1st International Conference on Science and Technology, ICOST 2019, 2-3 May, Makassar, Indonesia

Research Article

Detection of Industrial Machine Work Errors using LVQ Neural Network

Download591 downloads
  • @INPROCEEDINGS{10.4108/eai.2-5-2019.2284705,
        author={Muh. Rafli Rasyid and Zulkifli  Tahir and Syafaruddin  Syafaruddin},
        title={Detection of Industrial Machine Work Errors using LVQ Neural Network},
        proceedings={1st International Conference on Science and Technology, ICOST 2019, 2-3 May, Makassar, Indonesia},
        publisher={EAI},
        proceedings_a={ICOST},
        year={2019},
        month={6},
        keywords={computer vision image processing learning vector quantization (lvq) neural network},
        doi={10.4108/eai.2-5-2019.2284705}
    }
    
  • Muh. Rafli Rasyid
    Zulkifli Tahir
    Syafaruddin Syafaruddin
    Year: 2019
    Detection of Industrial Machine Work Errors using LVQ Neural Network
    ICOST
    EAI
    DOI: 10.4108/eai.2-5-2019.2284705
Muh. Rafli Rasyid1,*, Zulkifli Tahir2, Syafaruddin Syafaruddin1
  • 1: Department of Electrical Engineering, Faculty of Engineering, Universitas Hasanuddin Makassar, Indonesia. 92119
  • 2: Department of Informatics, Faculty of Engineering, Universitas Hasanuddin Makassar, Indonesia. 92119
*Contact email: rasyidmr17d@student.unhas.ac.id

Abstract

In the world of industry, the utilization of technology machinery industry is one of the most important factors to facilitate the employment of human. However, an industrial machine does not work regardless of fault that can inhibit the production process and cause harm to the industry. This paper aims to detect errors with the industrial machine work to analyze the movement of industrial machinery in a video, at this stage of the process of preprocessing, image resizes, do segmentation method thresholding, and the morphological operations with the opening operation. The further step, the feature extraction performed by converting a binary image into vector data is used as input data in the classification process using Algorithm Learning Vector Quantization (LVQ) Neural Network version 1 and version 2. Research results obtained detection accuracy reached 100% for training using LVQ1 much higher than the results of the training using LVQ2 with an accuracy of only 67.59%.