Nokta Bulutu Verisi Kullanılarak Elma Bahçesinden Meyve Tespiti

Abstract

Meyve konumlarının güvenilir olarak tespit edilmesi hasat ve rekolte tahmini sürecini geliştirerek ekonomik, çevreci ve sürdürülebilir tarımın önünü açar. Meyvecilikte modern çözümler geliştirmek, meyve bahçelerinin karmaşık geometrisi nedeniyle zordur. Bu çalışmada fotogrametrik olarak elde edilen Fuji elma bahçesi nokta bulutu veri seti kullanılarak Fuji elmalarının mekânsal konumlarının belirlenmesi için yeni bir çerçeve önerilmiştir. Önerilen çerçevede en uygun komşuluğun belirlenmesi için omnivaryans tabanlı bir yaklaşım kullanılmıştır. En uygun komşuluk sayısı belirlendikten sonra her bireysel noktadan 30 adet 2 boyutlu ve 3 boyutlu geometrik özellik çıkarılmıştır. Ardından, veri setini en iyi temsil eden özellikler Minimum artıklık maksimum ilgililik yöntemi kullanılarak seçilmiştir. Farklı özelliklerin elma belirleme üzerine etkisinin incelenmesi için ilgili özellikler ağırlık düzeyine göre altı farklı gruba ayrılarak istatistiksel ve görsel karşılaştırmaları gerçekleştirilmiştir. Destek vektör makine kullanılarak yapılan sınıflandırma işleminin sonuçlarına göre 25 özelliğin kullanılması (%95.81 doğruluk ve %93.20 kesinlik) en yüksek sınıflandırma performansını sağlamıştır. Bütün veya sınırlı sayıda özelliklerin kullanılması sınıflandırma performansını azalttığı belirlenmiştir. Ayrıca, 2 boyutlu özelliklerin 3 boyutlu özellikler kadar etkili olduğu görülmüştür.

Keywords

• Elma belirleme
• Nokta bulutu
• özellik çıkarımı
• özellik seçimi
• sınıflandırma

Fruit Detection from Apple Orchard Using Point Cloud Data

Abstract

Reliable fruit location detection enhances harvest and yield estimates, paving the way for cost-effective, ecologically beneficial, and sustainable agriculture. Developing modern solutions in orchards is difficult due to the complex geometry of orchards. In this study, a new framework is proposed for the spatial location of Fuji apples using the photogrammetrically obtained Fuji apple orchard point cloud dataset. In the proposed framework, an omnivariance-based approach was used to determine the most suitable neighborhood. After determining the most suitable size of neighborhoods, 30 2D and 3D geometric features were extracted from each individual point. Then, the features that best represent the data set were selected using the minimum redundancy maximum relevance method. In order to examine the effects of different features on apple detection, the related features were divided into six different groups according to their weight level and statistical and visual comparisons were made. According to the results of the classification process using a support vector machine, the use of 25 features (95.81% accuracy and 93.20% precision) provided the highest classification performance. It has been determined that the use of all or a limited number of features reduces the classification performance. In addition, 2D features were found to be as effective as 3D features.

Keywords

• Apple detection
• Point cloud
• feature extraction
• feature selection
• classification

References

1. [1] Burunkaya, M., "Hassas tarım uygulamaları için yeni nesil damla sulama sistemi tasarımı ve gerçekleştirilmesi", Politeknik Dergisi, 22(3), 785-792, (2019).
2. [2] Nowak, B., "Precision Agriculture: Where do We Stand? A Review of the Adoption of Precision Agriculture Technologies on Field Crops Farms in Developed Countries", Agricultural Research, 1-8, (2021).
3. [3] Häni, N., Roy, P., Isler, V., "A comparative study of fruit detection and counting methods for yield mapping in apple orchards", Journal of Field Robotics, 37(2), 263-282, (2020).
4. [4] Lawal, O.M., "YOLOMuskmelon: quest for fruit detection speed and accuracy using deep learning", IEEE Access, 9, 15221-15227, (2021).
5. [5] Gan, H., Lee, W.S., Alchanatis, V., Abd-Elrahman, A., "Active thermal imaging for immature citrus fruit detection", Biosystems Engineering, 198, 291-303, (2020).
6. [6] Bulanon, D., Burks, T., Alchanatis, V., "Study on temporal variation in citrus canopy using thermal imaging for citrus fruit detection", Biosystems Engineering, 101(2), 161-171, (2008).
7. [7] Sa, I., Ge, Z., Dayoub, F., Upcroft, B., Perez, T., McCool, C., "Deepfruits: A fruit detection system using deep neural networks", Sensors, 16(8), 1222, (2016).
8. [8] Okamoto, H., Lee, W.S., "Green citrus detection using hyperspectral imaging", Computers electronics in agriculture, 66(2), 201-208, (2009).

