ICTACT Journal on Image and Video Processing

R. Obulakonda Reddy ¹, B. Eswara Reddy ¹, E. Keshava Reddy ²

Affiliations
1 Department of Computer Science and Engineering, JNTUA College of Engineering, IN
2 Department of Mathematics, JNTUA College of Engineering, IN

Abstract

Texture extraction and classification is the key feature that is used in pattern recognition and classification. Binary patterns are very powerful discrimination operators that are able to extract texture features irrespective of its illumination changes. This paper mainly focuses on extraction of fabric texture patterns that are used in discriminating the defects and the non-defects. A ternary pattern is a powerful tool for extracting the microstructures of the images, used for feature extraction that has robustness towards the illumination invariance. On the other hand, a Zernike moment which is simultaneously invariant to similarity transformation and rotation is also explained. Experimental analysis is conducted both on standard texture images and fabric images. The performance of the proposed approach is evaluated using SVM, KNN and Bayes classifiers.

Keywords

SVM Classification, Texture Extraction, Ternary Patterns, Moment of Invariants.

Full Text

     

K. Sundeep Kumar ¹, C. F. Jacob ², B. Eswara Reddy ³

Affiliations
1 Department of Information Science and Engineering, SEA College of Engineering & Technology, IN
2 Department of Computer Science and Engineering, CMR Institute of Technology, IN
3 Department of Computer Science and Engineering, JNTU College of Engineering, IN

Abstract

The purpose of our research is to introduce two unique approaches for the unsupervised segmentation of wound images. The first method of segmentation is by using the texture of the image and is performed using the multi-channel filtering hypothesis for the processing of visual or optical data during the preliminary stages of the Human Visual System. We obtain the different channels from an image by filtering the image using a Gabor Filter Bank. The textural features are obtained from each filtered image and the final segmented image is acquired by reconstructing the original input image from these filtered images. The second method of segmentation was performed using parametric kernel graph cuts. Using a kernel function we transform the image data implicitly such that a piecewise constant model of the graph cut interpretation is now applicable. The objective function comprises of an original data term in order to assess the deviance of the transformed data from the initial input data within each partition. This method avoids sophisticated modelling of the input image data while availing of the computational advantages of graph cuts. By using a conventional kernel function, the energy minimization boils down to image partitioning via graph cut iterations and assessments of region parameters by means of fixed point calculations. The efficacy and flexibility of both the methods are established by carrying out investigations on real wound images.

Keywords

Gabor Filter Banks, Graph Cuts, Radial Basis Function (RBF), Wound Healing, Human Visual System (HVS).

Full Text

     

P. Chandra Sekhar Reddy ¹, B. Eswara Reddy ², V. Vijaya Kumar ³

Affiliations
1 Deprtment of Computer Science and Engineering, Nalla Narasimha Reddy Education Society’s Group of Institutions, IN
2 Deprtment of Computer Science and Engineering, JNTUA College of Engineering, IN
3 Deprtment of Computer Science and Engineering, Anurag Group of Institutions, IN

Abstract

Today face recognition capability of the human visual system plays a significant role in day to day life due to numerous important applications for automatic face recognition. One of the problems with the recent image classification and recognition approaches are they have to extract features on the entire image and on the large grey level range of the image. The present paper overcomes this by deriving an approach that reduces the dimensionality of the image using Shape primitives and reducing the grey level range by using a fuzzy logic while preserving the significant attributes of the texture. The present paper proposed an Image Dimensionality Reduction using shape Primitives (IDRSP) model for efficient face recognition. Fuzzy logic is applied on IDRSP facial model to reduce the grey level range from 0 to 4. This makes the proposed fuzzy based IDRSP (FIDRSP) model suitable to Grey level co-occurrence matrices. The proposed FIDRSP model with GLCM features are compared with existing face recognition algorithm. The results indicate the efficacy of the proposed method.

Keywords

GLCM Features, Preprocessing, Grey Level Range, Significant Image Features, Dimensionality Reduction.

Full Text

     

Pushpalatha S. Nikkam ¹, B. Eswara Reddy ²

Affiliations
1 Department of Computer Science Engineering, Jawaharlal Nehru Technological University, Anantapur, IN
2 JNTUA College of Engineering,, IN

Abstract

Image based content recognition and retrieval is critical in many applications. Existing mechanisms for content based image retrieval lack in terms of performance. In this paper a hierarchical template tree based CBIR system is described. Content in image is represented using a combination of shape features and low level features. Comprehensive feature set definitions proposed enables in achieving better performance. Shape and low level features are considered as templates. Templates of similar categories are further decomposed to form a hierarchical template tree. Query image is converted into a query template and is decomposed. A part template based matching scheme and SVM classifier is used to retrieve visually similar images. Results presented in the paper prove superior performance of proposed technique when compared to recent existing mechanisms in place. An improvement of 10.45% and 9.69% in mean average precision and mean retrieval accuracy is reported using proposed approach.

Keywords

Part-Template, Hierarchical Template Tree, HOG, Shape, Tree-Formation, SVM Classifier.

Full Text

     

Anita Dixit ¹, Nagaratna P. Hegde ², B. Eswara Reddy ³

Affiliations
1 Department of Computer Science, Jawaharlal Nehru Technological University, Anantapur, IN
2 Department of Computer Science, Vasavi College of Engineering, IN
3 Department of Computer Science, JNTUA College of Engineering, IN

Abstract

Remote sensing is one of the hottest topics of research, which intends to study or analyze a particular object in the topographic map. The monitoring and management is possible when it is possible to differentiate the objects in the satellite image. However, satellite image classification is not easy, as it consists of numerous minute details. In addition to this, the accuracy and faster execution of the classification system are significant factors. This article presents a satellite image classification system that is capable of differentiating between soil, vegetation and water bodies. To achieve the goal, we categorize the entire system into three major phases; they are satellite image pre-processing, feature extraction and classification. The initial phase attempts to denoise the satellite image by the adaptive median filter and the contrast enhancement is done by Contrast Limited Adaptive Histogram Equalization (CLAHE). As the satellite image possess many important features, this work extracts curvelet moments by applying curvelet transform. The feature vector is formed out of these curvelet moments and the ELM classifier is used to train these features. The performance of the proposed approach is observed to be satisfactory in terms of sensitivity, specificity, and accuracy.

Keywords

Remote Sensing, Satellite Image Classification, Feature Extraction.

Full Text