Method for detecting audio exceptional event based on environmental model

Abstract

The invention relates to a method for detecting an audio exceptional event based on an environmental model, comprising the following operating steps of: (1) analyzing audio data in a monitoring environment by utilizing a hidden Markova model to obtain the level of similarity of the current audio segment corresponding to each audio type, thus a similarity vector is formed; (2) sorting on the basis of the similarity vector to obtain a type detection vector; and (3) firstly building an environmental model when no exceptional event happens by utilizing a neural network-based method and adopting an offline training mode, and then judging whether the current audio is an exceptional event or not in an online testing way according to a trained model.

Description

technical field [0001] The present invention relates to a method for detecting abnormal events based on audio information, to be precise, relates to a method for detecting abnormal audio events based on an environment model, which belongs to the information security technology of audio signal analysis and its application in the audio monitoring system of multimedia information field. Background technique [0002] With the continuous development of multimedia technology and the improvement of people's safety awareness, the monitoring system based on audio and video plays an increasingly important role in people's work and life. For example, in a smart home, people will deploy cameras to monitor for illegal entry; in a smart office, people will deploy cameras and audio sensors to detect the occurrence of abnormal events; in a smart child monitoring system, people will deploy audio sensor nodes , See if the baby in the ward is crying, etc. [0003] With the popularity of surv...

AI Technical Summary

Problems solved by technology

[0003] With the popularization of surveillance systems, more and more deficiencies have been exposed in surveillance systems that rely solely on video

[0004] First, video surveillance systems are easily affected by light

In outdoor monitoring, the detection effect of the camera will be greatly reduced when it is cloudy; in indoor monitoring, the camera cannot work normally at night and other conditions that do not require lighting

[0005] Second, occlusions are prone to occur in video surveillance systems. For example, in traffic monitoring, the car we need to monitor is usually blocked by other vehicles, which affects the monitoring effect

In many cases, people's sorting out of abnormal events is not comprehensive and specific enough

Images