A block-based model for monitoring of human activity

Abstract

The study of human activity is applicable to a large number of science and technology fields, such as surveillance, biomechanics or sports applications. This article presents BB6-HM, a block-based human model for real-time monitoring of a large number of visual events and states related to human activity analysis, which can be used as components of a library to describe more complex activities in such important areas as surveillance, for example, luggage at airports, clients’ behaviour in banks and patients in hospitals. BB6-HM is inspired by the proportionality rules commonly used in Visual Arts, i.e., for dividing the human silhouette into six rectangles of the same height. The major advantage of this proposal is that analysis of the human can be easily broken down into regions, so that we can obtain information of activities. The computational load is very low, so it is possible to define a very fast implementation. Finally, this model has been applied to build classifiers for the detection of primitive events and visual attributes using heuristic rules and machine learning techniques.

Introduction

Nowadays it is very usual for public and private companies to have sophisticated surveillance systems that try to monitor the state of their “business” in order to detect anomalous situations and avoid adverse or undesirable situations. The final aim of an artificial vision system is to provide a task-focused scene description. Concentrating on the description of human activity, and in accordance with [25], [23], [14], the recognition of activities is considered as a problem of classifying spatial–temporal characteristics or, alternatively, as “queries” on higher-level constructions. In other words, activities are considered as complex events, which are defined by spatial–temporal composition of simpler events, and these simple events in turn are defined by other even simpler events and so on to form a hierarchy of events to link with primitive events, which are determined from changes in state of visual attributes.

This article presents BB6-HM, a model that can be used to monitor, in real time and with a minimum computational load, a large number of these primitive events related to humans, which can be used as components of a library to describe more complex activities in many surveillance tasks (luggage at airports, clients’ behaviour in banks, patients in hospitals, etc.). Finally, this model has been applied to build classifiers for the detection of primitive events and visual attributes using heuristic rules and machine learning techniques.

This article is organised as follows. The second section analyses other works related to the proposal. The third section describes the block-based human model. The fourth section describes some events and visual attributes detected using relations between the parameters of the human model and evaluates on some test sequences. Finally, the fifth section presents the conclusions and highlights future works.