Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SortScrews: A Dataset and Baseline for Real-time Screw Classification

Tianhao Fu, Bingxuan Yang, Juncheng Guo, Shrena Sribalan, Yucheng Chen

Abstract

Automatic identification of screw types is important for industrial automation, robotics, and inventory management. However, publicly available datasets for screw classification are scarce, particularly for controlled single-object scenarios commonly encountered in automated sorting systems. In this work, we introduce $\textbf{SortScrews}$, a dataset for casewise visual classification of screws. The dataset contains 560 RGB images at $512\times512$ resolution covering six screw types and a background class. Images are captured using a standardized acquisition setup and include mild variations in lighting and camera perspective across four capture settings. To facilitate reproducible research and dataset expansion, we also provide a reusable data collection script that allows users to easily construct similar datasets for custom hardware components using inexpensive camera setups. We establish baseline results using transfer learning with EfficientNet-B0 and ResNet-18 classifiers pretrained on ImageNet. In addition, we conduct a well-explored failure analysis. Despite the limited dataset size, these lightweight models achieve strong classification accuracy, demonstrating that controlled acquisition conditions enable effective learning even with relatively small datasets. The dataset, collection pipeline, and baseline training code are publicly available at https://github.com/ATATC/SortScrews.

SortScrews: A Dataset and Baseline for Real-time Screw Classification

Abstract

Automatic identification of screw types is important for industrial automation, robotics, and inventory management. However, publicly available datasets for screw classification are scarce, particularly for controlled single-object scenarios commonly encountered in automated sorting systems. In this work, we introduce , a dataset for casewise visual classification of screws. The dataset contains 560 RGB images at resolution covering six screw types and a background class. Images are captured using a standardized acquisition setup and include mild variations in lighting and camera perspective across four capture settings. To facilitate reproducible research and dataset expansion, we also provide a reusable data collection script that allows users to easily construct similar datasets for custom hardware components using inexpensive camera setups. We establish baseline results using transfer learning with EfficientNet-B0 and ResNet-18 classifiers pretrained on ImageNet. In addition, we conduct a well-explored failure analysis. Despite the limited dataset size, these lightweight models achieve strong classification accuracy, demonstrating that controlled acquisition conditions enable effective learning even with relatively small datasets. The dataset, collection pipeline, and baseline training code are publicly available at https://github.com/ATATC/SortScrews.
Paper Structure (21 sections, 4 figures, 2 tables)

This paper contains 21 sections, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Deep Learning for Image Classification
  4. Fine-Grained Visual Recognition
  5. Industrial Vision Datasets
  6. Dataset
  7. Data Collection Setup
  8. Dataset Overview
  9. Screw Categories
  10. Annotation Format
  11. Baseline Methods
  12. Network Architectures
  13. EfficientNet-B0.
  14. ResNet-18.
  15. Training Procedure
  16. ...and 6 more sections

Figures (4)

  • Figure 1: Overview of the SortScrews dataset and acquisition pipeline.
  • Figure 2: The device used for data collection, made in Myhal Fabrication Facility.
  • Figure 3: The six types of screws included in the SortScrews dataset. Their class indices are 1 to 6 from left to right.
  • Figure 4: Confusion matrices on the validation set.