Automated underwater 3D sonar image recognition has great potential to simplify many underwater tasks. Sonar modeling is necessary to recognize 3D object images. We propose a 3D sonar model that can predict what an object is based on by recognizing similarities to objects that pre-exist in a database. The sonar's displaying mechanism and characteristics of the 3D sonar image are studied. Due to the nature of acoustics, a sonar image is not necessarily always an accurate depiction of an object. The proposed model enables mapping of a 3D world onto a 2D sonar screen. This modeling framework enables the implementation of various optical vision techniques for recognition. Recognition experiments were conducted to evaluate the model's accuracy.

INTRODUCTION

Underwater, the sonar image and the optical camera are both feasible sensing methods for object recognition. They are essential to carrying out tasks such as object finding, environmental monitoring, and underwater structure installation and maintenance. The optical camera provides the highest underwater image resolution, but it has restricted applications due to its limited visibility (Sisman, 1982), especially since many of the tasks take place in shallow water with visibility distances of less than 1 m. The sonar image is also a feasible method for recognition. Recently, the quality of the sonar images has improved dramatically. The latest 3D sonar image such as DIDSON (Dual frequency IDentification SONar: Belcher, 2002; DIDSON website; Kim and Neretti, 2005; Negahdaripour et al., 2005) provides highresolution 3D images. Automatic sonar image recognition has great potential in many underwater fields such as mine detection, and maintenance and safety inspections. However, both automated sonar image recognition's displaying mechanism and the very nature of acoustics present challenges. We studied the sonar's characteristics and display mechanism and proposed this model to recognize objects from any viewpoint.

This content is only available via PDF.
You can access this article if you purchase or spend a download.