Motion Blur Image Restoration Variational Method

When acquires image by camera, motion blur often happens because of the motion between the camera and the object. Moreover, massive random noises are often introduced during the process of image acquisition and transmission. The motion blur and the noises cause the degradation of the image, and bring difficulty for further process like edge detect, image segmentation, feature extraction and the pattern recognition and so on. Total Variation (TV) model has its advantage comparing with traditional image restoration methods in the area of the motion blur image restoration. Based on fully studying on Total Variation model for image restoration, an improved adaptive TV model is proposed and is applied in restoration of real motion blur image.Firstly, Basic theories of motion blur image are introduced. Also, traditional image restoration methods are introduced and studied for the restoration of motion blur degraded image.Secondly, the basic Total Variation model for image restoration is introduced. Its character and numerical methods are studied. The adaptive TV model is analyzed, and is applied in restoration of motion blur degraded image. Better effects are got comparing with TV model.Thirdly, an improved adaptive TV model is proposed based on the analyzing of the advantages and disadvantages of the isotropic diffusion model and TV models. New model takes the advantages of the isotropic diffusion model and TV models by introducing two thresholds: the isotropic diffusion model is used if the gradient mold is less than thresholdβ₁, and TV model is used if the gradient mold is greater than the thresholdβ₂, and adaptive model is used if the gradient mold is between two thresholds. "Staircase" effect is reduced furthermore and better restoration effects are got by new model.Lastly, TV model and its improved model are applied in real motion blur image. The optimized window method is adopted to suppress the ring effects for real image, and finally idea restore image is got.