Geosciences and Remote Sensing

In our present topic, ant colony optimization (ACO) and particle swarm optimization (PSO) based optimization techniques have been applied to perform satellite image classification with fewer amounts of discontinuity, conflicts and constraint of imprecise knowledge and evaluation of data. Land cover analysis by virtue of image classification are always associated with certain amount of vagueness, uncertainty and ambiguity during the classification from the remotely sensed data. In the present scenario, we are able to presents a hybrid ACO/PSO technique that are extracted through expert knowledge for a more focused EOS satellite image classification. This abstract investigates the principle of traditional rule mining, which will produce more non-supplementary candidate sets when it reads data into candidate items. Especially, when it deals with massive data, if the minimum support and minimum confidence are relatively small, combinatorial explosion of common item sets will occur and computational power and storage space required are likely to exceed the limits of machine. ACO/PSO optimization algorithm based on conventional ant-miner and swarm optimization algorithm is proposed and is used in rules mining for supervised clustering of digital number (DN) values

in satellite images. Measurement formula of effectiveness of the rules is improved and pheromone concentration update strategy is also carried out. The experiment results show that execution time of proposed algorithm is lower than traditional algorithm and has better execution time and accuracy for EOS image.