Artificial vision - ITCL

Computer Vision

 

 

ITCL develops research and development projects in the field of Computer Vision.

Using images captured by conventional or industrial cameras, Advanced Machine Vision Algorithms can simplify, automate or improve a wide range of tasks:

Pattern detection and recognition

  • Defect detection in the inspection of manufactured parts
  • Product quality on the production line
  • OCR
  • Classification
  • Surveillance and Biometrics

3D data capture

  • High resolution facial reconstruction
  • 3D model capture
  • Mapping of buildings and residences

Motion tracking

  • Augmented Reality
  • Indoor Tracking and Environmental Recognition
  • Visual odometry (autonomous robots, self-driven vehicles)

Using Artificial Vision technology ITCL has developed several projects, especially in the field of Industrial Inspection, Augmented Reality, Product Quality, Assisted Systems, etc.

Working lines

3D Reconstruction using Structured Lighting:

tructured light cameras allow the capture of 3D models by projecting an invisible pattern (by infrared light or laser projector) onto the scene. Cameras of this type, such as Microsoft’s Kinect, Asus Xtion or Intel REALSENSE allow for real-time capture and recognition of objects, people and gestures.

This powerful technology can be used for other applications such as real-time mapping or reconstruction of large scenes, such as an office or residence.

One of the applications developed by ITCL is the 3D reconstruction of complete residences and health care centers for assisted living, in the NACODEAL project.

Stereo Vision:

reconocimiento_escenas_vision_artificial_va

A single camera allows a Machine Vision System to detect and track objects, but does not provide any information about their depth, geometry or position.

By combining two or more cameras in a stereo vision system, it is possible to determine the geometry of the scene without using dedicated lighting or projectors. This versatility makes this approach ideal for outdoor and uncontrollable environments, and is often used by industrial robots and vehicles.

Stereo photometric:

estereo_Fotometrico_3D_vision_artificial_secuencia

Several machine vision methods rely on specialized equipment or a set of two or more cameras to capture the 3D surface of an object. Stereo-photometric is a high-resolution approach that allows the 3D surface of objects to be recovered using a single camera, and 3 or more light sources. By analyzing the images obtained with different illuminations under the same point of view, not only the geometry of the object is reconstructed, but also its real color is obtained, without illumination or shading effects.

This is very useful in applications where the level of detail is important (dermatology, fracture analysis or facial recognition). The data can also be used for classification and recognition task.

3D camera tracking:

3D tracking is a necessity in, for example, locating and tracking people in a controlled environment. In several places this is not feasible with the use of conventional equipment, such as Wifi triangulation, RFID beaconing, etc.

By using this technology, ITCL has developed a novel indoor location that allows for the display of Augmented Reality to users. This application was used to help people with slight memory loss in the NACODEAL project.