Remote Sensing Analysis Method Research Based On Ant Colony Algorithm

Crops remote sensing image processing is the most important core and the key point of the field of remote sensing technology, and how to deal with the remote sensing image better, faster and more accurate has great significance in crops planning, production and control. Conventional image processing methods have become increasingly unsuited to the development of remote sensing data source, so it is necessary to introduce some new ways to adapt to technological development, the ant colony algorithm is a good choice.Ant colony algorithm is a new kind of bionics group intelligence algorithm, which has the characteristics of highly parallel, self-organizing, self-adaption, collaborative and so on. This paper firstly introduces the development status of crops remote sensing, and secondly introduces the principle and the model of ant colony algorithm, take10meter resolution’s ALOS phantoms as the examples, then lead the ant colony algorithm into the field of remote sensing image processing and combine the ant colony algorithm with fuzzy classification on crops remote sensing, and finally propose a fuzzy classification method based on ant colony algorithm, as well as a multi-scale segmentation and classification mothed based on ant colony algorithm, and in this foundation, has focused on the research of ant colony algorithm based on multiscale classification method. The important role which is played by segmentation and classification mothed based on ant colony algorithm in crops remote sensing is proved by a large number of experiments.The innovative significance of this paper is:firstly, propose a fuzzy classification method based on ant colony algorithm for crops remote sensing, and this algorithm has a higher accuracy and convergence compared to traditional algorithm; secondly, propose a multi-scale classification method based on ant colony algorithm, and provide new ideas and ways for the research of different scale surface features. The corresponding experimental results show that the ant colony algorithm has a good feasibility and effectiveness in crops remote sensing image processing and parsing problems, and the ant colony algorithm has a good application prospects in this field.