Visual classifier construction method with consideration of characteristic reliability

Abstract

The invention provides a visual classifier construction method with the consideration of characteristic reliability. The method includes the steps that video characteristics of each video sample in a video sample set are extracted to obtain a video characteristic set; a label is endowed to each video sample to indicate that the video sample belongs to a first category or a second category; reliability assessment is carried out on each video sample to obtain a reliable factor of each video sample; based on the video characteristic set, the label of each video sample and the reliable factor of each video sample, a weighted support vector machine algorithm is utilized to obtain a visual classifier. The visual classifier construction method can be applied to services of internet harmful video filtration, video supervision and the like to maintain safety of contents of the internet.

Description

technical field [0001] The invention relates to the field of computer application technology, in particular to a method for constructing a video classifier considering feature reliability. Background technique [0002] With the rapid development of Internet technology, various multimedia applications are constantly emerging, and digital libraries, distance education, video on demand, digital video broadcasting, interactive TV, etc. all generate and use a large amount of multimedia data. Even without leaving home, people can learn knowledge, consult information, and enjoy various entertainment activities through the Internet. However, in addition to useful information for people's work, study and life, due to the openness of the Internet, many harmful information are also disseminated through the Internet. Harmful information on the Internet has had a serious impact on society, especially the adverse impact on minors is often seen in newspapers. The harm caused by bad infor...

AI Technical Summary

Problems solved by technology

However, online video data is usually more complex. From the three modalities of text, vision and audio, some videos have rich text around them, while others have little; some videos have high visual quality, while others have low ;Some videos have a very clear audio signal, others are very noisy

Features extracted from poor quality modalities usually do not truly reflect the characteristics of the video due to their low reliability

The current recognition methods based on multimodal feature fusion do not consider the issue of feature reliability, resulting in the inability to achieve accurate and reliable video recognition and classification

Images