Adaptive Lighting for Curved and Nonuniform Objects in Optomechanical Inspection Systems

Abstract

Visual inspection is omnipresent and critical in precision manufacturing. However, complex geometries of parts hinder uniform illumination, and high reflectivity challenges accurate focusing for digital visual data collection. This research provides a novel adaptive illuminance distribution for consistent lighting to facilitate quality imaging over complex-shaped, highly reflective surfaces. The central approach entails using arrays of independently controlled light sources to reliably generate different lighting patterns, structures, and colors. Such results consider the geometry, the 3-D pose of parts in the environment, and the surface topography of the workpiece to be inspected, hence amplifying the capabilities of an image capturing system. In this article, we discuss the mathematical problem formulation, analytic solution, optimality of the proposed lighting, and experimental results in imaging curved parts common in aerospace manufacturing. The efficacy of the resulting defect identification is tested using a deep neural network.