Qualification: 
M.Tech
d_bharathi@cb.amrita.edu

Bharathi D. currently serves as Assistant Professor at Department of Computer Science and Engineering, School of Engineering, Coimbatore Campus. Her areas of research include Image Processing.

Publications

Publication Type: Journal Article

Year of Publication Publication Type Title

2018

Journal Article

K. Srunitha and D. Bharathi, “Mango leaf unhealthy region detection and classification”, Lecture Notes in Computational Vision and Biomechanics, vol. 28, pp. 422-436, 2018.[Abstract]


Diseases in any plant decrease the productivity and quality of product. Identification of plant leaf diseases by naked human eye is very difficult. Image processing techniques can identify the diseased leaf by preprocessing and classifying leaf unhealthy regions. This paper delivers an implementation on Mango leaf unhealthy region detection and classification. In the Proposed work Multiclass SVM is used for diseases classification and segmentation through k-means. The experimental results show the effectiveness of the proposed method in recognizing the diseases affected mango leaf.

More »»

2016

Journal Article

Malathi P., D. Bharathi, and Vinodhini, R. E., “Illumination Invariant Face Recognition using Fisher Linear Discriminant Algorithm (FLDA)”, International Journal of Control Theory and Applications, vol. 9, no. 10, pp. 4201-4210, 2016.[Abstract]


In image processing domain biometrics is an emerging field, in which matching the face images of optical and infrared is a tough toil. Since the optical and infrared images are captured by two disparate devices there exists a great diversity between one and the other kinds of images. A classy method supported by Feature discriminant analysis[1], which uses fisher linear discriminant algorithm (FLDA) is proposed in this paper. This approach has two steps to minimize this chaos and to maximize the performance of optical-infrared face recognition. In first step, extract all the common discriminant features from heterogeneous (infrared and optical) face images using FLDA. In second step, k-Nearest Neighbors (k-NN) algorithm is used on the result to conclude whether they match or not. To show that the algorithm works better than the existing ones, experiments are conducted on optical and infrared datasets.

More »»

2015

Journal Article

A. Loganathan and D. Bharathi, “Sparsification of graph laplacian for image indexing using multidimensional spectral hashing”, International Journal of Imaging and Robotics, vol. 15, pp. 43-56, 2015.[Abstract]


Many multimedia applications need an algorithm to search similar images from a large scale database. Hashing based techniques are used for searching similar images in practice. The existing sparsification of graph laplacian with spectral hashing for similarity search is suitable only for smaller neighborhoods. But in the most of the cases, multimedia applications requires an algorithm for larger neighborhoods. This creates a research potential to develop a novel approach for generating optimal binary code to retrieve larger neighborhoods. This paper proposes multidimensional spectral hashing that uses sparsification of graph laplacian. Multidimensional spectral hashing uses outer product Eigen functions to improve the codes. The exponential growth of outer product functions are handled using kernel-trick. This makes our proposed algorithm to achieve storage-efficient multi-dimensional spectral hashing. The performance analysis of our proposed algorithm shows better result in terms of binary code generation, true positives and retrieving neighbor’s images.

More »»

Publication Type: Conference Proceedings

Year of Publication Publication Type Title

2016

Conference Proceedings

P. Sathish and D. Bharathi, “Automatic Road Sign Detection and Recognition Based on SIFT Feature Matching Algorithm”, Proceedings of the International Conference on Soft Computing Systems, Part of the Advances in Intelligent Systems and Computing book series (AISC), vol. 38. Springer India, New Delhi, 2016.[Abstract]


The paper presents a safety and comfort driving assistance system to help driver in analyzing the road sign boards. The system assists the drivers by detecting and recognizing the sign boards along roadside from a moving vehicle. Signboards are detected using color and shape detection techniques, and they are recognized by matching the extracted scale invariant features with the features in the database. The proposed method based on SIFT feature detects 90 {%} of road sign boards accurately. Experimental analysis carried out using C/C++ and Open CV Libraries shows better performance in detecting and recognizing sign boards when real-time video frames are given as input.

More »»
Faculty Research Interest: 
207
PROGRAMS
OFFERED
6
AMRITA
CAMPUSES
15
CONSTITUENT
SCHOOLS
A
GRADE BY
NAAC, MHRD
8th
RANK(INDIA):
NIRF 2018
150+
INTERNATIONAL
PARTNERS