Abstract:Derin öğrenme, makine öğrenmesi ve görüntü işleme teknikleri tarımsal işlerin kolaylaştırılmasında ve üretim aşamsındaki farklı problemlere çözümler geliştirilmesinde kullanılan önemli birer araç haline gelmişlerdir. Bu çalışma kapsamında AugeLab Studio’da derin öğrenme mimarilerinden CNN kullanılarak, eş zamanlı nesne tespiti için genelde tercih edilen YOLO algoritmasıyla bir tohum sayısı ve türünün tespit uygulaması geliştirilmiştir. Çalışma sonucunda 3000 iterasyonla ortalama kayıp 0.417 civarına düşürülmüş… Show more
The remote sensing technique is of great importance in agriculture in determining vegetation cover, monitoring its development, classification, and yield estimation. Various sofwares, mathematical algorithms, and statistical approaches are used to make satellite images meaningful in remote sensing. In this study, it is aimed to determine the rice plant plots and areas by using the Augelab Studio sofware, which is a new approach in artificial intelligence-supported image processing techniques. Using the RGB image covering an area of 2.5 km2 obtained from Google Earth Pro, the classification of paddy rice fields and the calculation of these areas were made. Rice fields from parcels with different plant patterns were separated using Augelab Studio artificial intelligence image processing software using filtering blocks. The real areas of the other rice parcels were determined by the coefficient created by taking the pixel area values of some of the parcels whose total area is known as a reference. It is found that total areas of rice parcels in Augelab Studio and Google Earth Pro programs to be 798 and 801 decares, respectively. It has been observed that the areas of the paddy rice parcels can be determined with high accuracy by using Augelab Studio.
The remote sensing technique is of great importance in agriculture in determining vegetation cover, monitoring its development, classification, and yield estimation. Various sofwares, mathematical algorithms, and statistical approaches are used to make satellite images meaningful in remote sensing. In this study, it is aimed to determine the rice plant plots and areas by using the Augelab Studio sofware, which is a new approach in artificial intelligence-supported image processing techniques. Using the RGB image covering an area of 2.5 km2 obtained from Google Earth Pro, the classification of paddy rice fields and the calculation of these areas were made. Rice fields from parcels with different plant patterns were separated using Augelab Studio artificial intelligence image processing software using filtering blocks. The real areas of the other rice parcels were determined by the coefficient created by taking the pixel area values of some of the parcels whose total area is known as a reference. It is found that total areas of rice parcels in Augelab Studio and Google Earth Pro programs to be 798 and 801 decares, respectively. It has been observed that the areas of the paddy rice parcels can be determined with high accuracy by using Augelab Studio.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.