Document Type



Spatial association rule mining is a kind of spatial data mining to carry some interesting and implicit knowledge about spatial associations from spatial databases. Moreover, many excellent studies on Remote Sensed Image (RSI) have been conducted for potential relationships of crop yield. Lee and Chen [13] proposed a two-phase algorithm by creating Histogram Generators for fast generating coarse-grained spatial association rules, and further mining the fine-grained spatial association rules w.r.t the coarse-grained frequently patterns obtained in the first phase. However during the image processing, partitioning the image into parts can improve its efficiency; to this point, the concept of image blocking is incorporated onto the coarse-grained two-phase data mining of spatial association rules. Therefore, an adaptive two-phase spatial association rules mining method is proposed in this paper to improve two-phase method in terms of efficiency. The proposed adaptive method conducts the idea of partition on an image for efficiently quantizing out non-frequent patterns and thus facilitate two-phase process. Such adaptive two-phase approache saves much computations and will be shown by lots of experimental results in the paper.