANATHAN T，PANNEERSELVAM A，KUMARASWAMY S K.Object detection and multi-object tracking based on optimized deep convolutional neural network and unscented Kalman filtering[J].Concurrency Computat Pract Exper，2022，34（25）：e7245.
[21] MEI X，SUN X，ZHOU M C，et al.On building an accurate stereo matching system on graphics hardware[C]//2011 IEEE International Conference on Computer Vision Workshops（ICCV Workshops），2011：467-474.