Enhancing Acoustic Data Mobilization: A User-Friendly Data Platform for the Asian Soundscape Monitoring Network

Biodiversity Information Science and Standards, Journal Year: 2024, Volume and Issue: 8

Published: Oct. 8, 2024

Passive Acoustic Monitoring (PAM) has emerged as a crucial tool in monitoring efforts to track environmental changes and evaluate conservation measures response the biodiversity crisis (Sugai et al. 2018). PAM now offers long-term continuous insights into biodiversity, using acoustic indices correlating with deep-learning tools aid detection identification of animal sounds (Sueur 2014, Kahl 2021, Wu 2022). However, managing mobilizing original recordings remains challenging due their large size lack structure, hindering broader applications these valuable data. Established 2014 by Academia Sinica Taiwan Forestry Research Institute, Asian Soundscape Network amassed over 20 million minutes audible ultrasonic from diverse landscapes, including forests, wetlands, urban parks farmlands across Malaysia, Thailand, Vietnam. To enhance data mobilization utilization, we developed user-friendly platform enabling browsing, searching, visualization, exploration retrieval (Fig. 1), well derived such species occurrences, associated weather records 2) The also allows users visualize temporal dynamics soundscape identify events at sites examining spectrograms time series 3). Detailed information about recorder deployment is provided. Each recording tagged Creative Commons license, unique Archival Resource Key assigned every for persistent identification, facilitating reuse. Additionally, containing human voices are identified restricted protect privacy. Our aim this streamline data, foster applications, overall value

Language: Английский

---

A Comprehensive Survey of Animal Identification: Exploring Data Sources, AI Advances, Classification Obstacles and the Role of Taxonomy

International Journal of Intelligent Systems, Journal Year: 2024, Volume and Issue: 2024(1)

Published: Jan. 1, 2024

With the rapid development of entity recognition technology, animal has gradually become essential in modern society, supporting labour‐intensive agriculture and husbandry tasks. Severe problems such as maintaining biodiversity can also benefit from identification technology. However, certain invasive systems have resulted permanent harm to animals, while noninvasive methods exhibit drawbacks. This paper conducts a systematic literature review (SLR), presenting comprehensive overview various technologies their applications. Specifically, it examines methodologies deep learning, image processing acoustic analysis used for different characteristics purposes. The contribution machine learning feature extraction is highlighted, emphasising its significance taxonomy wild species monitoring. Additionally, this addresses challenges limitations current technologies, including data scarcity, model accuracy computational requirements, suggests opportunities future research overcome these obstacles.

Language: Английский

---

Deep Learning for Recognizing Bat Species and Bat Behavior in Audio Recordings

Published: June 1, 2023

Monitoring and mitigating the continuous decline of biodiversity is a key global challenge to preserve existential basis human life. Bats as one most widespread species among terrestrial mammals are excellent indicators for hence health an ecosystem. Typically, bats monitored by analyzing ultrasonic sound recordings. Stateof-the-art deep learning approaches automatic bat detection recognition commonly rely on audio spectrogram classification models based fixed time segments, lacking exact call boundaries. While great progress has been made using echolocation calls, little attention paid behavior that provides valuable additional information about populations. In this paper, we present novel end-to-end approach neural network object detection. contrast state-of-the-art approaches, presented model accurate It recognizes 19 distinguishes between three different behaviors: orientation (echolocation calls), hunting (feeding buzzes), social (social calls). Our experiments with two data sets show our method clearly outperforms previous recognition, achieving up 86.2% mean average precision. also performs very well reaching 98.4%, 98.3%, 95.6% precision recognizing feeding buzzes, respectively.

Language: Английский

---

Evaluating community-wide temporal sampling in passive acoustic monitoring: A comprehensive study of avian vocal patterns in subtropical montane forests

F1000Research, Journal Year: 2023, Volume and Issue: 12, P. 1299 - 1299

Published: Oct. 11, 2023

Background: Passive acoustic monitoring (PAM) has become a popular tool for bird monitoring, with vocal activity rate (VAR) being key metric to gauge populations. However, the effective temporal sampling design at community level representative VAR data remains underexplored. Methods: In this study, we used vocalizations extracted from recordings of 12 species, taken 14 PAM stations situated in subtropical montane forests over four-month period, assess impact on across three distinct scales: seasonal, diel, and hourly. For seasonal analysis, employed hierarchical clustering analysis (HCA) coefficient variation (CV). Generalized additive models (GAMs) were utilized diel determined average difference values per minute hourly analysis. Results: We identified significant day species-specific fluctuations. The survey season was divided into five segments; earliest two showed high variability are best avoided surveys. Data days heavy rain strong winds reduced should be excluded Continuous spanning least seven days, extending is optimal minimizing variance. Morning chorus effectively capture majority vocalizations, frequent, shorter intervals aligns closely continuous recording outcomes. Conclusions: While our findings context-specific, they highlight significance strategic avian optimizing resource utilization enhancing breadth efforts.

