Ecological Informatics最新文献

{"title":"Spatiotemporal evolution and driving mechanism of Dongting Lake based on 2005–2020 multi-source remote sensing data","authors":"Mingzhe Fu ,&nbsp;Yuanmao Zheng ,&nbsp;Changzhao Qian ,&nbsp;Qiuhua He ,&nbsp;Yuanrong He ,&nbsp;Chenyan Wei ,&nbsp;Kexin Yang ,&nbsp;Wei Zhao","doi":"10.1016/j.ecoinf.2024.102822","DOIUrl":"10.1016/j.ecoinf.2024.102822","url":null,"abstract":"<div><div>As one of the largest inland lakes in China, Dongting Lake has attracted widespread attention owing to its rich natural resources, unique geographical landscape, and important ecological functions. Recently, Dongting Lake has experienced phenomena such as an early dry season and backflow during the flood season. Multi-source remote sensing data and the normalised difference water index (NDWI) threshold method were used to systematically analyse the water area of the lake from 2005 to 2020. Additionally, it employed a centre of gravity migration model and a geographic detector model to investigate the lake's evolution patterns and driving mechanisms. The research identified notable fluctuations in Dongting Lake's water area during this period, with a particularly sharp decline in 2006—from 1509.74 km² to 815 km², marking a decrease of 694.74 km² and a shrinkage rate of 46.01 %. Spatial analysis indicated that the centre of gravity of these water areas changed primarily between Nandashan Town, the Dongting Lake Management Committee, Wanzihu Township, and Qingtan Township, underscoring their significant influence on lake dynamics, including runoff, surface water availability, sediment deposition, and precipitation, all of which displayed strong positive correlations (Pearson coefficients of 0.57, 0.68, and 0.63, respectively), whereas population density showed a negative correlation (Pearson coefficient of −0.56). Furthermore, the study highlighted the substantial impact of the Digital Elevation Model (DEM) and its interaction with slope and aspect on Dongting Lake's evolution, with Q values of 0.537 and 0.543, respectively, emphasising their critical roles in shaping lake area changes and providing a crucial scientific basis for enhancing the understanding and effective management of water resources in the Dongting Lake Basin through comprehensive analysis of its spatiotemporal evolution and driving mechanisms.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003649/pdfft?md5=7c5aa5f56347f8489f910ec55f75d4d6&pid=1-s2.0-S1574954124003649-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142315036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Harnessing artificial intelligence for efficient systematic reviews: A case study in ecosystem condition indicators","authors":"Isabel Nicholson Thomas ,&nbsp;Philip Roche ,&nbsp;Adrienne Grêt-Regamey","doi":"10.1016/j.ecoinf.2024.102819","DOIUrl":"10.1016/j.ecoinf.2024.102819","url":null,"abstract":"<div><p>Effective evidence synthesis is important for the integration of scientific research into decision-making. However, fully depicting the vast mosaic of concepts and applications in environmental sciences and ecology often entails a substantial workload. New Artificial Intelligence (AI) tools present an attractive option for addressing this challenge but require sufficient validation to match the vigorous standards of a systematic review. This article demonstrates the use of generative AI in the selection of relevant literature as part of a systematic review on indicators of ecosystem condition. We highlight, through the development of an optimal prompt to communicate inclusion and exclusion criteria, the need to describe ecosystem condition as a multidimensional concept whilst also maintaining clarity on what does not meet the criteria of comprehensiveness. We show that, although not completely infallible, the GPT-3.5 model significantly outperforms traditional literature screening processes in terms of speed and efficiency whilst correctly selecting 83 % of relevant literature for review. Our study highlights the importance of precision in prompt design and the setting of query parameters for the AI model and opens the perspective for future work using language models to contextualize complex concepts in the environmental sciences. Future development of this methodology in tandem with the continued evolution of the accessibility and capacity of AI tools presents a great potential to improve evidence synthesis through gains in efficiency and possible scope.