Semi-automatic recognition and vectorization of regions in remote sensig images

Author(s)

Jefersson Alex dos Santos

• Document type: Master's Dissertation
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Instituto de Computação
• Defense date: 2009-02-27

Examining board members

Ricardo da Silva Torres; Hemerson Pistori; Hélio Pedrini

• Advisor: Ricardo da Silva Torres

Abstract

The use of remote sensing images as a source of information in agrobusiness applications is very common. In these applications, it is fundamental to know how the space occupation is. However, the identification and recognition of crop regions in remote sensing images are not trivial tasks yet. Although there are automatic methods proposed to that, users prefer sometimes to identify regions manually. That happens because these methods are usually developed to solve specific problems, or, when they have a general purpose, they do not yield satisfying results. This work presents a semi-automatic method to vectorize regions from remote sensing images using relevance feedback based on genetic programming (GP). Relevance feedback is a technique used in content-based image retrieval (CBIR). Its objective is to agregate user preferences to the search process. The proposed solution consists in using image descriptors to encode texture and spectral features from the images, applying relevance feedback based on GP to combine these features with information obtained from the users interactions and, finally, segment the image. Finally, segmented image (raster) is converted into a vector representation. The main contributions of this work are: comparative study of image vectorization techniques; extension of a recently proposed relevance feedback approach for dealing with image regions; extension of the relevance feedback model for region recognition in remote sensing images; parcial implementation of the semi-automatic and vectorization system of remote sensing images regions; proposal a validation methodology. (AU)

• FAPESP's process: 07/53607-9 - Semi-automatic recognition and vectorization of regions in remote sensing images
• Grantee: Jefersson A dos Santos
• Support Opportunities: Scholarships in Brazil - Master