Abstract
One of the most important tasks for ensuring life safety is monitoring objects and tracking the behavior of people in urban infrastructure from an unmanned aerial vehicle (UAVs). Difficulties arise when the UAVs is used for surveillance in areas that are difficult or unsuitable for humans to access, and then there is a risk of accidents. Therefore, analysis, reconstruction and tracking of objects can help in solving existing cases. As a critical technology in computer vision and video processing, video object segmentation has far-reaching pragmatic and application implications. In this paper, we propose an algorithm for segmenting video objects based on background reconstruction for extracting moving objects from a video sequence taken from a UAVs. In the theoretical part of the work, methods for detecting and segmenting moving objects are considered based on background subtraction algorithms, the optical flow of the Kalman filter. An exact mask for detecting changes is proposed, taking into account the estimate of the threshold of noise characteristics, as well as an algorithm for detecting movements. A qualitative analysis of the background reconstruction error was carried out using a statistical criterion. A modified algorithm for obtaining a change detection mask (MDM) using a threshold estimated using noise characteristics is proposed. The results of segmentation of objects, calculation of the root-mean-square error of background reconstruction for the proposed method when extracting moving objects are shown. The developed method allows you to effectively recognize and extract moving objects in a video sequence. This algorithm is suitable for the case when the UAVs freezes in the air. After the background reconstruction, there are small defects that can be eliminated using morphological operations.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: 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.
