The invention discloses an aAutomobile whistling classification method based on a support vector machine

Abstract

The invention provides an automobile whistling classification method based on a support vector machine. Aan air sonar sensor is used for receiving a sound signal; p; performing framing and windowing processing on the received signal; e; extracting five types of acoustic signal characteristics including sub-band energy, sub-band energy variance, frequency band total kurtosis, sub-band kurtosis meanvalue and short-time average zero-crossing rate, then constructing a decision function, repeating the training samples, dividing the training samples into whistling sounds and environmental noises through hyperplanes, and training an optimal hyperplane classifier; and finally, actual measurement signals are received, and an optimal hyperplane classifier is used for classifying the automobile whistling sound and the environmental noise. According to the invention, the sound signals of the automobile whistling on the road can be accurately classified and identified.

Description

technical field [0001] The invention belongs to the field of signal processing and relates to theories such as machine learning, sonar beam forming, and acoustic signal detection. Background technique [0002] Noise is one of the main pollution in modern cities, and it is also an important indicator for evaluating the environmental performance of automobiles. With the continuous and rapid development of China's economy and society, the number of motor vehicles continues to maintain a rapid growth trend. As of the end of 2017, the number of motor vehicles in the country reached 310 million, including 217 million cars. Vehicle noise has posed a hazard to the health and life of urban people, and it is imperative to control the level of vehicle noise. Traffic police departments across the country have invested a lot of energy to rectify the problem of illegal honking of cars, but the governance effect is not satisfactory. The reason is that the traffic police department inves...

AI Technical Summary

Problems solved by technology

Traffic police departments around the world have devoted a lot of energy to rectify the problem of illegal honking of cars, but the governance effect is not satisfactory

The reason is that the traffic police department investigates and punishes illegal whistle mainly by manual identification. It is more difficult to manually identify on noisy roads, and it is even more difficult to collect evidence. It is of great significance to use modern intelligent means to classify and identify illegal whistle vehicles

Images