ChickenCheck - Camera-Based Quality Control Systems

The ChickenCheck systems developed by CLK GmbH are camera-based quality control systems for the poultry industry. By utilizing state-of-the-art image processing technologies, they enable objective, standardized, and consistent assessment of poultry at the slaughter line.

Our systems are already in use in numerous poultry processing plants across Europe and make an important contribution to quality assurance as well as the monitoring of relevant animal welfare indicators. The systems within the ChickenCheck product family detect and classify injuries, abnormalities, and diseases. In addition, they enable the precise determination of total carcass weight as well as the weight of individual parts.

In addition to our existing systems, our experts are continuously developing new solutions to provide customized quality control systems tailored to individual customer requirements.

Scientific Validation: We are also a partner in the European research project aWish. As part of the project, a study validating our ChickenCheck system was published. The results demonstrate a very high level of agreement between assessments performed by human experts and the automated detection carried out by ChickenCheck.

The full article can be found here: Validation of an automated system for at-slaughter assessment of footpad dermatitis and hock burn in broiler chickens - ScienceDirect

Quality Control System: Chicken

Task

Analysis of the knees, detection of black skin changes and classification into individually adjustable quality levels

Assessment

Individual placement in up to 3 classes
Evaluation is based on the average value of the left and right knee; other valuation methods can also be set

Rate

approx. 4 pairs/second
Speed approx. 0.5m/s

Documentation

Storage of the database records with the evaluation data as well as the captured images
Each picture has detailed rating
Evaluation contains information about: time of pick-up, determined quality level, size of the defect, associated herd number and the line on which product is located
Daily report is generated about each herd of the current

Task

Analyzes and determines the size of black discoloration on the balls of the feet
System can be used for chickens, turkeys and ducks

Assessment

Individual placement in up to 5 classes
Evaluation is based on the average value of the left and right foot
System detects twisted, double & tilted feet and takes them out of the rating

Rate

approx. 5 pairs/second (18.000/hour)

Documentation

Storage of the database records with the evaluation data as well as the captured images
Each picture has detailed rating
Evaluation contains information about: time of pick-up, determined quality level, size of the defect, associated herd number and the line on which the product is located
Daily report is generated about each herd

Task

Contactless and precise weight measurement of breast fillet, leg, wing, as well as the total carcass weight at the start of the slaughter line
Evaluated independently of breed and gender
Optimization of production control

Assessment

Two 3D cameras are used to capture the front and back of the poultry
The images are combined into a 3D model
The model is then analyzed in detail using robust statistical methods as well as artificial intelligence

Rate

Analysis of up to 18,000 chickens per hour

Documentation

Storage of database records containing analysis data and captured images
Statistics for the current batch: average weight, number of measured and unmeasured animals
Automatic generation of daily reports

Task

Examination for the possible presence of crops, so that verifiable evaluations including image credits can be created for the fattener

Assessment

By means of a score value, the broiler chickens are classified into two classes: OK and GOITER
Option of individual classification available

Implementation

Detection: Use of color and 3D cameras
Algorithm/neural network analyzes thickenings in the neck area

Rate

Band speed 15000 bhp / 4.2 fps

Task

Automated analysis of damage to chest, thighs, and wings
The system detects hematomas and broken wings
Also: Differentiation between damage caused by catching or slaughtering

Assessment

Individual classification and evaluation of detected hematomas
Integration into 180° curve
The centrifugal forces occuring in the curve cause the wings to open
Open wings are automatically captured by cameras

Task

Analysis of ascites and deep dermatitis and classification into different classes

Assessment

Color space transformation of the image makes yellowish discoloration visible. Yellowish areas then become visible in the transformed image as large dark spots.
Based on the 3D image information, further geometric features and protrusions are determined
Classification into 3 classes: healthy, sick & undecidable

Options

Basic system: 3D color camera records the front of the poultry. Detection of abdominal ascites & dermatitis
Additional module: Additional colour camera records the back of the poultry and thus enables the detection of deep dermatitis in the back area

Rate

approx. 300 chickens/minute

Documentation

System is ideal for pre-sorting and veterinary support

Quality Control Systems: Ducks and turkeys

Task

Analyzes the breast skin for bumps and detects skin changes using a 3D volume measurement

Assessment

Placement in up to 5 classes

Documentation

Storage of the database records with the evaluation data as well as the captured images
Each picture has detailed rating
Evaluation contains information about: time of pick-up, determined quality level, size of the defect, associated herd number and the line on which the product is located
Daily report is generated about each herd

Task

Analyzes and determines the size of black discoloration on the balls of the feet
System can be used for chickens, turkeys and ducks

Assessment

Individual placement in up to 5 classes
Evaluation is based on the average value of the left and right paddle
System detects twisted, double & tilted paddle and takes them out of the rating

Rate

approx. 5 pairs/second (18.000/hour)

Documentation

Storage of the database records with the evaluation data as well as the captured images
Each picture has detailed rating
Evaluation contains information about: time of pick-up, determined quality level, size of the defect, associated herd number and the line on which the product is located
Daily report is generated about each herd

Features of our systems

Highly efficient evaluation with database connectivity
Customizable to individual needs
Capture unit with lighting, protected for harsh environments
Detailed automatic reporting
Autonomous inspection & warning system for system conditions
100% inspection
Optional: Automatic generation of daily reports & integration of reports on tablet/smartphone

Improvement of final product quality
Adherence to quality standards
Enhancement of animal welfare and animal health
Increase in food safety
boosted customer confidence
Cost savings through improved control of incoming goods & process management

Joint control

Footpad control

Rendite system

Crop detection

Catch damage

ChickenVet

Breast skin assessment turkeys

Paddle control ducks

Characteristics

Advantages