Generating Robust Output Image Data from Derivative Image Data

摘要

Output image data is generated by obtaining derivative data (120) relating to a reference image (e.g. obtained by camera 110), obtaining a constraint for output image data (130), and generating the output image data from the derivative data relating to the reference image in dependence on the constraint (140). The constraint may be requiring that the output image is a linear combination of a set of smooth functions, or comprises minimizing the sum of the difference between a derivative of the output image data and the derivative data for the reference image and a penalty term. This method can be used to recover a robust output image from the derivative of an input image. Applications include object recognition, dynamic range compression, navigation, tracking and image compression.