| Machine vision is a non-contact automatic identification technology,it is a kind of technology that computer can do judgment and measurement for our eyes through image processing and its related technical.Converting Intake target to image signals,then transmitting it to special image processing system,according to the pixel distribution and brightness,color feature and so on,changing into digital information signal.The image system do all kinds of operations for these signals to extracting features of our target,according to the results we can control the equipment action of our site.with the rapid development of information technology,digital image processing techonology has been widely used in our world.In order to meet needs of attendance statistics of public places such as classrooms, halls and others, the paper describes two kinds of statistical methods based on the machine vision. There are two ways to make the statistics based on the number of machine vision,the one is monitoring the installation of cameras at the entrance,and according to tracking the targets'feature to determine the direction,which is based on the number of entrances and exits out; the other is making a general statistics directly from the situations of the public. at first,this paper made the appropriate background and significance of the corresponding introduction of statistics on the number of machine vision to achieve the feasibility study.and it designed the whole of the systems, primarily making the design of software systems.and then,this paper described two statistical methods separately.ing Firstly, target tracking based on OpenCV we can calculate the number statistics, through the camera installed at the entrance, we can calculate the number statistics According to the export orientation and the target size.the other way is to calculating general statistics on the whole Occasions,through the error compensation we can also achieve a high accuracy. Finally, we make the total analysis and outlooks. |
PDF Full Text Request