9. [9] Gené-Mola, J., Sanz-Cortiella, R., Rosell-Polo, J.R., Morros, J.-R., Ruiz-Hidalgo, J., Vilaplana, V., Gregorio, E., "Fruit detection and 3D location using instance segmentation neural networks and structure-from-motion photogrammetry", Computers Electronics in Agriculture, 169, 105165, (2020).
10. [10] Liu, X., Chen, S.W., Aditya, S., Sivakumar, N., Dcunha, S., Qu, C., Taylor, C.J., Das, J., Kumar, V., "Robust fruit counting: Combining deep learning, tracking, and structure from motion", IEEE/RSJ International Conference on Intelligent Robots and Systems, 1045-1052, (2018).
11. [11] Stein, M., Bargoti, S., Underwood, J., "Image based mango fruit detection, localisation and yield estimation using multiple view geometry", Sensors, 16(11), 1915, (2016).
12. [12] Gongal, A., Silwal, A., Amatya, S., Karkee, M., Zhang, Q., Lewis, K., "Apple crop-load estimation with over-the-row machine vision system", Computers Electronics in Agriculture, 120, 26-35, (2016).
13. [13] Gené-Mola, J., Gregorio, E., Cheein, F.A., Guevara, J., Llorens, J., Sanz-Cortiella, R., Escolà, A., Rosell-Polo, J.R., "Fruit detection, yield prediction and canopy geometric characterization using LiDAR with forced air flow", Computers Electronics in Agriculture, 168, 105121, (2020).
14. [14] Gené-Mola, J., Gregorio, E., Guevara, J., Auat, F., Sanz-Cortiella, R., Escolà, A., Llorens, J., Morros, J.-R., Ruiz-Hidalgo, J., Vilaplana, V., "Fruit detection in an apple orchard using a mobile terrestrial laser scanner", Biosystems engineering, 187, 171-184, (2019).
15. [15] Weinmann, M., Urban, S., Hinz, S., Jutzi, B., Mallet, C., "Distinctive 2D and 3D features for automated large-scale scene analysis in urban areas", Computers Graphics, 49, 47-57, (2015).
16. [16] Vanegas, C.A., Aliaga, D.G., Benes, B., "Automatic extraction of Manhattan-world building masses from 3D laser range scans", IEEE transactions on visualization computer graphics, 18(10), 1627-1637, (2012).
17. [17] Boyko, A., Funkhouser, T., "Extracting roads from dense point clouds in large scale urban environment", ISPRS Journal of Photogrammetry Remote Sensing, 66(6), S2-S12, (2011).
18. [18] Gorte, B., Elberink, S.O., Sirmacek, B., Wang, J., "IQPC 2015 Track: Tree separation and classification in mobile mapping lidar data", The International Archives of Photogrammetry, Remote Sensing Spatial Information Sciences, 607, (2015).
19. [19] Günen, M.A., Beşdok, E., "Comparison of point cloud filtering methods with data acquired by photogrammetric method and RGB-D sensors", International Journal of Engineering Geosciences, 6(3), 125-135, (2021).
20. [20] Kurban, T., Beşdok, E., "3 Dimensional Point Cloud Filtering Using Differential Evolution Algorithm", IEEE Jordan International Joint Conference on Electrical Engineering and Information Technology (JEEIT), 653-657, (2019).
21. [21] Lee, I., Schenk, T.J., "Perceptual organization of 3D surface points", International Archives of Photogrammetry Remote Sensing Spatial Information Sciences, 193-198, (2002).
22. [22] Filin, S., Pfeifer, N., "Neighborhood systems for airborne laser data", Photogrammetric Engineering Remote Sensing, 71(6), 743-755, (2005). [23] Weinmann, M., Reconstruction and analysis of 3D scenes, Springer, (2016).
