Fault Detection and Inventory Management in Manufacturing with Plastic Bricks: A Dataset

Dataset for Smart Manufacturing

The dataset contains images of plastic bricks showcasing various colors, shapes, and minor surface damages, designed to represent the use-cases of quality classification and inventory management in manufacturing. The use cases are separated and include 3 categories for quality classification and 24 categories for inventory classification. The data is analyzed in the publication titled "Demonstrating Computer Vision to Small- and Medium-sized Enterprises in Manufacturing: Towards Overcoming Costs and Implementation Challenges". The associated research explores the development of a simple computer vision demonstrator and its demonstration to small- and medium-sized enterprises. 

Structure of Files

quality_classification (224 images)

    defect

    defect_free

    empty

inventory_classification (2732 images)

   beige_large

   beige_small

   blue_bright_large

   21 further categories (colour_shape)