| The television has become an integral part of social life. However due to the illegal singal interference, the safety of the television programs broadcast became the issues of concern. So looking for a reliable and efficient video surveillance is imperative.This thesis is focused on the extraction method of the logo and image matching algorithms. First analyze the design of the television monitoring as well as the television monitoring system program in domestic. Describe the simple method based on the difference of frame to extract logo,and logo detection method based on the brightness information, also describe the basic methods of image matching.This paper presents a new logo extraction method,that is Otsu method of global optimal threshold segmentation to extrac logo template. The premise of this method is due to the characteristic of the logo and background,the superposition of the frames will form a frame that distinguish the background and the logo obviously. The logo and the background distribute in the left and the right sides of a value,then use the Otsu method to extract logo template.Put the logo template to the template library,then use distance transformation of real-time video image to get a distance image in which the value of a pixel represents the distance to the object image.According to the information provided by the logo template..calculate the edge distance to distinguish whether the image is matching or not.The logo detection system proposed in this paper has a high matching efficiency,noise immunity and better robustness. |
PDF Full Text Request