考虑类别概率信息的基于图的半监督高光谱成像番茄成熟度鉴别

摘要：准确、无损地鉴别番茄的成熟度对番茄的收获和保存具有重要意义。由于中间相邻成熟度阶段之间的颜色差异不 大，获取大量精确的成熟度标签非常耗时，仅靠视觉特征难以区分多个成熟度阶段。本研究设计了一种基于高光谱成像 技术的半监督方法，利用少量标记样本进行番茄成熟度鉴别。首先，利用高光谱成像系统提取番茄样品的高光谱数据， 并通过多重散射校正算法进行预处理。然后，通过标记样本的稀疏编码来描述未标记样本的类别概率信息。接下来，设 计了一种基于拉普拉斯得分和光谱信息散度（（ SIDLS）的半监督算法，利用类别概率信息构建图，以选择具有代表性的波 段子集。最后，建立基于类别概率信息（（ CSR）的稀疏表示模型，构建连接图，并采用标签传播算法对番茄成熟度进行判 别。实验结果表明，SIDLS 算法在特征选择上优于基于拉普拉斯评分（SSLS）和半费舍尔评分（SFS）算法的半监督算 法，CSR模型在构建更具判别性的图方面也优于其他图构建方法。该方法的判别准确率可达96.78%，在番茄成熟度判别 中具有良好的应用前景。

关键词：番茄成熟度判别 .半监督学习 .特征选择 .类别概率信息 .稀疏表示 .高光谱成像

期刊：Food Analytical Methods

DOI：https://DOI.org/10.1007/s12161-020-01955-5

设备：GaiaField

Discrimination of Tomato Maturity Using Hyperspectral Imaging Combined with Graph-Based Semi-supervised Method Considering Class Probability Information

Abstract：Accurate and nondestructive maturity discrimination of tomatoes is significant for harvest and preservation. Color differences between intermediate adjacent maturity stages are not significant, so it is time-consuming to obtain a large amount of precise maturity labels and difficult to discriminate multiple maturity stages by visual features merely. This study designed a semi supervised method based on hyperspectral imaging technology for tomato maturity discrimination by using a small amount of labeled samples. Firstly, the hyperspectral data of tomato samples was extracted by the hyperspectral imaging system and pre processed by the multiple scattering correction algorithm. Then, the class probability information of unlabeled samples was described by the sparse coding of labeled samples. Next, a semi-supervised algorithm based on Laplacian score and spectral information divergence (SIDLS), which used class probability information to construct graphs, was designed to select a repre sentative waveband subset. Finally, the sparse representation model based on class probability information (CSR) was established to construct a connection graph, and label propagation algorithm was used to discriminate tomato maturity. Experimental results demonstrated that SIDLS algorithm had an advantage in the feature selection over semi-supervised algo rithm based onLaplacianscore (SSLS) and Semi_Fisher Score(SFS) algorithm,and CSRmodelwas alsosuperiortoother graph construction methods in constructing a more discriminative graph. The discrimination accuracy of this method could reach 96.78%, which shows a promising prospect in tomato maturity discrimination.

Keywords：Tomatomaturitydiscrimination .Semi-supervisedlearning .Featureselection .Classprobabilityinformation .Sparse representation . Hyperspectral imaging