Reconstruction of Digital Images by Combining Multiple Gradients

Up to now, many algorithms have been introduced for computing reconstruction in grayscale images. But the execution time is required by the known grayscale reconstruction algorithms make their practical use rather cumbersome on conventional computers. A new algorithm is introduced to bridge this gap and reconstruct the image in better way. This is based on the notion of regional maxima, regional minima, most significant value in the region and regional sorted Meddle values and uses different gradient algorithms constructed based on regional maxima, regional minima and regional sorted Meddle values for reconstruction. The existing study shows that the watershed by foreground markers is able to segment real and simple images containing few irregularities in a better way than the standard watershed segmentation algorithms. This method is based on markers and simple morphology, which allows a regularization of the watersheds. But it is not a flexible approach for further optimization parameters. The algorithm is able to segment or extract desired parts of only simple gray-scale images. To overcome these draw backs here new method is propose for real and multiple scale images by combining multiple gradient operators.