The follow video shows the main features and the power of our vision system ArtificialInspection.
Below there are brief explanations for introducing the features of the ArtificialInspection vision system.

Controller MC500.

Compact x86-based flexible system, fast and reliable. It covers most of the applications of small and medium size.

The vision system interface with the cameras of the major brands.

USB, USB3 and GigE connection.

Our system supports both color and black and white cameras.

3D laser scanning system for reconstruction by triangulation of laser profiles.

Used for 3D robot guidance, 3D object measurement and dimensional comparison.

Multiview system for stereo-visual reconstruction.

Used for 3D robot guidance.

The continuous development in the field of artificial vision allows us to constantly improve the software by implementing more sophisticated algorithms that can solve the problems that our customers require.

Main features of the system:

  • Ability to enable only the categories of functions needed for the application to reduce the cost of the final system.
  • Connecting to a variety of cameras (monochrome and color). Use with any type of interface such as camera-link, FireWire, USB2.0, USB3Vision, GIGEvision.
  • Custom access levels and passwords for access to system functionality.
  • Flexibility in programming based on VbScript macros, using a simple graphical interface to use for the end user.
  • Image Sampling and Backup / Restore Functions of the Program. Allows remote assistance via e-mail, allowing you to scan images and loaded programs remotely.
  • Remote assistance through Internet. Possibility to connect the controller to the network and to have remote support.

All common bar codes can be read with any orientation, even with a width of just 1.5 pixels. The ECC 200, QR, QR Micro, Aztec and PDF417 codes can also be recognized. A robust recognition allows reading data on distorted images, different surfaces, and under variable lighting conditions.

You can read texts on different surface types: this function is useful for recognizing objects or verifying the marking of certain characters on the final product.

Identification of objects without encoding is possible through an algorithm based on the learning of identification samples. It is able to distinguish a large number of objects: you can recognize objects based on features such as color or texture, thus eliminating the need to use special encodings such as bar codes or data codes for identifying such objects. It also works with deformed objects or different perspective views of the object.

You can acquire a flat or round surface such as a jar, then perform a check on the serigraphy to determine its quality.

Detectable defects:

  • Absence or excess of printings
  • Surface defects
  • Incorrect printing position
  • Tone defects
  • Dirt spots

Fast and accurate controls can be performed, allowing you to inspect 100% of the most varied details during the production phase.

We use different techniques to inspect the surfaces to recognize any defects such as dirt, scratches, breaks or deformations.

Location of objects that deforms irregularly, such as a pastry bag.

Sophisticated algorithms of pattern matching detect the position of the known objects allowing their handling with articulated robots.It is also possible to localise deformed objects according to the perspective.

Using a CAD file of an object, it is possible to define a 3D position of the objects in a scene.

Using the 3D reconstruction, it is possible to detect the position of the known objects using the CAD model or 3D scan in order to guide the robot to the grab from a pile or a chest.

The measurement of objects is one of the most used controls. It verifies the tolerance of the component in the production line. It is possible to reach 1 micron meter precision.

Using the 3D reconstruction, it is possible to define the measure of the object along the three dimensions.

The robot guidance is one of the areas where we have begun to do experienced since 1999. Our vision system is characterized by the simplicity of use and speed in creating new programs with a few steps of self-learning.
New 3D acquisition technology allows to make measurements and controls that until a few years ago it was not possible to perform.