Station logo recognition method

Abstract

The present invention relates to a station logo recognition method. The method comprises the processes of station logo extraction and station logo recognition. The station logo extraction process is conducted mainly based on the probability statistics characteristics of a video, wherein a clean and complete station logo image is extracted in combination with the space-time invariance of the station logo region. The station logo recognition process mainly aims at the extracted station logo image. During the station logo recognition process, firstly, SIFT features are calculated and the coarse delete matching is conducted based on the SIFT features. Secondly, the affine transformation and correction is conducted based on SIFT feature points. Thirdly, in combination with the characteristics of the station logo, the color of the station logo and the spatial distribution features of the station logo, the matching is further conducted. Finally, the template matching is conducted according to the pattern of the station logo, the channel name of the station logo and the channel number of the station logo. The above structure-based station logo recognition method solves the problems in the prior art that: the images of existing translucent station logos can not be effectively extracted; the backgrounds in the extracted images of station logos are not clear and are provided with noise points; similar local station logos cannot be effectively distinguished and identified and the like. Meanwhile, the method can be used for processing station logos that deform at a certain degree, and enables the real-time input of station logo templates.

Description

technical field [0001] The invention relates to the field of computer software, in particular to the technical field of image recognition, and specifically refers to a method for recognizing a station logo. Background technique [0002] The station logo is a special logo used to represent a TV station or a TV channel. It is the condensation and concentrated expression of the media image, and the agglomeration and carrier of media goals, media philosophy, and media spirit. It is also one of the important semantic sources for video analysis and retrieval. [0003] With the rapid development of TV technology, at present, the entire TV system in our country has become more and more large and complex, and there are hundreds of TV programs at all levels. However, some illegal TV signals have never given up trying to infiltrate into normal TV channels. , it can be seen that it is very necessary to monitor the video signal. How to effectively monitor the video signal, the station ...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the above-mentioned shortcoming of the prior art, provide a kind of solution that can't effectively extract the station logo image of translucent station logo, the background of the station logo image that extracts is not clean, exists noise point, and its similar place The Taiwan logo recognition method for problems such as the station logo cannot be effectively distinguished and recognized

Images