Computer Analysis of Visual Textures

1. Front Matter

   Pages i-xi

   PDF

2. Introduction

   • Fumiaki Tomita, Saburo Tsuji

   Pages 1-11

3. Statistical Texture Analysis

   • Fumiaki Tomita, Saburo Tsuji

   Pages 13-36

4. Image Segmentation

   • Fumiaki Tomita, Saburo Tsuji

   Pages 37-55

5. Shape Analysis

   • Fumiaki Tomita, Saburo Tsuji

   Pages 57-70

6. Structural Texture Analysis

   • Fumiaki Tomita, Saburo Tsuji

   Pages 71-82

7. Grouping

   • Fumiaki Tomita, Saburo Tsuji

   Pages 83-97

8. System and Evaluation

   • Fumiaki Tomita, Saburo Tsuji

   Pages 99-114

9. Object Recognition

   • Fumiaki Tomita, Saburo Tsuji

   Pages 115-135

10. Shape from Texture

    • Fumiaki Tomita, Saburo Tsuji

    Pages 137-162

11. Back Matter

    Pages 163-173

    PDF

This book presents theories and techniques for perception of textures by computer. Texture is a homogeneous visual pattern that we perceive in surfaces of objects such as textiles, tree barks or stones. Texture analysis is one of the first important steps in computer vision since texture provides important cues to recognize real-world objects. A major part of the book is devoted to two-dimensional analysis of texture patterns by extracting statistical and structural features. It also deals with the shape-from-texture problem which addresses recovery of the three-dimensional surface shapes based on the geometry of projection of the surface texture to the image plane. Perception is still largely mysterious. Realizing a computer vision system that can work in the real world requires more research and ex­ periment. Capability of textural perception is a key component. We hope this book will contribute to the advancement of computer vision toward robust, useful systems. vVe would like to express our appreciation to Professor Takeo Kanade at Carnegie Mellon University for his encouragement and help in writing this book; to the members of Computer Vision Section at Electrotechni­ cal Laboratory for providing an excellent research environment; and to Carl W. Harris at Kluwer Academic Publishers for his help in preparing the manuscript.

• Image segmentation
• Textur
• cognition
• computer
• computer vision
• object recognition
• perception

• Electrotechnical Laboratory, Tsukuba, Japan

  Fumiaki Tomita

• Osaka University, Toyonaka, Japan

  Saburo Tsuji

Policies and ethics