摘要
本文结合目前电视广告检测系统的实际情况,基于感知哈希算法、视频特征、模板匹配等相关技术,设计并实现了一套智能电视广告监播系统。对比基于语音特征的识别系统,本文设计的系统有效提高了匹配速度和识别准确率,降低了人工工作量,提高了监管效率。
The detection system of TV AD plays an important role in promoting the healthy and orderly development of radio and television industry.According to the current situation of TV AD detection system,this paper designs and implements a set of smart TV AD monitoring system based on the related technologies such as perceptual hash algorithm,video feature and template matching.Compared with the recognition system based on audio features,the system proposes in this paper effectively improves the matching speed and recognition accuracy,and it reduces the manual workload and improves the supervision effciency.
作者
李厦
Li Sha(Academy of Broadcasting Planning,NRTA,Beijing 100866,China)
出处
《广播与电视技术》
2019年第6期137-141,共5页
Radio & TV Broadcast Engineering
关键词
广告监播系统
感知哈希算法
视频特征
图像匹配
AD monitoring system
Perceptual Hash algorithm
Video features
Image matching
