Color, texture and shape information have been the primitive image descriptors in content based image retrieval systems. This paper presents a novel framework for combining all the three i.e. color, texture and shape information, and achieve higher retrieval efficiency. The image is partitioned into non- overlapping tiles of equal size. The color moments and moments on Gabor filter responses of these tiles serve as local descriptors of color and texture respectively. This local information is captured for two resolutions and two grid layouts that provide different details of the same image. An integrated matching scheme, based on most similar highest priority (MSHP) principle and the adjacency matrix of a bipartite graph formed using the tiles of query and target image, is provided for matching the images. Shape information is captured in terms of edge images computed using gradient vector flow fields. Invariant moments are then used to record the shape features. The combination of the color, texture and shape features provide a robust feature set for image retrieval. The experimental results demonstrate the efficacy of the method.

Content Based Image Retrieval Using Color, Texture and Shape Features

The increased need of content based image retrieval technique can be found in a number of different domains such as Data Mining, Education, Medical Imaging, Crime Prevention, Weather forecasting, Remote Sensing and Management of Earth Resources. This paper presents the content based image retrieval, using features like texture and color, called WBCHIR (Wavelet Based Color Histogram Image Retrieval).The texture and color features are extracted through wavelet transformation and color histogram and the combination of these features is robust to scaling and translation of objects in an image. The proposed system has demonstrated a promising and faster retrieval method on a WANG image database containing 1000 general-purpose color images. The performance has been evaluated by comparing with the existing systems in the literature.

/pdf/content-based-image-retrieval-using-color-and-texture-1ajzlvqmbl.pdf

Content Based Image Retrieval using Color and Texture

Identification of local regions from where optimal discriminating features can be extracted is one of the major tasks in the area of pattern recognition. To locate such regions different kind of region sampling techniques are used in the literature. There is no standard methodology to identify exactly such regions. Here we have proposed a methodology where local regions of varying heights and widths are created dynamically. Genetic algorithm (GA) is then applied on these local regions to sample the optimal set of local regions from where an optimal feature set can be extracted that has the best discriminating features. We have evaluated the proposed methodology on a data set of handwritten Bangla digits. In the present work, we have randomly generated seven sets of local regions and from every set, GA selects an optimal group of local regions which produces best recognition performance with a support vector machine (SVM) based classifier. Other popular optimization techniques like simulated annealing (SA) and hill climbing (HC) have also been evaluated with the same data set and maximum recognition accuracies were found to be 97%, 96.7% and 96.7% for GA, SA and HC, respectively. We have also compared the performance of the present technique with those of other zone based techniques on the same database.

A genetic algorithm based region sampling for selection of local features in handwritten digit recognition application

This paper proposes a novel quantum genetic algorithm (NQGA),whose core is that the strategies of updating quantum gate using qubit phase comparison approach and adjusting search grid adaptively are introduced.NQGA is characterized by rapid convergence,good global search capability and short computing time.Test results of complex functions and application example demonstrate NQGA is superior to conventional genetic algorithms in quality and efficiency.

A Novel Quantum Genetic Algorithm and Its Application

An efficient and computationally fast method for segmenting text and graphics part of document images based on textural cues is presented. We assume that the graphics part have different textural properties than the nongraphics (text) part. The segmentation method uses the notion of multiscale wavelet analysis and statistical pattern recognition. We have used M-band wavelets which decompose an image into M/spl times/M bandpass channels. Various combinations of these channels represent the image at different scales and orientations in the frequency plane. The objective is to transform the edges between textures into detectable discontinuities and create the feature maps which give a measure of the local energy around each pixel at different scales. From these feature maps, a scale-space signature is derived, which is the vector of features at different scales taken at each single pixel in an image. We achieve segmentation by simple analysis of the scale-space signature with traditional k- means clustering. We do not assume any a priori information regarding the font size, scanning resolution, type of layout, etc. of the document in our segmentation scheme.

Document image segmentation using wavelet scale-space features

This paper proposes a new image retrieval scheme using visually significant features. Clusters of points around significant curvature regions (high, medium, weak type) are extracted to obtain a representative image. Illumination, viewpoint invariant color features are computed from those points for evaluating similarity between images. Relative importance of the features is evaluated using a fuzzy entropy based measure computed from relevant and irrelevant set of the retrieved images marked by the users. The performance of the system is tested using different set of examples from general purpose image database. Robustness of the system has also been shown when the images have undergone different transformations.

