6533b870fe1ef96bd12d0407

RESEARCH PRODUCT

Vision system for defect imaging, detection, and characterization on a specular surface of a 3D object

Patrick GorriaDenis AluzeChristophe DumontFred Merienne

subject

Computer scienceMachine visionbusiness.industrySpecular surfaceLighting systemVision computingObject (computer science)Characterization (materials science)Signal ProcessingComputer visionComputer Vision and Pattern RecognitionDefect sizeArtificial intelligencebusiness

description

Abstract A vision system capable of imaging, detecting, and characterizing defects onto highly reflective, non-plane surfaces, is presented in this paper. Defects are typically dust, and hair located under the metallic layer of packaging products used in cosmetic industries. The vision system comprises an innovative lighting solution to reveal defects onto highly reflective non-plane surfaces. Several image acquisitions are performed to build a synthetic image, where defects clearly appear white on a mid-gray background. Our lighting system allows imaging defects on various-shaped objects. The vision system measures the defect size to make a decision on the product rejection. The authors assess system performance by conducting series of tests.

https://doi.org/10.1016/s0262-8856(02)00046-x