| 摘要: |
| 为深入系统地了解近红外光谱技术（Near-infrared spectroscopy，NIRS）在农业废弃物堆肥研究中的研究进展与应用现状，基于Web of Science核心数据库与CNKI数据库，以“近红外光谱”“农业废弃物”“堆肥”和“好氧发酵”等为关键词进行检索，共计筛选出58篇相关文献，并从堆肥基础特性检测、过程监测和质量评估等3个方面对现有研究工作进行归纳总结。结果表明：1）在堆肥基础特性检测中，提高模型精度需增加样本、提取相关波段和适配更多算法；2）NIRS对堆肥进行过程监控实现精细化管理需优化通用模型、加强硬件开发；3）进一步完善堆肥NIRS评价系统可通过迁移学习、多特性同步预测和质量分级提高评价的准确性与可靠性。未来NIRS应与机器学习、深度学习、计算机视觉和高光谱成像等新技术融合，从而为农业领域数据密集型科学创造新的机遇，为农业废弃物堆肥的现场监测与质量控制提供参考。 |
| 关键词: 近红外光谱技术 农业废弃物 堆肥 过程监测 质量评估 |
| DOI：10.11841/j.issn.1007-4333.2024.08.15 |
| 投稿时间：2023-12-16 |
| 基金项目:国家重点研发计划（2022YFD2002103）；国家现代农业产业技术体系资助（CARS36）；教育部“创新团队发展技术”项目（IRT1293） |
|
| Research progress on NIRS-based detection of agricultural waste composting |
|
LUO Li1, SUN Huiqi1, SHI Su’an1*, ZHANG Hongmei2,3, YANG Zengling1, HAN Lujia1
|
| (1.College of Engineering, China Agricultural University, Beijing 100083, China;2.College of Mechanical and Electrical Engineering, Tarim University, Alar 843300, China;3.Modern Agricultural Engineering Key Laboratory at Universities of Education Department of Xinjiang Uygur Autonomous Region, Alar 843300, China) |
| Abstract: |
| In order to gain an in-depth and systematic understanding of the research advances and current applications of near-infrared spectroscopy （NIRS） in the study of agricultural waste composting， a literature search was conducted based on the Web of Science core database and the CNKI database. ‘Near-infrared spectroscopy’，‘agricultural waste’，‘composting’ and ‘aerobic fermentation’ were used as keywords. A total of 58 relevant literature articles were screened through this search. Subsequently， a synthesis and summary of existing research work were undertaken， which focused on three aspects of composting fundamental characteristic detection， process monitoring and quality assessment. The results are as follows： 1）To improve the accuracy of models in detecting basic composting features， it’s important to increase the number of samples， identify relevant spectral bands， and adapt different algorithms； 2）Effective real-time monitoring of the composting process using NIRS requires refining universal models and improving hardware to ensure greater sensitivity and reliability； 3）Further improvements in the NIRS evaluation system for composting can be achieved through methods like transfer learning， predicting multiple characteristics simultaneously， and grading quality. These contribute to better accuracy and reliability in assessments. Looking ahead， the integration of NIRS with emerging technologies such as machine learning， deep learning， computer vision and hyperspectral imaging is expected to create new opportunities for data-intensive scientific exploration in agriculture and provide practical guidance for on-site monitoring and quality control in agricultural waste composting. |
| Key words: near-infrared spectroscopy agricultural waste compost process monitoring quality assessment |
