Renk Momentleri ve Destek Vektör Makineleri Kullanarak Haploid Mısır Tohumlarının Tanımlanmasında Renk Uzaylarının Sınıflandırma Performansına Etkisinin Karşılaştırılması

Öz

Katlanmış haploid mısır ıslah tekniği gerek ıslah süresini kısaltması, gerekse %100 homozigot hatlar elde edilmesi açısından, geleneksel ıslah yöntemleri ile karşılaştırıldığında önemli avantajlar sağlar. Katlanmış haploid mısır ıslah yönteminde önemli işlerden biri melezleme sonucunda elde edilen haploid ve diploid tohumların ayrıştırılmasıdır. Günümüzde bu işlemin elle yapılıyor olması, emek ve zaman kaybının yanı sıra yüksek sınıflandırma hatalarına neden olmaktadır. Bu çalışmada, haploid mısır tohumlarının tanımlanması için bilgisayar görmesine dayalı bir yöntem önerilmiştir. 3000 mısır tohumundan oluşan veri seti üzerinde yürütülen çalışmada, k-ortalamalar kümeleme yöntemi ile mısır tohum embriyoları bölütlenmiştir. RGB, HSV ve Lab renk uzaylarında her renk kanalı için ilk dört dereceden renk momentleri çıkarılmıştır. Böylece her renk uzayı için 12 öznitelik kullanılmıştır. Elde edilen öznitelikler destek vektör makinesi ile sınıflandırılmıştır. Daha genel sonuçlar elde edebilmek için 10-kat çapraz doğrulama yöntemi 20 kez tekrar edilmiştir. Deneysel sonuçlara göre en iyi performans Lab renk uzayında  %87,03 doğruluk, %80,15 duyarlılık ve %93,00 özgünlük değerleriyle ölçülmüştür.

Anahtar Kelimeler

• mısır,haploid tanıma,renk uzayları,renk momentleri,destek vektör makineleri

Comparison of the Effect of Color Spaces on Classification Performance in Identification of Haploid Maize Seeds Using Color Moments and Support Vector Machines

Öz

Doubled haploid maize breeding technique provides significant advantages over conventional breeding methods both in terms of shortening breeding time and obtaining 100% homozygous lines. One of the important works in doubled haploid maize breeding method is the separation of haploid and diploid seeds obtained as a result of hybridization. Today, this process is performed by hand, as well as loss of labor and time causes high classification errors. In this study, a computer vision based method for identification of haploid maize seeds was proposed. In the study conducted on the dataset of 3000 maize seeds, seed embryos were segmented by k-means clustering method. In the RGB, HSV and Lab color spaces, the first four degrees of color moments are extracted for each color channel. Thus, 12 features were used for each color space. Obtained features were classified with support vector machine. To obtain more general results, the 10-fold cross-validation method was repeated 20 times. According to the experimental results, the best performance was measured with 87,03% accuracy, 80,15% sensitivity and 93,00% specificity in Lab color space.

Anahtar Kelimeler

• maize,haploid identification,color spaces,color moments,support vector machines

Kaynakça

1. Y. Altuntaş, Z. Cömert, and A. F. Kocamaz, “Identification of haploid and diploid maize seeds using convolutional neural networks and a transfer learning approach,” Comput. Electron. Agric., vol. 163, p. 104874, Aug. 2019.
2. B. M. Prasanna, V. Chaikam, and G. Mahuku, Doubled Haploid Technology in Maize Breeding: Theory and Practice. Mexico: CIMMYT, 2012.
3. H. H. Geiger, “Doubled Haploids,” in Handbook of Maize, New York, NY: Springer New York, 2009, pp. 641–657.
4. C. Chidzanga, F. Muzawazi, J. Midzi, and T. Hove, “Production and use of haploids and doubled haploid in maize breeding : A review,” vol. 4, no. 4, pp. 201–213, 2017.
5. F. K. Röber, G. A. Gordillo, and H. H. Geiger, “In vivo haploid induction in maize - Performance of new inducers and significance of doubled haploid lines in hybrid breeding,” Maydica, vol. 50, pp. 275–283, 2005.
6. A. E. Melchinger, W. Schipprack, H. Friedrich Utz, and V. Mirdita, “In Vivo Haploid Induction in Maize: Identification of Haploid Seeds by Their Oil Content,” Crop Sci., vol. 54, no. 4, p. 1497, 2014.
7. H. H. Geiger, G. Andrés Gordillo, and S. Koch, “Genetic Correlations among Haploids, Doubled Haploids, and Testcrosses in Maize,” Crop Sci., vol. 53, pp. 2313–2320, 2013.
8. A. Smelser, M. Blanco, T. Lübberstedt, A. Schechert, A. Vanous, and C. Gardner, “Weighing in on a method to discriminate maize haploid from hybrid seed,” Plant Breed., vol. 134, no. 3, pp. 283–285, Jun. 2015.

