Approche biométrique pour l'identification automatique des produits bois

Abstract : The concept of traceability has been introduced to our everyday life due to recurrent problems in the agricultural and pharmaceutical industries. However, other industrial areas subject to minimum safety restrictions are also involved in the technologies related to this concept. Thus, they are trying to meet the growing demand of clients in terms of quality, delays and personalization of products in an ever-competitive market through traceability. The environment of this work is related to wood industries for which the ecocertification and the improvement of the production process performances are two motivating factors for the start up of traceability. In fact the characteristics of products and industry processes make difficult the synchronization of product flow with information flow and, consequently, the access to information for all actors of the transformation chain. The problem can be summarized in two points : (1) The lack of means of automatic identification, for an individual follow up of products. Commonly used systems like bar codes and electronic labeling show limitations in certain parts of the life cycle. (2) The lack of an information system able to offer in-line information necessary for decision making and able to ensure afterwards the usefulness associated to traceability. Then this work deals with these two complementary problems. First, the start up of a non-destructive control sensor is proposed for setting a biometrical identification of products. The sensor, based on microwave technology, allows the acquisition of representative signals of the inner structure of wood. These signals are treated and then delivered to an algorithm of pattern recognition. A proposition for an information system model is shown for answering the second problem. Such model focused on the product takes into consideration all transformations that may be performed during the fabrication process. Moreover, the modelling makes evident the different automatic identification systems that are possible to use through all the life cycle of the product.
Document type :
Theses
Complete list of metadatas

https://hal.univ-lorraine.fr/tel-01748083
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 11:23:59 AM
Last modification on : Thursday, April 12, 2018 - 1:57:47 AM
Long-term archiving on: Thursday, September 13, 2018 - 5:28:20 PM

File

SCD_T_2005_0133_FUENTEALBA.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-01748083, version 1

Collections

Citation

Cecilia Fuentealba. Approche biométrique pour l'identification automatique des produits bois. Autre. Université Henri Poincaré - Nancy 1, 2005. Français. ⟨NNT : 2005NAN10133⟩. ⟨tel-01748083⟩

Share

Metrics

Record views

131

Files downloads

82