Deep learning and machine learning methods based on the NIRS dataset for rapid determination of the nutrients content and quality of oat hay

IF 7.7 1区农林科学 Q1 AGRICULTURE, MULTIDISCIPLINARY

Computers and Electronics in Agriculture Pub Date : 2025-04-22 DOI:10.1016/j.compag.2025.110428

Li Wang , Tao Wang , Luming Dai , Fei Li , Tao Guo , Fadi Li , Zhiyuan Ma , Kaidong Li , Hui Xu , Maimaiti Reshalaitihan

{"title":"Deep learning and machine learning methods based on the NIRS dataset for rapid determination of the nutrients content and quality of oat hay","authors":"Li Wang , Tao Wang , Luming Dai , Fei Li , Tao Guo , Fadi Li , Zhiyuan Ma , Kaidong Li , Hui Xu , Maimaiti Reshalaitihan","doi":"10.1016/j.compag.2025.110428","DOIUrl":null,"url":null,"abstract":"<div><div>Oat hay is characterized by a high content of neutral detergent fiber, elevated sugar levels, and exceptional palatability, rendering it an ideal forage option for ruminant animals. This study investigates the rapid classification of oat hay quality grades under different standards, utilizing a combination of 2DCOS and deep learning methods. The 2DCOS images distinctly exhibit the spectral discrepancies among oat hay of diverse qualities within the 1100–1800 nm range. The deep learning model demonstrated a 100 % accuracy rate in identification under different standards. Moreover, MPLS and SSA-Lasso were employed to predict the contents of dry matter (DM), neutral detergent fiber (NDF), acid detergent fiber (ADF), crude protein (CP), Ash, ether extract (EE), water soluble carbohydrates (WSC), calcium (Ca), phosphorus (P) and kalium (K) in oat hay. The MPLS effectively predicted the content of DM, NDF, ADF, CP, Ash, WSC, Ca, P and K, with an RPD of ≥ 2.00. With an RPD of 2.01, the SSA-Lasso-based EE prediction model produced the best results. The successful outcomes demonstrated that machine learning applied to NIRS data is a suitable method for rapidly verifying the nutrient content and quality of oat hay.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"236 ","pages":"Article 110428"},"PeriodicalIF":7.7000,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers and Electronics in Agriculture","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0168169925005344","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRICULTURE, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

Abstract

Oat hay is characterized by a high content of neutral detergent fiber, elevated sugar levels, and exceptional palatability, rendering it an ideal forage option for ruminant animals. This study investigates the rapid classification of oat hay quality grades under different standards, utilizing a combination of 2DCOS and deep learning methods. The 2DCOS images distinctly exhibit the spectral discrepancies among oat hay of diverse qualities within the 1100–1800 nm range. The deep learning model demonstrated a 100 % accuracy rate in identification under different standards. Moreover, MPLS and SSA-Lasso were employed to predict the contents of dry matter (DM), neutral detergent fiber (NDF), acid detergent fiber (ADF), crude protein (CP), Ash, ether extract (EE), water soluble carbohydrates (WSC), calcium (Ca), phosphorus (P) and kalium (K) in oat hay. The MPLS effectively predicted the content of DM, NDF, ADF, CP, Ash, WSC, Ca, P and K, with an RPD of ≥ 2.00. With an RPD of 2.01, the SSA-Lasso-based EE prediction model produced the best results. The successful outcomes demonstrated that machine learning applied to NIRS data is a suitable method for rapidly verifying the nutrient content and quality of oat hay.

Abstract Image

查看原文本刊更多论文

基于近红外光谱数据集的深度学习和机器学习方法快速测定燕麦干草的营养成分和质量

燕麦干草的特点是中性洗涤纤维含量高，糖含量高，适口性好，是反刍动物理想的饲料选择。本研究利用2DCOS和深度学习相结合的方法，研究了不同标准下燕麦干草质量等级的快速分类。在1100 ~ 1800 nm范围内，不同品质燕麦干草的光谱差异明显。深度学习模型在不同标准下的识别准确率均达到100%。此外，利用MPLS和SSA-Lasso预测了燕麦干草中干物质（DM）、中性洗涤纤维（NDF）、酸性洗涤纤维（ADF）、粗蛋白质（CP）、灰分、粗脂肪（EE）、水溶性碳水化合物（WSC）、钙（Ca）、磷(P)和钾(K)的含量。MPLS能有效预测DM、NDF、ADF、CP、Ash、WSC、Ca、P、K的含量，RPD值≥2.00。基于ssa - lasso的EE预测模型的RPD为2.01，预测结果最好。成功的结果表明，将机器学习应用于近红外光谱数据是快速验证燕麦干草营养成分和质量的合适方法。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Computers and Electronics in Agriculture 工程技术-计算机：跨学科应用

CiteScore

15.30

自引率

14.50%

发文量

800

审稿时长

62 days

期刊介绍： Computers and Electronics in Agriculture provides international coverage of advancements in computer hardware, software, electronic instrumentation, and control systems applied to agricultural challenges. Encompassing agronomy, horticulture, forestry, aquaculture, and animal farming, the journal publishes original papers, reviews, and applications notes. It explores the use of computers and electronics in plant or animal agricultural production, covering topics like agricultural soils, water, pests, controlled environments, and waste. The scope extends to on-farm post-harvest operations and relevant technologies, including artificial intelligence, sensors, machine vision, robotics, networking, and simulation modeling. Its companion journal, Smart Agricultural Technology, continues the focus on smart applications in production agriculture.