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003613/pdfft?md5=a2a00c40d3636d32055ec22bbf0011ce&pid=1-s2.0-S1574954124003613-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142173240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reliable and efficient integration of AI into camera traps for smart wildlife monitoring based on continual learning","authors":"Delia Velasco-Montero ,&nbsp;Jorge Fernández-Berni ,&nbsp;Ricardo Carmona-Galán ,&nbsp;Ariadna Sanglas ,&nbsp;Francisco Palomares","doi":"10.1016/j.ecoinf.2024.102815","DOIUrl":"10.1016/j.ecoinf.2024.102815","url":null,"abstract":"<div><p>In this paper, we comprehensively report on an efficient approach for the integration of artificial intelligence (AI) processing pipelines in camera traps for smart on-site wildlife monitoring. Our work covers hardware, software, and algorithmics. We have built two prototypes of smart camera trap on a maximum bill of materials of 100$. We have also built two datasets, made publicly available, comprising over 17 k images, many of them notably challenging even for humans. Leveraging our broad expertise on embedded systems, specialized software libraries and toolchains, and AI techniques such as transfer learning, explainable AI, and, most importantly, continual learning, we achieve more reliable inference on-site - specifically 10 % higher F1-score - than MegaDetector run off-site on a desktop computer. The paper includes many practical details on system realization and on-site training in addition to a vast set of lab and experimental results.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003571/pdfft?md5=eb0fce560598e6780765baf89fb57705&pid=1-s2.0-S1574954124003571-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142173771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An enhanced chlorophyll estimation model with a canopy structural trait in maize crops: Use of multi-spectral UAV images and machine learning algorithm","authors":"Gaurav Singhal ,&nbsp;Burhan U. Choudhury ,&nbsp;Naseeb Singh ,&nbsp;Jonali Goswami","doi":"10.1016/j.ecoinf.2024.102811","DOIUrl":"10.1016/j.ecoinf.2024.102811","url":null,"abstract":"<div><p>Leaf chlorophyll concentration (LCC) is a key indicator of leaf nitrogen (N) and changes in canopy structure, particularly the leaf area index (LAI), play a significant role in estimating LCC. Spectral prediction models for chlorophyll are a useful tool for timely nutritional management, particularly in precision agriculture. However, the accuracy of LCC estimation is influenced by the LAI. Considering LAI data as input in the spectral prediction model is still inadequate for improving LCC estimation. This study tested the hypothesis that LCC estimate accuracy could be enhanced by using LAI as an input using high-resolution (5 cm) multi-spectral images from an unmanned aerial vehicle (UAV). For this, maize was grown as a test crop under different nutrient management in the hilly ecosystem of Meghalaya. LCC was measured using laboratory destruction methods from ground sampling that coincided with UAV flights. Machine learning algorithms such as random forest (RF), support vector machine (SVM), and kernel ridge regression (KKR) were employed to develop the LCC estimation model, utilizing band reflectance, vegetation indexes, and measured chlorophyll. The model was assessed for its sensitivity to LCC estimation using LAI data. KKR outperformed other two algorithms (RF and SVM) in accuracy of LCC estimation by &gt;11.0 to 19.0 %. The KKR-derived LCC estimation model was significantly improved by the inclusion of LAI (R² increased from 0.785 to 0.928 and RMSE decreased from 0.065 to 0.053 mg g⁻¹). The model's reliability was proven on multiple UAV flights for maize crops that are healthy and nutrient-stressed. Thus, LCC models derived from multispectral UAV images using KKR algorithms could benefit the adoption of precision agriculture at field scale in mountain ecosystems.