Processing and advanced image analysis applications for image characterization of atomic force microscopy

Author(s)

Carlos Alberto Rodrigues

• Document type: Doctoral Thesis
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Instituto de Física de São Carlos (IFSC/BT)
• Defense date: 2003-04-22

Examining board members

Luciano da Fontoura Costa; Andre Luiz Battaiola; Mônica Alonso Cotta; Marcelo de Assumpcao Pereira da Silva; Jorge Stolfi

• Advisor: Luciano da Fontoura Costa; Roberto Mendonça Faria

Abstract

This thesis address the application of advanced techniques of processing and analysisof images in original problems involving images of atomic force microscopy. For this, a series of algorithms for characterization and understanding of process of formation of new polymeric materials was developed and implemented, with perspectives of many technological applications. The analysis was applied to the determination of orientation of the morphology of substrates for alignment of liquid crystals, counting and estimative of radiuses of granules in layer-by-layer polymer films, analysis of enhancement of photoluminescence in PPV cast films, as well as the study of curvature spontaneous of macromolecules. The principal algorithms included are determination of inclination of eigenvectors of matrix of covariance of coordinates of points of shape,application of regional maxima technique and Voronoi diagrams, passband filters 2D through Fourier Transform and curvature multiscale. The implementation of these algorithms involved a series of image analysis algorithms such as squeletonization, exact dilations and extracting of contour of shapes. The principal contribution of this work was to develop a software called SPIA (Scanning Probe Image Analysis) that includes tools for analysis of processing of images including that were used in this work. This software was developed in Delphi under object orientation paradigm to Windows NT/9X/2000/XP. It has a friendly interface similar to other image processing softwares. All this techniques were tested extensively and the results that corroborate the robustness of the algorithms are included throughout the thesis (AU)