Application and Implementation of Improve Image Processing and Feature Recognition Applied to Full-Field Measurements

The understanding of crowd behavior in semi-confined spaces, it is important to design a new pedestrian facility, for major layout modifications in daily management sites that are subjected to crowd traffic. Conventional manual measurement techniques are not suitable for data collection of patterns of site occupation and movement. Real-time monitoring is tedious and tiring, but safety-critical. Also sometimes the probability of this measure will yield a false alarm and efficient methods for estimating this probability at run time. This paper presents some image processing techniques which, using existing closed-circuit television systems can support both data collection and on-line monitoring of crowds. The paper describes techniques to perform efficient and accurate crowd recognition in difficult domains. In order to accurately model small, irregularly shaped targets; the crowd objects and image are represented by their edge maps, with a local orientation associated with each edge pixel. Three-dimensional objects are modeled by a set of two-dimensional (2-D) views of the object by Translation, rotation, and scaling of the full three-dimensional (3-D) motion of the object. And this information can be used to maintain a low false alarm rate or to rank competing hypotheses based on their likelihood of being a false alarm. The application of these methods could lead to a better understanding of crowd behavior, improved design of the built environment and increased pedestrian safety.