Astronomical School’s Report, 2018, Volume 14, Issue 1, Pages 30–34

https://doi.org/10.18372/2411-6602.14.04
Download PDF
UDC 528.8+504.064+535.361.2

Methods of increasing the accuracy of classification of agrophytocenoses by satellite images

Semeniv O.V.1, 2, 3, Pidgorodetska L.V.1

1Space Research Institute NASU & SSAU, 03187, Hlushkova Avenue 40, Kyiv, Ukraine
2Svitla Systems, Inc., 02002, Yevhena Sverstyuka St. 2A, Kyiv, Ukraine
3Thermo Fisher Scientific, Inc., 3380 Central Expressway, Santa Clara, CA 95051, USA

Abstract

The article is devoted to the problem of the agrophytocenoses classification with satellite observation data and noise filtration methods. The model of agricultural crop types determination with multispectral data and the support vector machine method was conducted. The modified algorithm for agricultural crops classification approbation was carried out on the satellite imageries of the “Steppe” farm of the Kamyansko-Dniprovsky district of the Zaporizhzhya region. Ground-based reference data on crop types, normative base of agricultural development phases for heavy-loamy and medium-loamy soils characteristics of the southern Ukraine steppe zone and remote sensing data of the studied territory gathered by Landsat ТМ and ETM+ satellites for 2001 and 2003 were used for digital processing. It was investigated that a significant number of false-detected pixels (>5%) belong to the interface zones (field boundaries) and chaotic scattered segments of incorrectly classified crops within the field (≈12). This error is mainly due to spatial variations in soil moisture, plant health, type of projective surface, cultures spectral similarity, phenological stage of plants development, time of satellite observation. Median and rank filtration methods were used to minimize the effect of the above-mentioned problems on the final classification result. The method of median filtration allowed to increase the classification accuracy up to 3%, and the rank filtration method up to 2% compared with the simple procedure of the support vector machine method. The classification error and the disadvantages of filtration are largely due to the interface zones of the different crops, the fields boundaries. Since the spatial resolution is 30×30 meters, it would be advisable to reject the pixels of the transition (on the boundary) from the training sample to identify more accurate classification model.

Keywords: remote sensing; estimation; vegetation state; classification; filtration

References

  1. Vizilter Y.V., Zheltov S.Y., Knyaz V.A., Hodarev A.N., Morzhen A.V. (2009). Image analysis and processing with LabVIEW IMAQ Vision. M.: DMK Press. 464 p.
  2. Lyalko V.I., Popov M.O. (2006). Multispectral remote sensing in nature management. Kyiv: Naukova dumka. 360 p.
  3. Semeniv O.V. (2014). Estimation of biochemical components in vegetation based on statistical learning methods and remote sensing data. Astronomical School's Reports, 10, No. 1, 70–74. https://doi.org/10.18372/2411-6602.10.1070
  4. Shadchyna T.M. (2001). Scientific fundamentals of crops state remote monitoring. Kyiv: Phytosociocenter. 220 p.
  5. Astola J., Haavisto P., Neuvo Y. (1990). Vector median filters. Proc. of the IEEE., 78(4), 678–689. https://doi.org/10.1109/5.54807
  6. Bo Chen, Fang Qiu, Bingfang Wu, Hongyue Du. (2015). Image segmentation based on constrained spectral variance difference and edge penalty. Remote Sens, 7, 5980–6004. https://doi.org/10.3390/rs70505980
  7. Bruzzone L., Pieto D.F. (2002). A partially unsupervised cascade classifier for the analysis of multitemporal remote-sensing images. Pattern Recognition Letters, 23, 1063–1071. https://doi.org/10.1016/s0167-8655(02)00053-3
  8. Egbert S.L., Park S., Price K.P. (2002). Using conservation reserve program maps derived from satellite imagery to characterize landscape structure. Compute Electron Agric, 37. – R.141–156. https://doi.org/10.1016/s0168-1699(02)00114-x
  9. Foody G.M., Mathur A. (2004). Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification. Remote Sensing of Environment, 93, 107–117. https://doi.org/10.1016/j.rse.2004.06.017
  10. Kogan F.N. (2001). Operational space technology for global vegetation assessment. Bull. Am. Meteorol. Soc., 82(9), 1949–1964. https://doi.org/10.1175/1520-0477(2001)082<1949:ostfgv>2.3.co;2
  11. Lim J.S. Two-dimensional signal and image processing. Englewood Cliffs, NJ, Prentice Hall, 1990, P.469–476.
  12. Podgorodetskaya L.V., Prutsko Y.V., Semeniv O.V. (2016). Farmland state estimation based on agrophytocenoses classification and remote sensing satellite data. Journal of Automation and Information Sciences, 48(5), 55–63. https://doi.org/10.1615/jautomatinfscien.v48.i5.50
  13. Semeniv O.V., Pidgorodetska L.V. (2014). Biophysical Parameters of Water Ecosystem Estimation Using Satellite Images and Optimization Techniques. Journal of Automation and Information Sciences, 46(9), 68–77. https://doi.org/10.1615/jautomatinfscien.v46.i9.60
  14. Schmedtmann J., Campagnolo M.L. (2015). Reliable crop identification with satellite imagery in the context of common agriculture policy subsidy control. Remote Sens, 7, 9325–9346. https://doi.org/10.3390/rs70709325
  15. Yan L., Roy D.P. (2014). Automated crop field extraction from multi-temporal Web Enabled Landsat Data. Remote Sensing of Environment, 144, 42–64. https://doi.org/10.1016/j.rse.2014.01.006
  16. Xi Chen, Tao Fang, Hong Huo, Deren Li (2015). Measuring the effectiveness of various features for thematic information extraction from very high resolution remote sensing imagery. IEEE Transactions on geoscience and remote sensing, 53(9), 4837–4851. https://doi.org/10.1109/tgrs.2015.2411331

Download PDF