Blurred Image Classification Based on Adaptive Dictionary

Two frameworks for blurred image classification based on adaptive dictionary are proposed. Given ablurred image, instead of image deblurring, the semantic category of the image is determined by blurinsensitive sparse coefficients calculated depending on an adaptive dictionary. The dictionary is adaptiveto an assumed space invariant Point Spread Function (PSF) estimated from the input blurred image. In oneof the proposed two frameworks, the PSF is inferred separately and in the other, the PSF is updatedcombined with sparse coefficients calculation in an alternative and iterative manner. The experimentalresults have evaluated three types of blur namely defocus blur, simple motion blur and camera shake blur.The experiment results confirm the effectiveness of the proposed frameworks.