EVK ALPHA Real-Time Image Processing
The EVK ALPHA is a powerful data processing platform to rapidly and securely process spectral image data in real time. The analysis and classification of materials takes place separately from the sensor or camera and is therefore particularly flexible in its use.
Thanks to the modular concept, advanced algorithms analyze data from well established sensors. Another advantage of the modular concept is that existing machines and systems can be upgraded cost-effectively. This provides companies with an economical and efficient way to update existing machines and benefit from the advantages of state-of-the-art technology.
The EVK ALPHA Real Time Image Processing platform captures data from hyperspectral cameras and processes it in real time, allowing for a respons in milliseconds. Sorting and inspection operations can be performed at high precision and speed.
Combined with EVK HELIOS hyperspectral cameras, it is used effectively in numerous industries. For seamless integration into existing systems, the edge computing platform is compatible with the industry standards GigE Vision and GenICam.
Fast Processing of Spectral Data
in the Production Line
In industry, hyperspectral camera systems are used, for example, to detect foreign bodies in products and provide sorting information accordingly. However, to initiate a reaction in real time, information must be acquired and processed very quickly. Using the real-time image processing platform EVK ALPHA, spectral data can be processed quickly and securely, directly in the production line. Sorting commands are transmitted to e.g. valves or sorting flaps in real time.
EVK HELIOS EC & EVK ALPHA - A Powerful Duo
The EVK ALPHA real-time spectral data processing platform is fully compatible with various EVK products. Together they forma powerful system for your industrial application.
EVK HELIOS EC Hyperspectral Camera
This hyperspectral core camera operates in the short-wave infrared (SWIR) range and provides non-contact measurement of material properties beyond simple color information. It reliably and continuously monitors and gives reliable sorting commands - the perfect solution in a wide range of industries. The camera captures data and forwards it to the real-time data processing platform for classification.
Features and supported algorithms
Separates of different materials into material grades. Up to 100 material spectra are supported by the algorithm.
Converts spectral information into color information. Typically used to determine the subtle chemical differences in inspected products.
Combines features and delivers sorting decisions based on a relation between physical and chemical features of a good or bad product.
Quantitatively analyzes up to three chemical constituents, measuring the concentration levels of chemical compounds in products.
Regardless of the algorithm chosen, the spectra of a hyperspectral data cube can be classified in just a few milliseconds.
What is Edge Computing?
Compared to cloud computing, where data is sent over the internet to remote servers, edge computing involves data processing, analysis and application directly at the edge, which means locally and close to the data source.
The primary reason for using edge computing technologies is to reduce latency and rapidly respond to real-time data. Processing data in close proximity to where it originates or is consumed can eliminate delays that sometimes occur when data is transmitted to remote data centers. This is particularly vital for applications that require a response in real time - such as industrial sorting or monitoring tasks.
Real-Time Image Processing:
What is Real-Time?
When we talk about real-time in edge computing data processing, it`s important to establish what real time actually means. The meaning of real time in industrial applications is defined in the standard DIN ISO/IEC 2382. This describes an IT system that must react within a certain, prescribed time frame. Literally, the standard states:
"Real time refers to operating a computing system in such a way that programs for processing accruing data are constantly ready for operation and the results are available within a predetermined period of time. Depending on the application, the data may accrue randomly or at predetermined points in time."
EVK systems are capable of processing data so fast that they are able to transmit the relevant information in real time. All material fractions of a recorded line are determined in just a few milliseconds. The time required to classify a line remains constant and does not change.
Why is it Important to Classify Materials Rapidly?
The faster that material fractions can be classified, the faster a facility`s conveyor belts can run and the greater the possible material throughput. The speed and size of the machine directly impacts the operator`s return on investment (ROI). By using EVK`s real-time data processing platform, our customers not only increase the efficiency of their operations, they also improve their profitability to secure decisive competitive advantages.
The real-time data processing platform EVK ALPHA combined with the EVK HELIOS hyperspectral core camera and the EVK SQALAR software are ideal for:
- Detecting foreign bodies in the food industry
- Determining the material type in plastics sorting
- Measuring material qualities
- Analyzing material concentrations quantitatively in production processes in the line