Making the home safer and more secure through visual surveillance

A. Prati², R. Cucchiara¹ and R. Vezzani¹

¹Department of Information Engineering, University of Modena and Reggio Emilia, Modena, Italy
²Department of Sciences and Methods in Engineering, University of Modena and Reggio Emilia, Modena, Italy

Video surveillance has a direct application in intelligent home automation or domotics (from the Latin word domus, that means ‘home’, and informatics). In particular, in-house video surveillance can provide good support for people with some difficulties (e.g. elderly or disabled people) living alone and with limited autonomy. New hardware technologies for surveillance are now affordable and provide high reliability. Problems related to reliable software solutions are not completely solved, especially concerning the application of general-purpose computer vision techniques in indoor environments. Indeed, assuming the objective is to detect the presence of people, track them, and recognize dangerous behaviors by means of abrupt changes in their posture, robust techniques need to cope with nontrivial dificulties.

In particular, luminance changes, shadows and frequent posture changes must be taken into account. Long-lasting occlusions are common due to the proximity of the cameras and the presence of furniture and doors that can often hide parts of a person’s body. For these reasons, we developed computer vision techniques based on probabilistic and appearancebased tracking, particularly conceivable for people tracking and posture classi.cation.

Despite its effectiveness for long-lasting and large occlusions, this approach tends to fail whenever the person is monitored with multiple cameras and he appears in one of them already occluded. Different views provided by multiple cameras can be exploited to solve occlusions by warping known object appearance into the occluded view. To this aim, this paper describes an approach to posture classi.cation based on projection histograms, reinforced by HMM for assuring temporal coherence of the posture. The single camera posture classification is then exploited in the multi-camera system to solve the cases in which the occlusions make the classi.cation impossible.

These above-mentioned problems are analyzed and solutions based on background suppression, appearance-based probabilistic tracking, and probabilistic reasoning for posture recognition are described. Experimental results with different amounts and types of occlusions are reported and demonstrate that the proposed approach is capable of correctly classifying the posture even in the case of large occlusions.

Paper presented at Measuring Behavior 2005 , 5^th International Conference on Methods and Techniques in Behavioral Research, 30 August - 2 September 2005, Wageningen, The Netherlands.

© 2005 Noldus Information Technology bv