Spectral Index Derivation Based on Multi-Spectral Satellite Imagery for Apple and Pear Crop Classification

Hae Chan Jeon; Byeong Heon Min; Seung Wook Song; Jung Gun Cho; Il Ryong Kweon

doi:10.7235/HORT.20250006

All Issue

2025 Vol.43, Issue 3 Next Page

Research Article

Spectral Index Derivation Based on Multi-Spectral Satellite Imagery for Apple and Pear Crop Classification 다분광 인공위성 영상 기반 분광 지표 도출에 따른 사과, 배 작목 분류

30 June 2025. pp. 388-398

PDF XML

Abstract

In the agricultural sector, crop classification serves as a crucial indicator for evaluating the adaptability of crops influenced according to factors such as climate, soil conditions, and altitude. In this context, this study aims to analyze the accuracy of crop classification models based on spectral index extraction methods using multi-spectral satellite imagery for apple and pear crops. For this purpose, multi-spectral images were acquired for Nonsan city in Chungcheongnam-do, and apple and pear crops were classified according to the main spectral index extraction methods. The results showed that a machine learning algorithm, support vector machine (SVM), yielded higher accuracy when using the top 50% of the spectral indices (R² = 0.7487) compared to the use of all spectral indices. Additionally, a deep learning model, multi-layer perceptron (MLP), showed relatively little variation in accuracy based on the spectral index extraction method (R² = 0.8902). Therefore, the findings of this study present an efficient and scientific spectral index extraction method for classifying various crops, including apples and pears, across a wide range.

Keywords

crop monitoring

image processing

multi-layer perceptron

spectral index extraction

support vector machine

농업 분야에서 작목 분류는 기후, 토양 환경, 고도 조건 등 요인들에 영향을 받는 작목의 적응성을 평가하기 위한 중요한 지표로 활용한다. 이러한 차원에서 이 연구는 다분광 인공위성 영상을 활용하여 사과, 배 작목의 주요 분광 지표 도출 방안에 따른 작목 분류 모델의 정확도를 분석하는 데 목적이 있다. 이를 위해, 충청남도 논산시를 대상으로 다분광 영상을 취득하고 주요 분광 지표 도출 방안에 따라 사과, 배 작목을 분류하였다. 그 결과, 기계학습 알고리즘인 서포트 벡터 머신(Support Vector Machine, SVM)은 모든 분광 지표를 사용할 때 보다 상위 50%의 분광 지표를 사용했을 경우 정확도가 높게 도출되었다(R² = 0.7487). 또한, 딥러닝 모델인 다층 퍼셉트론(Multi-Layer Perceptron, MLP)의 경우, 분광 지표 도출 방안에 따른 정확도의 차이가 상대적으로 적었다(R² = 0.8902). 따라서, 이 연구 결과는 광범위한 영역에서 사과와 배를 포함한 다양한 작목을 효율적이고 과학적으로 분류할 수 있는 분광 지표 도출 방안을 제시한다.

키워드

작물 모니터링

이미지 처리

다층 퍼셉트론

분광 지표 도출

서포트 벡터 머신

References

Cho SH, Lee GS, Hwang JW (2020) Drone-based Vegetation Index Analysis Considering Vegetation Vitality. Journal of the KAGIS 23:21-35. https://doi.org/10.11108/kagis.2020.23.2.021

10.11108/kagis.2020.23.2.021

Environmental Geographic Information Service (2023) Land Cover Map. Available via https://egis.me.go.kr/. asp Accessed 1 August 2023

Ghassemi B, Dujakovic A, Żółtak M, Immitzer M, Atzberger C, Vuolo F (2022) Designing a European-Wide Crop Type Mapping Approach Based on Machine Learning Algorithms Using LUCAS Field Survey and Sentinel-2 Data. Remote Sens 14:541. https://doi.org/10.3390/rs14030541

10.3390/rs14030541

Ji S, Zhang C, Xu A, Shi Y, Duan Y (2018) 3D Convolutional Neural Networks for Crop Classification with Multi-Temporal Remote Sensing Images. Remote Sens 10:75. https://doi.org/10.3390/rs10010075

10.3390/rs10010075

Jin H, Eklundh L (2014) A physically based vegetation index for improved monitoring of plant phenology. Remote Sens. Environ 152:623-525. https://doi.org/10.1016/j.rse.2014.07.010

10.1016/j.rse.2014.07.010

Karki S, Basak JK, Paudel B, Deb NC, Kim NE, Kook JH, Kang MY, Kim TH (2024) Classification of strawberry ripeness stages using machine learning algorithms and colour spaces. Hortic. Environ. Biotechnol 65:337-354. https://doi.org/10.1007/s13580-023-00559-2

10.1007/s13580-023-00559-2

Lillesand T, Kiefer WR, Chipman J (2015) Remote Sensing and Image Interpretation, Ed 7, Wiley, New York

Ministry of Agriculture, Food and Rural Affairs (2022) Agricultural Crop Production Statistical Report, report, 11-1543000-000128-10, Republic of Korea

Ministry of Agriculture, Food and Rural Affairs (2023) Agricultural Management Registration Information Inquiry Service, Nonsan 2022 crop cultivation distribution map. Available via https://uni.agrix.go.kr/docs7/biOlap/dashBoard.do. asp Accessed 1 August 2023

Ministry of Environment (2023) Environmental Geographic Information Service, Subdiviision Land Cover Map, Abailable via https://egis.me.go.kr/intro/land.do. asp Accessed 1 August 2023

Nonsan City Hall (2023) General status. Available via https://nonsan.go.kr/. asp Accessed 1 August 2023

Segarra J, Buchaillot ML, Araus JL, Kefauver SC (2020) Remote Sensing for Precision Agriculture: Sentinel-2 Improved Features and Applications. Agronomy 10:641. https://doi.org/10.3390/agronomy10050641

10.3390/agronomy10050641

Sonobe R, Yamaya Y, Tani H, Wang X, Kobayashi N, Mochizuki KI (2018) Crop classification from Sentinel-2 derived vegetation indices using ensemble learning. J. Appl. Remote Sens 12:026019. https://doi.org/10.1117/1.JRS.12.026019

10.1117/1.JRS.12.026019

The Ministry of Land, Infrastructure and Transport (2023) knowledge space. Available via http://www.nsdi.go.kr/lxportal/?menuno=4064. asp Accessed 1 August 2023

Yao J, Wu J, Xiao C, Zhang Z, Li J (2022) The Classification Method Study of Crops Remote Sensing with Deep Learning, Machine Learning, and Google Earth Engine. Remote Sens https://doi.org/10.3390/rs14122758

10.3390/rs14122758

Zhang X, Sun Y, Shang K, Zhang L (2016) Crop Classification Based on Feature Band Set Construction and Object-Oriented Approach Using Hyperspectral Images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 9:4117-4128. https://doi.org/10.1109/JSTARS.2016.2577339

10.1109/JSTARS.2016.2577339

Information

Publisher :KOREAN SOCIETY FOR HORTICULTURAL SCIENCE
Publisher(Ko) :한국원예학회
Journal Title :Horticultural Science and Technology
Journal Title(Ko) :원예과학기술지
Volume : 43
No :3
Pages :388-398
Received Date : 2023-10-04
Revised Date : 2024-06-11
Accepted Date : 2024-06-12
DOI :https://doi.org/10.7235/HORT.20250006

Horticultural Science and Technology 원예과학기술지 ISSN:1226-8763(Print) 2465-8588(Online)

All Issue