The generation of a time correlated 2D random process for ocean wave motion

摘要

Increasingly, image analysis is concerned with image sequences as opposed to individual image frames. The need to generate model data for such image analysis problems is therefore apparent. One class of data which is well suited to the techniques discussed is that of the motion of ocean waves and time sequences of these images. This problem occurs in sensor modelling and analysis, for example, in sonar studies, the problem of studying the backscatter characteristics from waves is essential for understanding the sonar process and its uses in the analysis of sonar images. Similarly, the analysis of infrared sensor data for glint from ocean scenes presents unique problems. Both types of data involve wave motion and its effect on the analysis of such sequences presents particularly unique problems. This paper looks at the modelling of time sequences of ocean waves with application to the analysis of images, synthesised from the sensors mentioned previously.