Study on Strawberry Sorting System with Precooling Function

具有预冷功能的草莓分选系统的研究

• 批准号: 08456129
• 负责人: NAGATA Masateru
• 金额: $ 4.93万
• 依托单位: Miyazaki University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B)
• 财政年份: 1996
• 资助国家: 日本
• 起止时间: 1996 至 1997
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-08456129/
• 关键词: strawberry; Sorting Machine; image proceing; neural network; automatic; mechanizating; post-harvesting; 自動選別

Production of fresh strawberry in greenhouses is very high earnings. After harvesting, strawberry is sorted into three grades based on shape and five grades based on size. This process is tedious and takes long working hours. Moreover, given its soft and sensitive/delicate surface, a slight pressure rendered from sorting, in addition to harvesting could cause damages to the fruit, lowering its commercial value. Hence, amechanized system of sorting is deemed necessary.This research the advances on the development of strawberry sorting system. Machine vision technique based on image processing techology and multiple-valued neural network theory (MVNN) were utilized in the design of the system and the controlling computer software.1.Study on strawberry sorting system(1)Development of sorting mechanismThis prototype consists of the feeding conveying, image processing, judging, and separating units. The circular pans are fed to the conveyor belt from a hollow cylinder located at the extreme … More end. When the strawberry comes just below the CCD camera, the program gets actuated to judge its shape. The separation of judged strawberries is camied out by the air cylinders, moving across the belt. The strawberries are separated into A,B and C channels within width of the belt.(2)Development of software for grade judging of strawberryComputer software based on the multiple-valued neural network. The sorting system must be able to recognize a strawberry when it "sees" one and must decide what grade or class the fruit should belong based on its shape and size. This is essentially the function which the judging component has to accomplish by employing an appropriate computer software. Percent accuracy ranges from 83 to 95%, very close to human judging ability.2.Study on precooling system for keeping qualityQuality of strawberry after harvesting and sorting could kept by precooling of 1ﾟC and controlling of CO_2 It was recognized a hardness of strawberry surface increased by precooling system with CO_2.This system was useful for decreasing a damage of strawberry fruits. Less

在温室中生产新鲜草莓的收益非常高。收获后，草莓按形状分为三等，按大小分为五等。这个过程很繁琐，需要很长时间的工作。此外，由于其表面柔软而敏感/细腻，除了收获外，在分拣过程中产生的轻微压力可能会对水果造成损害，从而降低其商业价值。因此，一种机械化的分类系统被认为是必要的。本文对草莓分选系统的发展进行了研究。基于图像处理技术和多值神经网络理论（MVNN）的机器视觉技术在系统设计和控制计算机软件中得到了应用。草莓分选系统的研究(1)分选机构的研制本样机由进料输送、图像处理、判断、分离四个单元组成。圆形平底锅从位于最末端的空心圆筒送入传送带。当草莓出现在CCD相机下方时，程序就会启动来判断它的形状。通过在传送带上移动的气缸来完成对草莓的分离。草莓在传送带的宽度内被分成A、B和C三个通道。(2)草莓品级判定软件的开发——基于多值神经网络的计算机软件。分拣系统必须能够在“看到”一颗草莓时就识别出来，并且必须根据草莓的形状和大小来决定它应该属于哪个等级或类别。这基本上是判断组件必须通过使用适当的计算机软件来完成的功能。准确率在83%到95%之间，非常接近人类的判断能力。预冷系统对草莓收获分选后品质保持的研究1‰的预冷和CO_2的控制可以保持草莓收获分选后的品质，CO_2预冷系统可以提高草莓表面硬度。该系统可有效降低草莓果实的危害。少

项目成果

NAGATA, M.: "Basic Stydy on Strawberry Sorting system in Japan" Proceedings of the 1997 ASAE Anual International Meeting. 973095 (1997)

NAGATA, M.：“日本草莓分选系统的基本研究”1997 年 ASAE 国际年会记录。

永田雅輝他: "イチゴの選別システムに関する研究-ベルト式による選別機構の開発-" 第56回農業機械学会年次大会講演要旨. 17-18 (1997)

Masaki Nagata等：“草莓分选系统的研究-带式分选机构的开发-”第56届日本农业机械学会年会摘要17-18（1997）。

曹其新・永田雅輝他: "マシンビジョンによる果菜の等級判定に関する研究(第2報)" 植物工場学会誌. 8巻4号. 228-236 (1996)

曹启新，Masaki Nagata，等：“利用机器视觉进行水果和蔬菜分级的研究（第二次报告）”植物工厂学会杂志，第8卷，第4期。228-236（1996）。

Cao, Q., Nagata.M., et.al.: "An Approach to Profile Pattern Judgment of Strawberry Imege by Using Neurai Network" Journal of Kyushu Branch of the Japanese society of Agricultural Machinery. Vol.45. 26-30 (1996)

曹Q.，Nagata.M.，等：“利用神经网络进行草莓图像轮廓图案判断的方法”，日本农业机械学会九州分会学报。

M.Nagata et.al: "Basic Stydy on Strawberry Sorting system in Japan" Proceedings of ASAE,. (Paper No.973095). (1997)

M.Nagata 等人：“日本草莓分选系统的基本研究”ASAE 论文集，。