EngVisTech Kft - AI-based image processing

AI-based image processing

In the field of industrial image processing, AI-based solutions enable the automation of manufacturing processes, improve quality inspection, and reduce human error. The following technologies are among the most commonly used methods.

More information
EngVisTech Kft - Technologiesfelső alsó design elementje
EngVisTech Kft - Technologiesblokk alsó design elementje

Technologies

Segmentation

Segmentation breaks an image into distinct, meaningful regions. This makes it possible to precisely separate individual objects or even defective areas within a single object. In industrial environments, it is especially useful for locating surface defects such as scratches and cracks, as well as inspecting complex-shaped components. The technology enables highly accurate, detailed analysis even under varying environmental conditions. It often serves as a foundation for other inspection methods.
More information
Segmentation - EngVisTech Kft

Object Detection

Object detection not only identifies objects but also determines their exact location within the image. This capability is critical in robotic systems, where knowing the position of components is essential for picking or assembly tasks. It can also be used to detect missing or incorrectly placed parts. Typical applications include in-line inspection and automated material handling. The technology is also suitable for real-time operation.
More information
Object Detection - EngVisTech Kft

Anomaly Detection

Anomaly detection aims to identify patterns that deviate from the normal state, i.e., to detect defects. The system learns the characteristics of defect-free products and treats any deviation as a potential defect. In industrial applications, it is particularly advantageous where many different types of defects may occur, or where they cannot be predefined. It is for example effective in inspecting painted, machined, or cast surfaces. It improves the reliability of quality inspection and helps reduce scrap rates.
More information
Anomaly Detection - EngVisTech Kft

Classification

During classification, the system assigns an image or object to one of several predefined categories. In industrial environments, this can be used for product quality grading (acceptable / defective) or for identifying different types of components. The method provides a fast and efficient solution for automated inspection of high-volume production. It is often used in combination with other technologies, such as following object detection. It plays an important role in optimizing manufacturing processes.
More information
Classification - EngVisTech Kft

OCR – character recognition

OCR (Optical Character Recognition) enables the automatic extraction of text information from images. In industrial applications, it is typically used to verify serial numbers, production dates, part numbers, and other markings. Modern AI-based solutions can also handle distorted, low-quality, or hard-to-read text. This is particularly important for laser markings or inkjet prints. OCR plays a key role in ensuring traceability and detecting incorrect labeling.
More information
OCR – character recognition - EngVisTech Kft

Request a free consultation!

Let’s discuss your application and find the best technical solution. After a short consultation, we will provide clear directions and concrete recommendations for implementation.