Traffic road sound monitoring and abnormal sound recognition system

Abstract

The invention discloses a traffic road sound monitoring and abnormal sound recognition system. The traffic road sound monitoring and abnormal sound recognition system comprises sound acquisition endsand a server end which are connected through a network; each sound acquisition end comprises a sound pickup, a sound card, a GPS positioning module, a data processing module and a wireless communication module; and the sound pickups and the data processing modules can be arranged and installed on designated road sections according to the requirements; traffic road sounds are monitored in real time, and when an abnormal sound is detected, the abnormal sound data is transmitted to the server end for further recognition, thereby greatly reducing the data transmission amount; various eigenvalues of an abnormal sound signal are extracted, and recognition and classification are performed by using a neural network; when recognition and classification are performed on the abnormal sound data, a deep convolutional neural network (CNN) is used and is very suitable for sound recognition and classification, and the network can greatly improve the training efficiency and the recognition accuracy, and efficiently recognize abnormal sounds generated in traffic roads.

Description

technical field [0001] The invention relates to the technical field of sound monitoring and recognition, in particular to a traffic road sound monitoring and abnormal sound recognition system. Background technique [0002] Abnormal sounds refer to sounds that should not be produced in a normal environment. Abnormal sounds in public places usually include explosions, collisions, screams, gunshots, etc. The abnormal sound on the traffic road can reflect the occurrence of traffic accidents and emergencies. By monitoring the sound of the traffic road, you can understand the traffic conditions of a certain road. When an abnormal situation occurs, you can analyze the abnormal sound by identifying the abnormal sound. the nature of the exception. In addition, road sections that are prone to traffic accidents or traffic jams can be judged based on the monitoring situation. [0003] The road monitoring system is an important part of intelligent transportation. In the existing traffi...

AI Technical Summary

Problems solved by technology

The manual active recognition method is cumbersome and time-consuming, and the image processing recognition method requires a large amount of calculation and complex calculation algorithms

Images