Study On Forest Fuel Survey Streaming Submodualr Set Cover Based On Visual Image

For the large area of autophytic forest and wild forest,it is necessary to effectively detect the representative combustible areas and carry out targeted fire monitoring.At present,the effective means of forest fuel survey is to obtain forest image data set by aerial vehicle shooting.However,the acquired data set is huge,and it is necessary to select representative areas of fuel for monitoring,which will also bring complicated problems in data processing.The problem of selecting a small part of representative data from a data set can be called the data summary problem,which is usually solved by the traditional submodular set cover method.The computational complexity caused by large data sets can be solved by introducing streaming algorithm.Based on this,this paper combines the streaming algorithm with the submodular set cover algorithm to make the selected feature set stable when the elements of the data set change..The main work of this paper is as follows:(1)An approximate submodular set cover algorithm of the streaming algorithm is proposed.In this paper,the problem of selecting a small number of representative elements in a large data set is summarized as a data summary problem,and the submodular set cover method is adopted to solve the problem.However,due to the sequential element selection scheme of the submodular set cover algorithm,it cannot be extended to large data sets well.In this paper,the streaming algorithm is introduced into the submodular set cover of data set,and the pointer jumping algorithm is used to simplify the solving of the submodular set cover problem,so as to reduce the communication complexity.Experimental results show that the submodular set cover algorithm based on streaming approximation improves the cover performance of dominant set and vertex cover problems by 18.2% and 32.4%,respectively,compared with random selection.(2)A streaming submodular set cover algorithm with robustness judgment is proposed.In view of the situation that some elements in the set change or some elements in the set are removed,the selected set is expected to maintain its stability and representativeness.In this paper,based on the cover of streaming submodular set,the robustness judgment is added to it to enhance its anti-interference performance.The boundary and communication complexity of the proposed algorithm are compared with those of the previous algorithms.Experiments show that the robust streaming submodular set cover algorithm can not only avoid accessing a large amount of memory efficiently,but also improve the set stability by more than 10% compared with the normal streaming submodular algorithm after deleting some elements.(3)Application of robust streaming submodular set cover algorithm for forest fuel census.The method in this paper can also be used in the sensor layout of forest combustible fire warning.In this paper,the forest iamge data set shot by the UAV was used for experiments.The selected areas with vegetation coverage rate of more than 90% can be used as key areas to arrange sensors and fire alarm monitoring.