This book mainly carries out the uncertainty research of each stage of object-based image analysis (OBIA) of remote sensing, including the object-based classification and object-based change detection, and explores the efficient classification technology and optimization models for segmentation objects. This book systematically studied the uncertainty of object-based image analysis of remote sensing. The results of OBIA are discussed in detail, including the effects of different processing processes on OBIA results, parameters and methods. A method of extracting unsupervised farmland information from object-based high-resolution remote sensing image is proposed. Using the context information provided by the segmentation object, this book also studies the extraction methods of typical cultivated land information to overcome the influence of uncertain factors on classification. A series of object-based classification optimization models based on machine learning are introduced in this book. On the basis of the uncertainty analysis and cognition of OBIA, combined with the advantages of machine learning, different optimization models were proposed for different stages of OBIA, including sampling, feature selection, classification methods, and so on.

Table of Contents

Preface

Chapter 1 Introduction

Chapter 2 Multiscale Segmentation Uncertainty and Segmentation

Optimization

Chapter 3 Analysis of Object-Based Features and Scale Effects

Chapter 4 Uncertainty Research of Feature Selection Method

Chapter 5 Study on the Uncertainty of Object-Based Supervision

Classification Method

Chapter 6 Study on the Uncertainty of Object-Based Change Detection

Chapter 7 Exploration of the Object-Based Unsupervised Classification

Method

Chapter 8 Accuracy Assessment Methods of Object-Based Image Analysis

College

…

Reference