摘要
含水率影响着花生的质量、储藏时长与出油率。本研究针对当前花生含水率测量效率低、有损检测、无法适应大规模检测等问题,探索基于高光谱成像技术的花生含水率无损快速检测方法。测量并建立了300份不同种类花生的高光谱原始图像及光谱数据集,并利用小波变换、多元散射校正(MSC)和一阶导数对数据进行预处理,结合PLS、XGBoost、BO-XGBoost算法建立花生含水量无损检测模型。通过实验对比得出,利用小波变换对原始光谱数据进行预处理后的光谱数据建立的BO-XGBoost模型最优,预测模型决定系数R^(2)=0.9539,均方根误差RMSE=0.8065。实验表明,高光谱成像技术结合BO-XGBoost能够对花生含水率进行快速、准确、无损检测,且对其他农作物水分含量检测具有一定的借鉴意义。
Moisture content affects the quality,storage time and oil yield of peanuts.In view of the inefficiency of peanut moisture content measurement,the destructive detection and the inability to adapt to large-scale detection,this paper studied a nondestructive rapid detection method of peanut water content based on near infrared hyperspectral technique.Three hundred near infrared hyperspectral original images and spectral datasets of different peanut species were measured and established.The data were preprocessed by using wavelet transform,multivariate scatter correction(MSC)and first-order difference.A nondestructive detection model of peanut water content was established by combining PLS,XGBoost and BO-XGBoost algorithms.Through experimental comparison,the BO-XGBoost model established by pre-processing the original spectral data with the wavelet transform was the best,with the model determinant=0.9539 and the root mean square error RMSE=0.8065.The experimental results revealed that the near infrared hyperspectral technology combined with BO-XGBoost can detect the water content of peanut quickly,accurately and non-destructively,and has some reference significance for other crops.
作者
黄琦
沈建国
蒋敏兰
冯昌广
方小生
张长江
石小威
Huang Qi;Shen Jianguo;Jiang Minlan;Feng Changguang;Fang Xiaosheng;Zhang Changjiang;Shi Xiaowei(School of Physics and Electronic Information Engineering,Zhejiang Normal University,Jinhua321004;School of Electronic and Information Engineering,Taizhou University,Taizhou318000;Hangzhou Hikvision Digital Technology Co.,Ltd.,Hangzhou310000)
出处
《中国粮油学报》
CSCD
北大核心
2023年第5期135-140,共6页
Journal of the Chinese Cereals and Oils Association
基金
国家自然科学基金(42075140)
浙江省教育厅一般科研项目(Y202045743)
浙江省大学生新苗人才计划(2021R404065)。
