Spectral covariance and fuzzy regions for image indexing

Localization Based on Building Recognition

“…Commonly used global descriptors, which provide some invariance to occlusions and clutter proposed in the past include responses to banks of filters [15] and multi-dimensional histograms [4]. In [12], the authors suggested to improve the discriminant power of plain color indexing technique with encoding of spatial information, by dividing the image into 5 partially overlapping regions. Local feature based techniques have recently become very effective in the context of different object recognition problems.…”

A Comparison and Analysis of Soft Computing Techniques for Content based Image Retrieval System

“…Fuzzy logic is proposed for the computation of fuzzy colour histogram as well as posing the queries in CBIR. This paper reviews the prominent CBIR systems along with fuzzy logic based techniques and proposes a technique based on fuzzy logic and neural networks for retrieving the images using natural language query for colour and texture features fuzzy logic based similarity is proposed between the two images [5]. The weights are assigned to fuzzy colour content during the calculation of similarity between the two images.…”

Fuzzy color histogram and its use in color image retrieval

“…Each of these regions is efficiently represented by a binary color set and its location information as well. Stricker and Dimai's method [12] tessellates each image into five partially overlapping fuzzy regions and extracts the first two color moments of each region both weighted by the membership functions of the region, respectively, to form a feature vector for the image.…”

“…With a suitable membership matrix , the FCHs of images and can be computed by (4), respectively. On the other hand, the squared Euclidean distance between their -bin FCHs is (12) Compared with (11), the simple squared Euclidean distance between two -bin FCHs is equivalent to the quadratic histogram distance between their -bin CCHs. Note that the computationally intensive matrix multiplication in computing quadratic distance of CCHs (11) is incurred at online retrieval stage.…”