探索基于高光谱成像和转移学习预测蓝溶性固体含量的通用模型,以应对空间异质性挑战
Guoliang Chen1, Mianqing Yang1, Guozheng Wang1
1College of Computer and Control Engineering, Northeast Forestry University, Harbin, Heilongjiang 150040, China.
Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy
|February 22, 2025
概括
这项研究开发了一种使用超光谱成像,残余多层感知子和转移学习的通用模型,以准确预测蓝的可溶性固体含量 (SSC),尽管果实放置和生物差异的变化.
科学领域:
- 农业科学 农业科学
- 频谱学是一种光谱学.
- 机器学习 机器学习
背景情况:
- 准确的可溶性固体含量 (SSC) 评估对于蓝的质量至关重要.
- 生物异质性和生产线中的随机放置扭曲了光谱数据,影响了SSC预测模型.
- 需要强大的建模来克服这些挑战在现实世界的场景.
研究的目的:
- 开发一种通用模型,用于预测蓝SSC在各种安置情况.
- 为了最大限度地减少由蓝部分异质性引起的光谱扭曲的影响.
- 提高SSC预测模型的概括能力和稳定性.
主要方法:
- 从1150个蓝样本获得的超谱成像 (HSI) 数据和SSC值.
- 利用剩余的多层感知子,为不同的蓝表面构建本地SSC预测模型.
- 应用转移学习微调本地模型,创建一个通用预测模型.
主要成果:
- 优化的模型在不同表面上显著提高了预测准确度.
- 基于赤道表面数据 (增强-赤道-1) 的模型表现出色.
- 外部验证产生了高的预测相关系数 (rp > 0.92) 和低误差 (RMSEP < 0.62%),其中RPD值表明模型可靠性良好.
结论:
- 结合剩余的多层感知和转移学习,有效地减轻了蓝部分异质性对SSC预测的影响.
- 开发的方法增强了对生物变异的模型稳定性.
- 这种方法提高了在各种生产线条件下蓝SSC检测的准确性.
更多相关视频
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
01:32UV–Vis Spectroscopy of Conjugated Systems
6.8K
Organic compounds with conjugated double bonds show strong absorption features in the UV–visible region of the electromagnetic spectrum attributed to π → π* electronic excitations. Generally, a UV–vis absorption spectrum is recorded as a plot of absorbance vs wavelength. The wavelength of maximum absorbance, which manifests as a peak in the absorption spectrum, is denoted as λmax.
One of the factors influencing λmax is the extent...
One of the factors influencing λmax is the extent...
6.8K