将光谱转换为图像可通过2D-CNN改进植物性状检索
Turning spectra into images improves plant trait retrieval with 2D-CNNs
浏览论文内容
中文总结 AI 辅助
本研究将一维光谱转为二维图像，用EfficientNet-B0在GreenHyperSpectra数据集上实验，发现直接重塑为二维网格的方法提升了植物多性状预测精度，且2D光谱图像的表征优势是性能提升的关键。
中文摘要 AI 辅助
高光谱反射光谱技术可实现植物功能性状的非破坏性估算，但当前深度学习方法将光谱作为一维序列处理，限制了其捕获长程波段间依赖关系的能力。本研究探究将一维光谱转换为二维图像表示是否可通过卷积神经网络（CNN）改进多性状预测。我们在GreenHyperSpectra数据集（含7897条标注光谱、8个性状，波长范围400-2450nm）上使用EfficientNet-B0对比了9种转换方式，以同一划分下已发表的一维CNN结果为基准。从零开始训练时，最简单的转换方式——将光谱直接重塑为二维网格，表现最佳（R²=0.684±0.001），且优于最先进的一维基线（R²=0.587，提升0.097）。随后，我们在139000张未标注光谱图像上预训练二维掩码自编码器（MAE-2D），冻结编码器仅训练多层感知机头的线性探测达到R²=0.646，超过所有一维自监督对应模型，包括微调后的MAE-1D（R²=0.641）。跨数据集评估下，所有模型的精度大幅下降，无模型显著优于一维基线。为识别驱动各预测的波长，我们应用了集成梯度法和Grad-CAM，将波段重要性展开回光谱轴。蛋白质（r=0.45）和叶片水分（r=0.33）与PROSAIL辐射传输模型模拟的敏感性一致，而类胡萝卜素（r=0.06）和叶面积指数（r=-0.11）则不一致，表明模型读取具有尖锐吸收特征的性状的已确定叶片化学信息。2D光谱图像的表征优势，而非架构复杂性或ImageNet预训练，驱动了其相较于一维方法的性能提升。
英文摘要
Hyperspectral reflectance spectroscopy enables non-destructive estimation of plant functional traits, yet current deep learning approaches process spectra as one-dimensional sequences, which limits how they capture long-range inter-band dependencies. We asked whether transforming 1D spectra into 2D image representations improves multi-trait prediction with convolutional neural networks (CNN). We compared nine transformations using EfficientNet-B0 on the GreenHyperSpectra dataset (7,897 labeled spectra, eight traits, 400-2450 nm), benchmarked against published 1D CNN results on the same split. Trained from scratch, the simplest transformation, a direct Reshape of the spectrum into a 2D grid, performed best ($R^2 = 0.684 \pm 0.001$) and improved on the state-of-the-art 1D baseline ($R^2 = 0.587$, $+0.097$). We then pretrained a 2D masked autoencoder (MAE-2D) on 139,000 unlabeled spectral images. Linear probing, which freezes the encoder and trains only a multilayer perceptron head, reached $R^2 = 0.646$ and exceeded every 1D self-supervised counterpart, including the fine-tuned MAE-1D ($R^2 = 0.641$). Under cross-dataset evaluation all models lost most of their accuracy and none beat the 1D baseline significantly. To identify which wavelengths drive each prediction, we applied Integrated Gradients and Grad-CAM and unfolded band importance back to the spectral axis. Protein ($r = 0.45$) and leaf water ($r = 0.33$) agreed with sensitivities simulated by the PROSAIL radiative-transfer model, while carotenoids ($r = 0.06$) and leaf area index ($r = -0.11$) did not, showing that the model reads established leaf chemistry for traits with sharp absorption features. The representational advantage of 2D spectral images, rather than architectural complexity or ImageNet pretraining, drives the gain over 1D approaches.
发表机构
- Adolfo Ibáñez University（阿道夫·伊瓦涅斯大学）
- University of Chile（智利大学）
- University of Freiburg（弗莱堡大学）
- Leipzig University（莱比锡大学）
机构由 AI 辅助整理，请以论文原文为准。
