Sie sind auf Seite 1von 31

1

M. Antonio Ramrez Ramrez


PH_D. student 1st Semester

Dr. Ricardo Barrn Fernndez


ADVISOR

Schedule:

Background Proposal Justification State of the art Research goals Contributions Strategy Work Plan

The word Pattern, derives from the french term patron that means role model. A pattern, is an entity that can be named and is represented by a set of measurable properties, (Watanabe 1985).

Micro-patterns are small patterns used for scale or multilevel analysis and for analyzing the relationship between individuals and the environment in which they operate.

The use of micro-patterns allows us to separate the role of each component of a complex structure, in order to improve the knowledge of the situation to each of them.

The micro-patterns modeling can be used to describe elements in the spatial context of an image, such as borders, edges, lines, focus, corners and all that is typically obtained with conventional patterns.

Micro-patterns are similar to conventional patterns, but the micro-patterns are also unique since built-software point of view, due that they are mechanically recognizable, since each such micro-pattern can be expressed as a formal condition on the structure of a class.

On fingerprints recognition is apply the micro-pattern identification stablishing over a binary image, the directions of lines located giving it orientation labels.

90

45

Another way to work with the micro-patterns contained in a fingerprint image, is considering the position and quantity of it.

The micro-pattern in one image, could be a element that represent an unit of recognition by itself.

A different way of take the micro-patterns is like Structural Syntactic Approach, It is based on the descriptions of objects in terms of their own parts.

10

Authors
Xie Zhihua Liu Guodong

Paper name Weighted Local Binary Pattern Infrared Face Recognition

Method or application area Face recognition based on Webers law

Date 2011

Sanqiang Zhao, Yongsheng Gao and Baochang Zhang

Performance Evaluation of Micro Pattern Micro-Pattern Representation on Gabor Features for Representation on four Face Recognition forms of Gabor features Fingerprint Recognition Using Directional Micro-Pattern Histograms and LVQ Networks Micro-Pattern Micro-PatternEvolution Evolution Modeling Micro-patterns for Feature Extraction From micro-patterns to nano-structures by controllable colloidal aggregation at air-water interface. Fingerprint enhancement and binarization Micro-Pattern Micro-Pattern Extraction Extraction on Java Code on Java Code Feature Extraction from Micro-structure Modeling An approach to construct nano-structures by Colloidal Aggregation.

2010

Sheng-De Wang and Chih-Jen Lee Sunghun Kai & SunghunKim, Kim, KaiPan Pan & E. James E. JamesWhitehead, Whitehead,Jr. Jr. Qiong Yang, Dian Gong and Xiaoou Tang. Xiaorong Xiong, Kenvin Wang and Karl F. Bohringev

2008

2006 2006 2005

2004

11

Micro-pattern Evolution is a continued work about the idea of decomposing design patterns into recurring elements has emerged from the context of forward and reverse engineering. Such elements are called fragments, motifs, mini-patterns, elemental design patterns, sub-patterns or micro-patterns. One of the main benefits of decomposing design patterns into subcomponents is the reduction of errors over forward engineering .

12

Presents a catalog of 27 micro-patterns defined on Java classes and interfaces.

The catalog captures a wide spectrum of common programming practices.


Use of inheritance, immutability, data management, emulation of procedural-, modular-, and even functional- programming paradigms with object oriented constructs. Together, the patterns present a set of prototypes after which a large portion of all Java classes and interfaces are modeled. A statistical analysis of occurrences of micro-patterns in a large software corpus, spanning some 70,000 Java classes The micro-pattern occurrence raise the hope of improve a software design and code quality.

13

Classification of textures

Local binary pattern

Face detection Expression analysis Face recognition

Extraction of code
Micro-patterns Representation

Sobel operator
Real part Imaginary part Magnitude Phase

Gabor features magnitudes

Local derivative pattern

14

This project will focus on the strategies to define a new method for micro-pattern recognition in the context of digital images. This method will be useful to solve some problems currently still on research in the AI area, as unsupervised recognition over satellite imagery.

15

Although pattern recognition (PR) has been used since the early sixties, it is still one of the main tools in artificial intelligence.

The handling, registering and use of the information obtained from micro-patterns is becoming a necessary aid in several areas, like in the nanotechnology field. A micro-pattern is a small information unit that by itself or together with other similar ones, can be used, where appropriate, to analyze and describe object characteristics.

PR has been used in many studies as well as in image classification and all kinds of signals. All those studies have led to face identification systems, as well as fingerprints, voice, and even human tissue recognition systems. Science still faces some limitations in the recognition area. Extended techniques and methods from micro-pattern recognition, could mean a significant advance in image analysis.

17

PR is a widely used tool in some AI applications such as face recognition.

The analysis, study and application of micro-patterns will mean major methodological changes and an increase in computational efficiency.

18

The use micro-patterns as more complex elements or those that require more thorough analysis, could represent a heavier work.

Another example is the localization of smaller features or the identification of specific elements in the aerial images.

20

We would like to have a method for recognize a 3D form viewed from any angle or distance

21

General.
Design, develop and test a method to recognize micropatterns on digital imagery.

Specifics:
1.

Study, program and test different algorithms known to use micro-patterns.

2.

Design a new algorithm, strategies to help improve and streamline of some AI applications, using micro-patterns such as searching, sorting or computer vision.

22

Formalizing the term micro-pattern, from a computational point of view. Develop a method that allows us to determine the shape and size suitable for the use of micro-patterns involved, depending on the application or study area to investigate. Demonstrate the usefulness of the method developed by applying the proposed algorithms to local binary pattern technique.

23

24

Digital topology. Interest points.

Wavelets over finite fields.

25

The modeling of extraction of features with micro-patterns can improve the performance of the identification and recognition tasks.

26

1st Semester 2nd Semester 3rd Semester 4th Semester 5th Semester 6th Semester Optional course I Seminary I State of the art (rev) Divulgation paper Optional course II Seminary II Code tests Paper report Optional course III Seminary III Experimentation Paper report Optional course IV Pre-Ph D. Exam Main paper report Thesis work Optimization tests Paper report Ph D. degree exam.

Total course
27

28

29

A12
Curso Opcional 1 Seminario de Titulacin Revisin del "Estado del Arte" Planeacin de la estrategia Estructurar el plan de trabajo Reporte en Artculo Presentacin ante el Comit

Enero

Febrero

Marzo

Abril

Mayo

Junio

Primer semestre

30

El operador Sobel es utilizado en procesamiento de imgenes, especialmente en algoritmos de deteccin de bordes. Tcnicamente es un operador diferencial discreto que calcula una aproximacin al gradiente de la funcin de intensidad de una imagen. Para cada punto de la imagen a procesar, el resultado del operador Sobel es tanto el vector gradiente correspondiente como la norma de ste vector. Operador de Moravec, este operador calcula el valor de cambio de la intensidad de la imagen cuando desplazamos ligeramente esta ventana, seleccionando puntos con textura en diferentes direcciones, lo que hace que puedan ser localizados con precisin en la imagen. Una esquina viene caracterizada por elevados cambios de intensidad. Es altamente sensible a ruido.

31

Das könnte Ihnen auch gefallen