Abstract: A method and apparatus for identifying and classifying pixel defects, and in particular Mura defects using digital processing techniques. The present method includes steps of acquiring an image with a Mura defect, and performing a Laplacian convolution on the image to enhance the Mura defect against background illumination. A step of thresholding the Mura defect against the background illumination is also provided. The thresholded Mura defect is compared against the original Mura defect to define statistical parameters of the original Mura defect. An annular region is defined around the periphery of the Mura defect. Statistics of the annular region defines statistics for background illumination as compared to the original Mura defect. The statistics from the Mura defect are then compared to the background illumination statistics for Mura defect classification and analysis.

Abstract: A method for inspecting a substrate having a plurality of output pixels using an image sensing device having a plurality of input pixels includes the steps of capturing an input image of a plurality of groups of output pixels with a plurality of groups of input pixels, each group of input pixels capturing a group of output pixels, each input pixel in a group of input pixels having a position, inhibiting modulation contributions from input pixels in the input image, forming a plurality of images from the plurality of groups of input pixels in the input image, each image including input pixels from a similar position in each group of input pixels, detecting defects in each of the plurality of images, and determining defects in sub-pixels of the substrate in response to the defects detected in each of the plurality of images.