An Airborne Image Stabilization Method Based on Projection and the Gaussian Mixture Model

In this paper, an image stabilization method based on the adaptive Gaussian mixture model (GMM) is presented in order to smooth down the airborne image vibration. Firstly, the image edge is enhanced and then projection is adopted to estimate motion; secondly, GMM parameter is obtained after analyzing characteristics of the first n images; finally, a stable image sequence is achieved after GMM motion filter operates on the motion parameter. The experimental results show that the method has the advantage of fast speed and effectively smooth unwanted vibration of image sequences.

1. Introduction

When a camera is mounted on an unsteady or moving platform, serious vibration will appear in screen images. In this case, the image vibration can make our observers feel tired, affect their differentiation, and probably lead to big errors in their later image processing work. Therefore, how to stabilize images becomes quite important. So far, there are three means for image stabilization: mechanical stabilization, optical stabilization and digital image stabilization [1]–[3]. The former two means require complex equipment and expensive cost. With advances of electronic technology, digital image stabilization develops rapidly. Due to its high precision and low price, it is being widely used in military and civil areas. Its fundamental principle is to use motion estimation algorithm to get displacement vector of the current frame image and then to compensate the vector until a steady image is to be achieved. The general approach is shown in figure 1: Figure 1. Digital Image Stabilization system