9. V. Chaikam, L. Martinez, A. E. Melchinger, W. Schipprack, and P. M. Boddupalli, “Development and validation of red root marker-based haploid inducers in maize,” Crop Sci., vol. 56, no. 4, pp. 1678–1688, 2016.
10. D. K. Nanda and S. S. Chase, “An Embryo Marker for Detecting Monoploids Of Maize (Zea Mays L.)1,” Crop Sci., vol. 6, no. 2, p. 213, 1966.
11. A. E. Melchinger, W. Schipprack, T. Würschum, S. Chen, and F. Technow, “Rapid and accurate identification of in vivo-induced haploid seeds based on oil content in maize,” Sci. Rep., vol. 3, p. 2129, Jul. 2013.
12. B. W. Boote, D. J. Freppon, G. N. De La Fuente, T. Lübberstedt, B. J. Nikolau, and E. A. Smith, “Haploid differentiation in maize kernels based on fluorescence imaging,” Plant Breed., vol. 135, no. 4, pp. 439–445, 2016.
13. J. Lin, L. Yu, W. Li, and H. Qin, “Method for Identifying Maize Haploid Seeds by Applying Diffuse Transmission Near-Infrared Spectroscopy,” Appl. Spectrosc., vol. 72, no. 4, pp. 611–617, 2018.
14. G. N. De La Fuente, J. M. Carstensen, M. A. Edberg, and T. Lü bberstedt, “Discrimination of haploid and diploid maize kernels via multispectral imaging,” Plant Breed., vol. 136, no. 1, pp. 50–60, Feb. 2017.
15. Y. Altuntas, A. F. Kocamaz, R. Cengiz, and M. Esmeray, “Classification of haploid and diploid maize seeds by using image processing techniques and support vector machines,” in 2018 26th Signal Processing and Communications Applications Conference (SIU), 2018, pp. 1–4.
16. Y. Altuntas, A. F. Kocamaz, Z. Comert, R. Cengiz, and M. Esmeray, “Identification of Haploid Maize Seeds using Gray Level Co-occurrence Matrix and Machine Learning Techniques,” in 2018 International Conference on Artificial Intelligence and Data Processing (IDAP), 2018, pp. 1–5.
17. S. Sergyán, “Color Content-based Image Classification,” in 5th Slovakian-Hungarian Joint Symposium on Applied Machine Intelligence and Informatics, 2007, pp. 427–434.
18. V. N. Vapnik, The nature of statistical learning theory. Springer, 2000.
19. Z. Cömert, A. F. Kocamaz, and V. Subha, “Prognostic model based on image-based time-frequency features and genetic algorithm for fetal hypoxia assessment,” Comput. Biol. Med., vol. 99, pp. 85–97, Aug. 2018.
20. Z. Cömert and A. F. Kocamaz, “Comparison of Machine Learning Techniques for Fetal Heart Rate Classification,” Acta Phys. Pol. A, vol. 132, no. 3, pp. 451–454, Sep. 2017.
21. Z. Cömert and A. F. Kocamaz, “Open-access software for analysis of fetal heart rate signals,” Biomed. Signal Process. Control, vol. 45, pp. 98–108, Aug. 2018.