Dataset of Annotated Rice Panicle Image from Bangladesh
Description
This dataset focuses on drone-based rice panicle detection in Gazipur, Bangladesh, offering valuable visual data to researchers in agricultural studies. Captured using an advanced drone with a 4K resolution camera, the dataset comprises 2193 high-resolution images of rice fields and 5701 images after augmentation. All the images are annotated with precision to aid in automated rice panicle identification. Its main purpose is to support the development of algorithms and systems for critical agricultural tasks like crop monitoring and yield estimation, as well as disease identification and plant health evaluation. The dataset's creation involved extracting frames from drone-recorded video footage and meticulously annotating them with manual and deep learning algorithms using a semi-automatic approach.