新型混合转移神经网络用于小麦作物生长阶段识别,使用现场图像进行识别
Aisha Naseer1, Madiha Amjad2, Ali Raza3
1Institute of Information Technology, Khwaja Fareed University of Engineering and Information Technology, Rahim Yar Khan, 64200, Pakistan.
Scientific reports
|April 8, 2025
概括
本研究介绍了MobDenNet,这是一种混合深度学习模型,用于准确识别小麦生长阶段. 该模型获得了99%的F1得分,有望提高精准农业和农业生产率.
科学领域:
- 农业科学 农业科学
- 计算机科学 计算机科学
- 机器学习 机器学习
背景情况:
- 准确识别小麦生长阶段对于精准农业和优化作物产量至关重要.
- 现有的方法在区分小麦发育阶段方面面临挑战,影响农业效率.
- 这项研究解决了对可靠的小麦作物阶段识别的需求.
研究的目的:
- 开发和评估一种创新的方法,MobDenNet,用于实时识别小麦作物阶段.
- 将MobDenNet的性能与已建立的深度学习和转移学习模型进行比较.
- 通过改进作物管理决策,提高农业生产率.
主要方法:
- 在七个生长阶段收集了4496张小麦图像的多样化数据集.
- 应用严格的预处理和数据增强技术.
- 开发并测试了一种混合深度学习模型,MobDenNet,合并了MobileNetV2和DenseNet-121架构.
- 我们将MobDenNet与MobileNetV2,DenseNet-121,NASNet-Large,InceptionV3和一个CNN进行了比较.
主要成果:
- 混合MobDenNet模型实现了99%的精度,回忆和F1得分.
- 个别模型的准确性各不相同:MobileNetV2 (95%),DenseNet-121 (94%),NASNet-Large (76%),InceptionV3 (74%) 和CNN (68%),这些模型的准确性各不相同.
- K-fold交叉验证证实了MobDenNet方法的稳定性.
结论:
- 该MobDenNet模型为小麦生长阶段检测提供了一个高度准确的解决方案.
- 这项技术在提高农业生产率和优化资源管理方面具有重大潜力.
- 赋予农民以数据驱动的洞察力,以便在小麦农业中做出明智的决策.
相关概念视频
02:16Light Acquisition
8.4K
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.
8.4K
01:59Plant Breeding and Biotechnology
18.5K
Crop cultivation has a long history in human civilization, with records showing the cultivation of cereal plants beginning at around 8000 BC. This early plant breeding was developed primarily to provide a steady supply of food.
18.5K