Garden Flower Rhododendron Identification Image Dataset

Currently, in the field of agriculture, forestry, and fisheries, particularly in horticultural management, accurate variety identification is a key step. However, the diversity of flowers and similarity in morphology pose challenges to identification work. Existing manual identification methods are inefficient and prone to human factors. This dataset aims to assist in building more precise automated identification models by providing high-quality Rhododendron images, meeting the actual needs of the horticulture industry. Data collection is conducted in various natural environments, using high-resolution cameras to record flower images under different lighting and angles. Quality control includes multiple rounds of expert annotation, machine-assisted inspection, and manual review to ensure the consistency and accuracy of the annotations. The annotation team consists of experts in botany and computer vision, with more than 20 people, ensuring the professionalism of the dataset. The data preprocessing process includes image cropping, color standardization, and label encoding. Data is stored in JPG format and organized by flower species.