Antarctic Sea ice distribution detection based on improved ant colony algorithm

The changes in the Antarctic sea ice area are directly related to the changes in the atmosphere and oceans. Determining the Antarctic sea ice distribution is of great significance to the global climate change analysis. The ant colony algorithm adopts a positive feedback mechanism to continuously converge the search process and ultimately approaches the optimal solution, making it easy to find the optimal segmentation threshold for detecting the sea ice distribution. However, the ant colony algorithm has the problems of high computational complexity and easy getting stuck in local optima. In order to better apply the ant colony algorithm to sea ice distribution detection, an improved ant colony algorithm was proposed, which improves the selection of initial clustering centers and the update of pheromone volatilization factors in the ant colony algorithm. We compared the improved ant colony algorithm with iterative algorithm, maximum entropy algorithm, and basic global threshold algorithm, and the results showed that the proposed algorithm is feasible. To further validate the accuracy of the improved ant colony algorithm, we compared the results obtained from MODIS data with the improved ant colony algorithm, iterative algorithm, maximum entropy algorithm, and basic global threshold algorithm for sea ice detection, and the results showed that the accuracy of the proposed algorithm was 4.99%, 3.66%, and 5.46% higher than the other three algorithms, respectively.