Model to assist the identification of numerical code using Paraconsistent Annotated Logic: Case study in containers

Wellington Tuler Moraes, Mauricio Conceição Mario, Maurício Fontoura Blos, João Inácio da Silva Filho

Abstract


In this paper we describe the construction of a software model that will be used to identify numerical codes extracted from digital images of containers; the images were obtained at the port of Santos (SP). To take the decision of the identification of the numerical codes, Paraconsistent Annotated Logic (PAL), was used as an alternative to the use of traditional of Artificial Intelligence techniques. The way of data treatment used in this article was the use of four different networks analysis Paraconsistent Annotated Logic with annotation of two values (PAL2v), which are able to provide satisfactory and sufficient results to identify the figures based on patterns of images of numerical codes extracted from containers. The software presented in this dissertation makes it possible to choose one of four different networks analysis Paraconsistent Annotated Logic with annotation of two values. Networks can be trained with number of numerical code patterns ranging from three to five, and with the images of extracted from containers numbers that are references, generate results based on the comparison of the tests with the values obtained in the network analysis Paraconsistent Annotated Logic with annotation of two values. The presented results achieved up to 100% of assertiveness in the identification of numerical codes using the best network configurations for each digit, and will serve as a basis for adding this technique to customs processes of container terminals.

 

Keywords: Container. Numeric Code. Paraconsistent Logic. Paraconsistent Annotated Logic

Full Text: PDF

Refbacks

  • There are currently no refbacks.