Publication Type : Conference Paper
Publisher : 2017 International Conference on Intelligent Computing and Control
Source : 2017 International Conference on Intelligent Computing and Control (I2C2), IEEE (2018)
Url : https://ieeexplore.ieee.org/document/8321839
Keywords : fruit and veggies, Haar wavelet features, image classification, voting based classification approach, Watershed transform
Campus : Mysuru
School : School of Arts and Sciences
Department : Computer Science
Year : 2018
Abstract : Massive increase in the degree of data in online data repositories desires the need of efficient data retrieval and management. The current state of art algorithms are adept in information retrieval concerned with text type data and to the success rate of almost 99%. However the information retrieval based on multimedia data like images needs more revisions so as to obtain the expected outcome. In this paper, the research objective is to classify the images based on the shape and regional characteristics. The categories of images considered include fruits and vegetables. The proposed technique for image classification employs watershed transform for segmentation and from which the Haar wavelet features are computed and are directed for classification using SVM, KNN and Naïve Bayes classifier. A voting based technique is employed for classification of images and the overall accuracy of the system is about 90%. © 2017 IEEE.
Cite this Research Publication : Manisha Shivaji Pawar, Louis Perianayagam, and Shobha Rani N., “Region based image classification using watershed transform techniques”, in 2017 International Conference on Intelligent Computing and Control (I2C2), 2018.