用于识别地中海植物的人工视觉模型:四个生态系统的分析
Parminder Kaur1, Anna Grassi2, Federica Bonini2
1Department of Computer Science, Durham University, Durham, United Kingdom.
PloS one
|September 5, 2025
概括
这项研究评估了六种物体检测模型,用于在自然息地识别植物物种. 这项研究对YOLOv8n和Faster RCNN等不同数据集进行了微调,为生态监测提供了见解.
科学领域:
- 生态学
- 计算机视觉
- 生物多样性监测
背景情况:
- 对象识别对于许多应用至关重要,但在自然环境中识别植物物种仍未得到充分研究.
- 自然息地对生物多样性和生态系统健康至关重要,需要有效的监测策略.
- 识别关键植物物种有助于保护这些关键生态系统.
研究的目的:
- 量化评估六种用于野生植物物种识别的流行物体检测模型的性能.
- 评估各种自然息地的模型有效性,包括,沙丘,草原和森林.
- 研究微调预训练模型对定制数据集的影响,以改善生态应用.
主要方法:
- 从四个不同的息地收集了野生植物物种的数据集,使用的是人类操作员和ANYmal C四足机器人.
- 选择了六种物体检测模型:两个单阶段 (RetinaNet,YOLOv8n),两个双阶段 (Faster RCNN,Cascade RCNN) 和两个基于变压器的 (DETR,可变形DETR).
- 预先训练的模型在收集的数据集上进行了微调,实验涉及类平衡和超参数调整.
主要成果:
- 该研究提供了对不同植物物种数据集的六种物体检测模型的定量性能评估.
- 结果突出了微调模型在不同自然息地识别各种植物物种的有效性.
- 在生态数据的背景下,对一阶段,两阶段和基于变压器的探测器的性能差异进行分析.
结论:
- 对象检测模型在自然息地自动识别植物物种方面具有前景,对生物多样性和生态研究至关重要.
- 微调预先训练的模型显著提高了这种特定生态应用的性能.
- 这些发现为开发强大的息地监测和保护工作提供了宝贵的见解.
更多相关视频
12:26Integrating Remote Sensing with Species Distribution Models; Mapping Tamarisk Invasions Using the Software for Assisted Habitat Modeling SAHM
Published on: October 11, 2016
13.4K
06:28Author Spotlight: Unraveling Plant Responses to Abiotic Stresses Using the PlantScreen Robotic Platform
Published on: June 7, 2024
2.0K
相关概念视频
02:16Light Acquisition
8.6K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
8.6K
02:22Introduction to Plant Diversity
45.7K
From Water to Land
45.7K
02:00Photoreceptors and Plant Responses to Light
24.2K
Light plays a significant role in regulating the growth and development of plants. In addition to providing energy for photosynthesis, light provides other important cues to regulate a range of developmental and physiological responses in plants.
24.2K
01:28Methods of Classification and Identification
181
Bacterial identification relies on a diverse array of techniques to classify and understand microorganisms, each tailored to uncover specific characteristics. Traditional morphological approaches, while still valuable, are limited for closely related or structurally simple organisms. Modern methods integrate biochemical, serological, genetic, and advanced molecular tools to achieve greater accuracy.Morphological and Biochemical TechniquesMorphological characteristics, such as cell shape and...
181