說明
資料紀錄
此資源出現紀錄的資料已發佈為達爾文核心集檔案（DwC-A），其以一或多組資料表構成分享生物多樣性資料的標準格式。 核心資料表包含 13,880 筆紀錄。
亦存在 1 筆延伸集的資料表。延伸集中的紀錄補充核心集中紀錄的額外資訊。 每個延伸集資料表中資料筆數顯示如下。
此 IPT 存放資料以提供資料儲存庫服務。資料與資源的詮釋資料可由「下載」單元下載。「版本」表格列出此資源的其它公開版本，以便利追蹤其隨時間的變更。
版本
以下的表格只顯示可公開存取資源的已發布版本。
如何引用
研究者應依照以下指示引用此資源。:
D'hondt B, Steen F, Neukermans A, Adriaens T, Dalby L, Gillespie M A, Mellerup K A, Høye T T, Bonnet P, Joly A, Govaert S (2026). IAS pilot - Observations of alien plant species using a vehicle-mounted camera system in Flanders (Belgium). Version 1.1. Research Institute for Nature and Forest (INBO). Occurrence dataset. https://ipt.inbo.be/resource?r=ias-pilot-be&v=1.1
權利
研究者應尊重以下權利聲明。:
此資料的發布者及權利單位為 Research Institute for Nature and Forest (INBO)。 To the extent possible under law, the publisher has waived all rights to these data and has dedicated them to the Public Domain (CC0 1.0). Users may copy, modify, distribute and use the work, including for commercial purposes, without restriction.
GBIF 註冊
此資源已向GBIF註冊，並指定以下之GBIF UUID: 9c85f295-cc92-473c-969a-bb5d3eccfd0b。 Research Institute for Nature and Forest (INBO) 發佈此資源，並經由Belgian Biodiversity Platform同意向GBIF註冊成為資料發佈者。
關鍵字
Occurrence; CamAlien; invasive species; alien species; non-native species; IAS; plants
聯絡資訊
- 元數據提供者 ●
- 出處 ●
- 連絡人
https://orcid.org/0000-0002-1330-1457
- 出處
- 出處
- 出處
- 出處
- 元數據提供者 ●
- 出處
地理涵蓋範圍
Flanders, Belgium
| 界定座標範圍 | 緯度南界 經度西界 [50.705, 4.216], 緯度北界 經度東界 [51.072, 4.983] |
|---|
分類群涵蓋範圍
6 non-native species of vascular plants
| Genus | Reynoutria sp. |
|---|---|
| Species | Ailanthus altissima, Impatiens glandulifera, Robinia pseudoacacia, Rosa rugosa |
時間涵蓋範圍
| 起始日期 / 結束日期 | 2024-07-18 / 2024-09-09 |
|---|
計畫資料
Biodiversa+ monitoring pilots aim to facilitate the harmonisation of biodiversity monitoring at a transnational level, by testing common protocols and new technologies. The Biodiversa+ IAS pilot project (2023-2027) implemented the monitoring of invasive alien plants using a vehicle-mounted camera system in various countries across Europe.
| 計畫名稱 | Monitoring Invasive Alien Species with image-based methods |
|---|---|
| 辨識碼 | IAS pilot |
| 經費來源 | Biodiversa+, the European Biodiversity Partnership supporting excellent research on biodiversity with an impact on policy and society. Biodiversa+ was jointly developed by BiodivERsA and the European Commission as part of the EU Biodiversity Strategy 2030. |
| 研究區域描述 | Europe |
| 研究設計描述 | During several field seasons of the project, the CamAlien camera system (AI Lab, Denmark) was mounted on vehicles to capture images of the vegetation along linear transport infrastructure. Typically, this system was mounted on a car for the monitoring of roadside vegetation. The system was also tested on other types of vehicles (e.g., boats, trains). Plant species were then identified from the images using species recognition algorithms (Pl@ntNet). The focus was specifically on invasive alien plants, in the context of large-scale surveys, rapid detection, and support of species management. This pilot study was conducted in 11 countries across Europe. |
參與計畫的人員:
- 作者
取樣方法
The CamAlien system essentially is a camera mounted on a moving car. The system continuously takes photos while driving, even at highway speeds. Images are anonymised for GDPR compliance and analysis is powered by the Pl@ntNet API, which returns species confidence scores. The software applies adaptive recording based on speed, and a human observer is able to tag plant sightings on the go, helping to build a better training set for the image recognition models.
| 研究範圍 | The study aimed to detect a set of predetermined invasive plant species in roadside verges on a regional scale during the field season of several consecutive years. |
|---|---|
| 品質控管 | The species recognition algorithm (Pl@ntNet) returned a confidence value for each identification of a plant species in an image. A threshold value above which a confidence score is considered accurate was determined through expert validation and statistical modelling. Expert botanists annotated a subset of images with a range of confidence scores for presence/absence of the classified species. Presence/absence was then modelled as a binary response in a logistic GLM (General Linear Model) against confidence score. The confidence score providing a 50% probability of presence was then selected as threshold value. The images resulting from this filtering process were not all individually reviewed. |
方法步驟描述:
- Partners from the project consortium defined the set of target species (invasive alien plant species of interest), and agreed on a generic sampling plan for the coming field season.
- The CamAlien camera system was mounted on a vehicle.
- The vehicle was driven along the road infrastructure, with the CamAlien system running.
- The resulting images were shared with the lead partner (Aarhus University), for storage and first-level edits.
- The images were transferred to Pl@ntNet, which returned classifications (identifications) of the target species in each of the images.
- The returned data were processed, by filtering on the confidence level threshold, or any other step deemed necessary to increase reliability.
引用文獻
- Dyrmann M, Skovsen SK, Christiansen PH et al. (2024) High-speed camera system for efficient monitoring of invasive plant species along roadways. F1000Research 2024, 13:360. https://doi.org/10.12688/f1000research.141992.2
- Espitalier V, Lombardo JC, Goëau H, Botella C, Høye T T, Dyrmann M, Bonnet P, Joly A (2025). Adapting a global plant identification model to detect invasive alien plant species in high-resolution road side images. Ecological Informatics 89. https://doi.org/10.1016/j.ecoinf.2025.103129
額外的詮釋資料
| 替代的識別碼 | 9c85f295-cc92-473c-969a-bb5d3eccfd0b |
|---|---|
| https://ipt.inbo.be/resource?r=ias-pilot-be |