Spatial Data Mining using Novel Neural Networks for Soil Image Classification and Processing

The emergent field of spatial data mining uses spatial dependency that is prevalent on spatial data sets, which can be modeled and incorporated into data mining process. Spatial relations are modeled during a data pre- processing step, consisting of the density analysis and vertical view approach, after which an exploration with visual data mining follows. In this paper we implemented, spatial image processing mining for soil classification using diversified domains like Digital Image Processing, Neural Networks, and Soil fundamentals. The three most important algorithms used in implementation are Back Propagation Network (BPN), Adaptive Resonance Theory 1 (ART) and Simplified Fuzzy ARTMAP for soil classification as well as spatial image recognition. Further we are working on our research by combining the visual data mining with spatial data mining algorithms, such as spatial clustering, spatial association rules, a self-organizing map etc. in order to try to detect patterns in the data in an even more effective way.