Enhancing the Image Contrast of Defect Patterns in Molded Plastic Cases of Mobile Phones with a Digital Image Process

In this work, we describe a scatterometer, which is devised to detect defect patterns in the molded plastic cases of mobile phones. Fourier transform is applied to the scattered image of the defect pattern to calculate the distribution of spatial frequency of the pattern. By examining the spatial frequency distribution, which meets the defect patterns, we can greatly enhance the image contrast of the defect by separating the defect and background patterns. For practical purposes, a dielectric-layer coating containing micrometer-sized TiO2-coated urethane balls is proposed for the molded case to reduce the visibility of defect patterns. The resulting changes in visibility are analyzed to investigate the effect of the weight ratio of the microballs in the dielectric layer. The visibility decreases by a factor of 2.7 with 5% weight ratio of the microball.