1
|
Graves SJ, Marconi S, Stewart D, Harmon I, Weinstein B, Kanazawa Y, Scholl VM, Joseph MB, McGlinchy J, Browne L, Sullivan MK, Estrada-Villegas S, Wang DZ, Singh A, Bohlman S, Zare A, White EP. Data science competition for cross-site individual tree species identification from airborne remote sensing data. PeerJ 2023; 11:e16578. [PMID: 38144190 PMCID: PMC10749090 DOI: 10.7717/peerj.16578] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/25/2023] [Accepted: 11/13/2023] [Indexed: 12/26/2023] Open
Abstract
Data on individual tree crowns from remote sensing have the potential to advance forest ecology by providing information about forest composition and structure with a continuous spatial coverage over large spatial extents. Classifying individual trees to their taxonomic species over large regions from remote sensing data is challenging. Methods to classify individual species are often accurate for common species, but perform poorly for less common species and when applied to new sites. We ran a data science competition to help identify effective methods for the task of classification of individual crowns to species identity. The competition included data from three sites to assess each methods' ability to generalize patterns across two sites simultaneously and apply methods to an untrained site. Three different metrics were used to assess and compare model performance. Six teams participated, representing four countries and nine individuals. The highest performing method from a previous competition in 2017 was applied and used as a baseline to understand advancements and changes in successful methods. The best species classification method was based on a two-stage fully connected neural network that significantly outperformed the baseline random forest and gradient boosting ensemble methods. All methods generalized well by showing relatively strong performance on the trained sites (accuracy = 0.46-0.55, macro F1 = 0.09-0.32, cross entropy loss = 2.4-9.2), but generally failed to transfer effectively to the untrained site (accuracy = 0.07-0.32, macro F1 = 0.02-0.18, cross entropy loss = 2.8-16.3). Classification performance was influenced by the number of samples with species labels available for training, with most methods predicting common species at the training sites well (maximum F1 score of 0.86) relative to the uncommon species where none were predicted. Classification errors were most common between species in the same genus and different species that occur in the same habitat. Most methods performed better than the baseline in detecting if a species was not in the training data by predicting an untrained mixed-species class, especially in the untrained site. This work has highlighted that data science competitions can encourage advancement of methods, particularly by bringing in new people from outside the focal discipline, and by providing an open dataset and evaluation criteria from which participants can learn.
Collapse
Affiliation(s)
- Sarah J. Graves
- Nelson Institute for Environmental Studies, University of Wisconsin-Madison, Madison, Wisconsin, United States
| | - Sergio Marconi
- Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, United States
| | - Dylan Stewart
- Department of Electrical and Computer Engineering, University of Florida, Gainesville, Florida, United States
| | - Ira Harmon
- Department of Computer and Information Sciences and Engineering, University of Florida, Gainesville, Florida, United States
| | - Ben Weinstein
- Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, United States
| | - Yuzi Kanazawa
- Artificial Intelligence Laboratory, Fujitsu Laboratories Ltd., Kawasaki, Kanagawa, Japan
| | - Victoria M. Scholl
- Earth Lab, Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado at Boulder, Boulder, Colorado, United States
- Department of Geography, University of Colorado at Boulder, Boulder, Colorado, United States
| | - Maxwell B. Joseph
- Earth Lab, Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado at Boulder, Boulder, Colorado, United States
| | - Joseph McGlinchy
- Earth Lab, Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado at Boulder, Boulder, Colorado, United States
| | - Luke Browne
- Yale School of the Environment, Yale University, New Haven, Connecticut, United States
| | - Megan K. Sullivan
- Yale School of the Environment, Yale University, New Haven, Connecticut, United States
| | | | - Daisy Zhe Wang
- Department of Computer and Information Sciences and Engineering, University of Florida, Gainesville, Florida, United States
| | - Aditya Singh
- Department of Agricultural & Biological Engineering, University of Florida, Gainesville, Florida, United States
| | - Stephanie Bohlman
- School of Forest, Fisheries, and Geomatics Sciences, University of Florida, Gainesville, Florida, United States
| | - Alina Zare
- Department of Electrical and Computer Engineering, University of Florida, Gainesville, Florida, United States
- Informatics Institute, University of Florida, Gainesville, Florida, United States
- Biodiversity Institute, University of Florida, Gainesville, Florida, United States
| | - Ethan P. White
- Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, Florida, United States
- Informatics Institute, University of Florida, Gainesville, Florida, United States
- Biodiversity Institute, University of Florida, Gainesville, Florida, United States
| |
Collapse
|
2
|
Besson M, Alison J, Bjerge K, Gorochowski TE, Høye TT, Jucker T, Mann HMR, Clements CF. Towards the fully automated monitoring of ecological communities. Ecol Lett 2022; 25:2753-2775. [PMID: 36264848 PMCID: PMC9828790 DOI: 10.1111/ele.14123] [Citation(s) in RCA: 21] [Impact Index Per Article: 10.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/22/2022] [Revised: 08/09/2022] [Accepted: 09/06/2022] [Indexed: 01/12/2023]
Abstract
High-resolution monitoring is fundamental to understand ecosystems dynamics in an era of global change and biodiversity declines. While real-time and automated monitoring of abiotic components has been possible for some time, monitoring biotic components-for example, individual behaviours and traits, and species abundance and distribution-is far more challenging. Recent technological advancements offer potential solutions to achieve this through: (i) increasingly affordable high-throughput recording hardware, which can collect rich multidimensional data, and (ii) increasingly accessible artificial intelligence approaches, which can extract ecological knowledge from large datasets. However, automating the monitoring of facets of ecological communities via such technologies has primarily been achieved at low spatiotemporal resolutions within limited steps of the monitoring workflow. Here, we review existing technologies for data recording and processing that enable automated monitoring of ecological communities. We then present novel frameworks that combine such technologies, forming fully automated pipelines to detect, track, classify and count multiple species, and record behavioural and morphological traits, at resolutions which have previously been impossible to achieve. Based on these rapidly developing technologies, we illustrate a solution to one of the greatest challenges in ecology: the ability to rapidly generate high-resolution, multidimensional and standardised data across complex ecologies.
