A Video Image Segmentation System for the Fruit-trees in Multi-stage Outdoors Orchard under Natural Conditions

Abstract

Segmentation is an important part of each machine vision system that has a direct relationship with the final system accuracy and performance. Outdoors segmentation is often complex and difficult due to both changes in sunlight intensity and the different nature of background objects. However, in fruit-tree orchards, an automatic segmentation algorithm with high accuracy and speed is very desirable. For this reason, a multi-stage segmentation algorithm is applied for the segmentation of apple fruits with Red Delicious cultivar in orchard under natural light and background conditions. This algorithm comprises a combination of five segmentation stages, based on: 1- L*u*v* color space, 2- local range texture feature, 3- intensity transformation, 4- morphological operations, and 5- RGB color space. To properly train a segmentation algorithm, several videos were recorded under nine different light intensities in Iran-Kermanshah (longitude: 7.03E; latitude: 4.22N) with natural (real) conditions in terms of both light and background. The order of segmentation stage methods in multi-stage algorithm is very important since has a direct relationship with final segmentation accuracy. The best order of segmentation methods resulted to be: 1- color, 2- texture and 3- intensity transformation methods. Results show that the values of sensitivity, accuracy and specificity, in both classes, were higher than 97.5%, over the test set. We believe that those promising numbers imply that the proposed algorithm has a remarkable performance and could potentially be applied in real-world industrial case.

Keywords

• background,daylight,machine vision,natural condition,texture,color space

References

1. Aquino A, Diago M.P, Millan B & Tardaguila J (2017). A new methodology for estimating the grapevineberry number per cluster using image analysis. Biosystem Engineering 156: 80-95.
2. Arroyo J, Guijarro M & Pajares G (2016). An instance-based learning approach for thresholding in crop images under different outdoor conditions. Computers and Electronics in Agriculture 127: 669–679.
3. Bai X, Cao Z, Wang Y, Yu Z, Hu Z, Zhang X & Li C (2014). Vegetation segmentation robust to illumination variations based on clustering and morphology modelling. Biosystem Engineering 125: 80-97.
4. Behroozi-Khazaei N & Maleki M.R (2017). A robust algorithm based on color features for grape cluster segmentation. Computers and Electronics in Agriculture 142: 41–49.
5. Camargo A & Smith J.S (2009). An image-processing based algorithm to automatically identify plant disease visual symptoms. Biosystem Engineering 102: 9–21.
6. Dorj U.-O, Lee M & Yun S.-s (2017). An yield estimation in citrus orchards via fruit detection and counting using image processing. Computers and Electronics in Agriculture 140: 103–112.
7. Gonzalez R.C, Woods R.E & Eddins S.L (2004). Digital Image Processing Using MATLAB. Prentice Hall.
8. Hernández-Hernández J.L, García-Mateos G, González-Esquiva J.M, Escarabajal-Henarejos D, Ruiz-Canales A & Molina-Martínez J.M (2016). Optimal color space selection method for plant/soil segmentation in agriculture. Computers and Electronics in Agriculture 122: 124–132.

9. Kataoka T, Kaneko T, Okamoto H & Hata S (2003). Crop growth estimation system using machine vision, IEEE/ASME Int. Conf. Adv. Intell. Mechatronics (AIM 2003), pp. 1079–1083.
10. Li Y, Cao Z, Lu H, Xiao Y, Zhu Y, Cremers A.B (2016). In-field cotton detection via region-based semantic image segmentation. Computers and Electronics in Agriculture 127: 475–486.
11. Liu X, Zhao D, Jia W, Ruan C, Tang S, Shen T (2016). A method of segmenting apples at night based on color and position information. Computers and Electronics in Agriculture 122: 118-123.
12. Montalvo M, Guerrero J.M, Romeo J, Emmi L, Guijarro M, Pajares G (2013). Automatic expert system for weeds/crops identification in images from maize fields. Expert Systems with Applications 40: 75-82.
13. Onyango C.M & Marchant J.A (2003). Segmentation of row crop plants from weeds using colour and morphology. Computer and electronoc in agriculture 39: 141-155.
14. Slaughter, D.C., Giles, D.K., Downey, D., 2008. Autonomous robotic weed control systems: a review. Computer and Electronoc in Agriculture 61, 63–78.
15. Wisaeng K (2013). A Comparison of Decision Tree Algorithms For UCI Repository Classification. International Journal of Engineering Trends and Technology 4: 3393-3397.
16. Zhao C, Lee W.S, He D (2016). Immature green citrus detection based on colour feature and sum of absolute transformed difference (SATD) using colour images in the citrus grove. Computers and Electronics in Agriculture 124: 243-253.