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003534/pdfft?md5=369473b281e167821f894f52594fa7a2&pid=1-s2.0-S1574954124003534-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142162810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of ChatGPT's potential as an innovative tool in searching for information on wild mammals","authors":"Jenner Rodas-Trejo ,&nbsp;Paola Ocampo-González","doi":"10.1016/j.ecoinf.2024.102810","DOIUrl":"10.1016/j.ecoinf.2024.102810","url":null,"abstract":"<div><p>In November 2022 OpenAI launched ChatGPT, an Artificial Intelligence model capable of processing, retrieving, and organizing large amounts of data and identifying patterns, thereby generating text on various topics and contexts. In recent months ChatGPT has gained wide attention and adoption in the academic and scientific fields, so its use is being widely evaluated and discussed. In this connection, an evaluation of the performance of ChatGPT was carried out on the accuracy of the answers on general knowledge about wild mammals and the specific knowledge of 30 species. A descriptive study was carried out using three chats where twenty-one questions on general knowledge and three questions on specific knowledge were asked. The questions considered information on the taxonomy, natural history, and conservation status of each species, as well as concepts and indices commonly used in the study of wild mammals. The answers were compared with scientific literature and a value was assigned to later obtain the percentage of precision. The results showed a high precision in the specific knowledge of the species, with an average of 88 % correct answers. Precision varied by species, with species scoring close to 100 % and others scoring as low as 65 %. The taxonomy question had 100 % correct answers, the natural history questions 90 %, and the conservation status question 56 %. In the precision of the general knowledge answers in the study of wild mammals, a moderate precision of 73.54 % was obtained. The study shows that ChatGPT has high precision, so it can be a helpful tool in the search for information in research on wild mammals. On the other hand, concerns are raised about its applicability in the academic field, due to the risk of producing unreliable or biased results and generating inaccurate or misleading content, so it is important to take into account the limitations and risks associated with its use. It is suggested that further research and insight into accuracy be done to explore the full potential of ChatGPT.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003522/pdfft?md5=d5025382c641c9f60b6256f7ed5d09b2&pid=1-s2.0-S1574954124003522-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142162309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatio-temporal variability of turbidity derived from Sentinel-2 in Reloncaví sound, Northern Patagonia, Chile","authors":"Wirmer García-Tuñon ,&nbsp;Elizabeth D. Curra-Sánchez ,&nbsp;Carlos Lara ,&nbsp;Lisdelys González-Rodríguez ,&nbsp;Esther Patricia Urrego ,&nbsp;Jesús Delegido ,&nbsp;Bernardo R. Broitman","doi":"10.1016/j.ecoinf.2024.102814","DOIUrl":"10.1016/j.ecoinf.2024.102814","url":null,"abstract":"<div><p>Turbidity is associated with the loss of water transparency due to the presence of particles, sediments, suspended solids, and organic or inorganic compounds in the water, of natural or anthropogenic origin. Our study aimed to evaluate the spatio-temporal variability of turbidity from Sentinel-2 (S2) images in the Reloncaví sound and fjord, in Northern Patagonia, Chile, a coastal ecosystem that is intensively used by finfish and shellfish aquaculture. To this end, we downloaded 123 S2 images and assembled a five-year time series (2016–2020) covering five study sites (R1 to R5) located along the axis of the fjord and seaward into the sound. We used Acolite to perform the atmospheric correction and estimate turbidity with two algorithms proposed by Nechad et al. (2009, 2016 Nv09 and Nv16, respectively). When compared to match-up, and <em>in situ</em> measurements, both algorithms had the same performance (R² = 0.40). The Nv09 algorithm, however, yielded smaller errors than Nv16 (RMSE = 0.66 FNU and RMSE = 0.84 FNU, respectively). Results from true-color imagery and two Nechad algorithms singled an image from the austral autumn of 2019 as the one with the highest turbidity. Similarly, three images from the 2020 austral autumn (May 20, 25, 30) also exhibited high turbidity values. The turbid plumes with the greatest extent occurred in the autumn of 2019 and 2020, coinciding with the most severe storms and runoff events of the year, and the highest turbidity values. Temporal trends in turbidity were not significant at any of the study sites. However, turbidity trends at sites R1 and R2 suggested an increasing trend, while the other sites showed the opposite trend. Site R1 recorded