Reflectance-based material classification for printed circuit boards

This paper describes a method for classifying object materials on a raw circuit board based on surface-spectral reflectance. First we introduce a multi-spectral imaging system for observing tiny objects and capturing their spectral data. The imaging system is composed of a liquid-crystal tunable filter, a monochrome CCD camera, macro-lens and a personal computer. We describe how we can estimate the spectral reflectance functions of object surfaces by using the multi-spectral imaging system. We show that dielectric materials like plastics can be distinguished from metals based on the reflectance difference in changing illumination geometries. Then an algorithm is presented for classifying the objects into several circuit elements based on the estimated spectral-reflectances. Region segmentation results of the circuit board are demonstrated in an experiment using a real board. The performance of the proposed imaging system and algorithms is examined in comparison with the RGB-based methods using a normal color camera.