In Senegal, farmers in general and rice growers particularly are still facing many issues such as climatic hazards and water-scare environments in their daily life. A very acute and challenging problem for rice crops remains, however, their destruction by pest birds. These latter attack the rice crops when they are mature, leaving the farmers in disarray and without solution. Indeed, such an attack results in a drastic reduction in yields during harvest. Over time, many repellent techniques like scarecrow have been used, but show their limitations. In this paper, we tackle this problem and propose a pest birds detection approach in rice crops using pre-trained YOLOv4 detector and transfer learning. To show the efficiency of our model we conduct experiments on a real bird dataset, exhibiting a mean average precision of 96 %.
CITATION STYLE
Diakhaby, I., Ba, M. L., & Gueye, A. D. (2022). Pest Birds Detection Approach in Rice Crops Using Pre-trained YOLOv4 Model. In Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST (Vol. 449 LNICST, pp. 223–234). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-23116-2_19
Mendeley helps you to discover research relevant for your work.