An Efficient Method for Similarity Measure in Independent PCA based Classification
Journal Title: INTERNATIONAL JOURNAL OF COMPUTERS & TECHNOLOGY - Year 2013, Vol 9, Issue 3
Abstract
The paper presents a new approach of finding nearest neighbor in image classification algorithm by proposing efficient method for similarity measure. Generally in supervised classification, after finding the feature vectors of training images and testing images, nearest neighbor classifier does the classification job. This classifier uses different distance measures such as Euclidean distance, Manhattan distance etc. to find the nearest training feature vector. This paper proposes to use Mean Squared Error (MSE) to find the nearness between two images. Initially Independent Principal Component Analysis (PCA),which we discussed in our earlier work, is applied to images of each class to generate Eigen coordinate system for that class. Then for the given test image, a set of feature vectors is generated. New images are reconstructed using each Eigen coordinate system and the corresponding test feature vector. Lowest MSE between the given test image and new reconstructed image indicates the corresponding class for that image. The experiments are conducted on COIL-100 database. The performance is also compared with distance based nearest neighbor classifier. Results show that the proposed method achieves high accuracy even for small size of training set.
Authors and Affiliations
Dr. H. B. Kekre, Dr. Tanuja K. Sarode, Jagruti K. Save
Keywords
Related Articles
- EP ID EP650140
- DOI 10.24297/ijct.v9i3.3335
- Views 82
- Downloads 0
In-Silico Methodologies for Cancer Multidrug Optimization
Drug combinations is considered as an effective strategy designed to control complex diseases like cancer. Combinations of drugs can effectively decrease side effects and enhance adaptive resistance. Therefore, increasin...
Integration of IR-UWB Services into Single- and Multi-Channel Optical Coherent OFDM Network
Distribution of wireless communication signals over implemented optical networks is a challenged task since the presence of wireless services may affect the performance of the existing wired ones. This paper addresses th...
Hybrid ANFIS Controller for 6-DOF Manipulator with 3D Model
This paper proposes a hybrid ANFIS (Adaptive Neuro-Fuzzy Inference System) controller with DMRAC (Direct Model Reference Adaptive Control) and mathematical modeling of the kinematic and dynamic solutions. The controller...
Finite-Difference Time-Domain (FDTD) Simulation of Novel Terahertz Structures
Previous work on compact, variable, efficient, high brightness radiation sources is extended by calculating the radiated power and angular distributions for different configurations and drive sources. Figures of merit ar...
Formal Methods in Information Security
Formal methods use mathematical models for analysis and verification at any part of the program life-cycle. The use of formal methods is admitted, recommended, and sometimes prescribed in safety-and security-related stan...