Jul 2026· e-Journal of Nondestructive Testing· Vol 31· 0 citations
TL;DR
Validation at demonstrator scale confirmed the system’s ability to meet the initially defined specifications, demonstrating reliable detection of both defects and organic residues, including those caused by impacts and insect accumulation on aircraft surfaces.
Abstract
The inspection of aeronautical components requires compliance with extremely high safety and quality standards, where early detection of defects and verification of surface conditions are critical for aircraft reliability. The use of robots equipped with computer vision enables the automation of these inspections, improving coverage, consistency, and safety compared to manual methods, while enhancing overall operational efficiency.
The incorporation of artificial intelligence further enhances these systems by enabling automatic defect detection, classification according to severity, and real-time inspections, reducing human errors and streamlining the inspection process.
This work presents a technically viable solution not only for defect detection but also for evaluating the cleanliness or contamination of aeronautical surfaces, using non-destructive methods based on computer vision, profilometry, and artificial intelligence, grounded in image acquisition and analysis. Validation at demonstrator scale confirmed the system’s ability to meet the initially defined specifications, demonstrating reliable detection of both defects and organic residues, including those caused by impacts and insect accumulation on aircraft surfaces.
This approach combines automation, precision, and traceability, representing a significant step forward toward advanced and safe inspection practices in the aerospace industry.
This article examines the application of machine vision systems for automated quality control at manufacturing enterprises. It analyzes the current state of the market, the hardware structure of visual inspection systems, and the technological capabilities of 2D, 3D, hyperspectral, and intelligent machine vision for defect detection, dimensional measurement, assembly verification, marking inspection, and packaging control. A classification of production tasks based on the feasibility of machine vision implementation is proposed. Special attention is paid to the selection of cameras, lenses, lighting, image processing algorithms, and integration with programmable logic controllers, MES, and ERP systems. The article presents accuracy indicators, methods for evaluating algorithm performance, and recommendations for reducing false rejects and missed defects. Based on the analysis of practical experience from Russian and international enterprises, recommendations are formulated for choosing an optimal machine vision architecture for different types of production, and promising directions for further development of the technology in the context of industrial digital transformation are identified.
Ya.A. Gaidukov, M. Naumov· Glavnyj mekhanik (Chief Mech...· 0 citations
Infrastructure assets such as bridges, roads, highways, tunnels, dams, power transmission lines, and high-rise buildings require regular inspection to ensure safety and reliability. Conventional visual inspection methods are time-consuming, expensive, and often expose inspectors to hazardous environments. Autonomous monitoring drones (UAVs) provide a faster, safer, and more cost-effective alternative by using advanced sensors, GPS, cameras, and AI-based technologies to collect and analyze structural data. This study proposes an AI-enabled autonomous drone framework for infrastructure inspection that integrates intelligent flight planning, automated data collection, computer vision-based defect detection, and condition assessment. The system accurately identifies cracks, corrosion, and other structural defects while optimizing flight paths for improved efficiency. The proposed framework enhances inspection accuracy, operational safety, and scalability compared to traditional methods, demonstrating significant potential for modern infrastructure management. Future developments may integrate swarm intelligence, edge computing, and digital twins to enable fully autonomous infrastructure monitoring and maintenance.
Yuki Nakamura· International Journal of Mod...· 0 citations
Addressing the limitations of traditional electrical equipment defect detection methods, including low efficiency, high missed detection rates, strong subjectivity, and insufficient real-time performance, this paper investigates the application of computer vision technology and corresponding strategies for improving recognition accuracy. Reliable defect identification is essential for ensuring the operational safety and electromagnetic reliability of modern power equipment and intelligent electromagnetic infrastructures. First, based on image processing, deep learning, and pattern recognition theories, a computer vision framework for electrical equipment defect detection is established, covering image acquisition, preprocessing, feature extraction, and defect classification. Second, through field investigations and experimental analysis, the visual characteristics of typical defects in transformers, circuit breakers, and insulators are systematically analyzed, while key challenges including complex environmental interference, limited small-sample recognition capability, and inadequate feature extraction are identified. Finally, an experimental platform is constructed using three representative categories of electrical engineering equipment, and comparative experiments are conducted to evaluate multiple optimization strategies. The results demonstrate that the proposed approach effectively improves defect recognition performance and provides a practical solution for intelligent electromagnetic equipment inspection, condition monitoring, and reliability enhancement in advanced power systems.
Industrial vision inspection systems increasingly combine machine vision, industrial metrology, artificial intelligence, industrial communication, and manufacturing automation within integrated cyber-physical production environments. Although extensive research addresses individual technologies, comparatively limited methodological support is available for coordinating engineering decisions across the complete development lifecycle of such systems. This paper presents the Engineering Framework for Intelligent Vision Inspection Systems (EF-IVIS), a domain-specific systems engineering approach organized into seven interconnected engineering domains: inspection requirements, image acquisition, illumination engineering, measurement strategy, artificial intelligence, industrial integration, and validation and continuous improvement. EF-IVIS was developed through a bottom-up synthesis of established engineering principles and four complementary industrial investigations addressing AI-based defect detection, measurement strategies, illumination conditions, and PLC-supported integration. The empirical assessment distinguishes direct experimental evidence, direct industrial evidence, cross-cutting retrospective evidence, and methodological support. Direct evidence supports selected aspects of image acquisition and the illumination engineering, measurement strategy, artificial intelligence, and industrial integration domains, whereas inspection requirements and validation and continuous improvement are supported primarily through cross-study synthesis and established systems engineering principles. The framework introduces cross-domain dependencies, application-specific Decision Gates, and feedback mechanisms for iterative engineering assessment. A worked industrial-integration example demonstrates how reported technical evidence can be translated into Decision Gate criteria without treating the reported values as universal thresholds. The present study supports the internal consistency and practical relevance of EF-IVIS, while its usability, repeatability, transferability, and complete lifecycle effectiveness require independent prospective validation.
Unknown authors· Italian National Conference...· 0 citations
Experimental evaluation demonstrates significant improvements in inspection accuracy, fault classification, decision-making, maintenance prediction, reduced downtime, and lower human intervention, providing a foundation for next-generation intelligent industrial automation.
O. Dahl, K. Nygaard· International Journal of Mod...· 0 citations
Autonomous robotic surface inspection combines intelligent robotics with computer vision to enable accurate, real-time, and contactless detection of surface defects such as cracks, corrosion, scratches, dents, and coating degradation. Unlike manual inspection, it improves consistency, enhances safety, reduces inspection time, and lowers operational costs. By integrating AI, deep learning, edge computing, IIoT, and Digital Twin technologies, autonomous robots can navigate complex industrial environments, capture high-resolution images, and perform automated defect analysis with minimal human intervention. The proposed framework integrates robotic navigation, visual sensing, image processing, defect classification, and maintenance decision support to achieve reliable inspection across diverse industrial sectors. This approach supports predictive maintenance, improves quality assurance, and advances smart manufacturing in Industry 4.0.
Arvind Kumar Singh, Lakshmi Narayanan· International Journal of Int...· 0 citations