Fault Detection and Diagnosis of Cyber-Physical System Using the Computer Vision and Image Processing

Abstract

In the techno world, Corporate Business applies new technologies for manufacturing and production with numerous cyber-physical system strategies. This makes the process depend upon multiple computers, machines, and applications with varying specifications, efficiency, and latency. These technological strategies are extremely diverse on cyber-physical systems, from an extensive range of processing technologies is available. The currently available technologies are not well adapted to these processes, which require information management regarding fault detection and diagnosis at a complexity level separated from technology. In this article, the Image Processing assisted Computer Vision Technology for Fault Detection System (IM-CVFD) is suggested to resolve such issues in industrial cyber-physical systems. Group Activation Mapping Algorithm is presented for efficient information collection from the processed output, simplifying the managing of fault details with different needs. Besides achieving the optimized information concerning latency, efficiency, the Uncertainty Reduction algorithm is introduced. In a suitable processing environment, a detailed simulation is conducted. The empirical findings indicate the high efficiency of the IM-CVFDwitha with a minimum error rate, energy usage, and minimized delay with high service. In contrast with conventional methods, the IM-CVFD obtains a better result efficiently.

Artificial Intelligence-Based Decision-Making Algorithms, Internet of Things Sensing Networks, and Deep Learning-Assisted Smart Process Management in Cyber-Physical Production Systems

“…As underlying mechanisms between cyber-physical production networks, deep learningassisted smart process planning, product decision-making information systems, artificial intelligence-based decision-making algorithms, and cognitive automation in sustainable manufacturing Internet of Things are comprehended to a limited extent in the current literature, certain empirical studies [81][82][83][84][88][89][90][91][92][93][97][98][99][103][104][105]108,164] systematically indicate that smart connected devices have heterogeneous processing and manufacturing capabilities and optimization operation mechanisms. Manufacturing process monitoring systems have advanced as decentralized reconfigurable networked entities by use of cutting-edge intelligent machines.…”

“…The convergence of standard automation systems within CPPSs, together with serviceoriented designs and fog, edge, and cloud computing technologies [103][104][105], are developing sustainable manufacturing Internet of Things and cyber-physical process monitoring systems. For the purpose of ensuring robust manufacturing processes, any disruptions throughout the system have to be monitored by operational technology and big data services.…”

Image Data-Centric Visual Feature Selection on Roll-to-Roll Slot-Die Coating Systems for Edge Wave Coating Defect Detection

“…Studies focusing on vision image-based defect detection methods have primarily focused on the structure of deep neural networks to facilitate efficient learning [ 25 , 26 , 27 , 28 , 29 ]. Many of these studies have proposed modifications to learning algorithms or structural layers within neural networks to achieve high-accuracy detection [ 30 , 31 , 32 , 33 , 34 ]. However, the proposed PCS method selects the input data for utilization by the algorithm or model.…”