back

IMAGE RECOGNITION AND FEATURE EXTRACTION

Image Processing

Approximation of point sets with geometric shapes

The objective of this investigation is to build a learning shape recognition engine which we call ShapeWiz (from shape wizard). The system approximates a set of two-dimensional input points, either as an ordered set of input strokes (known as online-case) or as an unorderd set of simple points (known as offline-case) with one or several geometric shapes, where new shapes are learned with the help of the feedback of the user. If the appoximation is sufficiently close to some known shape according to some similarity measure, the input is said to be recognized. Depending on the domain certain restrictions on the recognition process might be applied and possibly additional domain specific knowledge can be used. Applications of such a system are, among others, general vectorization task, automatic image description based on edge/border interpretation, content based image recognition, user friendly graphical input devices, and educational software.

We have already a quite versatil version of ShapeWiz with the following properties:

  • It handles both the online- and the offline-case for a single shape.
  • It uses almost parameter free deterministic algorithms.
  • A demonstration version with an HTML5 user interface supporting both mouse and touch input is available at http://pcformella.ei.uvigo.es/sw/shapewizdemo.html

 

Automatic reading of displays

Calibrating and certifying measuring instruments is a time comsuming, manually executed task. The operator has to read several imes during an often lengthly procedure the display of the instrument to be certified and compare it to the reference value from the reference instrument. The instruments to be certified exhibit a large variety of forms, shapes, and display types. Reference instruments have a digital output that can be connected to a computer for reading, instruments to be certified mostly not.

     
    

We have developed a software system that helps to automize the calibration procedure by various means:

  • A digital camera is used to capture the displays of one or various instruments during the measuring procedure.
  • The display areas can be marked interactively by the operator.
  • The snapshots are taken according to the protocol specifications for this kind of instrument at the same time as the reference data is read.
  • Both the images (or only the marked areas) and the reference data are stored into a database.
  • The operators workload is reduced to set-up the system and to read the snapshots according to the established procedure.

 

To further improve the system, we have implemented

  • an automatic detection of the display areas and their rectanglur alignment (correctable interactively by the operator) and
  • an automatic reading of the digits of the display (correctable interactively by the operator).

 

The fully automaic reading has a precision of more the 98% tested on more than 1400 displays.

 

  • During the joint work with LOMG we developed the system SreC, which is capable of reading automatically digital displays.

 

Shape recognition of bones

 


Financed Projects


Publications


People Involved


Student Work


Press Articles



Partners

More about Laboratorio Oficial de Metrología de Galicia (LOMG)
Laboratorio Oficial de Metrología de Galicia (LOMG)
Ir a web
More about Trabeculae S.L
Trabeculae S.L
Ir a web
More about Grupo de Procesado de Imagen y Realidad Virtual
Grupo de Procesado de Imagen y Realidad Virtual
Ir a web
More about Laboratorio de Informática Aplicada
Laboratorio de Informática Aplicada
Ir a web
More about Centro de Investigación, Transferencia e Innovación (CITI)
Centro de Investigación, Transferencia e Innovación (CITI)
Ir a web