Analysis of Jumping Crowd on Stadium Stands Through Image Processing to Security Purposes

The goal of this paper is to introduce an innovative and quantitative measurement technique, useful for the estimation of the people motion on stadia stands, at least under a statistical point of view. The authors propose and qualify here image based people motion estimation: the idea is to measure the "motion level" of the people through the analysis of images grabbed with common video cameras or with an infrared thermal imager. Cameras working in the visible wavelength range usually have a higher resolution than thermal cameras and are much cheaper, due to this reasons they are often preferred for the estimation of people motion in case of events held during the day (i.e. with stable lighting conditions); nevertheless thermal cameras have various important advantages in some applications, as during concerts; in these cases, in fact, lighting is mainly projected on artists, leaving the audience in an almost completely dark atmosphere. In this scenario the use of a common camera often leads to poor quality images; for this reason in this research a thermal camera has also been used, in order to acquire sequences of infrared images of the jumping crowd. In this way more reliable results can be obtained even in case of very low lighting