Language: Английский

---

A novel deep learning‐based bioacoustic approach for identification of look‐alike white‐eye (Zosterops) species traded in wildlife markets

Ibis, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 1, 2024

The songbird trade crisis in East and South Asia has been fuelled by high demand, driving many species to the brink of extinction. This driven desire for songbirds as pets, singing competitions prayer animal release led overexploitation numerous introduction spread invasive alien diseases novel environments. ability identify traded efficiently accurately is crucial monitoring bird markets, protecting threatened enforcing wildlife laws. Citizen scientists can make major contributions these conservation efforts but may be constrained difficulties distinguishing ‘look‐alike’ markets. To address this challenge, we developed a deep learning‐based Artificial Intelligence (AI) bioacoustic tool enable citizen end, used three avian vocalization databases access data 15 morphologically similar White‐eye ( Zosterops ) that are commonly Asian Specifically, employed Inception v3 pre‐trained model classify ambient sound (i.e. non‐bird sound) using 448 recordings obtained. We converted into spectrogram image form) eight augmentation methods enhance performance AI neural network through training validation. found recall, precision F1 score increased amount increased, resulting up 91.6% overall accuracy an 88.8% identifying focal species. Through application bioacoustics learning, approach would law enforcement officials prohibited species, making important conservation.

Language: Английский

---

Toward Birds Conservation in Dry Forest Ecosystems Through Audio Recognition via Deep Learning

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 45 - 57

Published: Oct. 22, 2024

Language: Английский

---

A deep learning model for detecting and classifying multiple marine mammal species from passive acoustic data

Ecological Informatics, Journal Year: 2024, Volume and Issue: 84, P. 102906 - 102906

Published: Nov. 22, 2024

Language: Английский

---

Bioacoustic monitoring to determine addiction levels of primates to the human sphere: A feasibility study on Japanese macaques

American Journal of Primatology, Journal Year: 2023, Volume and Issue: 85(12)

Published: Oct. 2, 2023

Some nonhuman primate species, whose original habitats have been reclaimed by artificial activities, acquired boldness toward humans which is evident based on the diminished frequency of escape behaviors. Eventually, such species become regular users human settlements, and are referred to as "urban primates." Considering this, we developed a noninvasive technique bioacoustics provide transparent assessment troop addiction levels in anthropogenic environments, determined dependence agricultural crops living sphere for their diets daily ranging, respectively. We attempted quantify troops when raiding characterized "landscape fear" because presence predators. hypothesized that could be measured using two indices: events settlements amount time spent there. For hypothesis testing, devised an efficient method measure these indices sound cues (i.e., spontaneous calls) tracing movements obtainable throughout day from most (e.g., contact calls). conducted feasibility study this procedure, targeting Japanese macaques (Macaca fuscata). study, collected 346 recording weeks data autonomous recorders 24 with different during nonsnowy seasons. The results demonstrated reached threshold level, at radical interventions including mass culling members officially permitted, readily identified following behavioral characteristics: or three times per week mean event exceeding 0.4 h. Thus, bioacoustic monitoring valid option ensure objectivity policy judgment urban management.

Language: Английский

---

Suivis acoustiques de biodiversité : perspectives et défis en milieu continental terrestre

Naturae, Journal Year: 2023, Volume and Issue: 8

Published: Nov. 28, 2023

Les animaux produisent des sons pendant leurs activités ou pour assurer diverses fonctions biologiques comme la défense de territoires, l’attraction partenaires, dissuasion prédateurs. En enregistrant ces données acoustiques, les scientifiques obtiennent informations essentielles sur présence espèces. nouvelles technologies d’identification espèces sont plus abordables, efficaces et polyvalentes que méthodes classiques peuvent ainsi répondre au besoin urgent documenter biodiversité dans le contexte actuel crise. enregistreurs acoustiques automatisés en utilisés suivis faire face aux limites traditionnelles à l’émergence considérations déontologiques préconisant développement pièges non destructifs (i.e. létaux). Nous présentons ici outils d’acquisition milieu continental terrestre, gestion d’analyse classification automatique l’étude paysages sonores, avantages l’utilisation un objectif suivi terrestre.

---