الفهرس Only 14 pages are availabe for public view
 |  | from | 100 |  |  |
| | | from | 100 | | |
Abstract
The Principal objective of image enhancement is to process the image so that the processed image is better than the original one for a specific application or set of objectives. Image enhancement techniques can be generally classified into two broad categories: spatial domain methods, which operate directly on pixels, can be either based on point processing or mask processing, and frequency domain methods, which operate on the coefficients of the image transform, such as the Fourier, wavelet, and cosine transforms.
This thesis provides a new local image enhancement technique using a proposed contrast measure in the discrete wavelet transform. The contrast measure is defined as the ratio of high spatial-frequency content and low spatial-frequency content in the subbands of the discrete wavelet transform matrix. The proposed algorithm’s strategy is to raise up the ratio of strong features to faint features where an original image is decomposed into three levels of two dimensional discrete wavelet transform, modify the coefficients using the proposed algorithm and then reconstruct the image using the enhanced coefficients in order to get the enhanced image. The obtained enhanced images from the proposed technique have better visual quality when compared with previous discrete cosine transform-based enhancement technique and global histogram equalization. The objective evaluation criterion is taken to be the Edge Detection where the superiority of the proposed technique has been clearly seen.
The proposed image enhancement algorithm in straightforwardly extended to color images by applying it to each color component independently. The computer simulation results of applying the proposed contrast enhancement algorithm on both grey scale and colored images are satisfactory.