Algorithm for the Evaluation of Imperfections in Auto Bodywork Using Profiles from a Retroreflective Image

Autor: Valerie Zwilling, Miguel A. Salichs, Ramon Barber
Rok vydání: 2014
Předmět:
Zdroj: Sensors (Basel, Switzerland)
Sensors
Volume 14
Issue 2
Pages 2476-2488
Sensors, Vol 14, Iss 2, Pp 2476-2488 (2014)
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
instname
ISSN: 1424-8220
Popis: Nowadays the automobile industry is becoming more and more demanding as far as quality is concerned. Within the wide variety of processes in which this quality must be ensured, those regarding the squeezing of the auto bodywork are especially important due to the fact that the quality of the resulting product is tested manually by experts, leading to inaccuracies of all types. In this paper, an algorithm is proposed for the automated evaluation of the imperfections in the sheets of the bodywork after the squeezing process. The algorithm processes the profile signals from a retroreflective image and characterizes an imperfection. It is based on a convergence criterion that follows the line of the maximum gradient of the imperfection and gives its geometrical characteristics as a result: maximum gradient, length, width, and area. The authors gratefully acknowledge the funds from PSA PEUGEOT-CITRÖEN automobile group. Special thanks also to Jose Manuel Pequeño, for his support along this project, and to Yannick Bertret and Vincent Nalewajk, for the assistance during the work.
Databáze: OpenAIRE