Case study of pear flowering detection using UAV-based vegetation indices and SAM

Lak-Yeong Choi; Kyung-Do Lee; Meejeong Park; Jae-Hyun Ryu; Geun-Ho Kwak; Soo-Jin Lee; Seok Kyu Yun; Seul-Ki Lee

doi:10.22765/pastj.20260007

All Issue

2026 Vol.8, Issue 1 Preview Page

Research Article

Case study of pear flowering detection using UAV-based vegetation indices and SAM

31 March 2026. pp. 83-93

PDF XML

Abstract

Flowering is a critical stage in fruit trees that affects fruit yield, and climate change has increased risks such as early flowering and frost damage. Accurate monitoring of flowering rates is essential, yet field surveys are limited. This study aimed to develop a UAV-based approach for pear orchards using multispectral and RGB imagery. UAV data were collected at different flowering stages in Naju and Wanju, South Korea, and analyzed with vegetation indices and the Geometry Segment Anything Model (Geo-SAM). Results showed that Normalized Difference Vegetation Index(NDVI) had limited performance, while the Enhanced Bloom Index (EBI) clearly increased during full flowering and was more suitable for blossom detection. Based on Geo-SAM, segmentation achieved strong correlations with field observations (R²=0.91~0.94) and reduced background noise other than flowers, enabling flowering classification. Estimation of relative flowering rates using the EBI resulted in greater error (RMSE = 13.5~27.0%), whereas estimation using the SAM yielded higher accuracy (RMSE = 7.1~8.3%). These findings demonstrate that UAV imagery, particularly combining EBI and SAM, can provide reliable flowering monitoring across varieties and sites, offering potential for practical orchard management and yield prediction under variable climate conditions.

Keywords

UAV

Pear

Orchard

Flowering

Enhanced Bloom Index

SAM

References

Atanasov, A.Z., Evstatiev, B.I., Vladut, V.N. Biris, S.S. 2024. A novel algorithm to detect white flowering honey trees in mixed forest ecosystems using UAV-based RGB imaging. AgriEngineering 6: 95-112. https://doi.org/10.3390/agriengineering6010007

10.3390/agriengineering6010007

Chakraborty, M., Pourreza, A., Zhang, X., Jafarbiglu, H., Shackel, K.A., DeJong, T. 2023, Early almond yield forecasting by bloom mapping using aerial imagery and deep learning. Computers and Elecronics in Agriculture 212: 108063. https://doi.org/10.1016/j.compag.2023.108063

10.1016/j.compag.2023.108063

Chen, B., Jin, Y., Brown, P. 2019. An enhanced bloom index for quantifying floral phenology using multi-scale remote sensing observations. ISPRS Journal of Photogrammetry and Remote Sensing 156: 108-120. https://doi.org/10.1016/j.isprsjprs.2019.08.006

10.1016/j.isprsjprs.2019.08.006

Chen, J., Shen, M., Zhu, X., Tang, Y. 2009. Indicator of flower status derived from in situ hyperspectral measurement in an alpine meadow on the Tibetan Plateau. Ecological Indicators 9: 818-823. https://doi.org/10.1016/j.ecolind.2008.09.009

10.1016/j.ecolind.2008.09.009

Jang, H.I., Seo, H.H., Park, S.J. 2002. Strategy for fruit cultivation research under the changing climate. Korean Journal of Horticultural Science and Technology 20: 270-275 (in Korean).

Jeong, J.H., Han, J.H., Cho, J.G., Lee, D.Y., Lee, W.K., Jang, S.H., Ryu, S. 2023. Model evaluation for predicting the full bloom date of apples based on air temperature variations in South Korea’s major production regions. Journal of Bio Bio-Environment Control 32(4): 501-502 (in Korean). https://doi.org/10.12791/KSBEC.2023.32.4.501

10.12791/KSBEC.2023.32.4.501

Jordan, C.F. 1969. Derivation of leaf area index from quality of light on the forest floor. Ecology 50: 663-666. https://doi.org/10.2307/1936256

10.2307/1936256

Kim, J.H., Yun, E.J., Kim, D.J., Kang, D.G., Seo, B.H., Shim, K.M. 2020. Evaluation of regional flowering phenological models in Niitaka pear by temperature patterns. Korean Journal of Agriculture For Meteorol 22: 268-278 (in Korean).

