Garden Flower Salvia Identification Image Dataset

Currently, with the development of urban greening and the agriculture and forestry industry, accurate identification and management of garden vegetation have become important tasks. However, due to the large variety of vegetation, manual identification is inefficient and lacks consistency. At the same time, traditional plant identification methods are highly dependent on the environment and difficult to adapt to changes in different scenarios. This dataset aims to provide an efficient and accurate solution for Salvia identification to improve garden management efficiency. Data collection was performed using high-resolution camera equipment under different lighting and weather conditions to ensure sample diversity. Quality control involved multiple rounds of annotation, cross-validation, and expert review to ensure consistency and accuracy of the annotations. The annotation team consists of botany experts and professional data annotators with a scale of more than 30 people. Data preprocessing covers steps such as image normalization and noise reduction to enhance model training outcomes. Data storage adopts a hierarchical folder structure for effective organization and retrieval. The Garden Flower Salvia Identification Image Dataset has significant advantages in data quality, with an annotation accuracy exceeding 98% and consistency above 95%. A deep learning-based auto-annotation method was introduced to improve annotation efficiency. This dataset addresses plant identification issues in unmanned garden management, enhancing identification accuracy by 20%. Compared to traditional datasets, this dataset excels in breadth and data diversity, supporting seamless integration between heterogeneous platforms. Additionally, its dataset structure is easy to integrate and expand within company internal systems, possessing extensive industry applicability.