基于声纳图像处理的海底地貌分类研究

摘要

In this paper, the basic theory about MRF is introduced and Gauss-Rayleigh Mixture Model is presented according to the feature of sonar images. Comparing with the segmentation with traditional threshold method, the MRF method based on the Gauss-Rayleigh Mixture Model is suitable for the sonar image segmentation perfectly. Secondly, a scheme is proposed to extract the texture feature of sonar images using Gabor filters with optimal parameters. The Gabor filters are designed with constrained parameters to reduce the complexity and to improve the calculation efficiency. Meanwhile, at each orientation, the optimal Gabor filter parameters will be selected with the help of bandwidth parameters based on the Fisher criterion. This method can overcome some disadvantages of the traditional approaches of extracting texture features, and improve the recognition rate effectively. Finally, taking the scale or resolution factor into consideration, two more efficient classification algorithms are proposed. One is an information fusion method, which combines both Gabor filters and fuzzy fractal dimension, to extract features and to classify seabed. It can improve recognition rate effective. However, it wastes of time and memory. So another efficient classification algorithm, which is called Gabor wavelet, is applied to extract the features of sonar images. Concluding from the experiments, Gabor wavelet not only contains multi-resolution information of images but also indicates the multi-orientation feature of the images, which is useful to classify the sonar image.

目录

封面

声明

目录

1. Introduction

1.1 Foreword

1.2 Background of the acoustic vision research

1.3 Development of sonar image generation and processing

1.4 Purpose and significant contents of this thesis

2. Sonar image generating principles

2.1 Foreword

2.2 Principles of sonar and sonar equation

2.3 The oretical background of sonar imaging

2.4 One approaches to generate image:Beamforming

3. Sonar image preprocessing

3.1 Foreword

3.2 Image enhancement by reducing the speckle noise

4. Image segmentation

4.1 Foreword

4.2 The main approaches to segmentation of sonar image

4.3 Sonar Image Segmentation Based on Markov Gauss-Rayleigh Mixture

5. 2D classification of the seafloor based on sonar images

5.1 Foreword

5.2 Summary of the main approaches to classify the seafloor

5.3 Analysis of the problems in the sonar image generation

5.4 Seafloor classification using Gabor filter

6. Two more efficient classification algorithms of sonar images

6.1 Foreword

6.2 Sonar images classification based on the information fusion methods

6.3 Another efficient classification algorithm of seafloor using Gabor Wavelet

6.4 Conclusion

7. Conclusion and Future work

7.1 Conclusion

7.2 Future work

致谢

Main Work Achievement of the author

参考文献