Machine Vision

Machine vision deals with mathematical methods to extract information from images. Eliko develops custom machine vision systems for different applications from parking to industrial robotics.

We have broad expertise in software development and integrating different data capturing systems, including camera control and lighting. In image processing, Eliko’s engineers master feature extraction methods like speeded up robust features (SURF), histogram of gradients (HOG), and mixture of Gaussians (MOG).

Our machine vision algorithms are developed using our own proprietary software library. We design applications that run on PCs or single board computers.

Eliko has extensive experience in applying computer vision technology in manufacturing, medicine, and smart city applications. Our systems guide and identify objects in production, help guide vehicles and detect abnormalities in medical applications.

Why work with us

  • 5+ years of experience

  • Turnkey solutions

  • Reliable and low cost systems

  • Dedicated machine vision engineers

Selected projects

Part Identification System for Welding Robots

Eliko has developed a custom part identification system for industrial welding robots. The system forms an automated parts warehouse that allows the welding robot to work independently with minimal human assistance. The custom lighting and image capture system is designed specifically for the industrial environment.      

More info

The system is based on computer vision and machine learning techniques. It can learn to identify the objects fed into the welding robot and choose and initiate the appropriate welding program on the robot.

As a result, the welding robot can work more independently. This increases the robot’s working hours and reduces personnel costs. The automatic parts identification system is built as a custom solution for Norcar BSB for producing industrial machinery.

Norcar BSB Homepage

Parking Space Detection System

Eliko has developed a parking space detection system that makes it easier and quicker to find a free parking space in large parking lots or business districts. The demo system can use either visual or thermal camera images as an input.

More info

The system uses sophisticated image processing methods and Eliko’s custom machine vision algorithms to detect the presence of a car on a particular parking space. The information can then be displayed via a designated web application or on-site displays.

The automatic parking space detection system can be used in both open parking lots and garages. It is part of Eliko’s smart city applications.

ASAN Security Homepage

Read more

What is your biggest challenge?

Indrek Ruiso