PPIM: A point-by-point interpolation method for breakpoint connection used for load detection of belt conveyor

• Published in: Measurement : journal of the International Measurement Confederation 2022-11, Vol.204, p.112125, Article 112125
• Main Authors: Zhang, Mengchao, Zhang, Dongyue, Yuan, Chao, Li, Meixuan, Liu, Luxuan, Xue, Mingyuan, Hao, Nini, Zhang, Yuan
• Format: Article
• Language: English
• Subjects: Belt conveyor; Image processing; Load detection; Machine vision
• ISSN: 0263-2241; 1873-412X
• DOI: 10.1016/j.measurement.2022.112125

•An efficient and high-quality inspection method for conveying loads.•A new interpolation method used for connect breakpoint and breakline.•A novel laser-line thinning and continuous method. The method of vision with laser-assisted can better realize the measurement of conveyor load, which is the basis for realizing intelligent speed regulation and energy-saving operation of belt conveyors. Affected by factors such as the shape of the conveyed material and the shooting angle of the camera, the laser lines captured in the field of view of the camera are not continuous but have many breakpoints or broken lines instead, which is very unfavorable for subsequent image processing, then resulting in lower measurement accuracy and poor real-time performance. Aiming at the above problems, based on the accumulation form of the material carried by conveyors, this paper innovatively proposes a point-by-point interpolation method for breakpoint and broken line connection, named PPIM. The proposed method optimizes the steps of traversing the entire image when searching for breakpoints and lines in the existing method, thereby reducing the search area and improving the real-time detection; Then, the refinement operation for the laser line is also omitted, which effectively avoids the interference of the laser line width on the detection results; Meanwhile, the discrete points are interpolated in combination with the unique form of the material on the belt conveyor to realize the continuity of the laser line. Finally, the accuracy of the method proposed in this paper is obtained by comparing it with the data of the laser rangefinder. The results show that the proposed method can well fit the shape of the material section while considering both the detection accuracy and speed: compared with the existing alpha-shape method, the detection accuracy is increased by 11.67%, and the detection speed can reach 160 frames per second (FPS), which can provide technical support for the intelligent and energy-saving development of belt conveyors.