Landscape Flora Koelreuteria Identification Image Dataset

With the acceleration of global climate change and urbanization, the management and protection of plant diversity faces significant challenges. Traditional plant identification methods rely on manual differentiation by experts, which is not only time-consuming and laborious but also insufficient in large-scale applications. This dataset focuses on the identification of Koelreuteria in landscape flora, aiming to improve the precision and efficiency of automated identification. The dataset is collected from high-quality Koelreuteria images across different seasons and environments, addressing the shortcomings of existing image datasets in terms of geographic and seasonal coverage. Data collection was conducted using high-resolution cameras under various weather conditions to comprehensively cover all growth stages of the plant. Annotation was performed through multiple rounds and cross-validation by a team composed of botany experts and data engineers, totaling 20 people, to ensure consistency and accuracy. Data preprocessing includes image denoising, normalization, and classified annotation, finally stored in a hierarchical folder structure for easy access and expansion.