Abstract
This study presents a Bendlet Transform-based Object Detection (BTOD) system that recognizes an object in the image. Finding a specific object in images or videos is the goal of the field of object recognition. Though humans are able to identify a large number of objects, it is very difficult for computer vision systems in general. The appearance of the objects may change depending on the perspective, the size or scale, or translation and rotation. This work extracts Bendlet transform-based features from the images at different levels, and then the discriminant features are selected by employing genetic algorithms. The performance of the BTOD system is analyzed with different nearest neighbours for classifying objects in the Columbia Object Image Library (COIL-100) in terms of classification accuracy. It is observed from the results that the BTOD system with a one-nearest neighbour provides better performance than the two-nearest neighbour classifier. The former classifier gives 99.47% accuracy, whereas the later classifier gives 99.19%.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: INTERNATIONAL JOURNAL OF ADVANCES IN SIGNAL AND IMAGE SCIENCES
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.