Industrial Machine Vision

Food Grading

Customised Vision Applications

For automated food grading, one of the key drivers of innovation is the fourth industrial revolution, industry 4.0, and the advancement of smart factories.

With an increased focus on improving productivity within the industry, image-based technologies play an important role in improving product quality, production efficiency, food product safety, and freeing up human resources for meaningful work.

With safety and reliability at the centre of attention, we have developed a range of computer vision solutions for the food and beverage industry that provide affordable, high-quality results. When designing a vision solution, we also take the conditions of the production environment into account. These conditions include temperature, humidity, and exposure to various kinds of substances. Our camera units are often used in factory automation projects for 24-7-365 production lines, that are dependent on reliable products and support.

Are you ready to be positively surprised? Then look into the opportunities offered by automatic inspection and monitoring.

What is Your Smart Idea?

Talk to our product engineer and we will find a solution to meet your application requirements.

Contact now

Vision Application Cases

Fruit & Vegetable Grading

Potato grading, sorting , machine vision, Optical grading, industry 4.0, industrial inspection, quality control, AI

Technical Headlines

  • High Throughput
  • Object Tracking
  • Customised Lighting Solution

  • Neural Networks (Segmentation and Classification)
  • Real-Time Perspective Correction Functions
  • Sensor Integration to Monitor Products in Angled Positions

At Qtechnology, we develop customised computer vision solutions for food grading of production volumes up to 25 tons per hour, which requires analysing more than 250,000 products from around 7,500,000 images.

Throughout the integration of various vision systems for industrial inspection, we have obtained vast knowledge about various challenges related to mechanical integration, cleaning and handling of material in machines, and the restrictions this often poses to the visions systems in regards to speed, lighting, and image capturing in an industrial environment.

We have developed a range of customised machine vision systems for grading, weighing and packaging of fruit and vegetables, primary for Newtec A/S – a market-leading manufacturer. These solutions minimise scrap, ensure high product quality and prevent unsatisfactory products from reaching the end customer. With high-performance accuracy, our vision technology thus assists customers in reducing production and labor costs.

Moreover, we are developing novel x-ray technologies that efficiently inspect whole potatoes and other vegetables, detecting contaminants such as needles, nails, rocks, stones, glass, and other foreign objects as well as hollow hearts, which guarantee safety and quality.

Food Grading Solutions for Seafood Processing

Based on our knowledge within computer vision technology, we have developed a waterproof grading solution for onboard conveyor systems on commercial fishing vessels or onshore fish processing lines. This technology automates the seafood grading process, which is commonly done manually. Moreover, it eliminates human errors and inconsistencies, as it does not rely on human operators.

The seafood grading technology enables the grading of fresh seafood with more than 1mm accuracy, sorting of fish by size and type, while also detecting unrecognised seafood (unknown, defect and bent fish, etc.) and garbage (e.g. effects not detected as fish). The data from our cameras are sent to a programmable logic controller (PLC) system, that sorts fish and discharges foreign objects using robotics technology.

Optical grading, seafood grading, Machine vision, AI, neural network, deep learning

Technical Headlines

  • Integrated into Watertight Stainless Steel Box with Door
  • PLC Interface
  • Fish Species Detection (Sole, Flounder, Turbot, etc.)
  • Size Determination

Quality Control in Food Production

Quality control in food production, Machine vision, AI, neural network, deep learning

Technical Headlines

  • Detection of Various Defects in Different Products
  • Product Tracking on Production Lines
  • Communication with PLC’s, I/O Controllers, etc. With Different Protocols EtherCat, Profibus, ModBus TCP, Profinet
  • Sensor Interfacing Belt Encoders, GPS
  • Light Controllers

Our smart cameras have been applied for various quality control applications. An example of this is the production lines used for rice-based fast-food products in Asia. For this project, we applied computer vision and laser technology to develop solutions for detecting foreign objects and discoloration, sealing edges, vacuum bending and stress in plastic sealing.

In a similar project, we have also developed a computer vision solution for the detection of foreign objects in meat production.

Using smart cameras, data is collected, processed, and transferred to an ejection system further down the production line. Defects are identified on a conveyor and information of position is forwarded to the eject system that removes defective products. The system is thus decreasing the scrap rate and, most importantly, preventing unsatisfactory products from reaching end consumers.

Fruit & Vegetable Topping

In a computer vision system for a strawberry capping machine, we have applied shape analysis and digital images to locate the cutting line, enabling the machine to cap 300 strawberries per minute. Using machine vision technology, this automated capping solution also enables pre-sorting odd-shaped from well-shaped berries, determination of defects on the berries (rod, mold, unripe, etc.) and detection of foreign objects based on adjustable parameters.

For the cutting process, cameras installed above a roller bed are detecting the stem and cutting line of the strawberries. The cameras forward this information to the capping machine for high precision capping. At this stage, a rejection system also discards defective berries and recirculates berries that cannot be cut correctly.

In the same manner, the vision technology for strawberry capping may be customised for similar capping systems (e.g. other berries, fruits or vegetables), to ensure high product quality and productivity, by streamlining the grading and capping process.

Strawberry topping, Machine vision, AI, neural network

Technical Headlines

  • High Precision Determination of Cutting Lines
  • Pre-Sorting Based on Shape Analysis
  • Determination Product Defects
  • Detection of Foreign Objects

Photometry for Beer Analysis

Beer analysis, Machine vision, AI, neural network, deep learning

Technical Headlines

  • Observing the Wavelength of Electromagnetic Radiation
  • Instrument Design
  • Qt User Interface
  • Detection of Floating Particles

  • Classification of Particles
  • Research Cooperation with University of Copenhagen (KU) and Carlsberg Research Laboratory.

It has been observed that the protein and carbon-hydrates in beer sometimes combine into visible particles due to extended temperature changes and mechanical stress of the beer. This is an issue that is formed in the beer after the bottles are sealed and it thus requires inspection of particles inside individual bottles.

The beers are classified based on the number of detected particles and the distribution of big and small particles – these particles are called floaters. The particles are harmless but the bigger floaters pose a serious consumer concern, whereas the small floaters are invisible to the untrained eye.

Previously, only trained personnel could classify the beers in different categories but would be a subjective classification and dependent on many factors like light, bottle color, bottle condition (scratches, etc).

To solve this issue, we developed a camera solution that can classify beer based on size distribution and quantity of floaters in the beer. To see the floaters using vision technology, the bottle needs to be rotated and illuminated with a uniform backlight to observe the particles float fully unfolds.

The camera solution allows us to detect developments of impurities and haziness by calculating the particle size distribution and amount, which improves manual product control previously applied in the beer production process.