الفهرس 
Chapter 1 IntroductionOnly 14 pages are availabe for public view
 |  | from | 125 |  |  |
| | | from | 125 | | |
Abstract
In recent years, remote sensing technology has a vital role in a broad range of applications, especially in water resources monitoring, coastal zones, and maritime safety management. They can easily acquire detailed information on a large area about maritime and coastal zones. The sea-land segmentation (SLS) for remote sensing images (RSIs) is considered one of the most remote sensing applications. SLS is a key for many applications, such as ship detection, coastline extraction, maritime traffic control, ocean surveillance, and maritime safety management. Therefore, the implementation of automatic SLS approaches can provide precise results about the research cases. In this thesis, we proposed a model to segment RSIs, which mainly suffer from low contrast, intensity inhomogeneity with mixed pixels, and large image size with redundant information to sea and land regions. The proposed model is divided into three main phases, which are preprocessing, feature extraction, and segmentation phases. The first phase is the preprocessing phase, which includes two steps. The input image is firstly transformed from RGB to HSV color space because RGB is affected by illumination changes and low contrast. HSV is recommended because it is invariant to changes in illumination and brightness. Second, the image is over-segmented into superpixels using the Simple Linear Iterative Clustering (SLIC) method to reduce the information redundancy. The second phase is feature extraction. Multi-features include the spectral and texture, are extracted to encode each superpixel and characterize the sea and the land regions. The third phase is the segmentation phase, which includes two steps. First, Superpixel Fuzzy C-Means (SPFCM) employs local relationships among neighboring superpixels to cluster the superpixels based on their color and texture features. The purpose of the SPFCM result is to provide an automatic initial contour for a Modified Chan-Vese (MCV) model instead of manual initialization to improve the Chan-Vese (CV) model’s performance, reduce the number of iterations, and computational time of the CV model. In the last one, the CV model is modified by incorporation the color and texture features to produce final segmentation results. We applied the proposed technique to natural-colored remote sensing images from the USGS Global Visualization Viewer. Experimental results show that the proposed model achieved an average accuracy of 98.9%, an average Jaccard Similarity Coefficient (JSC) equals 97.1%, an average Disc Similarity Coefficient (DSC) equals 98.5%, and an average recall equals 99.3%. The proposed technique results are promising. It outperformed the results of other state-of-the-art sea-land segmentation methods. Moreover, the proposed strategy is tested on other application areas, such as skin lesion segmentation and natural color images. The experimental results prove that the proposed method works well in other applications and outperforms state-of-the-art methods, which poorly achieve segmentation results.