Performance Analysis of Feature Detection and Description (FDD) Methods on Accident Images

0Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Feature detection and description is a significant process in several computer vision tasks such as object recognition, detection, image classification and registration. The challenge relies on choosing appropriate feature detector and descriptor, concerning an application, regardless of the appearance and the content of an image. The objective of the performance analysis is to identify the suitable FDD method, in order to extract distinct and robust features from the accident images, possess crucial information for the analysis of accident severity. In this paper, the feature detection and description methods from the literature is applied to the accident images. However, the accident images captured in real-time through mobile cameras confronts motion blur, illumination, rotation and scale variations. Therefore, the robustness of the feature detectors and descriptors are evaluated under various image transformations and the results are compared and analyzed. The result shows, under feature detectors, CenSurE and SIFT performs better with reference to repeatability and matching score. SIFT and ORB are better in the category of feature descriptors for the analysis of accident images.

Cite

CITATION STYLE

APA

Beryl Princess, P. J., Silas, S., & Rajsingh, E. B. (2020). Performance Analysis of Feature Detection and Description (FDD) Methods on Accident Images. In Communications in Computer and Information Science (Vol. 1240 CCIS, pp. 384–396). Springer. https://doi.org/10.1007/978-981-15-6315-7_32

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free