the highest turbidity values, and the lowest values were recorded at R5 in the center of the sound. The month of May was characterized by the highest turbidity values. The application of algorithms from high-resolution satellite images proved to be effective for the estimation and mapping of this water quality parameter in the study area. The use of S2 imagery unraveled a predictable spatial and temporal structure of turbidity patterns in this optically complex aquatic environment. Our results suggest that the availability of <em>in situ</em> data and the continued evaluation of the performance of the Nechad algorithms can yield significant insights into the dynamics and impacts of turbid waters in this important coastal ecosystem.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S157495412400356X/pdfft?md5=e2b2b707e0bb8315edfdf66f5782072d&pid=1-s2.0-S157495412400356X-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142162308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems","authors":"Clara Oliva Gonçalves Bazzo ,&nbsp;Bahareh Kamali ,&nbsp;Murilo dos Santos Vianna ,&nbsp;Dominik Behrend ,&nbsp;Hubert Hueging ,&nbsp;Inga Schleip ,&nbsp;Paul Mosebach ,&nbsp;Almut Haub ,&nbsp;Axel Behrendt ,&nbsp;Thomas Gaiser","doi":"10.1016/j.ecoinf.2024.102813","DOIUrl":"10.1016/j.ecoinf.2024.102813","url":null,"abstract":"<div><p>Wet grasslands are crucial components of terrestrial ecosystems, known for their biodiversity and provision of ecosystem services such as flood attenuation and carbon sequestration. Given their ecological significance, monitoring biodiversity within these landscapes is of utmost importance for effective conservation and management strategies. This study, conducted in a wet grassland of Brandenburg, Germany, utilized unmanned aerial vehicles (UAVs) to facilitate the estimation of species richness by the integration of remotely sensed canopy features such as canopy height (CH), spectral data (Vegetation Indices, VI), and texture features (Gray-Level Co-occurrence Matrix, GLCM) using two machine learning methods (Partial Least Square regression (PLS) and Random Forest (RF)). Data was collected over two growing seasons under three different grass cutting regimes, employing multispectral sensors to capture detailed vegetation characteristics. The analysis revealed that the performance of the machine learning methods varied with the feature combinations. Models combining VI and GLCM features demonstrated the highest predictive accuracy, particularly in frequently cut grasslands, as indicated by higher R² values (up to 0.52) and lower root mean square errors (rRMSE as low as 34.9 %). RF models generally outperformed PLS models across different feature sets, with the CH + VI + GLCM combination yielding the best results. These findings underscore the potential of spectral and textural data to effectively capture the ecological dynamics of wet grasslands, providing valuable insights into biodiversity patterns.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003558/pdfft?md5=3e774375b1cc0d4f85b7005b06fab026&pid=1-s2.0-S1574954124003558-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142149320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Association of meteorological variables with leaf spot and fruit rot disease incidence in eggplant and YOLOv8-based disease classification","authors":"Arya Kaniyassery ,&nbsp;Ayush Goyal ,&nbsp;Sachin Ashok Thorat ,&nbsp;Mattu Radhakrishna Rao ,&nbsp;Harsha K. Chandrashekar ,&nbsp;Thokur Sreepathy Murali ,&nbsp;Annamalai Muthusamy","doi":"10.1016/j.ecoinf.2024.102809","DOIUrl":"10.1016/j.ecoinf.2024.102809","url":null,"abstract":"<div><p>Eggplant is one of the major vegetables consumed worldwide. Several fungal, bacterial, and viral diseases challenge the yield and quality of eggplant. The incidence of plant diseases is strongly influenced by weather factors such as temperature, humidity, rainfall, and wind speed. Mattu Gulla (MG) is a GI-tagged traditional variety of eggplant grown in Mattu village of the Udupi district in Karnataka state, India, with a cultural legacy of more than four centuries. In this study, we investigated the relationships between weather parameters and disease incidence in Mattu Gulla. Leaf spot (LS) and fruit rot (FR) are the major diseases affecting this plant variety. The influence of plant age and weather parameters on the modulation of the disease incidence (%) [DI (%)] of leaf spot and