Malaysian Journal of Analytical Sciences, Vol 27 No 3 (2023): 488 - 498

ANALYSIS OF HYPERSPECTRAL REFLECTANCE FOR DISEASE CLASSIFICATION OF SOYBEAN FROGEYE LEAF SPOT

USING KNIME ANALYTICS

(Analisis Refleksi Hiperspektra untuk Klasifikasi Penyakit Bintik Daun Frogeye

Kacang Soya Menggunakan Analisis KNIME)

Yuhao Ang, and Helmi Zulhaidi Mohd Shafri

Department of Civil Engineering and Geospatial Information Science Research Centre (GISRC),

Faculty of Engineering,

Universiti Putra Malaysia (UPM), 43400 Serdang, Selangor, Malaysia

*Corresponding author: helmi@upm.edu.my

Received: 10 October 2022; Accepted: 13 April 2023; Published: 23 June 2023

Abstract

The feasibility of classifying soybean frogeye leaf spot (FLS) has been investigated with the advance of hyperspectral technology. Hyperspectral reflectance data of healthy and FLS disease soybeans were used. The first step was to smooth out the data by using a filtering technique namely Savitzky-Golay to eliminate the noise of the spectrum. In order to select the most significant wavelengths, genetic algorithm (GA) was used as a forward feature selection technique. This analysis involved the implementation of machine learning (ML) algorithms, including decision trees, random forests, and stacking, to classify soybean FLS severity levels. Preprocessing ML steps including converting class numbers to strings, identifying and removing missing values, partitioning and normalizing data were implemented prior to the development of the model. Overall accuracy and the receiver operating characteristic curve measure were used to assess the performance of this analysis. All of these steps were carried out through KNIME analytical platform. Based on the results of the analysis, GA-stacking and random forest algorithms achieved the best overall accuracy of 85.9% and 84.3%, respectively. In terms of reproducibility, data flow control, data exploration, analysis and visualization, KNIME Analytics Platform provided great convenience in connecting tools graphically and ensuring the same results on different operating systems. The rapid implementation of workflow in KNIME Analytics Platform provided the opportunity to process hyperspectral reflectance data to classify crop diseases.

Keywords: hyperspectral reflectance, forward feature selection, genetic algorithms, machine learning, KNIME analytics platform

Abstrak

Kebolehlaksanaan mengklasifikasikan bintik daun frogeye kacang soya (FLS) telah disiasat dengan kemajuan teknologi hiperspektral. Data reflektan hiperspektral kacang soya yang sihat dan penyakit FLS telah digunakan. Langkah pertama ialah melicinkan data dengan menggunakan teknik penapisan iaitu Savitzky-Golay untuk menghapuskan hingar spektrum. Untuk memilih panjang gelombang yang paling ketara, algoritma genetik (GA) digunakan sebagai teknik pemilihan ciri ke hadapan. Analisis ini melibatkan pelaksanaan algoritma pembelajaran mesin (ML), termasuk pohon keputusan, hutan rawak, dan pemadatan, untuk mengklasifikasikan tahap keparahan FLS kacang kedelai. Langkah-langkah pra pemprosesan ML termasuk menukar nombor kelas kepada rentetan, mengenal pasti dan mengeluarkan nilai yang hilang, mempartisi dan menormalisasi data telah dilaksanakan sebelum pembangunan model. Ketepatan keseluruhan dan ukuran lengkung ciri operasi penerima digunakan untuk menilai prestasi analisis ini. Kesemua langkah ini telah dijalankan melalui platform analisis KNIME. Berdasarkan keputusan analisis, algoritma GA-stacking dan random forest masing-masing mencapai ketepatan keseluruhan terbaik iaitu 85.9% dan 84.3%. Dari segi kebolehulangan, kawalan aliran data, penerokaan data, analisis dan visualisasi, Platform Analisis KNIME memberikan kemudahan besar dalam menyambungkan alatan secara grafik dan memastikan hasil yang sama pada sistem pengendalian yang berbeza. Pelaksanaan pantas aliran kerja dalam Platform Analisis KNIME memberi peluang untuk memproses data reflektan hiperspektral dalam klasifikasi penyakit tanaman.

