Research And Application Of Computer Vision Technology

Comparing with Computer Vision,human eyes are belong to highly advanced biological vision system.As the visual signals are most predominant and most direct resource from which we obtain the information of surrounding environment,further more,in reality,there are more than80%of the sensorial data come from vision.Therefore,it is apparently easy for us to perceive the three dimensional structure’s information of the world around,meanwhile,as the eyes sense data,human brain rejects a great deal of redundant and irrelevant information and based on the regions of interests,only response to those we really care.Computer Vision is a subject that aims at human eyes system,combines with several other subjects and fields. It expects that computers with exceptional computational power and feature of error-proof to replace human eyes to take contorl of some specific jobs.These jobs tend to be impossible for naked eyes because of their working environments,such as:exploration and investigation during high pressure and temperature situations,doing some scientific experiments in some areas which is human unreachable,fulfilling complicated computation,boring and repetitive tasks.Just because of these requirements,Computer Vision does not only establish its importance and status,but also achieves a high-speed development.Video events dectection systems based on Computer Vision have been widely used in imporant fileds:media-health,industrial production, aeronautics and astronautics,military and civil service bussines.This article compares several mature implement methods and according to the hardware development fashion and the requirement of large-scale promotion during practical projects.Above these,a novel scheme was come up with,which is based on Computer Vision,and Digital Image Processing is implemented by high-performance GPU calculation. At last, using statistical method to fulfill analysis of gait. This scheme does not merely resolve the analysizing,processing and result-extraction of continuous video flow,but also prosess characteristics of flexible,efficient,low energy consumption and reliable.The process of implementation is guided by the principles of software engineering and take advantage of the multi-modules system.As a result,it can handle lots of variable factors during reality envirionment and has a great realistic meaning.