METHOD AND APPARATUS FOR REMOVING NON-UNIFORM MOTION BLUR USING MULTIFRAME

Improved processing speed, clear image quality with blur removed

This technology is about an image processing method and a way to remove blur from images.

Blur is one of the main causes of image quality deterioration. Often, when the exposure time is long, blur may occur in the acquired image due to the shaking of the image sensor. In order to improve this problem, this technology proposes a method for removing non-uniform motion blur using estimated non-uniform motion blur information and multi-frames using multi-frames containing non-uniform motion blur.

This technology estimates non-uniform motion blur information using the local area of the multi-frame image, and uses the estimated non-uniform motion blur information to remove the blur of the multi-frame of the original resolution. Not only can it achieve clear image quality, but it can also improve the speed of removing blur from images with large resolution.

‍

‍

‍

Key Features:

• The non-uniform motion blur information estimation unit estimates homography using Lucas-Kanade image registration
• Includes a step of acquiring the final restored image from multi-frames using non-uniform motion blur information
• Estimates homography for the multi-frame image and calculates weights
• The latent image acquisition unit estimates the latent image based on non-uniform motion blur information for some areas acquire video

‍

‍