An Improved Framework For Content Based Image Retrieval Based Chemical Reaction Optimization
Main Article Content
Abstract
With regards to the areas pertaining to image processing and machine learning, Content Based Image Retrieval (CBIR) system has become the main area of research in the recent years. Low-level attributes such as shape, color and texture are in general analyzed through CBIR systems. While describing a data collection, combination of features are selected among a given larger set through feature selection. While classifying and categorizing, valuations of feature performance and selection techniques are carried out. The Chemical Reaction Optimization technique has a better search capacity and can solve NP hard optimization problems. To be more specific, many problems with high efficiency are addressed through this technique. This work involves a novel classifier called modified AdaBoost and in this proposed technique, the number of classification trees and maximum depth per tree is optimized with the help of CRO.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.