Texture analysis on low resolution images using unsupervised segmentation algorithm with multichannel Local Frequency analysis

The Texture analysis is primarily concerned with the evaluation of characteristics where it materializes the channel deformation of image and its response. Wavelet is very useful for texture analysis and is widely adopted in the image segmentation. The channels are characterized by a bank of the spatial frequency domain. This paper deals with multi-channel filtering with Local Frequency using unsupervised segmentation algorithm on texture. It proposes a systematic filter selection scheme which is based on reconstruction of the input image from the filtered images. The image indexing and retrieval are conducted on textured images and natural images. Texture features are obtained by subjecting each selected filtered image to a nonlinear transformation and computing a measurement of energy in a window around each pixel. It incorporates a set of texture features under a segmentation work, based on the active contour without edges model with level set representation and a connected filtering strategy. The appearance of a surface texture is highly dependent on illumination. Surface texture classification methods require multiple training images captured under a variety of illumination conditions for each class if the surfaces are of uniform, smooth and the illumination is sufficiently far from the texture The performed result shows that it can be used for segmentation of multiple-textured images which allows the comparison between the segmented images versus its ground truth image.