基于高光谱成像技术的哈密瓜成熟度快速无损检测方法研究

项目名称： 基于高光谱成像技术的哈密瓜成熟度快速无损检测方法研究

项目编号： No.61263041

项目类型： 地区科学基金项目

立项/批准年度： 2013

项目学科： 自动化技术、计算机技术

项目作者： 马本学

作者单位： 石河子大学

项目金额： 45万元

中文摘要： 针对目前哈密瓜成熟度判别主要依据人的感官评定为主，判别标准不统一且对糖度、坚实度检测室有损检测的问题，提出一种基于高光谱成像技术的哈密瓜成熟度快速无损检测新方法。项目拟开展以下研究：根据哈密瓜分光反射特性建立适于哈密瓜成熟度无损检测的高光谱成像系统，采集哈密瓜高光谱图像，研究哈密瓜糖度、颜色、坚实度的光谱图像特性，揭示哈密瓜成熟度与糖度、颜色和坚实度之间内在机理，选择新疆典型哈密瓜产地的不同成熟度（过早采收、适宜采收、过晚采收）的哈密瓜为试验样本，对获取的高光谱图像数据进行特征提取，建立哈密瓜糖度、颜色、坚实度预测模型；对光谱信息和图像信息进行信息融合，建立哈密瓜成熟度最优判别模型，为确定哈密瓜的最佳采收期、保证其品质提供科学依据。上述研究方法和结论对进一步研究开发具有自主知识产权的基于多光谱成像的哈密瓜成熟度便携式检测仪提供理论基础，且可以借鉴于其它同类水果研究。

中文关键词： 高光谱成像；哈密瓜；成熟度；无损检测；方法

英文摘要： In view of the problem of criterion were uniformed and the detection of Hami cantaloupe's sugar content、firmness are detrimental, and the distinguishing the maturity of Hami cantaloupe was mainly base on human sensory evaluation，a new nondestructive testing method for maturity of Hami cantaloupe based on hyperspectral imaging has been proposed. The following research will be carried out：establish the hyperspectral imaging system for the nondestructive testing of maturity based on the spectral reflection characteristics of Hami cantaloupe;collect Hami cantaloupes' hyperspectral images in order to study the characteristics of the spectral image with sugar content, color, firmness; reveal the inherent mechanism of maturity with sugar content, color and firmness；establish the prediction model of sugar content, color and firmness through extracting the feature of hyperspectral image data by select the different maturity (premature harvesting, appropriate harvest, late harvest) as test samples that were planted in Xinjiang; establish the excellent discriminant model of Hami cantaloupes' maturity by the merge of the spectral information and image information;provide a scientific basis for determining the optimal harvest time of melon. The above-mentioned methods and conclusions provide a theoretical basis for further r

英文关键词： Hyperspectral imaging；Hami melon；Maturity；Non-destructive detection；Method