The automatic detection of best-before dates (BBD) is a central component of modern production and packaging lines. It is used for quality assurance, traceability and process automation.
However, reading this data places high demands on image processing systems: changing packaging materials, reflective surfaces, variable fonts and short cycle times make best-before date detection a real challenge.
With intelligent AI-based image processing systems, such as the Smart Camera B60 from wenglor, this task can now be solved reliably, quickly and stably – even under the most difficult conditions.
Challenge:
In highly automated production lines, processes run with very short cycle times. Whether during sorting, packaging or quality control – the best-before date and other product data such as batch or serial numbers must be recognized and processed in real time.
Any delay in text recognition can slow down the entire production flow or even lead to downtime.
Solution:
The Smart Camera B60 from wenglor with integrated Deep OCR algorithm enables fast and accurate text recognition even at high belt speeds.
The AI-based algorithm is able to reliably recognize varying fonts, faint print images and changing lighting conditions – in real time.
This keeps the product line efficient without sacrificing quality in data recognition.
Challenge:
Many packages are made of glossy or transparent films. These cause reflections that sometimes make the best-before date printed on them illegible.
In addition, wrinkles, stretching or deformation can lead to distorted characters and quickly push classic OCR systems to their limits.
Solution:
The combination of Machine Vision cameras, a Machine Vision controller or the Smart Camera B60 with targeted illumination technology ensures a clear view of the essentials.
Disturbing reflections are effectively reduced through the use of diffuse illumination and polarization filters.
The Deep OCR module in the uniVision 3 software also recognizes distorted or partially obscured characters – even on shiny, reflective or corrugated films.
The result: reliable recognition of date information even under challenging light and material conditions.
Challenge:
Best before dates are often printed on yoghurt pots, cans or plastic packaging on curved, glossy or color-changing backgrounds.
Conventional systems often fail due to weak contrasts, distorted lettering or changing print quality.
Solution:
The Smart Camera B60 with integrated Deep OCR function recognizes even difficult-to-read characters stably and reliably – regardless of font position, contrast or background color.
In combination with adapted illumination technology and individually adjustable exposure, the system automatically compensates for fluctuations in brightness.
This means that the best-before date is clearly and unambiguously recognized even on curved surfaces without affecting the production flow.
Challenge:
It is particularly difficult to identify batch numbers on textured or natural surfaces, such as wooden packaging or parquet materials.
Here, grains, pores or unevenness can overlay the print or distort the typeface.
However, clear identification of the production or expiration date is crucial for quality assurance, especially for materials that are further processed or stored.
Solution:
The Smart Camera B60 recognizes text even on irregular, low-contrast surfaces.
Thanks to powerful deep OCR algorithms and flexibly adjustable parameters in the uniVision 3 software, even faintly printed characters can be reliably captured.
The use of infrared illumination means that the wood grain is specifically masked out, as IR light minimizes deeper structures and reduces disruptive reflections.
The result: clear data recognition even on challenging, natural materials.
From glossy films to structured surfaces – the reliable detection of best-before dates places high demands on industrial image processing.
With the Smart Camera B60 and uniVision 3 software, wenglor offers a complete solution that combines speed, precision and flexibility:
This makes best-before date detection a robust, fully automated process that sustainably improves quality and efficiency in production.
Job title