This report describes a new procedure for tracking road segments and finding potential vehicles in imagery of approximately 1-3 feet per pixel ground resolution. This work is part of a larger effort by SRI International to construct an image understanding system for monitoring roads in aerial imagery. The overall effort is directed towards specific problems that arise in processing aerial photographs for such military applications as cartography, intelligence, weapon guidance, and targeting. A key concept is the use of a generalized digital map data base to aid in the interpretation of imagery. The primary objectives of the overall 'knowledge-based road expert system' are to analyze images to accomplish the following: (1) find road fragments in low- to medium-resolution images; (2) track roads in medium- to high-resolution images; (3) find anomalies on roads; and (4) interpret anomalies as vehicles, shadows, signposts, surface markings, etc. The road tracking algorithm is started by indicating the center and direction of a road fragment found in low- to medium- resolution images. The nominal road width is supplied either from the data base or by an image analysis function that can determine the width of a road fragment. The road tracker produces two forms of output: a point list describing the track of the road center, and a binary image of all points in the road that are anomalous and might belong to vehicles. In the complete road- expert system, this image will then be analyzed to screen out false alarms and interpret the remaining anomalies.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Cooperative methods for road tracking in aerial imagery

    McKeown, D.M. jun. / Denlinger, J.L. | Tema Archive | 1988


    COCOA: tracking in aerial imagery

    Ali, Saad / Shah, Mubarak | SPIE | 2006


    FAST ROAD DETECTION AND TRACKING IN AERIAL VIDEOS

    Zhou, H. / Kong, H. / Alvarez, J. et al. | British Library Conference Proceedings | 2014


    Road Pothole Detection Using Unmanned Aerial Vehicle Imagery and Deep Learning Technique

    Dana Mohammed Ali / Haval A.Sadeq | DOAJ | 2022

    Free access

    Efficient Road Detection and Tracking for Unmanned Aerial Vehicle

    Zhou, Hailing / Kong, Hui / Wei, Lei et al. | IEEE | 2015