Specific bird species oriented acoustic automated detection method

Abstract

The invention discloses a specific bird species oriented acoustic automated detection method. The method comprises the following steps: extracting potential song fragments of specific bird species based on an acoustic event detection process of a Gaussian mixture model, and completing robust detection and automatic segmentation by combining post processing procedures of energy, singing duration and frequency distribution based on a candidate acoustic event; performing adaptive signal noise reduction on various sound fragments, extracting corresponding Mel-frequency cepstral coefficient featureparameters from the enhanced sound fragments, and acquiring a potential song fragment feature set; and finally, completing acoustic detection of the specific bird species by combining the potential song fragment feature set and a support vector machine. The realization process of the method disclosed by the invention is convenient. In addition, with the adoption of adaptive signal enhancement processing based on a microphone array, the signal to noise ratio of the acoustic event can be obviously improved, so that the specific species identification accuracy is improved, the acoustic detectionof the specific bird species in a field natural environment is realized. The method has significances on ecological protection of wild rare bird species.

Description

technical field [0001] The invention belongs to the field of ecological monitoring and acoustic signal technology identification, in particular to an acoustic automatic detection method for specific bird species. Background technique [0002] Birds are a kind of vocalizing animals with a wide distribution. Compared with other animal communities, the living habits of birds are easier to be observed, and they can sensitively perceive small changes in the ecological environment. Therefore, birds are considered by most ecologists It is an ideal species for monitoring changes in the ecological environment. [0003] However, in recent years, due to the continuous expansion of human urbanization and the continuous destruction of the ecological environment by human activities, the number and species of birds have been reduced on a large scale. As an indicator species of woodland vegetation community, the sharp decline of birds will lead to the imbalance of ecological environment. ...

AI Technical Summary

Problems solved by technology

[0005] Most of the above-mentioned existing studies only added one kind of noise or did not suppress the noise in the experiment. However, the field acoustic environment is very complex, with a variety of environmental interference sources, and only single-channel audio enhancement front-end processing cannot meet the actual complex acoustics in the field. Requirements for acoustic monitoring tasks in environments

Images