Abstract

In this study, we present a framework for product quality inspection based on deep learning techniques. First, we categorize several deep learning models that can be applied to product inspection systems. In addition, we explain the steps for building a deep-learning-based inspection system in detail. Second, we address connection schemes that efficiently link deep learning models to product inspection systems. Finally, we propose an effective method that can maintain and enhance a product inspection system according to improvement goals of the existing product inspection systems. The proposed system is observed to possess good system maintenance and stability owing to the proposed methods. All the proposed methods are integrated into a unified framework and we provide detailed explanations of each proposed method. In order to verify the effectiveness of the proposed system, we compare and analyze the performance of the methods in various test scenarios. We expect that our study will provide useful guidelines to readers who desire to implement deep-learning-based systems for product inspection.

Highlights

  • Many manufacturing companies apply product inspection systems to detect product defects and to evaluate product quality

  • In this study, we propose connection schemes that efficiently link the edge servers with the existing inspection systems to improve the goals of the existing systems

  • We have proposed a unified framework for product quality inspection using deep learning techniques

Read more

Summary

Introduction

Many manufacturing companies apply product inspection systems to detect product defects and to evaluate product quality. The inspection systems examine the possibility of functional problems of the product and determine the location of the defects on the surface of the product. To this end, the inspection system generally uses several camera sensors to examine all or key parts of the products. Conventional methods have difficulty dealing with subtle changes in the environment (e.g., small changes in product location or illumination). They often fail to detect new types of defects owing to their simple criteria.

Objectives
Methods
Results
Conclusion
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call