Sound-spectrogram based automatic bird species recognition using MLP classifier

Pahuja, Roop; Kumar, Avijeet

doi:10.1016/j.apacoust.2021.108077

Cited by 40 publications

(21 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By treating spectrograms as images, previous studies have applied image processing techniques to extract bird vocalizations [10,59,60,83,[96][97][98], such as widely used median clipping [41,99,100] and frame-or acoustic event-based morphological filtering [66]. There are plenty of toolboxes available to extract acoustic traits [22], such as central frequency, highest frequency, lowest frequency, initial frequency, and loudest frequency and so on [10,59], which are basically time-frequency characteristics only. To the best knowledge, all these studies aimed to identify or classify one or several bird species specifically.…”

Section: Discussionmentioning

confidence: 99%

“…We developed an automated bird vocalization extraction approach based on OBIA, which followed a typical analysis workflow of bird vocalizations with three main steps [59]: preprocessing, automated extraction, and feature calculation. In line with this Recordings were obtained using Zoom H5 acoustic sensors (Zoom Inc., Tokyo, Japan, System 2.40) with XYH-5 X/Y microphones.…”

Section: Methodsmentioning

confidence: 99%

“…We developed an automated bird vocalization extraction approach based on OBIA, which followed a typical analysis workflow of bird vocalizations with three main steps [59]: preprocessing, automated extraction, and feature calculation. In line with this workflow, the processing details of each step of our approach are described in the following subsections.…”

Section: Methodsmentioning

confidence: 99%

“…Bird vocalizations of songs or calls are complex, non-stationary signals with a great degree of variation in intensity, pitch, and syllable patterns [71]. One of the proven methods for joint time-frequency domain analysis of non-stationary sound signals is STFT [59]. The STFT spectrogram is a two-dimensional convolution of the signal and window function [72]: the X-axis represents time, the Y-axis represents frequency, and the amplitude of a particular frequency at a particular time is represented by its color in the image [73].…”

Section: Short-time Fourier Transformation (Stft)-spectrogrammentioning

confidence: 99%

“…Considering their good performance with respect to feature analysis of syllables in previous studies, instantaneous frequency (IF) [59], and amplitude [84] were adopted in this work. Each parameter was further statistically analyzed to obtain more detailed descriptive values such as maximum (IF_MAX), minimum (IF_MI), mean, and central frequency.…”

Section: Feature Representation Of Extracted Syllablesmentioning

confidence: 99%

See 4 more Smart Citations

Diversity Monitoring of Coexisting Birds in Urban Forests by Integrating Spectrograms and Object-Based Image Analysis

Zhao

Yan

Jin

et al. 2022

Forests

View full text Add to dashboard Cite

In the context of rapid urbanization, urban foresters are actively seeking management monitoring programs that address the challenges of urban biodiversity loss. Passive acoustic monitoring (PAM) has attracted attention because it allows for the collection of data passively, objectively, and continuously across large areas and for extended periods. However, it continues to be a difficult subject due to the massive amount of information that audio recordings contain. Most existing automated analysis methods have limitations in their application in urban areas, with unclear ecological relevance and efficacy. To better support urban forest biodiversity monitoring, we present a novel methodology for automatically extracting bird vocalizations from spectrograms of field audio recordings, integrating object-based classification. We applied this approach to acoustic data from an urban forest in Beijing and achieved an accuracy of 93.55% (±4.78%) in vocalization recognition while requiring less than ⅛ of the time needed for traditional inspection. The difference in efficiency would become more significant as the data size increases because object-based classification allows for batch processing of spectrograms. Using the extracted vocalizations, a series of acoustic and morphological features of bird-vocalization syllables (syllable feature metrics, SFMs) could be calculated to better quantify acoustic events and describe the soundscape. A significant correlation between the SFMs and biodiversity indices was found, with 57% of the variance in species richness, 41% in Shannon’s diversity index and 38% in Simpson’s diversity index being explained by SFMs. Therefore, our proposed method provides an effective complementary tool to existing automated methods for long-term urban forest biodiversity monitoring and conservation.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

“…We developed an automated bird vocalization extraction approach based on OBIA, which followed a typical analysis workflow of bird vocalizations with three main steps [59]: preprocessing, automated extraction, and feature calculation. In line with this workflow, the processing details of each step of our approach are described in the following subsections.…”

Section: Methodsmentioning

confidence: 99%