Kata kunci: reflektan hiperspektral, pemilihan ciri ke hadapan, algoritma genetik, pembelajaran mesin, platform analisis KNIME

References

1. USDA Foreign Agricultural Service (2022). World Agricultural Production. Access from https://www.fas.usda.gov/data/world-agricultural-production.

2. Dashiell, K. E. and Akem, C. N. (1991). Yield losses in soybeans from frogeye leaf spot caused by Cercospora sojina. Crop Protection, 10(6): 465-468.

3. Mwase, W. F. and Kapooria, R. G. (2001). Incidence and severity of frogeye leaf spot and associated yield losses in soybeans in agroecological zone II of Zambia. Mycopathologia, 149(2): 73-78.

4. Mian, M. A. R., Missaoui, A. M., Walker, D. R., Phillips, D. V. and Boerma, H. R. (2008). Frogeye leaf spot of soybean: A review and proposed race designations for isolates of Cercospora sojina Hara. Crop Science, 48(1): 14-24.

5. George, A. P., Broadley, R. H., Nissen, R. J. and Ward, G. (2002). Effects of new rest-breaking chemicals on flowering, shoot production and yield of subtropical tree crops. Acta Horticulturae, 575(4): 835-840.

6. Bukhari, K. A. I. M., Nithya, K., Valluvaparidasan, V., Paranidharan, V. and Velazhahan, R. (2012). Detection of Colletotrichum falcatum causing red rot of sugarcane by enzyme-linked immunosorbent assay. Archives of Phytopathology and Plant Protection, 45(7): 823-830.

7. Clevers, J. G. P. W., Kooistra, L. and Schaepman, M. E. (2010). Estimating canopy water content using hyperspectral remote sensing data. International Journal of Applied Earth Observation and Geoinformation, 12(2): 119-125.

8. Peng, Y., Zhang, M., Xu, Z., Yang, T., Su, Y., Zhou, T., Wang, H., Wang, Y. and Lin, Y., 2020. Estimation of leaf nutrition status in degraded vegetation based on field survey and hyperspectral data. Scientific Reports, 10(1): 1-12.

9. Sonobe, R., Yamashita, H., Nofrizal, A. Y., Seki, H., Morita, A. and Ikka, T. (2021). Use of spectral reflectance from a compact spectrometer to assess chlorophyll content in Zizania latifolia. Geocarto International, 2021: 1-13.

10. Liu, Z. Y., Shi, J. J., Zhang, L. W. and Huang, J. F. (2010). Discrimination of rice panicles by hyperspectral reflectance data based on principal component analysis and support vector classification. Journal of Zhejiang University: Science B, 11(1): 71-78.

11. Sankaran, S. and Ehsani, R. (2010). Spectral evaluations for developing optical methods of Huanglongbing (HLB) detection in citrus orchards. Biosystems, 2021: 121-123.

12. Izzuddin, M. A., Seman Idris, A., Nisfariza, M. N., Nordiana, A. A., Shafri, H. Z. M. and Ezzati, B. (2017). The development of spectral indices for early detection of Ganoderma disease in oil palm seedlings. International Journal of Remote Sensing, 38(23): 6505-6527.

13. Vigier, B. J., Pattey, E. and Strachan, I. B. (2004). Narrowband vegetation indexes and detection of disease damage in soybeans. IEEE Geoscience and Remote Sensing Letters, 1(4): 255-259.

14. Zhang, N., Yang, G., Pan, Y., Yang, X., Chen, L. and Zhao, C. (2020). A review of advanced technologies and development for hyperspectral-based plant disease detection in the past three decades. Remote Sensing, 12(19): 1-34.

15. Wójtowicz, A., Piekarczyk, J., Czernecki, B. and Ratajkiewicz, H. (2021). A random forest model for the classification of wheat and rye leaf rust symptoms based on pure spectra at leaf scale. Journal of Photochemistry and Photobiology B: Biology, 223: 112278.

16. Chen, X., Dong, Z., Liu, J., Wang, H., Zhang, Y., Chen, T., Du, Y., Shao, L. and Xie, J. (2020). Hyperspectral characteristics and quantitative analysis of leaf chlorophyll by reflectance spectroscopy based on a genetic algorithm in combination with partial least squares regression. Spectrochimica Acta - Part A: Molecular and Biomolecular Spectroscopy, 243: 118786.