Image retrieval with visually prominent features using fuzzy set theoretic evaluation

We have worked on the development of a character recognition system in the soft computing paradigm. In this paper we present a genetic algorithm used for feature selection with a Feature Quality Index (FQI) metric. We generate feature vectors by defining fuzzy sets on Hough transform of character pattern pixels. Each feature element is multiplied by a mask vector bit before reaching the input of a multilayer perceptron (MLP). The genetic algorithm operates on the bit string represented by the mask vector to select the best set of features. The method has been tested with three benchmark data sets and the results show a fast convergence of the genetic algorithm.

/pdf/a-genetic-algorithm-for-feature-selection-in-a-neuro-fuzzy-52c0r8h9bf.pdf

A genetic algorithm for feature selection in a neuro-fuzzy OCR system

Geocasting is a special variant of multicasting, where a data packet is destined for a particular geographic region. For many applications in wireless ad hoc and sensor networks, geocasting is an important and widely used communication service. Many geocast routing protocols with different approachability have been developed. Amongst them the geometry driven approach addresses the usage of an optimum path through the Fermat point of a triangle for packet forwarding. However, the geometry driven scheme faces some practical problems while determining the Fermat point for a polygonal region. Moreover, the method proposed can not be generalized for triangles with an angle more than 120?. In this paper we discuss a global minima based scheme for finding the Fermat point of a n sided polygonal geographic region, which is more general in its approach and is free from some of the constraints present in the geometry based scheme for finding the Fermat point. Along with that, the present scheme outplays the geometry driven scheme when it comes to the distance traveled by a packet and the energy consumed therefore.

An Alternative Approach to Find the Fermat Point of a Polygonal Geographic Region for Energy Efficient Geocast Routing Protocols: Global Minima Scheme

Energy efficiency is a much talked about thing in the domain of geocast routing protocols for Wireless Ad Hoc and Sensor Networks (WASNs). Fermat point based protocols are capable of reducing the energy consumption of a WASN by reducing the total transmission distance in a multi hop-multi sink scenario. Presently, there are quite a handful of them but many of them have not considered the effect of changing propagation environment around the considered network while measuring the performance of the protocol. Congested environment around a WASN increases the chance of multipath propagation and it in turn introduces multipath fading. In this paper, the effects of both of these factors are considered on the performance of I-Min routing protocol designed for WASNs.

/pdf/effect-of-multipath-fading-and-propagation-environment-on-2q3yn6yiai.pdf

Effect of multipath fading and propagation environment on the performance of a fermat point based energy efficient geocast routing protocol

Presents a multi-stage character recognition system for an Indian script, namely Bengali (also called Bangla) using fuzzy features and multilayer perceptrons (MLP). The fuzzy features are extracted from the Hough transform of a character pixel pattern. We first define a number of fuzzy sets on the Hough transform accumulator cells. The fuzzy sets are then combined by t-norms to generate feature vectors from each character. A set of fuzzy linguistic vectors is next generated from these feature vectors. The MLPs used for the classification have the fuzzy features as inputs. The MLP outputs also represent the "belongingness" of an input pattern to different fuzzy character pattern classes. To improve the recognition accuracy of Bengali characters, we divide all the patterns into three distinct sets. Each set of characters is once again divided into a number of mutually exclusive character pattern classes. During recognition, the class of each pattern is first determined, followed by recognition of the actual character within that class. The recognition accuracy of the system is more than 98%.

/pdf/recognition-of-an-indian-script-using-multilayer-perceptrons-50g75dz049.pdf

Pradip K. Das

Papers

Image retrieval with visually prominent features using fuzzy set theoretic evaluation

A genetic algorithm for feature selection in a neuro-fuzzy OCR system

An Alternative Approach to Find the Fermat Point of a Polygonal Geographic Region for Energy Efficient Geocast Routing Protocols: Global Minima Scheme

Effect of multipath fading and propagation environment on the performance of a fermat point based energy efficient geocast routing protocol

Recognition of an Indian script using multilayer perceptrons and fuzzy features