Desion

Introduction

We offer camera-based inspection systems for complex objects to efficiently automate visual recognition processes. In addition to our many years of experience in the development of software products, we also use the latest research results. Thanks to our cooperation with Fraunhofer IGD, TU Darmstadt and our network of experts, we can find a solution even for complex challenges in development processes. In this way, we can enable or facilitate your project. Our technology offers rapid development of quality control systems based on artificial intelligence.

Our focus here is on High Variant Objects. These are deformable / highly variant objects that differ greatly in their appearance.

People

Soft Biometrics Attendance check People counter

Food

Fruits Vegetables Meat Canteen food

Liquids

Medical liquids Cosmetics Building materials Environmental analysis Food

Textiles

Cleaning cloths High-visibility clothing Recycled clothing Hotel linen Hospital linen

Homepage-HVO-en

Due to their elastic, viscous or fluid nature, information in highly variant objects can be hidden behind or within themselves. It may also be initially invisible and only recognisable in image sequences.

Our framework incorporates specially designed machine learning algorithms that are invariant to these problems. Within this framework, it is possible to automatically process image data, augment it, inject additional information and train it with the appropriate architecture. Currently, we have extensive data sets and models to recognise the following properties on High Variance Objects:

Size

Our algorithms enable the detection of the size of defects in a few milliseconds through depth images. This gives the possibility of using wear limits.

Colour

Our algorithms, with the help of our calibrated hardware, ensure reliable detection of colour and brightness differences, especially suitable for standard testing.

Garment Type

The visual differentiation of garment types is an important process in the sorting of garments. We are one of the first companies to solve this task reliably, camera-based.

Material Types

In addition to different types of material, our algorithms also distinguish between different weaves.

Rust

Detected rust stains can be removed with special detergents.

Stains

Only reliable detection of residual stains allows targeted treatment with special detergents and thus consistently high laundry quality.

Tears

Tears in textiles are often not detected by the naked eye. Our AI-based models allow for consistently high detection.

Holes

Whether they are caused by wear and tear, open seams, material defects or pests, holes often mark the end of a textile's use.

PORTFOLIO

ROLLGOODS-INSPECTION

Defis is a camera-based fabric inspection system. It detects visible defects in patterned and uniform fabrics using computer vision. Defis guarantees high and consistent product quality during production and optimises process efficiency.

MANTRAP

The secure separation of people when accessing high-security areas is becoming increasingly important. Mantrap portals are unmanned separation gates in which only one authorised person can pass through a transit room. One advantage lies primarily in the robustness against social engineering such as tailgating or piggybacking - that is, an authorised person bringing an unauthorised person into the secured area.

WARNCLOTH PROOF SYSTEMS

High-visibility clothing provides safety for many people in their everyday working lives. Our high-visibility clothing testing system is the first system that protects the textile service from liability risks. Our camera-based system tests the conformity of warning clothing to standards.

FREE-FALL QUALITY CONTROL

Our free-fall inspection systems enable the sorting and quality control of various objects. These are picked up from all sides in free fall. This improves speed and detection performance.