Description

The Enhancement of the texture images were more helpful in the identification of the visual characteristics of the images. The identification of the clear visual characteristics were helpful in the identification of the clear edges in the texture images. The images were decomposed into cartoon and texture component based on Morphological component analysis. The Textural characteristics of the images were calculated and analyzed with coarseness, contrast, Directionality and line likeliness. The identification of the textural characteristics were employed in high and low level of the images. The decomposed images were then recombined inorder to obtain the enhanced texture image. The input texture images were selected. The images were commonly resized to a common size. The images were divided into cartoon and the texture component. The decomposition of the images were done based on Morphological Component Analysis. In Morphological Component Analysis the residual information is obtained in the first step. Wavelet transformation is applied to the obtained residual information. Threshold is applied to the transformed images inorder to decompose the image. The same process is repeated with anyone of the component as constant. The process is done repeatedly till the stopping condition is achieved inorder to obtain the cartoon and the texture components. The images were manipulated and the textural characteristics of the images were calculated. The manipulated images were called as strong components and others is called as weak components. The characteristics measured were coarseness, contrast, Directionality and line likeliness were measured for both strong and weak components. For strong coarseness bilateral filter is employed and for weak coarseness wavelet thresholding is employed. For strong contrast Anisotropic Shrink is employed and for weak contrast Anisotropic diffusion is employed. For strong and weak Directionality SWT thresholding is employed. For strong and weak Line-likeness Gradient similarity based filtering is employed. The textural characteristics were well enhanced in the proposed method since the images were manipulated and the characteristics were studied. In the proposed system complete image is enhanced compared to the existing methods for enhancement of the images. Texture enhancement is more helpful in the analysis of the visual characteristics of the images. For different textural characteristics different filters and enhancement methods were employed and hence the calculated values were more accurate. The morphological component analysis produces more effective method for the decomposition of the images into cartoon and texture components. Morphological component analysis is a iterative process of identification of the residual images by keeping anyone of the components constant. The texture characteristics were then studied based on the calculation of the four components for the analysis of the texture in the images. The components were identified for the manipulated and unmanipulated images for the effective study of the texture characteristics of the images. The resting images were reconstructed to form the enhanced texture image. The values calculated for the characteristics were compared with the existing methods and they indicated that the proposed method is more efficient.