Generalised interpolation of segmented images using shape-independent orthogonal transforms

We develop a new method for object-oriented interpolation of images. It is an important tool in image processing, since using interpolation we can considerably decrease amount of data, necessary for image reconstruction. Application of this interpolation enables one to downsample object separately. The selected object can be processed at the different sampling level. This approach allows for example object oriented zoom. Moreover, the object-oriented approach is very novel idea that helps to understand the image content. The method is created from a cosine approximation implemented for coding with shape independent basis functions.