Toner Image Classification Dataset

The current retail e-commerce industry faces challenges such as a wide variety of products and consumer choice difficulties, especially in the cosmetics field. Toner, as a common product, still has issues in classification and recognition. Existing image classification solutions often rely on traditional machine learning methods, lack sufficient labeled data, and effective feature extraction algorithms, resulting in low classification accuracy. This dataset aims to provide rich toner images for machine learning and deep learning models, addressing the problems of low classification accuracy and insufficient data. Data collection uses high-resolution photography techniques, conducted in well-lit environments to ensure image clarity and authenticity. Multi-round annotation and expert review quality control measures are used during the labeling process to ensure data accuracy and consistency. The data will be stored in JPG format, organized by product category, facilitating subsequent model training and optimization.