Aiming at the characteristics of large amount of railway surveillance data and high timeliness of foreign intrusion detection, this paper proposed a method of abnormal perception of railway perimeter based on trackside surveillance video, by the pixel value change rate and gray projection. It consists of four steps: the color type of image is converted to HSV from original RGB in preprocessing; background modelling; abnormal pixel is identified according to the pixel value change rate; the location and size of abnormal area are extracted by gray projection, realizing abnormal perception of the railway perimeter. The small calculation of change rate can speed up the method which is suitable for large-scale camera numbers in the railway scene. Meanwhile, gray projection reflects relative position of abnormal area in the horizontal and vertical directions, obtaining effective information in the video data and getting rid of most useless information. The proposed method in this paper can achieve 0.964 accuracy and 35 FPS on CPU platform.
Abnormal Perception of Railway Perimeter Based on Trackside Surveillance Video
Lect. Notes Electrical Eng.
International Conference on Electrical and Information Technologies for Rail Transportation ; 2021 October 21, 2021 - October 23, 2021
Proceedings of the 5th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2021 ; Kapitel : 16 ; 149-157
2022-02-19
9 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Abnormal Perception of Railway Perimeter Based on Trackside Surveillance Video
British Library Conference Proceedings | 2022
|Europäisches Patentamt | 2022
|Europäisches Patentamt | 2018
|