欢迎访问《农学学报》,

农学学报 ›› 2025, Vol. 15 ›› Issue (1): 75-80.doi: 10.11923/j.issn.2095-4050.cjas2024-0148

• 农业工程 农业机械 生物技术 食品科学 • 上一篇    下一篇

基于综合因子载荷得分法的烤烟群体等级分级模型研究

蓝周焕1(), 杨美林1, 童德文1, 石三三1, 林博雅1, 陈添昌1, 王旭2, 江海东3()   

  1. 1 龙岩市烟草公司武平分公司,福建武平 364300
    2 山东中烟工业有限责任公司物资采购中心,济南 250000
    3 南京农业大学,南京 210095
  • 收稿日期:2024-07-24 修回日期:2024-09-03 出版日期:2025-01-20 发布日期:2025-01-13
  • 通讯作者:
    江海东,男,1968年出生,江苏江阴人,副教授,博士,主要从事作物栽培与作物信息技术研究。通信地址:210095 江苏省南京市卫岗1号南京农业大学,Tel:025-84395713,E-mail:
  • 作者简介:

    蓝周焕,男,1971年出生,福建上杭人,农艺师,学士,主要从事烟叶生产管理。通信地址:364300 福建省龙岩市武平县平川镇教育路1号 龙岩市烟草公司武平分公司,Tel:0597-4822399,E-mail:

  • 基金资助:
    福建省烟草公司龙岩市公司科技项目“初烤烟叶全杆智能等级分级系统的研发”(LK-202309)

Research on Tobacco Plant Population Grade Classification Model Based on Comprehensive Factor Loading Score Method

LAN Zhouhuan1(), YANG Meilin1, TONG Dewen1, SHI Sansan1, LIN Boya1, CHEN Tianchang1, WANG Xu2, JIANG Haidong3()   

  1. 1 Wuping Branch of Longyan Tobacco Company, Wuping Fujian 364300
    2 Material Purchasing Center, China Tobacco Shandong Industry Company Limited, Jinan 250000
    3 Nanjing Agricultural University, Nanjing 210095
  • Received:2024-07-24 Revised:2024-09-03 Online:2025-01-20 Published:2025-01-13

摘要:

为解决现有烤烟智能分级模型效率较低的问题,以烤烟群体数码图像为处理对象,以不同等级烤烟群体RGB颜色模型偏态参数、Lab颜色模型参数、HSV颜色模型参数、叶面纹理参数等4类31个表型参数为输入变量,构建基于贝叶斯分类算法的烤烟群体等级分级模型F1。在此基础上,提出并验证以综合因子载荷得分法获得的核心参数作为输入变量所构建的分级模型F2,F2整体准确度达到了82.24%,较F1模型提升了12.82%,且5个等级判定准确度都超过了70%,可为高效、实用烤烟智能化收购系统的开发提供应用基础理论。

关键词: 因子分析, 数码图像复合表型, 贝叶斯分类, 烤烟等级, 群体分级, 智能分级模型

Abstract:

To address the issue of low efficiency in existing intelligent tobacco grading models, this study focused on digital images of tobacco plant populations. Taking 31 phenotypic parameters from four main categories, including RGB color model skewness parameters, Lab color model parameters, HSV color model parameters, and leaf texture parameters, as input variables, a tobacco plant population grade classification model F1 based on the Bayesian classification algorithm was constructed. Furthermore, a core parameter-based classification model F2, utilizing the comprehensive factor loading score method, was proposed and verified. The overall accuracy of model F2 reached 82.24%, representing a 12.82% improvement compared to model F1, and the accuracy of all five grade judgments exceeded 70%. This study provides an applied theoretical basis for the development of an efficient and practical intelligent tobacco purchasing system.

Key words: factor analysis, digital image composite phenotype, Bayesian classification, flue-cured tobacco grade, population grading, intelligent grading model