Mobile QR Code
Title Target Identification Using a Moment Invariant Approach
Authors Muzhir Shaban Al-Ani;Aso Darwesh
DOI https://doi.org/10.5573/IEIESPC.2019.8.5.335
Page pp.335-346
ISSN 2287-5255
Keywords Object identification; Object recognition; Moments; Feature vectors; Digital image processing
Abstract A target identification process leads to identifying a specific object or target. There are many objects in nature; some of these objects have similar characteristics, but others are different. The main aim of this work is to design an efficient approach to object or target identification. The implemented approach tries to overcome the challenges that appear during processing. This approach is implemented via many steps (preprocessing, feature extraction, invariant feature application, and class calculations) and then compares the obtained results to decide the nearest distance of the specific object to an existing image in a database. The obtained results indicate that high-performance identification is achieved via application of this approach. Three orientations for images are tested (0degrees, 90degrees, and 180degrees), and in all these images, the obtained similarity is about 99%.