Target detecting and identifying method and target detecting and identifying system based on sound vibration signal

Abstract

The invention discloses a target detecting and identifying method and a target detecting and identifying system based on a sound vibration signal. The method comprises the steps of acquiring a vibration signal at a preset node position in a preset time period and storing the vibration signal into a vibration data matrix by means of M fiber sensors or equivalent M fiber sensors; performing preprocessing on the vibration data matrix by means of a pre-emphasis method, and selecting a bipolar data frame which comprises a target signal; extracting a speed characteristic, an energy distribution characteristic and a spectrum estimation characteristic of each row of each kind of signal from the bipolar data frame, acquiring a relatively long characteristic vector matrix according to each characteristic and a corresponding weighted value and performing dimension reduction on the relatively long characteristic vector matrix; and identifying the vibration signal kind of the bipolar signal data, the bipolar signal data frame and the relatively long characteristic vector after dimension reduction by means of a long-and-short-period selection memory neural network model. According to the target detecting and identifying method and the target detecting and identifying system, sound vibration signal identification is determined according to combined action of information in different steps, thereby improving target identification accuracy.

Description

technical field [0001] The invention relates to the field of audio technology, in particular to a target detection and recognition method and system based on acoustic vibration signals. Background technique [0002] The security system is a system that uses security products and other related products to form an intrusion alarm system, a video security monitoring system, an entrance and exit control system, and an explosion-proof security inspection system for the purpose of maintaining social public safety. At present, security systems usually use manual, video, and infrared monitoring methods. Due to high maintenance costs, low environmental sensitivity, and detection accuracy and range limited by hardware equipment, there are certain limitations in the scope of use of existing security systems. Therefore, how to further improve the reliability and robustness of the security system and reduce the monitoring cost has become one of the key issues to be studied. [0003] Whe...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the defects in the prior art, the present invention provides a target detection and recognition method and system based on acoustic vibration signals to solve the vibration target recognition system in the prior art when the actual environment is complex and changeable and the target types are various. About 80% of the recognition accuracy cannot meet the actual needs

Images