For image deblurring application with defocus blurring effect, a parameter estimation method based on Grayscale Mean Gradient (GMG) and Particle Swarm Optimization (PSO) algorithm was proposed to estimate the blurring parameter. First, a group of point spread functions with different blurring radius were randomly generated by PSO algorithm to process a blurred image with Wiener filtering algorithm, then a series of restored images were obtained and the corresponding GMG values were calculated. Secondly, concerning the property that the definition of an image is positively varied with its GMG value, which is shown by experimental results, the GMG values were taken as the fitness function values of the PSO algorithm, then a particle with maximum fitness was found, and the corresponding blurring parameter was taken as the final result of estimation. The experimental results show that the proposed algorithm outperforms spectral estimation method and cepstrum estimation method in estimation accuracy, especially in the case with large blur radius.

In general blind restoration algorithms, only the gray information of a color image is utilized to estimate the blurring kernel, and thus a restored image may be unsatisfactory if its size is too small or the salient edge in it is too little. Focused on the above mentioned problem, a new blind image restoration algorithm was proposed under a new tensorial framework, in which a color image was regarded as a third-order tensor. First, the blurring kernel was estimated utilizing the multi-scale edge information of blurred color image which could be obtained by adjusting the regularization parameter in tensorial total variation model. Then a deblurring algorithm based on tensorial total variation was adopted to recover the latent image. The experimental results show that the proposed algorithm can achieve obvious improvement on Peak Signal-to-Noise Ratio (PSNR) and subjective vision.

For haze removal in color image, a fast algorithm based on Principle Component Analysis (PCA) and atmospheric scattering model was proposed for color image haze removal. Firstly, the principal components of three color channels were extracted from original color image, and the three color channels were reconstructed by use of maximum principal component, and the Minimum Reconstruction Map (MRM) was obtained by taking the minimum gray value in three color channels. Then, the MRM was filtered by median filter to improve the accuracy of estimation of the global atmosphere light, then the global atmosphere light was estimated in MRM. Finally, according to the atmospheric scattering model to obtain media transmittance and the sence radiance of the haze removal image. The experimental results showed that the proposed algorithm achieved better visual recovery results, in comparison with dark channel prior haze removal algorithm and contrast limited adaptive histogram equalization algorithm. The results domonstrate that the proposed algorithm improves the operation efficiency, it is simple and easy to implement, and can quickly remove haze in color image.