Javier Mateos

This paper presents the investigation and results of the training and use of the popular object detectors YOLOv8 and YOLOv11 in the identification of anomalies during the painting process of wind turbines. Several data augmentation techniques have been performed on the original dataset to assess its impact during the training process and the evaluating metrics of the proposed object detectors. The classification process has been carried out through binary scenarios to determine how the data augmentation process can enhance the use of neural network behavior in real-time applications. The aim of this paper is to quantify and evaluate how the introduction of new data augmentation enhances the performance of existing state-of-the-art models relying on data augmentation and labelling techniques.

This paper presents the development of a dynamic control system designed to sort citrus fruits based on external defects identified through multispectral imaging. The proposed system combines PID controllers integrated through the OPC-UA protocol to facilitate real-time adjustments and optimize batch classification. By dynamically controlling classification thresholds, the system maximizes output within customer-specified tolerances, enhancing batch utilization and quality. Implementing Node-RED as an OPC-UA client enables seamless integration and real-time data processing, ensuring efficient and responsive control. Additionally, Docker containerization is employed to streamline deployment and scalability, enabling flexible and resource-efficient implementation across various operational environments. The system also introduces virtual classification tiers that dynamically adjust sorting thresholds, allowing for better utilization of each batch while maintaining adherence to strict quality specifications. Experimental evaluations demonstrate that the system significantly improves sorting quantities, optimizes classification efficiency, and reduces waste by adapting to varying conditions in real-time. These enhancements ensure high standards of quality and efficiency in citrus fruit classification while supporting sustainable agricultural and industrial practices through automated inspection.

The current paper addresses an event-driven application to compile the information required to deploy smart contracts in Hyperledger fabric blockchain for monetisation purposes. The communication between non-destructive inspection solutions installed on-premise in the customer facilities and the Marketplace is achieved through an opensource low-code platform. The application, named Traceability Gateway App, allows the non-destructive inspection solution to interact with the smart contract deployed in the blockchain. This integration layer is a key enabler to sustain an Inspection as a Service business model. The paper also presents a case study successfully carried out for the thermal inspection of the refrigerator's inner body parts manufactured by a thermoforming process. This allows the monitoring and continuous registration of the number of non-destructive inspections carried out by the solution, storing the uses in a private blockchain network and making them immutable and traceable for all smart contract users. The proposed framework enables the integration of non-destructive solutions with the smart contract deployed in a blockchain and linked with the Marketplace for monetisation purposes by employing lightweight event-driven applications.

This study aims to provide an overview of Zero Defect, Zero Waste, and non-destructive inspection technologies (NDITs), which play a crucial role in the early detection of defects and material consumption in industrial processes. Integrating Zero Defect and Zero Waste strategies with non-destructive inspection technologies supports Industry 4.0 by using advanced sensors, robotics, and AI to create smart manufacturing systems that optimise resources and improve quality. The analysis covers the main functionalities, applications and technical specifications of several NDITs to automate the inspection of industrial processes. It also discusses both the benefits and limitations of these techniques through benchmarking. Deploying inspection as a service solution based on NDITs with data-driven decision-making Artificial Intelligence for in-process or in-line inspection policies increases production control by reducing material waste and energy use, and by optimising the final factory cost. After a comprehensive assessment, this paper aims to examine and review recent developments in the Zero Defects and Zero Waste field due to emerging non-destructive inspection systems, and their combination with other technologies, such as augmented reality. Advances in sensors, robotics, and decision-making processes through Artificial Intelligence can increase Human–Robot Collaboration in the inspection process by enhancing quality assurance during production.