Overview

Acquisition programs are becoming increasingly sensitive to cost, schedule and performance risks. Maintenance efforts involved in corrosion inspection are significant across all branches of the armed forces, costing billions of dollars each year in manpower, equipment, and materials. By incorporating a system capable of identifying exposure to corrosive environments and corrosion of aircraft alloys, significant cost savings can be realized, not only in terms of minimized man-hours expended for inspection, but also in reducing aircraft downtime for scheduled maintenance.

Visual nondestructive inspections (NDI) for defects are critical for maintenance, repair, and overhaul (MRO) operations and are often the most economical method of detecting surface defects before they reach dangerous sizes. These inspections are conducted in a multitude of industries, including aerospace, infrastructure, and maritime. While technologies such as robotic borescopes and aerial drones have increased number and quality of available inspection data, humans remain the primary decision makers on the acceptability of a part or surface. This manual process is often costly, time-consuming, and subject to human errors that arise from mental fatigue or boredom. Computer vision (CV) and artificial intelligence (AI) algorithms have been shown to excel at image analysis tasks and have the potential to provide accurate and reliable defect recognition for automated visual inspection systems that can significantly reduce maintenance costs and improve inspection efficiency. Analatom's Automated Inspection using AI (AIDL) technology was developed as an automated, human-in-the-loop (HITL) solution for visual inspections, combining the power of Machine Learning algorithms with human expertise. The result is fast, highly-accurate inspections that are ideal for critical assets.

In addition to traditonal imaging, Analatom's AIDL takes NDI to the next level by leveraging hyperspectral imaging to enable to detection of subsurface defects in certain applications. Hyperspectral imaging is particularly effective for composite materials, including glass-fiber reinforced polymers (GFRP) and carbon-fiber composites (CFRP). This powerful combination allows for the detection of:

Subsurface defects such as cracks, voids, and internal damage.
Water ingression, which can compromise the integrity of the material.
Delamination, a common failure mode in composite structures.

Benefits of AIDL

Analatom's AIDL technology offers significant advantages for nondestructive inspections, enhancing accuracy, efficiency, and flexibility in maintenance processes:

Speed and Efficiency: Inspections are conducted quickly, reducing downtime for critical assets and increasing throughput to meet production goals. AIDL reduces the number of images requiring human review, allowing for faster, automated defect identification.
Ease of Use: Systems are intuitive and require limited training, enabling rapid adoption by maintenance teams and improving the overall effectiveness of users within the system.
Precision and Quality: Hyperspectral imaging captures detailed spectral data that, when paired with AIDL’s advanced AI algorithms, can detect surface and subsurface anomalies—such as cracks, voids, water ingression, and delamination—that standard visual methods often miss.
Improved Flexibility and Adaptability: AIDL’s AI algorithms can be rapidly retrained and tuned for new inspection tasks, offering unmatched flexibility for novel applications and ensuring adaptability to diverse maintenance systems.
Reduced Maintenance Costs: By automating defect recognition and improving inspection consistency, AIDL minimizes costly errors, reduces labor-intensive processes, and optimizes resource allocation. In one application at Robins AFB, one installment reduced inspection times by 70%, resulting in an estimated $70k per month savings in labor alone.
Continuous Improvement: Errors are tracked to correct misclassifications, and models are continuously updated, ensuring the system remains accurate, reliable, and up-to-date with evolving inspection requirements.

By combining cutting-edge hyperspectral imaging with AI-driven automation, Analatom provides a streamlined, cost-effective solution for high-precision inspections, empowering industries to enhance asset reliability, reduce operational costs, and optimize maintenance strategies.

Case Study

Borescope Inspection Automation reduced lead times by 70%

Previously, MRO technicians at Robins Air Force Base relied on a time-consuming process for borescope inspections. They inserted the borescope into the engine or component, captured dozens of images, and then painstakingly reviewed each one by hand, marking any defects on paper forms. With the introduction of AIDL software, this entire workflow was transformed. Using computer vision, AIDL automatically identifies and flags defects within the inspection images (Figure 1), allowing technicians to focus their attention on only the most critical or ambiguous cases. Their feedback on these edge cases is then fed back into the model, continuously improving its accuracy over time. The digital output of AIDL allows users to quickly export inspection results or generate a digital damage map that mirrors the manually created versions. By reducing the manual workload and improving accuracy, Robins AFB reported a remarkable 70\% reduction in inspection time, highlighting the efficiency gains and reduced labor cost achieved through this AI-powered solution.

Model identification of a defect from a borescope on a C-130 propeller blade installment.

Defect heatmap generated from a SWIR image on F-15 rudder. The AI model localizes and highlights the region of the image predicted to contain the defect.

Live demonstration of SWIR-AIDL detecting defects on composite material in near real-time

Defect heatmap generated from a SWIR image taken on an F-15 rudder. The AI model localizes and highlights the region of the image predicted to contain the defect.

Case Study #2

Unlocking AI-powered Subsurface Inspection

AIDL was expanded to utilize shortwave-infrared (SWIR) imaging under an SBIR Phase II contract for identification of defects on F-15 rudders. This system highlighted and automatically identified defects that were not visible to the human eye, capturing anomalies that would otherwise be left unseen by traditional inspection methods. The AI solution was deployed at Robins AFB as a web application within the Amazon Web Services ecosystem, creating a lightweight scalable system that delivers consistent and accurate defect detection capabilities across multiple platforms The AIDL system provided substantial resource and time savings by improving inspection efficiency through the use of convolutional neural networks (CNNs) to reduce the number of images that require manual review and increase the accuracy of inspections.

Key Industries