Sound localization-based video monitoring system

Abstract

The invention provides a sound localization-based video monitoring system. In the system, two or more microphones which face towards different directions in a plane or space serve as sound acquisition devices for acquiring sound; relative positions of the respective microphones in the acquisition devices are fixed, and are distinguished by labels; the spatial position of a sound source can be solely confirmed according to the time difference of acquiring the sound of different microphones and the difference of sound intensities; and a camera can be controlled to rotate to find a target and to track and photograph the target according to the position information. As the temperature has an influence on the propagation velocity of the sound, a correction can be provided for the system according to an acquired temperature value, so that the accuracy of the system is improved.

Description

Technical field [0001] The present invention relates to the fields of various monitoring, video recording and image collection such as security protection by means of video and images. Background technique [0002] At present, the known video surveillance is composed of an image acquisition part, a storage part and a transmission part. There are mainly two methods for the image acquisition part: one is to use a fixed camera; the other is to use a uniform rotation scanning method. There are also a small number of image processing methods to capture dynamic targets to achieve tracking and monitoring. These methods have their own disadvantages, wherein the first two methods can not achieve the dynamic tracking of the target; although the latter can achieve the tracking of the target, but because it cannot be monitored in all directions at the same time, there are two problems: one is: whether it can be The problem of finding the target at the beginning; the second is that the...

AI Technical Summary

Problems solved by technology

These methods have their own disadvantages, wherein the first two methods can not achieve the dynamic tracking of the target; although the latter can achieve the tracking of the target, but because it cannot be monitored in all directions at the same time, there are two problems: one is: whether it can be The problem of finding the target at the beginning; the second is that there is an image processing method to track the target, which requires high data processing equipment and increases the cost

Images