欢迎访问《中国农学通报》，
中国农学通报

中国农学通报 ›› 2025, Vol. 41 ›› Issue (5): 157-164.doi: 10.11924/j.issn.1000-6850.casb2024-0076

• 水产·渔业 • 上一篇    

基于高光谱成像技术的土壤含水量检测研究

马玲1(), 张祎洋1, 李亚娇1, 马思艳1, 王静1, 马燕1, 吴龙国1,2()   

  1. 1 宁夏大学葡萄酒与园艺学院，银川 750001
    2 宁夏现代设施园艺工程技术研究中心，银川 750001
  • 收稿日期:2024-02-04 修回日期:2024-10-30 出版日期:2025-02-15 发布日期:2025-02-11
  • 通讯作者:
    吴龙国，男，1988年出生，陕西人，讲师，博士，主要从事设施园艺产品无损检测方面的研究。通信地址：750021 宁夏银川市西夏区贺兰山西路489号 宁夏大学，E-mail：。
  • 作者简介:

    马玲，女，1997年出生，宁夏人，在读硕士研究生，研究方向为设施蔬菜栽培。E-mail：。

  • 基金资助:
    国家重点研发计划子课题专项“集约化育苗系列机械优化”(2021YFD1600302-3); 自治区级自然科学其他项目“红寺堡瓜菜产业提升关键技术集成创新”(2021BBF02019-03); 宁夏回族自治区科技创新团队项目(2024CXTD010)

Research on Soil Water Content Detection Based on Hyperspectral Imaging Technology

MA Ling1(), ZHANG Yiyang1, LI Yajiao1, MA Siyan1, WANG Jing1, MA Yan1, WU Longguo1,2()   

  1. 1 College of Enology and Horticulture, Ningxia University, Yinchuan 750001
    2 Ningxia Modern Protected Horticulture Engineering Technology Research Center, Yinchuan 750001
  • Received:2024-02-04 Revised:2024-10-30 Published:2025-02-15 Online:2025-02-11

RichHTML

5

PDF (PC)

7

摘要：

本研究旨在利用高光谱成像技术快速检测土壤含水量，以实现对番茄植株生长状况的及时监测。通过提取304个土壤样本的平均光谱反射率，采用异常值剔除、样本集划分、3种预处理方法对原始光谱进行预处理和优化处理。研究中运用了连续投影算法(successive projections algorithm，SPA)、无信息变量消除变换法(uninformation variable elimination，UVE)、迭代保留信息变量法(iterative retained information variable，IRIV)、遗传偏最小二乘算法(genetic partial-least-squares algorithm，GAPLS)4种方法提取特征波长，并基于这些波长建立偏最小二乘回归(partial-least-squares regression，PLSR)模型。进一步地，根据优选出的特征波长还构建了多元线性回归(Multiple linear regression，MLR)模型、主成分回归(Principal component regression，PCR)模型和卷积神经网络模型(convolutional neural network，CNN)。结果表明，在经过平均平滑法(moving average smoothing，MAS)对土壤含水量进行预处理后，使用IRIV法提取的特征波长所建立的土壤含水量定量预测模型表现最佳，其校正集相关系数Rc=0.7167，均方根误差RMSEc=0.0193；验证集相关系数Rp=0.6631，RMSEP=0.0272。特别是基于IRIV-CNN组合的模型展现出更优的性能，Rc=0.7655，RMSEc=0.0172。本研究不仅提供了一种有效的土壤水分监测手段，也为提高设施蔬菜产业中的水资源利用效率、促进番茄种植过程中科学合理的水分管理以及实现植株健康状态在线监控提供了强有力的技术支持。

关键词: 番茄, 土壤含水量, 高光谱成像, 检测, 特征波长选择, 番茄植株生长监测

Abstract:

In order to quickly detect the soil water content and achieve timely monitoring of tomato plant growth, the average spectral reflectance of 304 soil samples was extracted using hyperspectral imaging technology. The original spectra underwent preprocessing and optimization by removing outliers, dividing the sample set, applying three preprocessing methods, successive projections algorithm (SPA), uninformation variable elimination (UVE), iterative retained information variable (IRIV), genetic partial-least-squares algorithm (GAPLS) to extract the feature wavelengths. Following this, a partial-least-squares regression (PLSR) model was established based on the identified feature wavelengths. Utilizing these preferred feature wavelengths, multiple models were then constructed, including the PLSR model, multiple linear regression (MLR) model, principal component regression (PCR) model, and convolutional neural network (CNN) model. The results showed that: the preferred moving average smoothing (MAS) preprocessing of soil water content was applied, and the quantitative prediction model of soil water content, which was established using the characteristic wavelength extracted by the IRIV method, proved to have the best effect (Rc=0.7167, RMSEc=0.0193; Rp=0.6631, RMSEP=0.0272). Additionally, the model of soil water content based on IRIV-CNN also demonstrated good performance (Rc=0.7655, RMSEc=0.0172). This study holds great practical significance and usefulness for the development of water utilization efficiency in the facility vegetable industry, as well as the scientific management of water in tomato crops. It currently provides technical support for online monitoring of tomato plant growth.

Key words: tomato, soil moisture content, hyperspectral imaging, detection, feature wavelength selection, tomato plant growth monitoring