Contributions to OCR for unreadable characters in printed circuit boards by means of pattern matching and machine learning techniques (Registro nro. 236150)

MARC details
000 -LIDER
fixed length control field 02124nam a22002417a 4500
003 - CONTROL NUMBER IDENTIFIER
control field MX-MeUAM
005 - DATE AND TIME OF LATEST TRANSACTION
control field 20191002092058.0
008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION
fixed length control field 170131s2019 mx ||||fo||d| 00| 0 spa d
040 ## - CATALOGING SOURCE
Transcribing agency MX-MeUAM
Language of cataloging spa
050 00 - LIBRARY OF CONGRESS CALL NUMBER
Classification number TK7868 .P7
Item number N39 2019
100 1# - MAIN ENTRY--PERSONAL NAME
Personal name Nava Dueñas, Carlos Fábian
9 (RLIN) 18663
245 10 - TITLE STATEMENT
Title Contributions to OCR for unreadable characters in printed circuit boards by means of pattern matching and machine learning techniques
Medium [recurso electrónico] /
Statement of responsibility, etc. Carlos Fábian Nava Dueñas; director, Félix Fernando González Navarro
260 ## - PUBLICATION, DISTRIBUTION, ETC. (IMPRINT)
Place of publication, distribution, etc. Mexicali, Baja California,
Date of publication, distribution, etc. 2019
300 ## - PHYSICAL DESCRIPTION
Extent 1 recurso en línea,149 p. ;
Other physical details il. col.
500 ## - GENERAL NOTE
General note Maestría y Doctorado en Ciencias e Ingeniería
502 ## - DISSERTATION NOTE
Dissertation note Tesis (Doctorado)--Universidad Autónoma de Baja California, Instituto de Ingeniería, Mexicali, 2019.
520 ## - SUMMARY, ETC.
Summary, etc. In the last few decades, new computer vision technologies and image<br/>processing techniques have been very important in the improvement<br/>and automation of manual processes in many technical areas, e.g.,<br/><br/>in the semiconductor industry. In this thesis, we propose and com-<br/>pare several techniques in the areas of pattern matching and machine<br/><br/>learning to have optical character recognition (OCR) of damaged or<br/>unreadable numerical digit characters from images on printed circuit<br/><br/>boards (PCBs). We describe how the best machine learning algo-<br/>rithms are applied to extract the principal characteristics and fea-<br/>tures to compute, classify and find the correct numerical character<br/><br/>that corresponds to those features. We also present our work in the<br/>improvement of the image quality in the pre-processing stage to make<br/>pattern matching a good option over some specific conditions of PCBs<br/>damage.
650 #7 - SUBJECT ADDED ENTRY--TOPICAL TERM
Término temático o nombre geográfico como elemento de entrada Circuitos impresos
Fuente del encabezamiento o término lemb
Subdivisión de forma Tesis y disertaciones académicas
Subdivisión general Diseño y construcción
700 1# - ADDED ENTRY--PERSONAL NAME
Personal name González Navarro, Félix Fernando
9 (RLIN) 14607
Relator term dir.
710 2# - ADDED ENTRY--CORPORATE NAME
Corporate name or jurisdiction name as entry element Universidad Autónoma de Baja California.
Subordinate unit Instituto de Ingeniería
9 (RLIN) 3321
856 4# - ELECTRONIC LOCATION AND ACCESS
Uniform Resource Identifier https://drive.google.com/open?id=1xpi3ZhalG7X9WnNw23c0XaYezca0gryZ
Public note Tesis Digital
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Koha item type Tesis
Existencias
Estado de retiro Fuente de clasificación Colección Ubicación permanente Ubicación actual Fecha de ingreso Signatura topográfica Código de barras Date last seen Número de copia Tipo de material Categoría 1 de ítem Categoría 2 de ítem Categoría 3 de ítem
    Colección de Tesis Biblioteca Central Mexicali Biblioteca Central Mexicali 27/09/2019 TK7868 .P7 N39 2019 MXL122270 27/09/2019 1 Tesis Disco compacto Material adquirido por Donación Doctorado en Ciencias

Con tecnología Koha