A new approach to estimate image blur extent based on wavelet module maximum

摘要

An approach to estimate image blur extent based on image edge width and wavelet module maximum was proposed. According to the definition of wavelet module value, three-order spline wavelet was chosen to implement multiscale wavelet edge detection and wavelet edge module values were obtained. To select half of wavelet edge module maximum as threshold to segment blur image and its binary image was obtained. Area and girth of the blur edge of the objects in the blur image were calculated according to pixel statistic principle. Blur edge width of objects in the image was given by area and girth of the transition edge. Image blur extent was determined by blur edge width. Experiment results show that the algorithm is effective and performs better.