Garden Flower Balloon Flower Recognition Image Dataset

In the field of agriculture and forestry, automatic recognition and classification of horticultural flowers such as balloon flowers face challenges brought by image diversity and natural environmental changes. Many existing solutions rely on traditional manual identification, which is inefficient and prone to errors. The construction of this dataset aims to provide an efficient image recognition training set for smart agricultural management, supporting automated flower classification and detection, thereby improving industry operational efficiency and accuracy. Data collection uses high-resolution cameras to shoot under different lighting and climate conditions, ensuring scene diversity. The data undergo several rounds of expert annotation, consistency checks, and multi-level reviews to ensure high annotation accuracy and consistency. The annotation team consists of botany and image analysis experts, with a scale of 20 people. Images are pre-processed with noise filtering and edge enhancement and are organized in JPG format for quick reading and training.