10.5532/KJAFM.2020.22.4.268

Kim, D.J., Kim, J.H. 2018. An outlook of changes in the flowering dates and low temperature after flowering under the RCP8.5 projected climate condition. Korean Journal of Agricultural and Forest Meteorology 20(4): 313-320 (in Korean).

Kim, S.O., Kim, J.H., Chung, U., Kim, S.H., Park, G.H., Yun, J.I. 2009. Quantification of temperature effects on flowering date determination in niitaka pear. Korean Journal of Agricultural and Forest Meteorology 11(2): 61-71 (in Korean). https://doi.org/10.5532/KJAFM.2009.11.2.061

10.5532/KJAFM.2009.11.2.061

Kim, S., Jeon, J., Seok, S., Jung, J., Kim, T. 2025. Development of agricultural parcel detection model based on YOLO-seg using CAS500-1 satellites images. Journal of the Korean Society of Agricultural Engineers 67(2): 55-67(in Korean).

10.5389/KSAE.2025.67.2.055

Lee, K.D., Na, S.I., Baek, S.C., Park, K.D., Choi, J.S., Kim, S.J., Kim, H.J., Yun, H.S., Hong. S.Y. 2015. Estimating the amount of nitrogen in hairy vetch on paddy fields using unmaned aerial vehicle imagery. Korean Journal of Soil Science and Fertilizer 48(5): 384-390(in Korean). https://doi.org/10.7745/KJSSF.2015.48.5.384

10.7745/KJSSF.2015.48.5.384

Popescu, D., Ichim, L., Stoican, F. 2023. Orchard monitoring based on unmanned aerial vehicles and image processing by artificial neural networks: a systematic review. Frontiers in Plnt Science 14: 1237695. https://doi.org/10.3389/fpls.2023.1237695

10.3389/fpls.2023.123769538089793PMC10711071

RDA (Rural Development Administration), 2012. Agricultural science technology standards for investigation of research. RDA Jeonju Korea (in Korean).

Song, A. 2025. Assessment of the segment anything model’s applicability to coastal debris segmentation: A comparative analysis of prompt input methods. Korean Journal of Remote Sensing 41(3): 593-603(in Korean). https://doi.org/10.7780/kjrs.2025.41.3.8

10.7780/kjrs.2025.41.3.8

Tomas, J.R. Gausman, H.W. 1977. Leat reflectance vs. leaf chlorophyll and carotenoid concentrations for eight crops. Agronomy Journal 69(5): 799-802. https://doi.org/10.2134/agronj1977.00021962006900050017x

10.2134/agronj1977.00021962006900050017x

Zhang, C., Valente, J., Kooistra, L., Guo, L. Wang, W. 2021. Orchard management with small unmanned aerial vehicles: a survey of sensing and analysis approaches. Precision Agriculture 22: 2007-2052. https://doi.org/10.1007/s11119-021-09813-y

10.1007/s11119-021-09813-y

Zhao, Y., Li, Y., Xu, X. 2024. Object detection in high-resolution UAV aerial remote sensing images of blueberry canopy fruits. Agriculture 14: 1842. https://doi.org/10.3390/agriculture14101842

10.3390/agriculture14101842

Zhao, Z., Fan, C., Liu, L. 2023. Geo SAM: A QGIS plugin using Segment Anything Model (SAM) to accelerate geospatial image segmentation (1.1.0). Zenodo.

Information

Publisher :Korean Society of Precision Agriculture
Publisher(Ko) :한국정밀농업학회
Journal Title :Precision Agriculture Science and Technology
Journal Title(Ko) :정밀농업과학기술
Volume : 8
No :1
Pages :83-93
Received Date : 2026-02-12
Revised Date : 2026-03-23
Accepted Date : 2026-03-29
DOI :https://doi.org/10.22765/pastj.20260007

Precision Agriculture Science and Technology ISSN:2672-0086(Print) 2713-5632(Online) 정밀농업과학기술

All Issue