A Separating Method of Adjacent Apples Based on Machine Vision and Chain Code Information

Fruit location is an important parameter for apple harvesting robot to conduct picking task. However, it is difficult to obtain coordinates of each apple under natural conditions. One of the major challenges is detecting adjacent fruits accurately. Previous studies for adjacent detection have shortcomings such as vast computation, difficulty in implementation and over-segmentation. In this paper, we propose a novel and effective separating method for adjacent apples recognition based on chain code information and obtain the centroid coordinates of each fruit. Firstly, those valid regions of fruit are extracted by pre-processing the initial image. Secondly, chain code information is obtained by following the contour of extracted regions. Thirdly, through observing the changing law of chain code difference and adopting local optimum principle, concave points are found. Finally, the best point pairs are determined with different matching principles, and those adjacent apples are separated exactly. The experimental results show that the average rate of successful separation is greater than 91.2% with the proposed method, which can meet the requirements of applications in harvesting robots.

[1]  牧野 英二,et al.  米の貯蔵環境と米品質に関する研究 (第3報):貯蔵環境が米飯のテクスチャー物性値に及ぼす影響 , 2002 .

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

[3]  D. Stajnko,et al.  Modelling of Apple Fruit Growth by Application of Image Analysis , 2005 .

[4]  Radnaabazar Chinchuluun.  Citrus Yield Mapping System in Natural Outdoor Scenes using the Watershed Transform , 2006 .

[5]  J. Amiriparian,et al.  Apple Fruits Recognition Under Natural Luminance Using Machine Vision , 2010 .

[6]  Weihua Liu,et al.  Automatic segmentation of overlapping powder particle based on searching concavity points: Automatic segmentation of overlapping powder particle based on searching concavity points , 2011 .

[7]  Won Suk Lee,et al.  Recognition of partially occluded plant leaves using a modified watershed algorithm , 2004 .

[8]  Tatsuo Hiroma,et al.  A Color Model for Recognition of Apples by a Robotic Harvesting System. , 2002 .

[9]  Rafael C. González,et al.  Digital image processing using MATLAB , 2006 .

[10]  Liu Gang,et al.  Separating adjoined apples based on machine vision and information fusion. , 2009 .

[11]  Shi ZhiXing,et al.  Corn breed recognition based on support vector machine. , 2009 .

[12]  Jun Zhao,et al.  On-tree fruit recognition using texture properties and color data , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[13]  Luc Taerwe,et al.  Methods to quantify the colour development of concrete exposed to fire , 2011 .

[14]  Herbert Freeman,et al.  Computer Processing of Line-Drawing Images , 1974, CSUR.

[15]  Yang Li,et al.  Comparison of two algorithms based on mathematical morphology for segmentation of touching strawberry fruits , 2007 .

[16]  Liu Gang,et al.  Synthetical evaluation of agriculture machine navigation system. , 2010 .

[17]  Y.-C. Wang,et al.  Automatic segmentation of touching rice kernels with an active contour model , 2004 .

[18]  Wang Mulan,et al.  Cotton recognition based on randomized hough transform. , 2010 .

[19]  Jesús Angulo,et al.  Robust iris segmentation on uncalibrated noisy images using mathematical morphology , 2010, Image Vis. Comput..

[20]  Takashi Kataoka,et al.  Fruit detection system and an end effector for robotic harvesting of Fuji apples , 2010 .