Object tracking in a stereo and infrared vision system

Colantonio S., Di Bono M. G., Pieri G., Salvetti O., Benvenuti M.

Motion Prediction  Content-based Retrieval  Infrared Vision  Object Tracking  I.5 Pattern Recognition  I.5.4 Applications  H.3.3 Information Search and Retrieval 

In this paper we deal with the problem of real-time detection, recognition and tracking of moving objects in open and unknown environments using an infrared (IR) and visible vision system. This task is generally very challenging and automatic tools to identify and follow an object -target- are often subject to constraints regarding the environment under investigation, the characteristics of the target itself and its full visibility with respect to a background. The goal of our study is the identification, and the possible prediction, of the movements of a target operating in real-time using a vision system capable of stereo and IR vision. Multi-source information is acquired using a system composed of a thermo-camera and two stereo visible-cameras synchronized. We obtain a set of IR images, which make the system more robust and invariant to light changes in the scene, corresponding to stereo grey level images. Firstly, target detection is performed by extracting its characteristic features from the images and then by storing the computed parameters on a specific database; secondly, the tracking task is carried on using different algorithms. Two different computational approaches are followed for tracking: a Hierarchical Artificial Neural Network (HANN) is used during active tracking for the recognition of the actual target, while if occurring partial occlusions or masking, a database retrieval method is used to support the search of the correct target to follow. A robotized prototype has been designed with an IR camera and two visible stereo cameras mounted on it. The stereo cameras are used to obtain three-dimensional information about target with respect to its background. The prototype has been tested on case studies regarding the identification and tracking of animals moving at night in an open environment, and the surveillance of known scenes for unauthorized access control.

Source: Advanced Infrared Technologies and Applications- AITA, Rome, Italy, September 7-9,