Li Zeng-fang, Chu Bing-quan, Zhang Hai-liang, He Yong, Liu Xue-mei, Luo Wei
Guang Pu Xue Yu Guang Pu Fen Xi. 2016 Dec;36(12):4034-8.
Hyperspectral imaging technology is a rapid, non-destructive, and non-contact technique which integrates spectroscopy and digital imaging to simultaneously obtain spectral and spatial information. Hyperspectral images are made up of hundreds of contiguous wavebands for each spatial position of a sample studied and each pixel in an image contains the spectrum for that specific position. With hyperspectral imaging, a spectrum for each pixel can be obtained and a gray scale image for each narrow band can be acquired, enabling this system to reflect componential and constructional characteristics of an object and their spatial distributions. In this study, hyperspectral image technology is used to discuss the application of hyperspectral imaging detection technology of Jiangxi navel orange surface of different concentrations of pesticide residue changes with time relationship. The pesticide was diluted to 1 : 20, 1 : 100 and 1 : 1 000 solution with distilled water. A 1×2 matrix of dilutions was applied to each of 30 cleaned samples with different density pesticide residue. After 0, 4 and 20 d respectively, hyperspectral images in the wavelength range from 900 to 1 700 nm are taken. The characteristic wavelengths are achieved by using principal component analysis (PCA) and the PC-2 image based on PCA using characteristic wavelengths (930, 980, 1 100, 1 210, 1 300, 1 400, 1 620 and 1 680 nm) as the classification and recognition of image. Based on these 8 characteristic wavelengths for a second principal component analysis, the application of PC-2 image and appropriate image processing methods for different concentrations and different days of placing pesticide residues in non-destructive testing were applied. Using hyperspectral imaging technology to detect three periods a higher dilution of the fruit surface pesticide residues are more obvious. This research shows that the technology of hyperspectral imaging can be used to effectively detect pesticide residue on Gannan navel surface.
高光谱成像技术是一种快速、无损且非接触的技术，它将光谱学与数字成像相结合，以同时获取光谱和空间信息。高光谱图像由所研究样本的每个空间位置的数百个连续波段组成，图像中的每个像素都包含该特定位置的光谱。通过高光谱成像，可以获得每个像素的光谱，并获取每个窄带的灰度图像，使该系统能够反映物体的成分和结构特征及其空间分布。在本研究中，利用高光谱图像技术探讨了江西脐橙表面不同浓度农药残留随时间变化关系的高光谱成像检测技术的应用。将农药用蒸馏水稀释成1:20、1:100和1:1000的溶液。将1×2的稀释矩阵应用于30个具有不同密度农药残留的清洁样本中的每一个。分别在0、4和20天后，拍摄900至1700nm波长范围内的高光谱图像。通过主成分分析（PCA）获得特征波长，并基于PCA使用特征波长（930、980、1100、1210、1300、1400、1620和1680nm）作为图像的分类和识别得到PC-2图像。基于这8个特征波长进行第二次主成分分析，应用PC-2图像和适当的图像处理方法对不同浓度和不同放置天数的农药残留进行无损检测。利用高光谱成像技术检测三个时期发现，稀释度越高，果实表面农药残留越明显。本研究表明，高光谱成像技术可有效检测赣南脐橙表面的农药残留。
