A Hardware-Friendly Optical Flow-Based Time-to-Collision Estimation Algorithm.

Cong Shi,Shrinivas Pundlik,Zhuoran Dong,Gang Luo

doi:10.3390/s19040807

Abstract

This work proposes a hardware-friendly, dense optical flow-based Time-to-Collision (TTC) estimation algorithm intended to be deployed on smart video sensors for collision avoidance. The algorithm optimized for hardware first extracts biological visual motion features (motion energies), and then utilizes a Random Forests regressor to predict robust and dense optical flow. Finally, TTC is reliably estimated from the divergence of the optical flow field. This algorithm involves only feed-forward data flows with simple pixel-level operations, and hence has inherent parallelism for hardware acceleration. The algorithm offers good scalability, allowing for flexible tradeoffs among estimation accuracy, processing speed and hardware resource. Experimental evaluation shows that the accuracy of the optical flow estimation is improved due to the use of Random Forests compared to existing voting-based approaches. Furthermore, results show that estimated TTC values by the algorithm closely follow the ground truth. The specifics of the hardware design to implement the algorithm on a real-time embedded system are laid out.

Highlights

The concept of time-to-collision (TTC) is important for collision alerting/avoidance and finds applications in robot navigation [1,2], mobility assistive devices for visually impaired people [3,4], and driving safety [5]
We propose a novel hardware-friendly TTC estimation algorithm based on the field divergence of optical flow derived from motion energy features
In this paper we have shown that the proposed TTC estimation algorithm is accurate, hardware friendly, and can be potentially implemented on a smart video sensor hardware

Summary

Introduction

The concept of time-to-collision (TTC) is important for collision alerting/avoidance and finds applications in robot navigation [1,2], mobility assistive devices for visually impaired people [3,4], and driving safety [5]. Unlike in collision warning systems based on range sensors (radar, laser, structured-light depth sensor), or stereo cameras that measure the distance to the obstacles by binocular matching, TTC computation, requiring only a monocular camera, can be done without knowing the physical distance to the objects [1,2,3,4,5,6,7,8,9,10,11,12]. Low cost smart video sensors for collision avoidance can be made feasible by employing TTC-based approaches. We describe an algorithm for TTC computation with the intention of deploying it as a smart video sensor for collision avoidance. Some of the desirable qualities for the TTC algorithm for smart video sensors are low cost and good scalability for embedded implementation, real-time performance, and dense pixel-wise output

Results

Discussion

Conclusion