23. [24] Weinmann, M., Jutzi, B., Hinz, S., Mallet, C., "Semantic point cloud interpretation based on optimal neighborhoods, relevant features and efficient classifiers", ISPRS Journal of Photogrammetry Remote Sensing, 105, 286-304, (2015).
24. [25] Pauly, M., Keiser, R., Gross, M., "Multi‐scale feature extraction on point‐sampled surfaces", Computer graphics forum, Wiley Online Library, pp. 281-289, (2003).
25. [26] Mitra, N.J., Nguyen, A., "Estimating surface normals in noisy point cloud data", Proceedings of the nineteenth annual symposium on Computational geometry, 322-328, (2003).
26. [27] Lalonde, J.-F., Unnikrishnan, R., Vandapel, N., Hebert, M., "Scale selection for classification of point-sampled 3D surfaces", Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05), 285-292, (2005).
27. [28] Keskenler, M.F., Dal, D., Aydın, T., "Yapay Zeka Destekli ÇOKS Yöntemi ile Kredi Kartı Sahtekarlığının Tespiti", El-Cezerî Fen ve Mühendislik Dergisi, 8(2), 1007-1023, (2021).
28. [29] Tsoulias, N., Paraforos, D.S., Xanthopoulos, G., Zude-Sasse, M., "Apple shape detection based on geometric and radiometric features using a LiDAR laser scanner", Remote Sensing, 12(15), 2481, (2020).
29. [30] Günen, M.A., Atasever, U.H., Beşdok, E., "Analyzing the contribution of training algorithms on deep neural networks for hyperspectral image classification", Photogrammetric Engineering Remote Sensing, 86(9), 581-588, (2020).
30. [31] Yarğı, V., Postalcıoğlu, S., "EEG İşareti Kullanılarak Bağımlılığa Yatkınlığın Makine Öğrenmesi Teknikleri ile Analizi", El-Cezerî Fen ve Mühendislik Dergisi, 8(1), 142-154, (2021).
31. [32] Gulgezen, G., Cataltepe, Z., Yu, L., "Stable feature selection using MRMR algorithm", IEEE 17th Signal Processing and Communications Applications Conference, 596-599, (2009).
32. [33] Kurşun, O., Şakar, C.O., Favorov, O., Aydin, N., Gürgen, "Using covariates for improving the minimum redundancy maximum relevance feature selection method", Turkish journal of electrical engineering & computer sciences, 18(6), 975-989, (2010).
33. [34] Ding, C., Peng, H., "Minimum redundancy feature selection from microarray gene expression data", Journal of bioinformatics computational biology, 3(02), 185-205, (2005).
34. [35] Mountrakis, G., Im, J., Ogole, C., "Support vector machines in remote sensing: A review", ISPRS Journal of Photogrammetry Remote Sensing, 66(3), 247-259, (2011).
35. [36] Haala, N., Brenner, C., "Extraction of buildings and trees in urban environments", ISPRS journal of photogrammetry remote sensing, 54(2-3), 130-137, (1999).
36. [37] Vosselman, G., "Slope based filtering of laser altimetry data", International archives of photogrammetry remote sensing, 935-942, (2000).
37. [38] Cortes, C., Vapnik, V., "Support-vector networks", Machine learning, 20(3), 273-297, (1995).
38. [39] Laube, P., Franz, M.O., Umlauf, G., "Evaluation of features for SVM-based classification of geometric primitives in point clouds", Fifteenth IAPR International Conference on Machine Vision Applications (MVA), 59-62, (2017).
39. [40] Mallet, C., Bretar, F., Roux, M., Soergel, U., Heipke, C., "Relevance assessment of full-waveform lidar data for urban area classification", ISPRS journal of photogrammetry remote sensing, 66(6), S71-S84, (2011).