Scholarlink

In nowadays we are witnessing the rapid rise of information technologies within production systems. In this context, the image processing plays a major role. For application engineers, the market offers a wide range of vision systems with comprehensive vision tool libraries that are useful for most of the applications. Problem is the ambiguity of data, which can be caused by unstable input parameters. This has a decisive influence on reliability of the entire production system. For this reason, the use of machine learning algorithms would be beneficial for this purpose. This paper shows a simple practical example of using a neural network to process data from a vision system to improve the accuracy of data sent to the central Programmable Logic Controller (PLC) system via an Open Platform Communications Unified Architecture (OPC UA) server.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Improving Reliability of Image Processing Application using Feedforward Backpropagation Neural Network

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Patrik Gašparovič,
  • Martin Juhás,
  • Igor Halenár,
  • Milan Daňo,
  • Bohuslava Juhásová,
  • Fedor Burčiar

摘要

In nowadays we are witnessing the rapid rise of information technologies within production systems. In this context, the image processing plays a major role. For application engineers, the market offers a wide range of vision systems with comprehensive vision tool libraries that are useful for most of the applications. Problem is the ambiguity of data, which can be caused by unstable input parameters. This has a decisive influence on reliability of the entire production system. For this reason, the use of machine learning algorithms would be beneficial for this purpose. This paper shows a simple practical example of using a neural network to process data from a vision system to improve the accuracy of data sent to the central Programmable Logic Controller (PLC) system via an Open Platform Communications Unified Architecture (OPC UA) server.