Television channel background recognition method

Abstract

The invention discloses a television channel background recognition method which includes that a television station caption is intercepted according to time when a user switches channels, an initial television station caption template is established, television station caption templates in a television station caption base are updated dynamically, image analysis is conducted on television station captions transmitted by a client, simultaneously television station caption recognition is conducted, the television station captions are compared with the television station caption templates, and the television station captions which succeed in comparison are fed back to a data base and used as accordance of a television watching behavior habit of a user. Due to the fact that the television station captions of switched channels are intercepted when the user switches the channels and are compared with the dynamically updated television station caption templates in the television station caption base, calculation time is greatly saved, effects of a server on overall performance are avoided, television channel recognition is quickened, matching rate of a single television station caption is high, television channel recognition accuracy is improved, and the television station captions are suitable for being used as the accordance of the television watching behavior habit of the user.

Description

technical field [0001] The invention relates to the field of methods for identifying TV channels watched by users in Internet TV or digital TV, in particular to a TV channel background recognition method with high recognition accuracy and fast recognition speed. Background technique [0002] With the development of Internet TV and digital TV, how to effectively collect and accurately analyze the behavior habits of users watching TV programs, so that the operation department can carry out some functions such as customized advertisement push, message reminder and dynamic news release, and evaluate the network Work on TV favorite shows and more. [0003] However, the existing TV channel identification method takes a lot of time and affects the overall performance of the server, and the matching rate of a single TV station logo is not very high, and the accuracy of TV channel identification is relatively low, which is difficult to collect and analyze. The basis for the user's b...

AI Technical Summary

Problems solved by technology

[0003] However, the existing TV channel identification method takes a lot of time and affects the overall performance of the server, and the matching rate of a single TV station logo is not very high, and the accuracy of TV channel identification is relatively low, which is difficult to collect and analyze. Basis for users’ habits of watching TV programs

[0004] Therefore, the existing technology still needs to be improved and developed

Images