Kiwifruit recognition at nighttime using artificial lighting based on machine vision.

Most researches involved so far in kiwifruit harvesting robot suggest the scenario of harvesting in daytime for taking advantage of sunlight. A robot operating at nighttime can overcome the problem of low work efficiency and would help to minimize fruit damage. In addition, artificial lights can be used to ensure constant illumination instead of the variable natural sunlight for image capturing. This paper aims to study the kiwifruit recognition at nighttime using artificial lighting based on machine vision. Firstly, an RGB camera was placed underneath the canopy so that clusters of kiwifruits could be included in the images. Next, the images were segmented using an R-G color model. Finally, a group of image processing conventional methods, such as Canny operator were applied to detect the fruits. The image processing results showed that this capturing method could reduce the background noise and overcome any target overlapping. The experimental results showed that the optimal artificial lighting ranged approximately between 30-50 lx. The developed algorithm detected 88.3% of the fruits successfully. Keywords: Elliptic Hough transform, image capturing method, Kiwifruit, minimal bounding rectangle, optimal illumination intensity DOI: 10.3965/j.ijabe.20150804.1576 Citation: Fu L S, Wang B, Cui Y J, Su S, Gejima Y, Kobayashi T. Kiwifruit recognition at nighttime using artificial lighting based on machine vision. Int J Agric & Biol Eng, 2015; 8(4): 52－59.

[1]  Siddhartha S. Mehta,et al.  Vision-based control of robotic manipulator for citrus harvesting , 2014 .

[2]  Thomas F. Burks,et al.  A Machine Vision Algorithm Combining Adaptive Segmentation and Shape Analysis for Orange Fruit Detection , 2010 .

[3]  Zhou Quan-chen.  Research on the Object Extraction of Kiwifruit Based on Images , 2009 .

[4]  苏帅,et al.  A Method for Separation of Kiwifruit Adjacent Fruits Based on Hough Transformation , 2012 .

[5]  Thomas Rath,et al.  Novel image processing approach for solving the overlapping problem in agriculture , 2013 .

[6]  Wang Diqiong.  An Algorithm and Implementaion for Obtaining Minimum Exterior Rectangle of Image Region , 2004 .

[7]  Zhao Dean,et al.  System Design and Control of an Apple Harvesting Robot , 2020, ArXiv.

[8]  F Universit,et al.  Development and Experiment of End-effector for Kiwifruit Harvesting Robot , 2015 .

[9]  Giovanni Muscato,et al.  A prototype of an orange picking robot: past history, the new robot and experimental results , 2005, Ind. Robot.

[10]  Ying Zhang,et al.  Automatic recognition vision system guided for apple harvesting robot , 2012, Comput. Electr. Eng..

[11]  Kenta Shigematsu,et al.  Evaluation of a strawberry-harvesting robot in a field test , 2010 .

[12]  Liu Bin,et al.  Identification of fruit and branch in natural scenes for citrus harvesting robot using machine vision and support vector machine , 2014 .

[13]  Rory C. Flemmer,et al.  Development of an autonomous kiwifruit picking robot , 2000, 2009 4th International Conference on Autonomous Robots and Agents.

[14]  Takashi Kataoka,et al.  An image processing algorithm for detecting in-line potato tubers without singulation , 2010 .

[15]  Liu Yan-fei.  Study on Frame Type of Kiwifruit , 2012 .

[16]  S. Miller,et al.  Effects of leader pruning on vine architecture, productivity and fruit quality in kiwifruit (Actinidia deliciosa cv. Hayward) , 2001 .

[17]  F Universit.  Recognition of kiwifruit in field based on Adaboost algorithm , 2013 .

[18]  Lei Tian,et al.  Estimation of spatial soil moisture averages in a large gully of the Loess Plateau of China through statistical and modeling solutions , 2013 .

[19]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[21]  Zhensheng Kang,et al.  Identification and characterization of the causal agent of bacterial canker of kiwifruit in the Shaanxi province of China , 2013 .

[22]  Qiang Ji,et al.  A new efficient ellipse detection method , 2002, Object recognition supported by user interaction for service robots.

[23]  Wang Xiu,et al.  Design of structured-light vision system for tomato harvesting robot , 2014 .