Título del libro: Icmt 2012 - 16th International Conference On Mechatronics Technology
Título del capítulo: Contour object generation method for object recognition using FPGA
JUAN MARIO PEÑA CABRERA; VICTOR MANUEL LOMAS BARRIE; ROMAN VICTORIANO OSORIO COMPARAN;
Autores externos:
Idioma:
Inglés
Año de publicación:
2012
Palabras clave:
Automated manufacturing process; Contour information; Industrial environments; Manufacturing tasks; Neural network model; Process of learning; Robot vision systems; Simulation in matlabs; Algorithms; Field programmable gate arrays (FPGA); Flexible manufacturing systems; Manipulators; Manufacture; MATLAB; Object recognition; Pattern matching; Robots; Three dimensional computer graphics; Vision; Computer vision
Resumen:
The article presents a method for obtaining the contour of an object in real time from not binarized images and for objects that can be assembled on line in automated manufacturing processes. The contour information is integrated into a descriptive vector called [BOFnew], which is used by a neural network model of the type FuzzyARTMAP to test the feasibility of the method using the generated contour to learn of the object and then recognize it later. In this way, it is possible to obtain a process of learning the location and the recognition of parts for manufacturing purposes, and for its subsequent recognition and manipulation by a manipulator robot. To this end, it requires a fast and robust method to acquire, process and communicate to a robot the information about positioning and orientation of an object for assembly purposes. In this way, the robot as a central element can request tasks carried out within a manufacturing cell instrumented with a vision system that acquires images of the pieces and assembly objects to guide its movements. The article shows the explanation of the used algorithm and its simulation in MatLab 7.0 with specific instances of images of forms of objects. Robot vision systems can differentiate parts by pattern matching irrespective of part orientation and location. Some manufacturers offer 3D guidance systems using robust vision and laser systems so that a 3D programmed point can be repeated even if the part is moved varying its location, rotation and orientation within the working space. Despite these developments, current industrial robots are still unable to recognize objects in a robust manner; that is, to distinguish an object among equally shaped objects taking into account not only the object's contour but also its form and depth information. Having a robust and fast method for contour object generation for object recognition manufacturing tasks improves this methodology and allows the implementation of these algorithms in devices like FPGA's, which gives the object recognition applications in manufacturing cell a real possibility of fast and robust performance demanded by industrial environments.
Entidades citadas de la UNAM:
ISBN:
9787900769428
Editorial:
Tipo de producto:
Conference Paper
