基于激光图像的苹果品质分析与模型  

Modeling apple quality changes based on laser scattering image analysis

作　　者：陈育彦[1] 屠康[1] 任珂[1] 邵兴锋[1] 董庆利[1] 潘磊庆[1] 

机构地区：[1]南京农业大学食品科技学院,南京210095

出　　处：《农业工程学报》2007年第4期166-171,共6页Transactions of the Chinese Society of Agricultural Engineering

基　　金：教育部"新世纪优秀人才支持计划"资助(NCET05-0491)

摘　　要：该试验利用波长650 nm,功率25 mW的半导体激光及计算机视觉技术,初步探讨了激光图像的影响因素以及利用激光图像对苹果(嘎拉、红富士)品质进行无损检测的可行性。分别测定贮藏期间苹果的硬度、固酸比、果面底色、果型等品质指标,并获取图像参数及其像素个数(S1、S2、S3、S4)。试验结果显示,苹果的向阳面和背阳面对获取的激光图像的像素个数有显著的影响,而果型大小对其无显著影响。其中图像参数S3与各品质指标的皮尔逊相关系数最高。逐步回归表明,引入变量S3后模型的解释率最好(最高R2为0.99),且都为极显著水平(P≤0.01)。试验的结果为利用小功率激光图像对苹果品质进行无损检测提供了理论依据。650 nm, 25 mW semiconductor laser beam was applied as light source, and the quality of apples （c. v. Red Fuji and c.v. Gala） was monitored with a computer vision system. The laser scattering images of apples were analyzed. The maximum puncture force, CIE a value, hue angle, SSC/TA（soluble solid content/titrable acidity） ratio, and laser image parameters （S1, S2, S3, S4） were measured during the storage of apples. The results indicated that different sizes of the apples had no significant effects on the total number of pixels in the scattering image, while different sides （towards or against the sun） of apples made significant effects on the pixel numbers （P ≤ 0.05）. It was found that image parameter S3 had the highest Pearson correlations with other apple quality parameters. The stepwise regression analysis showed that the models had the best fitting （R^2=0.99） when the parameter S3 was used in the model. This trial laid a theoretical basis for applying low power laser beam to monitor apples＇ ripeness and quality in a nondestructive way.

关 键 词：激光 计算机视觉 苹果 品质 图像 

分 类 号：S123[农业科学—农业基础科学] S661