There are multiple web applications! How do we extract the web components of different sizes to compare the visual similarities between them?

Abstract

New visual similarity technology based on SSIM is developed as a tool for measuring visual similarities of multiple web applications of different sizes, which is the first technology model that targets web components of different sizes. This visual similarity technology takes screenshots of the web applications in the background and extracts the components needed for further study. For finding visual similarities between different components on the web applications, it is essential for the algorithms to be able to handle components of varying sizes, after components extractions have been performed by the visual similarity technology. This work describes a new algorithmic implementation for the detection of visual similarities of various web components developed. The new implementation extends upon the SSIM algorithm that measure the visual similarity of images of similar sizes. The program portrays successful detection of visually similar charts and maps of different sizes in the web applications. This implementation greatly expands the possibilities to investigate visual reuse between web applications' developers.