PSA Cathedra

Analyse d'aspect automatique

Main researcher: M.A. Salichs

Peugeot_logo

Description

The PSA Peugeot Citröen collaboration goal is to build an automatic
classification and quantification system to detect the defects on the metal
panels as consequence of the stamping process. Currently, this classification
is done the direct observation of the pieces. The final goal is to obtain
a classification that is similar, as much as possible, to the one obtained
by demerit.

The complete process starts in the image adquisition from the metal panels and leads to the defect classification.

image

Our work is organized as follows:

– Step 1: Study of the objective and possibilities analysis
– Step 2: Construction of a software to analysis
pictures
– Step 3 Validation of the software?s performance
o In terms of localization and determination of the geometrical
parameters
o In terms of capability to organize detected defaults in a hierarchy and define a criterium of quality.
– Step 4: Semi industrial validation (test with a large
range of parts)
– Step 5: Industrial unfurl
– Step 6: Production use
The PSA Peugeot Citröen collaboration goal is to build an automatic
classification and quantification system to detect the defects on the metal
panels as consequence of the stamping process. Currently, this classification
is done the direct observation of the pieces. The final goal is to obtain
a classification that is similar, as much as possible, to the one obtained
by demerit.

The complete process starts in the image adquisition from the metal panels and leads to the defect classification.

image

Our work is organized as follows:

– Step 1: Study of the objective and possibilities analysis
– Step 2: Construction of a software to analysis
pictures
– Step 3 Validation of the software?s performance
o In terms of localization and determination of the geometrical
parameters
o In terms of capability to organize detected defaults in a hierarchy and define a criterium of quality.
– Step 4: Semi industrial validation (test with a large
range of parts)
– Step 5: Industrial unfurl
– Step 6: Production use

Entries:
Tracking moving optima using Kalman-based predictions
Evolutionary Computation (ISSN: 1063-6560). num. 1 , vol. 16 , pages: 1 – 38 , 2008
J.C. Diaz M. Abderrahim
Speed-Sensorless Nonlinear Predictive Control of a Squirrel Cage Motor
WSEAS TRANSACTIONS on SYSTEMS and CONTROL, ISSN 1991-8763. num. 2 , vol. 3 , pages: 99 – 104 , 2008
M. Abderrahim Siller-Alcalá, I. Jaimes-Ponce, J. Alcántara-Ramirez, R.
Pose Estimation for Variable Configuration Objects:an Evolutionary Approach to Vision-based Navigation and Inspection
WSEAS Transactions on Information Science & Applications. num. 3 , vol. 3 , pages: 538 – 545 , 2006
J.C. Diaz M. Abderrahim

Entries:
Modified SoftPOSIT algorithm for 3D visual tracking
2007 IEEE International Symposium on Intelligent Signal Processing (WISP'2007), 2007, Alcala Henares, Spain
J.C. Diaz M. Abderrahim
Visual Inspection System for Autonomous Robotic On-Orbit Satellite Servicing
ASTRA2006: 9th ESA Workshop on Advanced Space Technologies for Robotics and Automation, 2006, Noordwijk, The Netherlands
J.C. Diaz M. Abderrahim
Automated Visual Inspection for Robotic On-Orbit Servicing
MX2006: The 10th Mechatronics Forum Biennial International Conference, 2006, Malvern, Pennsy, USA
J.C. Diaz M. Abderrahim
Evolutionary Model-Based Pose Estimation for Variable Configuration Objects
ISPRA 2006: 5th WSEAS International Conference on Signal Processing, Robotics and Automation, 2006, Alcala Henares, Spain
J.C. Diaz M. Abderrahim

Previous Project

next project