Scene analysis system based on deep learning technology and method thereof

Abstract

The invention discloses a scene analysis system and method based on a deep learning technology, and method thereof. The system comprises a data acquisition subsystem and a cloud AI platform. The data acquisition subsystem acquires images and voices; a face recognition module in the cloud AI platform carries out face recognition on the to-be-detected image according to a deep learning technology; the facial expression analysis module is used for analyzing and judging expressions on the face; the voice recognition module carries out voice recognition on the to-be-detected audio according to a deep learning technology; the voice analysis module analyzes and judges semanteme and intonation of the to-be-tested audio; and the comprehensive analysis module is used for comprehensively analyzing results obtained by the facial expression analysis module and the voice analysis module. According to the method, face and voice recognition can be met at the same time, recognition results of face expressions, voice semantics and intonations are obtained according to the deep learning technology, the recognition results are more accurate, the recognition speed is guaranteed, and the scene analysis technology is further enriched.

Description

technical field [0001] The present invention relates to the technical field of deep learning, and more specifically relates to a scene analysis system and method based on deep learning technology. Background technique [0002] With the continuous advancement of modern science and technology, the era of intelligence has come, and natural language processing and facial expression recognition technology have already become important research topics for those skilled in the art. [0003] However, on the one hand, due to the limitations of traditional shallow models, traditional natural language processing models need to use a large amount of linguistic knowledge to manually construct features, and these features are usually guided by specific applications, so they are not specific and widely applicable. However, if the specific task changes, new features must be constructed manually; [0004] On the other hand, the current face recognition technology is mainly implemented based...

AI Technical Summary

Problems solved by technology

[0003] However, on the one hand, due to the limitations of traditional shallow models, traditional natural language processing models need to use a large amount of linguistic knowledge to manually construct features, and these features are usually guided by specific applications, so they are not specific and widely applicable. If the specific task changes, new features must be constructed manually;

[0004] On the other hand, the current face recognition technology is mainly implemented based on hand-designed feature extraction algorithms. In the actual complex environment, face data is often affected by various factors, such as illumination, occlusion, posture, etc. In this case, the existing face recognition method based on hand-designed feature extraction algorithm has poor robustness and poor anti-interference ability to the above-mentioned influencing factors, and these uncontrollable factors make the face recognition method based on the current The face recognition performance of methods has dropped sharply, it is difficult to guarantee the effect of face recognition, and there is a problem of low accuracy of face recognition

[0005] Moreover, people are exploring the application of image recognition, speech recognition, and semantic analysis in different fields, but there are still few applications that combine natural language processing, face recognition, and facial expression recognition in scene analysis, which are still in the development stage and cannot be implemented. accurate identification

Images