TY - GEN
T1 - Detection and Positioning of Workpiece Grinding Area in Dark Scenes with Large Exposure
AU - Guo, Zhentao
AU - Zhao, Guiyu
AU - Bian, Jinyue
AU - Ma, Hongbin
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.
PY - 2023
Y1 - 2023
N2 - Workpiece grinding is a crucial process in the smart manufacturing chain. In order to meet the requirements of industrial precision and relieve heavy work, researchers have developed a vision-based grinding robot. However, the problem of workpiece grinding area detection and positioning is difficult to be solved in dark scenes with large exposure. This paper proposes a method that fuses technologies such as processing of the image, coordinate and point cloud, which can accurately detect and locate the workpiece grinding area. Firstly, A method based on YOLOv7 and improved image preprocessing is used to detect labels of the grinding area. Secondly, A model for the prediction of the coordinates based on multiple linear regression was used to predict the coordinates of missing labels for the same grinding area. Finally, Data processing of the point cloud and transformation of the system of coordinates are used to achieve the acquisition of the coordinates of positioning vertices in the grinding area and conversion from the camera coordinate system to the world coordinate system. We used several sets of data for evaluation in our experiments, and the experimental results show that our proposed method can effectively detect the workpiece grinding area. At the same time, our method can also predict the coordinates of missing labels, which provides a more stable and reliable guarantee for industrial production.
AB - Workpiece grinding is a crucial process in the smart manufacturing chain. In order to meet the requirements of industrial precision and relieve heavy work, researchers have developed a vision-based grinding robot. However, the problem of workpiece grinding area detection and positioning is difficult to be solved in dark scenes with large exposure. This paper proposes a method that fuses technologies such as processing of the image, coordinate and point cloud, which can accurately detect and locate the workpiece grinding area. Firstly, A method based on YOLOv7 and improved image preprocessing is used to detect labels of the grinding area. Secondly, A model for the prediction of the coordinates based on multiple linear regression was used to predict the coordinates of missing labels for the same grinding area. Finally, Data processing of the point cloud and transformation of the system of coordinates are used to achieve the acquisition of the coordinates of positioning vertices in the grinding area and conversion from the camera coordinate system to the world coordinate system. We used several sets of data for evaluation in our experiments, and the experimental results show that our proposed method can effectively detect the workpiece grinding area. At the same time, our method can also predict the coordinates of missing labels, which provides a more stable and reliable guarantee for industrial production.
KW - Bad scene
KW - Image pre-processing
KW - Multiple linear regression
KW - Point cloud
KW - Target detection
UR - http://www.scopus.com/inward/record.url?scp=85175976136&partnerID=8YFLogxK
U2 - 10.1007/978-981-99-6498-7_5
DO - 10.1007/978-981-99-6498-7_5
M3 - Conference contribution
AN - SCOPUS:85175976136
SN - 9789819964970
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 48
EP - 59
BT - Intelligent Robotics and Applications - 16th International Conference, ICIRA 2023, Proceedings
A2 - Yang, Huayong
A2 - Zou, Jun
A2 - Yang, Geng
A2 - Ouyang, Xiaoping
A2 - Liu, Honghai
A2 - Yin, Zhouping
A2 - Liu, Lianqing
A2 - Wang, Zhiyong
PB - Springer Science and Business Media Deutschland GmbH
T2 - 16th International Conference on Intelligent Robotics and Applications, ICIRA 2023
Y2 - 5 July 2023 through 7 July 2023
ER -