Infusion Rack Roller Small Target Detection Dataset

The current medical industry faces many challenges in the monitoring and management of infusion rack rollers, such as low efficiency in manual small target identification and susceptibility to omissions. Existing solutions largely rely on manual surveillance, lacking efficient and accurate automated methods, leading to wasted medical resources and safety hazards. This dataset aims to provide high-quality training data for target detection tasks to solve the technical problems of automatically identifying small targets of infusion rack rollers, meeting the business needs of automated detection. Data collection is mainly conducted in hospital environments using high-resolution cameras to ensure image clarity and the restoration of real scenes. To guarantee data quality, a multi-round annotation and expert review method is used for data processing, ensuring consistency and accuracy of annotation results. The data are stored in JPG format, organized such that each image contains relevant annotation information, facilitating subsequent training and testing.