Simplified Object Detection for Manufacturing: Introducing a Low-Resolution Dataset

Machine learning (ML), particularly within the domain of computer vision (CV), has established solutions for automated quality classification using visual data in manufacturing processes. Object detection as a CV method for quality classification provides a distinct advantage in enabling the assessment of items within the manufacturing environment, regardless of their location in images. However, substantial challenges remain regarding labeled data availability in manufacturing contexts, training examples, data imbalance, and the complexity of incorporating these methods into real-world applications. Furthermore, real-world datasets often lack adherence to FAIR principles, which limits their accessibility and interoperability, especially for small- and medium-sized enterprises (SMEs) working to integrate object detection into their manufacturing processes. In this article, we present a low-resolution 640x640 dataset based on plastic bricks for object detection, featuring two quality labels to identify minor surface defects as an example of quality classification. We analyze the dataset using a YOLOv5 model on three different dataset sizes, while accounting for class imbalance, to demonstrate the accuracy of an object detection model in a simple manufacturing use case. The mean Average Precision [email protected] for correctly identifying instances in our testing dataset ranges from 0.668 to 0.774, depending on dataset size and class imbalance. While our focus is on demonstrating object detection with low-resolution images and limited data availability, the generated data and trained model also adhere to FAIR principles.Therefore, these resources are made available with proper metadata to support their reuse and further investigation into object detection tasks for similar quality classification use cases in manufacturing.

NFDI4Ing Survey 2024 - Call - NFDI4Ing

Your feedback is important for the development of research data management in engineering. The ‘NFDI4ING Community Survey 2024’ has started!

A survey on the dissemination and usage of research data management and related tools in German engineering sciences

As the amount of collected and analysed data increases, a need for data management arises to ensure its usability. This also applies in research. This challenge can be addressed by Research Data Management (RDM), which brings clear focus on the reusability of data. To understand the status quo of the application of research data management in engineering sciences in Germany, as well as possible challenges and improvement chances, a survey was conducted over the last quartal of 2020. Over 168 (n=168) researchers from the engineering sciences in Germany provided their view via a questionnaire that contains 216 question items. The results give intel on the interviewees knowledge and perceived relevance of research data management in their daily research activities. For instance, the application of research data management related tasks, data sharing with third parties, usage of different tools, and the involvement of different file formats were part of the survey. The survey closed with questions regarding RDM specifications, support structures, and questions on reasons that could prevent researchers from adapting sustainable RDM. This paper presents the results of the study, providing an overview over the current RDM in engineering and pointing out possible measures and strategies to foster it, namely the integration of guidance and education for research data management. Along the paper, we publish the collected data set to enable further analysis and reuse (e.g. for extended statistical analysis).

The 2022 NFDI4ing Conference Special Issue is now completed

We are happy to announce that the 2022 NFDI4ing Conference Special Issue is now completed! For this issue, seven manuscripts underwent a successful open peer review process in our preprint …

Call for Papers: Ethical and Responsible Commitments for Sharing (FAIR) Data in Engineering Sciences

As Artificial Intelligence (AI) continues to evolve, researchers are increasingly grappling with its implications. For instance, psychologists and neuroscientists are working on explainable AI (XAI) to understand how AI systems …

ing.grid's first issue is now complete

We are happy to announce that the first issue of ing.grid is now complete. This first issue contains specimens of two types of ing.grid open peer-reviewed publications: a manuscript and …

ing.grid longlisted for Enter Award 2024

We are very happy to announce that ing.grid has been longlisted for the Enter Award 2024, funded by the Federal Ministry of Education and Research (BMBF) and powered by the …

All Preprints