Textural Feature Extraction of Natural Objects for Image Classification

Krishna, Vishal and Kumar, Ayush and Kishore, B (2015) Textural Feature Extraction of Natural Objects for Image Classification. International Journal of Image Processing, 9 (6). pp. 320-334.

[img] PDF
1855.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

The field of digital image processing has been growing in scope in the recent years. A digital image is represented as a two-dimensional array of pixels, where each pixel has the intensity and location information. Analysis of digital images involves extraction of meaningful information from them, based on certain requirements. Digital Image Analysis requires the extraction of features, transforms the data in the high-dimensional space to a space of fewer dimensions. Feature vectors are n-dimensional vectors of numerical features used to represent an object. We have used Haralick features to classify various images using different classification algorithms like Support Vector Machines (SVM), Logistic Classifier, Random Forests Multi Layer Perception and Naïve Bayes Classifier. Then we used cross validation to assess how well a classifier works for a generalized data set, as compared to the classifications obtained during training.

Item Type: Article
Uncontrolled Keywords: Feature Extraction, Haralick, Classifiers, Cross-Validation
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Depositing User: MIT Library
Date Deposited: 12 Jan 2017 13:15
Last Modified: 16 May 2017 06:43
URI: http://eprints.manipal.edu/id/eprint/148046

Actions (login required)

View Item View Item