Abstract
Wide Area Motion Imagery (WAMI) yields high resolution images with a large number of extremely small objects. Target objects have large spatial displacements throughout consecutive frames. This nature of WAMI images makes object tracking and detection challenging. In this paper, we present our deep neural network-based combined object detection and tracking model, namely, Heat Map Network (HM-Net). HM-Net is significantly faster than state-of-the-art frame differencing and background subtraction-based methods, without compromising detection and tracking performances. HM-Net follows object center-based joint detection and tracking paradigm. Simple heat map-based predictions support unlimited number of simultaneous detections. The proposed method uses two consecutive frames and the object detection heat map obtained from the previous frame as input, which helps HM-Net monitor spatio-temporal changes between frames and keeps track of previously predicted objects. Although reuse of prior object detection heat map acts as a vital feedback-based memory element, it can lead to unintended surge of false positive detections. To increase robustness of the method against false positives and to eliminate low confidence detections, HM-Net employs novel feedback filters and advanced data augmentations. HM-Net outperforms state-of-the-art WAMI moving object detection and tracking methods on WPAFB dataset with its 96.2% F1 and 94.4% mAP detection scores, while achieving a 61.8% mAP tracking score on the same dataset.
Highlights
O BJECT detection and tracking are broad and active subfields of computer vision
As a result of these novel features we have achieved state-of-the-art performance on both WAMIbased object detection and tracking, and our method is several times faster than recent Wide Area Motion Imagery (WAMI) detection and tracking algorithms
3-frame differencing The median background model is build for every 10 frame cycle Up to 5 prior frames are used by Convolutional Neural Networks (CNN) model for vehicle detection Up to 5 prior frames are used by CNN model for vehicle detection
Summary
O BJECT detection and tracking are broad and active subfields of computer vision. Object detection and tracking applications on Wide Area Motion Imagery (WAMI) have had its share of the attention, and today it still holds its place as a challenging computer vision task. WAMI deals with monitoring a large area, which is several kilometers in diameter, with an airborne sensor consisting of multiple cameras. Captured frames from cameras are stitched to obtain a high-resolution image of the target area. A registration algorithm is applied to consecutive images to compensate for the camera motion. Object detection and tracking with WAMI have numerous applications in both civilian and military domains [1]–[4]. Military domain applications include intelligence-gathering, reconnaissance, border security, and surveillance. Two stage CNN architecture (FoveaNet and ClusterNet). 3-frame differencing The median background model is build for every 10 frame cycle Up to 5 prior frames are used by CNN model for vehicle detection Up to 5 prior frames are used by CNN model for vehicle detection
Sign-in/Register to view highlights & summary 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.
