Training image-based simulation with complex training images using search tree partitioning

Using acomplex training image, such as an analog, for the SNESIM algorithm results in poor simulation since the training image contains trends and may not meet training image requirements. By pooling all the training image patterns in a single search tree but not recording the patterns’ relative locations, some critical features of these complex training images are lost. The search tree partitioning approach subdivides the large training image into imbricated, homogeneous, smaller images, called partition classes. Each of these partition classes have a corresponding search tree that can be utilized by the SNESIM algorithm. These partition classes are obtained by processing the training images with spatial filters that are pattern sensitive. The resulting filter scores are then clustered into partition classes. All patterns within a partition class are recorded by a search tree; there is one tree per partition class. At each pixel along the simulation path, the partition class is retrieved first and used to select the appropriate search tree. That search tree contains the patterns relevant to that partition