基于高光谱成像技术的红薯缺陷检测与分析

摘要：为了快速识别缺陷红薯，本研究利用高光谱成像技术对健康、冻伤和病害红薯进行分类。基于联合 x-y 距离的样 本集划分（（ SPXY）算法，按照3:1的比例划分训练集和预测集，然后对光谱数据进行标准正态变量（（ SNV）预处理。为 提高模型的运行速度和消除冗余变量分别采用蒙特卡洛无信息变量剔除法（（ MCUVE）、随机（（ RF）算法和连续投影算法 （（SPA）提取特征波长。最后采用偏最小二乘判别分析（（ PLS-DA）和线性判别分析（（ LDA）建立分类模型结果表明，SPA- LDA模型的分类性能最好，预测集总准确率达到99.52%。高光谱成像技术可以实现红薯缺陷的准确检测，为红薯的自动 分类提供了可行的思路。

关键词：红薯，品质，高光谱成像，缺陷；

期刊：Infrared Physics and Technology

DOI：https://DOI.org/10.1016/j.infrared.2022.104403

设备：GaiaSorter

Detection and analysis of sweet potato defects based on hyperspectral imaging technology

Abstract：In order to identify the defective sweet potato quickly, this study used hyperspectral imaging technology to classify healthy, frostbitten and diseased sweet potatoes. The training set and prediction set were divided ac cording to the ratio of 3:1 by sample set partitioning based on joint x-y distances (SPXY) algorithm, and then the standard normal variable (SNV) pretreatment was carried out for the spectral data. In order to improve the running speed of the model and eliminate redundant variables, Monte Carlo uninformative variables elimination (MCUVE), Random frog (RF) algorithm and successive projections algorithm (SPA) were used to extract char acteristic wavelengths. Finally, partial least square discrimination analysis (PLS-DA) and linear discrimination analysis (LDA) were used to establish classification models. The final results showed that the classification performance of the SPA-LDA model was the best, and the total accuracy of the prediction set reached 99.52%.hyperspectral imaging technology can accomplish the accurate detection of sweet potato defects, and provides feasible ideas for the automatic classification of sweet potatoes.

Keywords：Sweet potato Quality Hyperspectral imaging Defects