Automated detection of broadband clicks of freshwater fish using spectro-temporal features

Abstract

Large scale networks of embedded wireless sensor nodes can passively capture sound for species detection. However, the acoustic recordings result in large amounts of data requiring in-network classification for such systems to be feasible. The current state of the art in the area of in-network bioacoustics classification targets narrowband or long-duration signals, which render it unsuitable for detecting species that emit impulsive broadband signals. In this study, impulsive broadband signals were classified using a small set of spectral and temporal features to aid in their automatic detection and classification. A prototype system is presented along with an experimental evaluation of automated classification methods. The sound used was recorded from a freshwater invasive fish in Australia, the spotted tilapia (Tilapia mariae). Results show a high degree of accuracy after evaluating the proposed detection and classification method for T. mariae sounds and comparing its performance against the state of the art. Moreover, performance slightly improves when the original signal was down-sampled from 44.1 to 16 kHz. This indicates that the proposed method is well-suited for detection and classification on embedded devices, which can be deployed to implement a large scale wireless sensor network for automated species detection.

Passive acoustic monitoring as a potential tool to survey animal and ecosystem processes in freshwater environments

“…These methods are essential for the development of real time monitoring (Aide et al., ). Freshwater applications include automatic detection of an invasive fish (Kottege et al., ), detection of spawning events (Straight et al., ) and monitoring of endangered species (Dutilleux & Curé, ).…”

“…This method consists of recording and analysing the sounds emanating from an environment, in either air or water, to extract information about presence of particular species (e.g. Kottege, Jurdak, Kroon, & Jones, ; Ulloa et al., ) or about ecosystem conditions and dynamics (Fuller, Axel, Tucker, & Gage, ; Pieretti, Farina, & Morri, ; Sueur, Pavoine, Hamerlynck, & Duvail, ). Passive acoustic monitoring can be very accurate and cost‐effective for various applications in ecology such as detecting rare, invasive, or threatened species (Campos‐Cerqueira & Aide, ; Gasc, Anso, Sueur, Jourdan, & Desutter‐Grandcolas, ), contrasting differences in structure and condition of ecosystems (Duarte et al., ; Gasc, Sueur, Pavoine, Pellens, & Grandcolas, ), and continuously monitoring environments (Aide et al., ).…”

Freshwater ecoacoustics as a tool for continuous ecosystem monitoring

“… Species presence or absence: for instance, acoustic monitoring has been used in freshwater systems to detect invasive tilapia (Kottege et al . ); Abundance: to our knowledge, acoustic monitoring has not yet been used in freshwater systems to measure the population abundance of a single species or group of species, but it has been used extensively in marine surveys (eg long‐term monitoring of the endangered vaquita porpoise [ Phocoena sinus ]; Gerrodette et al . ); Habitat use (eg for feeding or nesting): for example, sediment disturbance by fish during spawning (eg river redhorse, Moxostoma carinatum ; Straight et al .…”

“…), only a single study has attempted to automatically detect calls of a freshwater fish species (that of Kottege et al . , who monitored spotted tilapia [ Pelmatolapia mariae ], an invasive fish in Australia, by means of their calls). Automated detection algorithms typically require some “training data” of known calls of the desired type to match against, but this presents a major challenge in many freshwater environments, because little is currently known about the structure and variety of sounds made by target biota in most aquatic systems.…”

“…• Species presence or absence: for instance, acoustic monitoring has been used in freshwater systems to detect invasive tilapia (Kottege et al 2015 Amorim et al [2015] for an overview) and were recently used to estimate fish abundance in a marine fish species, the Gulf corvina (Cynoscion othonopterus) (Rowell et al 2017); and • Movement: acoustic monitoring has been used in marine systems to characterize spatiotemporal variation in Atlantic cod (Gadus morhua) via sounds associated with spawning (Hernandez et al 2013).…”