SOURCE: IDRW.ORG.
The Indian Ministry of Defence (MoD) has issued a call to Indian industries, particularly MSMEs and startups, to develop Automatic Defect Recognition (ADR) software for weld radiography inspection, specifically for naval applications. This initiative is crucial as India’s defense forces are heavily involved in the fabrication of steel hulls for various naval vessels, and ensuring the integrity of these welds is critical for safety and performance.
Weld integrity is a vital aspect of naval construction, particularly in the fabrication of steel hulls that form the backbone of ships and submarines. The Indian Navy relies on nondestructive testing (NDT) methods to inspect these weldments, with Digital Radiographic Testing (DRT) being a widely adopted technique. DRT includes digitized film-based and computed radiography (CR)-based approaches that capture high-resolution images of welds for evaluation.
However, the manual analysis of these digital radiography images is time-consuming, requiring extensive effort from Quality Control (QC) inspectors to assess and approve or reject each weld. Moreover, the manual interpretation of these images demands a high level of expertise, a skill that is increasingly difficult to find. Recognizing these challenges, the MoD has identified the need for automated solutions that can accelerate the inspection process while improving reliability and accuracy.
To address the limitations of manual inspection, the MoD is encouraging the development of Automatic Defect Recognition (ADR) software that can autonomously evaluate radiographic images of welds. This intelligent software will not only reduce the time taken to analyze these images but also enhance the consistency and precision of defect detection.
The MoD envisions the ADR software to be based on a combination of rule-based and artificial intelligence (AI)-based algorithms. These two approaches complement each other, allowing the system to handle a wide range of defect scenarios while continuously improving its detection capabilities.
The proposed ADR software must incorporate advanced Deep Learning Convolutional Neural Networks (DL-CNN), a form of AI well-suited for image recognition and analysis. DL-CNN algorithms are particularly effective in identifying intricate patterns and defects in radiographic images, making them ideal for detecting flaws in weldments with high precision.
Additionally, rule-based algorithms will provide a structured approach to defect recognition, using predefined parameters to assess weld quality. This combination of AI and rule-based methods ensures that the software can adapt to a wide variety of welding environments and materials, increasing its versatility for naval applications.
The development of such ADR software will bring multiple benefits to India’s defense sector. By automating the inspection process, the naval shipbuilding industry can improve efficiency, reduce costs, and maintain high standards of safety. Furthermore, with the increasing complexity of naval vessels, this technology will become essential in ensuring that each weld meets stringent quality standards without delaying construction timelines.
NOTE : Article cannot be reproduced without written permission of idrw.org in any form even for YouTube Videos to avoid Copy right strikes. Websites doing illegal reproductions will get DMCA and Legal Notices.