fruit rot was recorded and analyzed via correlation and regression. Prediction equations for disease incidence was derived via regression. A significant negative correlation was observed between the leaf spot DI (%) and minimum temperature (Min. temp), and a positive correlation was observed between the DI (%) and fruit rot. In the case of FR, the DI (%) is also significantly positively correlated with wind speed (WS), temperature, maximum relative humidity (RH I), rainfall (RF), and wind speed (WS). An RH I of 86–87 % was favorable for the incidence of fruit rot in the field. Regression analysis revealed a significant association between Min. temp and leaf spot DI (%), and in the case of fruit rot DI (%), the association was with Min. temp and WS. An android application, “Leaf Guard,” has been developed for AI-based disease detection in eggplant. During testing, the accuracy of the trained model reached 98.2 %.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003510/pdfft?md5=74fceef23e01c58be36f0ae8087f1e59&pid=1-s2.0-S1574954124003510-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142158001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Utilizing deep transfer learning to discover changes in landscape patterns in urban wetland parks based on multispectral remote sensing","authors":"Chao Liu ,&nbsp;Xiuhe Yuan ,&nbsp;Guoqing Ni ,&nbsp;Yingjie Liu ,&nbsp;Yansu Qi ,&nbsp;Sheng Miao","doi":"10.1016/j.ecoinf.2024.102808","DOIUrl":"10.1016/j.ecoinf.2024.102808","url":null,"abstract":"<div><p>Urban wetland parks are essential for protecting ecosystems and alleviating urban heat island effects. Owing to the impact of urban sprawl and human activities, habitats in wetland parks have become increasingly fragmented, evoking an urgent need to accurately monitor and analyze such changes. In this study, a transfer learning-based ResNet-18 method was proposed to classify the landscape patterns of urban wetland parks by integrating the advantages of remote sensing technologies, i.e., long-time series of Gaofen-2 and high-accuracy data of unmanned aerial vehicle remote sensing. The proposed method solves the dual problems of low precision and sparse sample data in landscape pattern classification. By employing the proposed method, we realized long-time-series, high-accuracy analysis of landscape pattern changes in a national wetland park. Our results showed that the overall accuracy was 90.69–97.96 % and Kappa coefficient was stable between 0.865 and 0.968, fully verifying the effectiveness and reliability of our method. We revealed a shrinking trend in the area of water bodies along with an expanding trend in the area of other land types. Thus, our findings reflect the significant impact of urban sprawl on the landscape patterns of wetland parks.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003509/pdfft?md5=9045ae82681da8e0e191fb6cfa7a85b3&pid=1-s2.0-S1574954124003509-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142149317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Metadata augmented deep neural networks for wild animal classification","authors":"Aslak Tøn ,&nbsp;Ammar Ahmed ,&nbsp;Ali Shariq Imran ,&nbsp;Mohib Ullah ,&nbsp;R. Muhammad Atif Azad","doi":"10.1016/j.ecoinf.2024.102805","DOIUrl":"10.1016/j.ecoinf.2024.102805","url":null,"abstract":"<div><p>Camera trap imagery has become an invaluable asset in contemporary wildlife surveillance, enabling researchers to observe and investigate the behaviors of wild animals. While existing methods rely solely on image data for classification, this may not suffice in cases of suboptimal animal angles, lighting, or image quality. This study introduces a novel approach that enhances wild animal classification by combining specific metadata (temperature, location, time, etc) with image data. Using a dataset focused on the Norwegian climate, our models show an accuracy increase from 98.4% to 98.9% compared to existing methods. Notably, our approach also achieves high accuracy with metadata-only classification, highlighting its potential to reduce reliance on image quality. This work paves the way for integrated systems that advance wildlife classification technology.</p></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1574954124003479/pdfft?md5=53f932d7fdb006c6249ff8df70fc655e&pid=1-s2.0-S1574954124003479-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142149319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}