Author: Scanway
Publication date:
Poultry meat production is currently taking place under increasing pressure: increasing processing scale, high quality requirements and limited availability of skilled workers. At the same time, industry data show that 3-4% of all meat production is lost already at the processing stage due to technological errors, human errors and process problems. In high-volume conditions, this means real losses of raw material and the risk of reduced safety of the final product.

The scale of production and exports further reinforces the importance of the problem. According to preliminary data from the Ministry of Finance, in January-August 2025, the value of Poland’s meat product exports increased by 19% year-on-year and reached €8.4 billion, equivalent to about 22% of total agri-food export revenues. On such a scale, even single quality errors can have not only financial, but also image and regulatory consequences.
Vision system for inspection of poultry fillets
The vision system from the CUMULUS group is used for automatic inspection of raw material in the form of chicken fillets (breasts), transported on a two-track conveyor belt. The system works with a capacity of up to 130 pieces per minute, which allows integration into high throughput lines without limiting the production rate.
The main task of the system is:
- Detection of foreign bodies on the surface of meat,
- Identification of quality defects in the raw material,
- Unambiguous classification of the product into OK / NOK categories in real time.

Functional scope of the system
The system is capable of detecting a wide range of contaminants and quality defects that are important for food safety. As part of the foreign body inspection, the system identifies the presence of plastics such as PET, PE and PP, as well as metal components made of stainless steel and aluminum. Fragments of protective gloves, rubber and textile elements, as well as films – both colored and transparent – are also detected. In addition, the system allows detection of foreign bodies of natural and technological origin, such as wood, glass, paper, insects, and fragments of bone, feathers and cartilage. In parallel, quality defects in poultry fillets are inspected. The system detects blood stains and meat discoloration, fraying, and detects bone and skin residues that can affect the quality and safety of the final product.
Chicken fillet inspection is based on two complementary imaging technologies – RGB and hyperspectral imaging (HSI). Each technology covers a different aspect of product inspection: RGB imaging enables the detection of visible quality defects, while HSI is used for the surface detection of selected non-metallic foreign materials.
Detection range of vision system for quality control of poultry fillets
| TYPE OF DETECTION | DETECTED CONTAMINANTS/FOREIGN BODIES | TECHNOLOGY USED | MINIMUM DETECTABLE DEFECT SIZE |
| Foreign body detection | High density plastics | HSI | 10 x 10mm |
| Low density plastics | HSI | 10 x 10mm | |
| Dice | RGB/HSI | 10 x 10mm | |
| Quality defects | Blood stains | RGB | 5 x 5mm |
| Revision | RGB | control of the whole element | |
| Leather | RGB | whole element control | |
| Discoloration | RGB | inspection of the entire component | |
| Fraying (spaghetti) | RGB | control of the entire element | |
| Feathers | RGB | 5 x 5mm | |
| No coating | RGB/HSI | 5 x 5mm |
The system operates continuously and each piece of meat is inspected. The entire vision system, including cameras and illuminators, is housed in enclosures that meet the requirements of the food industry, made of stainless steel and rated IP69K.

Quality Defect Detection – RGB Imaging
The RGB module uses conventional colour imaging of the product surface. Its primary purpose is to detect visible quality defects in chicken fillets.
The inspection scope includes blood spots and bruises, shredded and damaged fillets, colour deviations and discolouration, spaghetti structure, as well as the presence of bones and cartilage. The expected detectable defect size is approximately 2–5 mm, depending on the type of defect and production conditions.
Foreign body detection – hyperspectral imaging
Plastic contamination detection is implemented using a hyperspectral camera. This technology captures the spectral signatures of materials, making it possible to clearly distinguish plastic from raw material – regardless of color, transparency or degree of dirtiness.
Analysis of spectral data and classification of detected objects is carried out using artificial intelligence algorithms (Machine Learning / Deep Learning), which allows adaptive adjustment of the system to the variability of raw material and production conditions.
The HSI module is designed for the surface detection of selected non-metallic foreign materials that cannot be reliably detected using RGB imaging or X-ray detectors. The inspection scope includes soft and hard plastics, film, rubber, wood, bones and cartilage. The expected detectable contaminant size is approximately 2–5 mm, depending on the type of material and production conditions.
Combining RGB and HSI imaging enables the system to cover two different inspection areas – visible product quality defects and selected non-metallic foreign materials. The target system is designed to integrate both modules into a single automated solution for chicken fillet inspection directly on the production line.


Request a quote
Send a request to our specialists. We will gladly advise and choose the best solution for your business.

Flawless
production
- 40 ms shortest time for processing one image
- 7 szt./s number of elements controlled per second
- 5 μm smallest defect we identify
- 16 m/s fastest moving object we control

