For example,Бобцов

IMAGE QUALITY ENHANCEMENT BY PROCESSING OF VIDEO FRAMES WITH DIFFERENT EXPOSURE TIME

Annotation

Subject of Research. We consider a method of image quality enhancement, when an image is registered with long exposure time under uncontrollable camera shake. Blur compensation is implemented by deconvolution computational algorithm with the point-spread function determining the image blur. Method. The main step of the deblurring algorithm consists in evaluation of a point-spread function involving the second underexposed noisy image frame as an initial approximation in iterative algorithm of point-spread function estimation. The subsequent blurred image deconvolution with the estimated point-spread function provides obtaining of enhanced image. Main Results. We have proposed new procedures for refinement of point-spread function estimates based on separation of values under hysteresis threshold application with the use of adopted Canny algorithm as well as modification in scale space. Practical Relevance. Obtained results can be used to enhance image quality in cases when camera moves during the image capture process including scientific research and computer vision systems.

Keywords

Articles in current issue