Real-Time Weed Location Estimation in Cassava Field Using an Aerial Multispectral Camera

Main Article Content

Pongsakon Bamrungthai
Nattagorn Pramona
Prasatporn Wongkamchang


This research presents a real-time weed location estimation system for precision agriculture in cassava planting. The widely employed vegetation index known as the normalized difference vegetation index (NDVI) was applied to identify and estimate the locations of the weed. The multispectral camera mounted on an unmanned aerial vehicle (UAV) with nadir orientation was used to capture the field images. The NDVI values were calculated in real-time using an onboard microcomputer and streamed weed locations in latitude/longitude format to the ground control station. The UAV with the attached camera was controlled to follow a predefined flight path using user-specified coordinates based on the configuration of the planting area. The flight altitude was set at 10 m above ground level. Experimental flight tests were conducted over a cassava field covering an approximate area of 2,500 m2. The results demonstrated that detected weed locations exhibited errors within the precision bounds of the GPS system. In the future, a spraying system could be implemented on the UAV to eliminate weeds and perform other planting operations.

Article Details

How to Cite
Bamrungthai, P., Pramona, N., & Wongkamchang, P. (2023). Real-Time Weed Location Estimation in Cassava Field Using an Aerial Multispectral Camera. NKRAFA JOURNAL OF SCIENCE AND TECHNOLOGY, 19(2), 1–8. Retrieved from
Research Articles


Tirawatsakul, M. (1983). Weed control in cassava. In: Cassava Monograph No.7, Department of Agriculture, Bangkok, Thailand. 163 p. (in Thai)

Nebiker, S., Annen, A., Scherrer, M. and Oesch, D., "A Light-Weight Multispectral Sensor for Micro UAV-Opportunities for very High Resolution Airborne Remote Sensing," The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 37, 1193-2000 (2008).

Martin D.B. et al., (2010). UAV-based environmental monitoring using multi-spectral imaging, Proc. SPIE 7668, Airborne Intelligence, Surveillance, Reconnaissance (ISR) Systems and Applications, Vol. 7, pp. 766811. Orlando, Florida, United States: SPIE

Y. A. Pederi and H. S. Cheporniuk. (2015). Unmanned aerial vehicles and new technological methods of monitoring and crop protection in precision agriculture. IEEE International Conference Actual Problems of Unmanned Aerial Vehicles Developments (APUAVD) (298–301). Kyiv, Ukraine: IEEE

Randal W. Beard and Timothy W. McLain. (2012). Small Unmanned Aircraft: Theory and Practice , Second Edition. New Jersey: Princeton University Press.

Louargant, M., Jones, G., Faroux, R., Paoli, J.-N., Maillot, T., Gée, C., & Villette, S. (2018). Unsupervised Classification Algorithm for Early Weed Detection in Row-Crops by Combining Spatial and Spectral Information. Remote Sensing, 10(5): 761

Sunil G C, Yu Zhang, Cengiz Koparan, Mohammed Raju Ahmed, Kirk Howatt, Xin Sun. (2022). Weed and crop species classification using computer vision and deep learning technologies in greenhouse conditions, Journal of Agriculture and Food Research, Volume 9: 1-15.

Corceiro, A., Alibabaei, K., Assunção, E., Gaspar, P. D., & Pereira, N. (2023). Methods for Detecting and Classifying Weeds, Diseases and Fruits Using AI to Improve the Sustainability of Agricultural Crops: A Review. Processes, 11(4): 1-40.