Off-line embedded abnormal sound detection system and method

Abstract

The invention provides an off-line embedded abnormal sound detection system. The off-line embedded abnormal sound detection system comprises a sound acquisition module, a sound audio feature extraction module and a neural network module. The sound audio feature extraction module processes sampling data obtained by the sound acquisition module through a digital microphone in a frequency domain by using fast Fourier transform, and inputs the sampling data to the neural network module to complete anomaly classification. The neural network module comprises a CNN feature extraction layer, an LSTM long-term and short-term memory layer, a full connection and classification layer and a trigger judgment layer. The number of network layers of the CNN feature extraction layer is dynamically adjustable, the network structure of the full connection and classification layer is dynamically variable, and a trigger decision layer is used for eliminating generalization errors generated by the neural network. The invention further comprises a method for carrying out anomaly detection by utilizing the off-line embedded abnormal sound detection system. The method works in an off-line environment, has less dependence on a network, is high in performance and reliable in work, and can adapt to a changing abnormal diagnosis working environment.

Description

technical field [0001] The invention relates to the field of embedded devices, in particular to an off-line embedded abnormal sound detection system and method. Background technique [0002] Sound is a convenient, efficient and fast way of conveying information. At present, abnormal detection of vehicle operation, fault detection of mechanical equipment such as compressors and motors, detection of abnormal room sounds, detection of children crying, etc. are mainly based on human judgment and rely too much on human subjective experience, which leads to large errors in locating these abnormalities. And the cost of consumption is high. [0003] In recent years, there have been some abnormal sound detection methods based on deep learning, which can show good practical application effects, but there are also some defects: [0004] 1. The system is complex, with a large amount of calculation, relying on complex calculation units and even GPU (Graphic Process Unit, video processi...

AI Technical Summary

Problems solved by technology

[0008] When the network transmits real-time audio, it is extremely easy to drop frames. When the audio stream drops frames, its spectral characteristics may change accordingly, resulting in failure of fault detection.

Images