Durian Plantation Epiphyte Moss Image Dataset

Current durian plantations face the challenge of identifying epiphytic moss, which affects plant health and yield. Existing solutions often rely on manual monitoring, which is time-consuming, labor-intensive, and lacks accuracy. This dataset aims to improve the detection efficiency and accuracy of epiphytic moss through automated image recognition technology. Data collection was carried out using drones and high-definition cameras in various environments and weather conditions, ensuring adequate daylight for optimal image quality. In terms of quality control, the data underwent multiple rounds of annotation, combined with reviews by agricultural experts to ensure accuracy and consistency. The annotation team consisted of more than 20 experts in botany and image processing. Data preprocessing included image cropping, noise filtering, and color correction, with storage organized in JPG format. The dataset is organized in a structured folder format for convenient retrieval and usage. The dataset maintains a high level of annotation accuracy with a 96% recognition rate. In terms of consistency and completeness, it has undergone multiple verifications to ensure reliable data quality. In technological innovation, multi-angle shooting and AI enhancement algorithms were introduced to improve recognition performance under various lighting conditions. This dataset helps enhance durian plantation management efficiency. Compared to other datasets on the market, this dataset is more targeted and practical, especially in terms of coverage and diversity for different growth stages. It achieved over a 20% improvement in moss recognition accuracy. The dataset design makes it suitable for current agricultural research and can be applied to other plant moss identification tasks, offering a high level of versatility.