Garden Plant Tamarisk Recognition Image Dataset

In agricultural and garden management, accurately identifying and classifying flowers is an important and challenging task. Traditional manual recognition methods are inefficient, easily influenced by the professional capability of the identifier, and difficult to promote in large-scale applications. Existing automated solutions, though somewhat effective, are often limited by insufficient data and inaccurate annotations. The construction of this dataset aims to improve model recognition accuracy and applicability through large-scale, high-quality image data. Data collection was conducted using professional cameras under different lighting, angles, and environmental conditions to ensure data diversity and authenticity. Quality control involves multiple rounds of annotation and consistency checking, with final review by plant identification experts. The annotation team consists of 10 members with backgrounds in botany. Preprocessing of the data includes image normalization, denoising, and enhancement, organized in JPEG format and stored by species for easy retrieval and use.