Laplacian Filter in Reconstruction of Images using Gradient-Based Algorithm

In this paper, we present a procedure for the reconstruction of images using a gradient-based algorithm, combined with the Laplacian filter as a noise-detection tool. Despite being commonly considered as an edge detection tool in the digital image processing, owing to its extensive noise sensitivity, the Laplacian can be efficiently used in the detection of noisy pixels. This paper deals with the noise corruption within the pixel intensity range. The reconstruction of detected pixels is performed using a gradient-based algorithm. The effectiveness of the method is verified through examples.