In modern industrial environments, the quality of surfaces is increasingly decisive for the safety, performance and reliability of entire systems. Whether micro-cracks in aerospace components, contaminants on wafers, material defects in medical technology or minimal structural deviations in high-precision components – many critical defects lie far beyond what can be reliably detected using conventional camera systems or the human eye.
It is precisely at this boundary that the technology from Opto GmbH comes into play with the solino system. The company is pursuing an approach that no longer primarily analyses the visible image of a surface, but rather its optical fingerprint.

From conventional image capture to reflection analysis

Conventional industrial image processing usually focuses on visible structures, colour deviations or geometric features. However, many critical material defects do not manifest themselves directly in the image itself, but in the way a surface reflects light.
It is precisely this physical effect that solino exploits. At the heart of the system is a computer-aided imaging approach that specifically analyses the interaction between light and material.
To this end, the test object is illuminated from various angles by 64 high-intensity LEDs. At the same time, a high-resolution 20-megapixel camera records each individual lighting scenario separately. The result is not a conventional photograph, but a multidimensional reflectance map of the surface.
This contains information on how materials react to light – including the finest variations in roughness, structure or material properties.

BRDF profiles make the invisible visible

Technologically, the process is based on so-called BRDF profiles (Bidirectional Reflectance Distribution Function). This involves a mathematical analysis of how light is reflected at different angles of incidence and observation.
Whilst conventional cameras primarily capture brightness and colour values, BRDF analysis allows for a significantly more detailed characterisation of surfaces.
Microscopic irregularities, minimal impurities, material stresses or the finest scratches produce specific reflection patterns that can be evaluated algorithmically. Many of these changes remain invisible to conventional machine vision systems or are difficult to detect reproducibly.
This capability is becoming increasingly important, particularly in highly sensitive industrial sectors. This is because modern manufacturing processes are increasingly operating within the realm of microscopic tolerances, where even the smallest surface deviations can have safety-critical implications.

AI and optical metrology are converging

The approach becomes particularly interesting through the combination of optical measurement technology and artificial intelligence. The so-called ‘anomaly map’ generated by solino can be integrated directly into machine learning workflows.
This creates a hybrid analysis system that combines physical light measurement with AI-supported pattern recognition.
In doing so, the AI processes not only visible image information but also highly complex reflection data. This opens up new possibilities for automated defect classification, predictive quality management and early defect detection.
In industrial practice, it is precisely this combination that is increasingly becoming the standard in modern quality control. This is because traditional rule-based machine vision systems often reach their limits when dealing with complex material surfaces – particularly when defect patterns are variable, difficult to define or occur statistically rarely.
AI-based systems, on the other hand, can detect subtle deviations, learn patterns across large volumes of data and continuously improve quality predictions.

Non-contact surface analysis is gaining strategic importance

Another advantage of modern optical methods lies in their non-contact operation. Sensitive materials or high-precision manufacturing processes, in particular, must often not be subjected to mechanical influence.
solino operates entirely non-contact and, at the same time, largely independently of ambient light. This makes the system suitable for both laboratory environments and industrial process integration directly into production lines.
This feature is particularly crucial in automated manufacturing environments. There, inspection systems must not only operate with precision but also deliver high throughput rates, process stability and reproducible results.
The ability to perform batch processing and real-time analysis via an intuitive user interface further supports integration into existing Industry 4.0 architectures.
Safety-critical industries are driving development
The fields of application for such systems now extend far beyond traditional industrial image processing. Industries with high regulatory or safety-critical requirements, in particular, are increasingly investing in high-precision surface analysis.
In aerospace, even the smallest material irregularities can affect the structural integrity of components. In semiconductor manufacturing, microscopic particles or defects determine the functionality of entire wafer batches. In medical technology, on the other hand, sterile, flawless and reproducible surfaces play a central role.
Added to this are applications in battery production, photonics, sensor technology and high-performance electronics.
Particularly in the security and KRITIS sectors, the ability to detect material conditions early and automatically is becoming increasingly relevant. This is because production quality, operational safety and failure protection depend ever more on high-precision condition monitoring.
The next stage in the evolution of industrial image processing
Overall, this development signals a fundamental shift within industrial image processing. Systems no longer analyse just visible images, but increasingly the physical properties of the materials themselves.
This shifts the focus from traditional inspection towards data-driven material intelligence.
Optical fingerprints, AI-supported anomaly detection and multidimensional reflection analyses could become a central component of modern quality and safety infrastructures in the coming years – particularly wherever the smallest deviations determine function, safety or service life.
The approach developed by Opto exemplifies how photonics, AI and industrial measurement technology are increasingly merging into intelligent sensor systems that go far beyond what the human eye can perceive.
‘Vision beyond the visible’ is thus becoming not only a technological guiding principle, but increasingly a reality in modern industrial quality control.