Akademik Veri Yönetim Sistemi
II. International Global Climate Change Congress, Çanakkale, Türkiye, 14 - 16 Eylül 2022, ss.1-3
One of the most important factors causing yield loss in paddy is weeds. Identifying weeds is quite
difficult, especially in places watered with a pan. In classical remote sensing studies, satellite images with a
certain resolution, which were usually obtained in certain periods, are used. However, it is both costly to
access these images and it is not always possible to get what belongs to the desired time. For this reason,
unmanned aerial vehicles (UAV) have started to be used both in weed detection and spraying today. In this
research, it is aimed to make the detection, classification and spatial distribution of weeds in the rice fields by
processing the images taken with the RGB (Red-Green-Blue) sensor quickly with the AugeLab Studio program.
The research was conducted in the district of Ipsala, which has 17% of the cultivated areas in Turkey in 2022.
Paddy production was carried out in 25 different randomly selected areas. During the study, images were
taken from these areas with a UAV every 10 days after a certain time after October. The images taken were
processed with artificial intelligence supported image processing technique in AugeLab Studio program. The
selected fields were entered and the weeds were separated according to their types and unique samples were
taken for each type. Thus, the relationship between the data of the two different methods obtained was
Deciphered. In the last of these (Echinochloa spp.) and Diplachne fusca (L.) P.Beauv , which pose an
important problem in rice cultivation, have been identified. Compared to the methods in question, weeds are
detected with high accuracy. As a result of this study, it is thought that significant drug and time savings can
be achieved in weed control by working in coordination with spraying UAVs used in the fight against weeds in
the rice fields.