Information loss degree as an assessment for the availability of remote sensing images

In satellite remote sensing images, cloud coverage causes information loss of the ground cover surfaces, thus affecting the availability of the images. To deal with this problem, a method to measure this information loss is proposed in this paper. It concerns two factors from both the clouds and the ground cover surfaces. The first is the Cloud Fragmentation degree, which describes how much the cloud clusters are scattered in the images and damage caused on the information integrity of the images. The second factor is the damage resistance degree of the ground cover surface, which determines the properties of the ground cover surfaces (i.e. effective areas and information amount) to resist the damage upon the information integrity of them. A formula is then given to combine these two factors in evaluating their interaction on the Information Loss Degree. With a comparison of the test results with the artificial classification of image availability, the validity of this model is demonstrated.