Abstract: Methods and systems are provided for automated validation of images generated by uncertified software. One method involves analyzing an unverified image and a reference image in parallel to identify sets of kernels associated with a layer of a convolutional neural network, creating a shared set of kernels associated with that layer in accordance with one or more kernel shortlisting criteria, obtaining a first feature map for the unverified image using that layer of the convolutional neural network and the shared set of kernels, obtaining a second feature map for the reference image using that layer of the convolutional neural network and the shared set of kernels, calculating a similarity score for the unverified image based at least in part on differences between the feature maps and a weighting factor associated with the first layer, and automatically validating the unverified image based at least in part on the similarity score.

Abstract: A method of verifying graphical objects using color invariant feature matching is provided. The method comprises determining, for each of a plurality of primary color channels, a color value for each inlier pixel in an actual object for that primary color channel; determining, for each color channel, a color difference factor (CDF) for an expected object by comparing, for each of the plurality of primary color channels, the color value for each inlier pixel in the expected object to the color value for each corresponding inlier pixel in the actual object; determining a false point value for the expected object from the CDF; and determining that the expected object matches the actual object when the false point value for the expected object is lower than the false point value for other expected objects.

Abstract: A method of verifying graphical objects using color invariant feature matching is provided. The method comprises determining, for each of a plurality of primary color channels, a color value for each inlier pixel in an actual object for that primary color channel; determining, for each color channel, a color difference factor (CDF) for an expected object by comparing, for each of the plurality of primary color channels, the color value for each inlier pixel in the expected object to the color value for each corresponding inlier pixel in the actual object; determining a false point value for the expected object from the CDF; and determining that the expected object matches the actual object when the false point value for the expected object is lower than the false point value for other expected objects.