Figure 4 - uploaded by Nirosha Priyadarshani
Content may be subject to copyright.
Source publication
According to the International Union for the Conservation of Nature Red Data List nearly a quarter of the world's bird species are either threatened or at risk of extinction. To be able to protect endangered species, we need accurate survey methods that reliably estimate numbers and hence population trends. Acoustic monitoring is the most commonly-...
Citations
... Unlike Fourier transforms commonly used in recognition applications, discrete wavelet transforms (DWT), particularly the Haar wavelet transform, provide an efficient acoustic signal analysis with significantly reduced mathematical operations required for feature extraction [15]. Although wavelet transforms, in computer-based animal sound recognition, have been commonly studied [2,10,[16][17][18], very few research work has addressed the application of DWT's features for acoustic animal recognition in WASN; see, for example, [10]. Despite the interesting results shown in this work [10], this approach requires applying complex algorithms for the classification and features optimization tasks, which cannot be executed at low-resources motes. ...
... Features derived from wavelet were adopted in several applications for the purpose of recognition in [2,10,[16][17][18]. The proposed approach in [17] had achieved 78% and 96% using SVM and MLP, respectively, for bird sounds classification based on wavelet packet decomposition (WPD) transform. ...
... The proposed approach in [17] had achieved 78% and 96% using SVM and MLP, respectively, for bird sounds classification based on wavelet packet decomposition (WPD) transform. Similarly, a recent efficient WPD-based approach for bird sounds recognition and denoising was proposed in [18]. In [2], WPD-based features were used to classify frog sounds. ...
Wireless acoustic sensor networks represent an attractive solution that can be deployed for animal detection and recognition in a monitored area. A typical configuration for this application would be to transmit the whole acquired audio signal through multi-hop communication to a remote server for recognition. However, continuous data streaming can cause a severe decline in the energy of the sensors, which consequently reduces the network lifetime and questions the viability of the application. An efficient solution to reduce the sensor's radio activity would be to perform the recognition task at the source sensor then to communicate the result to the remote server. This approach is intended to save the energy of the acoustic source sensor and to unload the network from carrying, probably, useless data. However, the validity of this solution depends on the energy efficiency of performing on-sensor detection of a new acoustic event and accurate recognition. In this context, this paper proposes a new scheme for on-sensor energy-efficient acoustic animal recognition based on low-complexity methods for feature extraction using the Haar wavelet transform. This scheme achieves more than 86% in recognition accuracy while saving 71.59% of the sensor energy compared with the transmission of the raw signal.
... Joshi et al. (2017) showed that such software is cost-effective in some situations but not others. Automatic detection of bittern calls has so far not been found to be cost-effective, either because expensive recording equipment was needed (Frommolt & Tauchert, 2014) or because it gave high false positive rates (21% precision rate; Priyadarshani, 2017). However, work on this is ongoing. ...
The inferences that can be made from any study are limited by the quality of the sampling design. By bad luck, when monitoring species that are difficult to detect (cryptic), sampling designs become dictated by what is feasible rather than what is desired. We calibrated and conducted a cost‐benefit analysis of four acoustic recorder options that were being considered as potential solutions to several sampling restrictions experienced while monitoring the Australasian bittern, a cryptic wetland bird. Such sampling restrictions are commonly experienced while monitoring many different endangered species, particularly those that are cryptic. The recorder options included mono and stereo devices, with two sound file processing options (visual and audible analysis). Recording devices provided call‐count data similar to those collected by field observers but at a fraction of the cost, which meant that “idealistic” sampling regimes, previously thought to be too expensive, became feasible for bitterns. Our study is one of the few to assess the monetary value of recording devices in the context of data quality, allowing trade‐offs (and potential solutions) commonly experienced while monitoring cryptic endangered species to be shown and compared more clearly. The ability to overcome challenges of monitoring cryptic species in this way increases research possibilities for data deficient species and is applicable to any species with similar monitoring challenges.
... In recent work Priyadarshani et al. (2016) and Priyadarshani (2017) have shown that wavelets can be effectively used to remove noise from field recordings collected with automatic sound recorders. In a field recording, while the birdsong is transient, a considerable amount of background noise, particularly the geophony, is nearly stationary. ...
Conservationists are increasingly using autonomous acoustic recorders to determine the presence/absence and the abundance of bird species. Unlike humans, these recorders can be left in the field for extensive periods of time in any habitat. Although data acquisition is automated, manual processing of recordings is labour intensive, tedious, and prone to bias due to observer variations. Hence automated birdsong recognition is an efficient alternative.
However, only few ecologists and conservationists utilise the existing birdsong recognisers to process unattended field recordings because the software calibration time is exceptionally high and requires considerable knowledge in signal processing and underlying systems, making the tools less user‐friendly. Even allowing for these difficulties, getting accurate results is exceedingly hard. In this review we examine the state‐of‐the‐art, summarising and discussing the methods currently available for each of the essential parts of a birdsong recogniser, and also available software. The key reasons behind poor automated recognition are that field recordings are very noisy, calls from birds that are a long way from the recorder can be faint or corrupted, and there are overlapping calls from many different birds. In addition, there can be large numbers of different species calling in one recording, and therefore the method has to scale to large numbers of species, or at least avoid misclassifying another species as one of particular interest. We found that these areas of importance, particularly the question of noise reduction, are amongst the least researched. In cases where accurate recognition of individual species is essential, such as in conservation work, we suggest that specialised (species‐specific) methods of passive acoustic monitoring are required. We also believe that it is important that comparable measures, and datasets, are used to enable methods to be compared.
Modern data analysis techniques include the use of artificial neural networks for classification, estimation, and prediction. An area in which data science can be helpful is in species identification and enumeration from images or sound recordings. In this chapter, we develop the necessary tools for using machine learning to identify bird species from recordings of bird calls. This includes a basic introduction to wavelet transforms and scalograms and the construction of convolutional neural networks for solving classification problems. We give some ideas for extending our results on bird species classification, as well as ideas for using related neural networks for broader application to image and audio classification.