17. Liu, Z. Y., Wu, H. F. and Huang, J. F. (2010). Application of neural networks to discriminate fungal infection levels in rice panicles using hyperspectral reflectance and principal components analysis. Computers and Electronics in Agriculture, 72(2): 99-106.

18. Zhu, H., Chu, B., Zhang, C., Liu, F., Jiang, L. and He, Y. (2017). Hyperspectral imaging for presymptomatic detection of tobacco disease with successive projections algorithm and machine-learning classifiers. Scientific Reports, 7(1): 1-12.

19. Wei, X., Johnson, M. A., Langston, D. B., Mehl, H. L. and Li, S. (2021). Identifying optimal wavelengths as disease signatures using hyperspectral sensor and machine learning. Remote Sensing, 13(14): 1314283.

20. Tian, L., Xue, B., Wang, Z., Li, D., Yao, X., Cao, Q., Zhu, Y., Cao, W. and Cheng, T. (2021). Spectroscopic detection of rice leaf blast infection from asymptomatic to mild stages with integrated machine learning and feature selection. Remote Sensing of Environment, 257(8): 112350.

21. Zhang, J., Cheng, T., Guo, W., Xu, X., Qiao, H., Xie, Y. and Ma, X. (2021). Leaf area index estimation model for UAV image hyperspectral data based on wavelength variable selection and machine learning methods. Plant Methods, 17(1): 1-14.

22. Al-Saddik, H., Simon, J. C. and Cointault, F. (2019). Assessment of the optimal spectral bands for designing a sensor for vineyard disease detection: The case of ‘Flavescence dorée.’ Precision Agriculture, 20(2): 398-422.

23. Fillbrunn, A., Dietz, C., Pfeuffer, J., Rahn, R., Landrum, G. A. and Berthold, M. R. (2017). KNIME for reproducible cross-domain analysis of life science data. Journal of Biotechnology, 261(July): 149-156.

24. Liu, S., Yu, H., Sui, Y., Zhou, H., Zhang, J., Kong, L., Dang, J. and Zhang, L. (2021). Classification of soybean frogeye leaf spot disease using leaf hyperspectral reflectance. PLoS ONE, 16(9): 1-23.

25. Ruffin, C. and King, R. L. (1999). The analysis of hyperspectral data using Savitzky-Golay filtering-theoretical basis (part 1). In IGARSS’99 Proceedings: Remote Sensing of the System Earth—A Challenge for the 21st Century: IEEE 1999 International Geoscience and Remote Sensing Symposium, 28 June–2 July 1999, Congress Centrum Hamburg: pp. 756-758.

26. Džeroski, S. and Ženko, B. (2004). Is combining classifiers with stacking better than selecting the best one? Machine Learning, 54(3): 255-273.

27. Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P. and Witten, I. H. (2009). The WEKA data mining software: An update. ACM SIGKDD Explorations Newsletter, 11(1): 10-18.

28. Gregory, P. H. (1968). Interpreting plant disease dispersal gradients. Annual review of Phytopathology, 6(1): 189-212.

29. Tsai, F. and Philpot, W. (1998). Derivative analysis of hyperspectral data. Remote Sensing of Environment, 66(1): 41-51.

30. Lu, J., Zhou, M., Gao, Y. and Jiang, H. (2018). Using hyperspectral imaging to discriminate yellow leaf curl disease in tomato leaves. Precision Agriculture, 19(3): 379-394.

31. Li, Y., Demetriades-Shah, T. H., Kanemasu, E. T., Shultis, J. K. and Kirkham, M. B. (1993). Use of second derivatives of canopy reflectance for monitoring prairie vegetation over different soil backgrounds. Remote Sensing of Environment, 44(1): 81-87.

32. Muhammed, H. H. (2005). Hyperspectral crop reflectance data for characterising and estimating fungal disease severity in wheat. Biosystems Engineering, 91(1): 9-20.

33. Cen, Y., Huang, Y., Hu, S., Zhang, L. and Zhang, J. (2022). Early detection of bacterial wilt in tomato with portable. Remote Sensing, 14(12): 2882.