« Back To Case Studies
senswork GmbH Logo

Member Since 2018

LEARN MORE

senswork is an expert in machine vision systems and specializes in optical inspection, industrial image processing and testing equipment manufacturing. Our ready-to-use camera technologies for automation and quality assurance are used every day by our renowned customers in numerous industries.

Content Filed Under:

Industry:
N/A

Application:
N/A

Inspection of Glass Vials

POSTED 08/16/2024

Inspection of Glass Vials - Error Detection for Reflective Surfaces with Deep Learning Fractured Glass Vial

Background

Glass vials are essential in various industries, particularly pharmaceuticals, where they are used to store and transport delicate substances. Ensuring the integrity and cleanliness of these vials is critical, as even minor defects like fractures or dirt can compromise the product's safety and effectiveness. However, the transparent and reflective nature of glass presents significant challenges for traditional inspection methods.

Problem Statement

The primary challenge in inspecting glass vials lies in differentiating between genuine defects (such as fractures) and benign irregularities (such as dirt or reflections). Traditional rule-based machine vision systems often struggle to make these distinctions, especially when defects or dirt are located on different layers of the glass (front or back). The high variability in lighting conditions and the inherent transparency of glass exacerbate these difficulties.

Solution: ViDi Detect

ViDi Detect, an AI-based deep learning tool, was implemented to address these challenges. Unlike traditional machine vision systems, ViDi Detect can be trained to recognize and distinguish subtle variations in the glass surface, enabling it to accurately identify defects even under challenging conditions.

Implementation

The process began by creating a reference set of "good" vials, capturing images of these under various lighting conditions and with different amounts of natural dirt. This reference set was then used to train the ViDi Detect system, teaching it to differentiate between acceptable variations and true anomalies.


 

Key tasks included:

  • Detection of Fractures in the Glass: The system was trained to recognize the specific patterns and characteristics of fractures, distinguishing them from other marks or reflections on the glass surface.
  • Detection of Dirt on the Glass: By learning the appearance of dirt in various forms and under different lighting, ViDi Detect was able to accurately identify dirt particles versus glass fractures.
  • Inspection Possible Despite Transparent Surface: ViDi Detect’s deep learning capabilities enabled it to effectively inspect glass vials, overcoming the challenges posed by transparency, such as distinguishing between layers and identifying defects without interference from reflections or refractions.
  • Inspection Despite High Fluctuations of Light Reflections: The system was trained under varying lighting conditions, allowing it to maintain consistent accuracy even when light reflections fluctuated significantly, a common issue with transparent and reflective surfaces like glass.

Results

After training, ViDi Detect demonstrated a remarkable ability to accurately detect fractures and dirt on glass vials and ensured that only genuinely defective vials were flagged for further inspection.

Conclusion

The implementation of ViDi Detect for the inspection of glass vials has proven to be a game-changer. By leveraging deep learning, it overcomes the limitations of traditional machine vision, providing a reliable and efficient solution for ensuring the quality and safety of glass vials in critical applications.