Ornamental Flowers Plum Recognition Image Dataset

In the current field of agriculture, forestry, and fisheries, plant recognition, especially the recognition of plum varieties, faces challenges of low efficiency and insufficient accuracy in manual recognition. Existing solutions largely depend on human experience and simple image retrieval, which are inadequate to meet the recognition needs under complex varieties and environments. This dataset aims to enhance the automation and accuracy of plum variety recognition through image classification and machine learning algorithms. The data is collected by high-definition cameras under natural light and outdoor environments, covering multiple garden and cultivation environments. Quality control is reinforced by multiple rounds of annotation to enhance consistency and clarity, and an expert team reviews the annotations to ensure professionalism. The annotation team consists of over 20 experts in botany and image processing. Data undergoes preprocessing such as image enhancement and noise removal to improve model training effectiveness and is stored and organized efficiently in JPG format. The integrated data processing workflow ensures efficient data utilization. This dataset is characterized by its high annotation accuracy and consistency, with an annotation accuracy rate of over 95% ensuring data reliability. Its innovation lies in new data augmentation methods and quality assessment systems, improving classification and management efficiency by over 30% when applied in practical garden management. Compared to similar datasets, this dataset offers a more comprehensive variety count, with over 50 varieties, making it an indispensable research resource due to its scarce data characteristics. Additionally, the data architecture design allows it to have good scalability and generality, easily expandable to other flower species recognition scenarios.