Fuzzy Image Retrieval: Recent Trends

Images have been playing very important role in human life since beginning of human civilization. Applications of images are found ranging from basic but very effective exchange of ideas and information to some advanced technologies in the industry, society, medical and military field. Image data has become an important field of research with rapid and huge growth of visual information in the number of large-scale and online image repositories. Image data is fuzzy in nature and imprecision and vagueness may exist in both image descriptions and query specifications. Techniques of Fuzzy set theory have been extensively applied to the representation and processing of imprecise and uncertain text and visual data. Soft computing has been applied in image retrieval and image analysis in numerous cases. For the fuzzy content-based image retrieval, fuzzy sets are applied for the extraction and representation of visual features like colors, shapes, textures, for similarity measures and indexing, for representing relevance feedback and for image retrieval. Fuzzy sets are also applied for fuzzy image query processing based on a defined database models and image data management. Image segmentation is used to extract information from complex medical images by partitioning an image into mutually exclusive regions such that each ROI is spatially contiguous and homogeneous; widely used homogeneity criteria being intensity, texture, color, range, surface normal and surface curvatures. In the field of medical imaging, soft computing based segmentation using Fuzzy-Neuro logic have been used to detect various tissues like white matter, gray matter; cerebral spinal fluid and tumor for a given magnetic resonance image data set. Fuzzy-Neuro logic segmentation algorithm is used to provide satisfactory results compared to K-means, Fuzzy C-Means, Neural Network and Fuzzy logic.