with the continuous promotion of the industrial upgrading of the manufacturing industry, the inspection efficiency requirements of the manufacturing industry for parts are becoming higher and higher. At present, the traditional measurement methods include gauge, caliper, profilometer, universal tool microscope, ray and other methods, but with the rapid development of industry, the traditional measurement methods can’t meet the needs of efficient measurement of industrial production line. Vision based measurement technology has the advantages of non-contact, high precision and easy to realize automation. The monocular vision measurement system is simple in structure and easy to operate, and has gradually become an important measurement method in the field of high-precision measurement [1] of part size. But for the detection of complex parts, the conventional 2D vision image is to extract the features of the measured object from the gray image and measure it in the X-Y plane. The method based on the standard scale is difficult to accurately measure the size. Aiming at the problems faced in the current complex parts detection, through the research of image theory, camera calibration theory and visual dimension measurement technology, a 3D depth vision [2] measurement system for complex part size is designed, which combines depth information with 2D vision measurement technology. The experimental results show that the system can meet the requirements of stability, reliability, high precision and strong anti-interference ability in measuring the size of complex parts.

Rui Hou, Hang Zhou. Complex parts measurement system based on 3D depth vision. Academic Journal of Computing & Information Science (2020), Vol. 3, Issue 4: 21-30. https://doi.org/10.25236/AJCIS.2020.030403.

[1] Ruoan X, Li Z, Shuangyan H U. High Accuracy Part Size Measurement System [J]. ence Technology and Engineer, 2005.
[2] Schreiber K, Crawford J D, Fetter M, et al. The motor side of depth vision [J]. Nature, 2001, 410 (6830): 819-822.
[3] Peremans K, Cornelissen B, Bossche B V D, et al. A REVIEW OF SMALL ANIMAL IMAGING PLANAR AND PINHOLE SPECT γ CAMERA IMAGING [J]. Veterinary Radiology & Ultrasound, 2010, 46 (2).
[4] Basseville M, Espiau B, Gasnier J. Edge detection using sequential methods for change in level--Part I: A sequential edge detection algorithm [J]. Acoustics Speech & Signal Processing IEEE Transactions on, 1981, 29 (1): 24-31.
[5] Mehrotra R, Nichani S, Ranganathan N. Corner detection [J]. Pattern Recognition, 1990, 23 (11): 1223-1233.
[6] Yi-Bo L, Jun-Jun L. Harris Corner Detection Algorithm Based on Improved Contourlet Transform [J]. Procedia Engineering, 2011, 15: 2239-2243.
[7] Xiong X, Junyuan W, Zhiqiang Z, et al. Study on the Size Measurement System Based on the CMOS Computer Vision [J]. Mechanical ence and Technology for Aerospace Engineering, 2013.
[8] Zuochang X, Subo T, Lili X, et al. The Measurement System Design of the Agricultural Machinery Hot Forging Parts Based on the Stereoscopic Vision [J]. Journal of Agricultural Mechanization Research, 2016.
[9] Chen, Qiyu, Wang, et al. Design and Implementation of Measurement System of Mechanical Parts with Multihole Based on Machine Vision [C]//The Third International Symposium on Mathematics and computer science, 2018.
[10] Xiaoming Zho, et al, 3D measurement of complex surface parts based on digital grating [J]. Electric locomotive and urban rail vehicle, 2016, 039 (004): 78-81.
[11] Feitin Zhu. Research on vehicle profile measurement based on laser point cloud 3D detection [J]. Technology Wind, 2019, 000 (003): 164-165.
[12] Xiuliang P, LV Jun, Xiaodan W, et al. A method for measuring solid geometric dimensions using 3D multi induction measuring instrument [J]. Dual use technology and products, 2014 (17)
[13] Camplani M, Mantecon T, Salgado L. Depth-Color Fusion Strategy for 3-D Scene Modeling With Kinect [J]. IEEE Transactions on Cybernetics, 2013, 43 (6): 1560-1571.