Video tracking system optimization using evolution strategies

Abstract

A video-based tracking system for airport surveillance, composed by modules performing vision tasks at different levels, is adapted for operational conditions by means of Evolution Strategies (ES). An optimization procedure has been carried out considering different scenes composed of representative trajectories, supported by a global evaluation metric proposed to quantify the system performance. The generalization problem (the search of appropriate solutions for general situations, avoiding over-adaptation to particular conditions) is approached considering evaluation of ES-individuals over combinations of trajectories to build the fitness function. In this way, the optimization procedure covers sets of trajectories representing different types of problems. Besides, alternative operators for aggregating partial evaluations have been analysed. Results show how the optimization strategy provides a sensitive tuning of performance related to input parameters at different levels, and how the combination of different situations improves the generalization capability of the trained system. The global performance final system after optimization is also compared with representative algorithms in the state of the art of visual tracking.

Examining the effect of road horizontal alignment on the speed of semi-automated vehicles

“…Foresti in [6] describes the low-level video processing requirements for the implementation of smart cameras. In [7][8][9], the tracking people problem is discussed and some applications are described. The existence of visual sensing nodes with increasing DSP processing capabilities, together with the possibility of being interconnected through wireless communication links opens up a new realm of intelligent vision-enabled applications.…”

Hybrid Modified Evolution Strategies and Linear Programming for Beef Cattle Feed Optimization

“…ES has been proven as a useful method for optimization. The experiment conducted by Garcia et al [16]for several scenarios of video tracking system shows that ES can improve the outcomes from the most particular case to general circumstances. Furthermore, based on Khuluqi et al [17], the ES can generate a better result than the manually evaluated amount generation process in the home textile industry for profit optimization.…”

Video Sensor Architecture for Surveillance Applications

“…Foresti in [ 6 ] describes the low-level video processing requirements for the implementation of smart cameras. In [ 7 – 9 ], the tracking people problem is discussed and some applications are described. The existence of visual sensing nodes with increasing DSP processing capabilities, together with the possibility of being interconnected through wireless communication links opens up a new realm of intelligent vision-enabled applications.…”