Abstract
Fast movement of objects and illumination changes may lead to a negative effect on camera images for object detection and tracking. Event cameras are neuromorphic vision sensors that capture the vitality of a scene, mitigating data redundancy and latency. This paper proposes a new solution to moving object detection and tracking using an event frame from bio-inspired event cameras. First, an object detection method is designed using a combined event frame and a standard frame in which the detection is performed according to probability and color, respectively. Then, a detection-based object tracking method is proposed using an event frame and an improved kernel correlation filter to reduce missed detection. Further, a distance measurement method is developed using event frame-based tracking and similar triangle theory to enhance the estimation of distance between the object and camera. Experiment results demonstrate the effectiveness of the proposed methods for moving object detection and tracking.
Highlights
Intelligent agents such as robots, unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and autonomous underwater vehicles (AUVs), are widely used in military and civilian fields [1–8]
Based on similar triangle theory, this paper proposes an event-frame-based distance measurement algorithm to measure the distance between the event camera and object
The third set of experiments compares the effects of the PnP (Perspective-n-Point) algorithm based on event frame with the similar triangle algorithm based on an event frame, and tests the ranging effect of the similar triangle algorithm at different distances
Summary
Intelligent agents such as robots, unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and autonomous underwater vehicles (AUVs), are widely used in military and civilian fields [1–8]. Object detection and tracking are very important to improve the autonomy of intelligent agents. For traditional object detection methods, Viola and Jones [10,11] propose an algorithm that realizes the real-time detection of human faces for the first time. Their algorithm is called the VJ detector. The VJ detector uses a sliding window for detection and greatly improves the detection speed by using three technologies: the integral map, feature selection, and detection cascades [9–11]. Dalal et al propose a histogram of oriented gradient (HOG) feature descriptor [12]. HOG is an important improvement to scale-invariant feature transformation [13,14] and shape contexts [15]. The HOG detector is an important foundation for many object detectors. Felzenszwalb et al [16] propose the deformable parts model (DPM)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