Collapse
Affiliation(s)
- Marc Besson
- School of Biological SciencesUniversity of BristolBristolUK,Sorbonne Université CNRS UMR Biologie des Organismes Marins, BIOMBanyuls‐sur‐MerFrance
| | - Jamie Alison
- Department of EcoscienceAarhus UniversityAarhusDenmark,UK Centre for Ecology & HydrologyBangorUK
| | - Kim Bjerge
- Department of Electrical and Computer EngineeringAarhus UniversityAarhusDenmark
| | - Thomas E. Gorochowski
- School of Biological SciencesUniversity of BristolBristolUK,BrisEngBio, School of ChemistryUniversity of BristolCantock's CloseBristolBS8 1TSUK
| | - Toke T. Høye
- Department of EcoscienceAarhus UniversityAarhusDenmark,Arctic Research CentreAarhus UniversityAarhusDenmark
| | - Tommaso Jucker
- School of Biological SciencesUniversity of BristolBristolUK
| | - Hjalte M. R. Mann
- Department of EcoscienceAarhus UniversityAarhusDenmark,Arctic Research CentreAarhus UniversityAarhusDenmark
| | | |
Collapse
|
3
|
Zhang C, Zhou L, Xiao Q, Bai X, Wu B, Wu N, Zhao Y, Wang J, Feng L. End-to-End Fusion of Hyperspectral and Chlorophyll Fluorescence Imaging to Identify Rice Stresses. PLANT PHENOMICS 2022; 2022:9851096. [PMID: 36059603 PMCID: PMC9394116 DOI: 10.34133/2022/9851096] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 03/19/2022] [Accepted: 07/03/2022] [Indexed: 11/07/2022]
Abstract
Herbicides and heavy metals are hazardous substances of environmental pollution, resulting in plant stress and harming humans and animals. Identification of stress types can help trace stress sources, manage plant growth, and improve stress-resistant breeding. In this research, hyperspectral imaging (HSI) and chlorophyll fluorescence imaging (Chl-FI) were adopted to identify the rice plants under two types of herbicide stresses (butachlor (DCA) and quinclorac (ELK)) and two types of heavy metal stresses (cadmium (Cd) and copper (Cu)). Visible/near-infrared spectra of leaves (L-VIS/NIR) and stems (S-VIS/NIR) extracted from HSI and chlorophyll fluorescence kinetic curves of leaves (L-Chl-FKC) and stems (S-Chl-FKC) extracted from Chl-FI were fused to establish the models to detect the stress of the hazardous substances. Novel end-to-end deep fusion models were proposed for low-level, middle-level, and high-level information fusion to improve identification accuracy. Results showed that the high-level fusion-based convolutional neural network (CNN) models reached the highest detection accuracy (97.7%), outperforming the models using a single data source (<94.7%). Furthermore, the proposed end-to-end deep fusion models required a much simpler training procedure than the conventional two-stage deep learning fusion. This research provided an efficient alternative for plant stress phenotyping, including identifying plant stresses caused by hazardous substances of environmental pollution.
Collapse
Affiliation(s)
- Chu Zhang
- School of Information Engineering, Huzhou University, Huzhou 313000, China
| | - Lei Zhou
- College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing, China
| | - Qinlin Xiao
- College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
- Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou 310058, China
| | - Xiulin Bai
- College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
- Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou 310058, China
| | - Baohua Wu
- College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
- Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou 310058, China
| | - Na Wu
- College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
- Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou 310058, China
| | - Yiying Zhao
- Institute of Digital Agriculture, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China
| | - Junmin Wang
- Institute of Crop Science and Nuclear Technology Utilization, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China
| | - Lei Feng
- College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
- Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou 310058, China
| |
Collapse
|
4
|
Musinsky J, Goulden T, Wirth G, Leisso N, Krause K, Haynes M, Chapman C. Spanning scales: The airborne spatial and temporal sampling design of the National Ecological Observatory Network. Methods Ecol Evol 2022. [DOI: 10.1111/2041-210x.13942] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
Affiliation(s)
- John Musinsky
- National Ecological Observatory Network, Battelle Boulder CO USA
| | - Tristan Goulden
- National Ecological Observatory Network, Battelle Boulder CO USA
| | | | | | - Keith Krause
- National Ecological Observatory Network, Battelle Boulder CO USA
| | - Mitch Haynes
- National Ecological Observatory Network, Battelle Boulder CO USA
| | - Cameron Chapman
- National Ecological Observatory Network, Battelle Boulder CO USA
| |
Collapse
|