Open AccessAnomaly Detector for Manufacturing Industries using Lab view
Author(s) -
Dhanush Kumar. M,
Julio C. Tafur S.,
Vijayanand S*
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.a1138.109119
Subject(s) - anomaly detection , artificial intelligence , computer science , template matching , computer vision , matching (statistics) , process (computing) , grayscale , detector , image (mathematics) , pattern recognition (psychology) , mechanical engineering , engineering drawing , engineering , mathematics , telecommunications , statistics , operating system
In assembling enterprises the extra parts can arrive in a wide scope of various sizes and shapes, yet the essential creation process is large and continues with its different stages. It begins by manufacturing steel wire into the correct shape, trailed by warmth treatment to enhance the quality and surface treatment to enhance strength, before the packaging procedure. Splits or anomaly on the extra parts like bolts are one of the serious issues in the assembling enterprises which lead to parcel of issues when utilized in any machine. By manual investigation it is hard to discover the breaks. As a solution for this problem we have designed, anomaly detector for manufacturing industries using LabVIEW to detect the defected bolts which may cause serious issues in running machines like electromagnetic interference and unnecessary vibrations. In this proposed system, the shapes are detected using geometric matching and the defects are identified by varying the threshold levels. Also, the colour matching is used to find the erosion. The proposed system. The image is converted into gray scale to compare with template image using color plane extraction and the defects are identified comparing the two images i.e., the template and the acquired image using match pattern where the patterns are matched for both the images. The image is taken in real time and compared with template image using web cam and my Rio.