Parameters selection for Blurred Image Matching









Abstract

Blurred Image Matching (BIM) is a features comparison technique based on image pre-processing and blobs detection. BIM is designed to perform comparison on images presenting a strong level of noise. The technique's previously published results display excellent robustness, speed, and unique characteristics when compared to existing techniques, leading to its implementation in the industry. This article investigates the process BIM is based on, and offers precision over its parameters, in order to optimize their selection. The article investigates the technique's performances when using various blurring and thresholding parameters, two elements that are central to BIM's process. Threw the experiments it was determined that certain combinations of parameters were increasing the probability of finding relevant blobs for comparison, while also highlighting the importance of images colorimetry in the processing. The results obtained are approximations of ideal parameters for the functioning of BIM, resulting in an improvement of the technique's performances.


Modules


Algorithms


Software And Hardware

• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL