Comparison of Several Features of Building Detection in Remote Sensing Image

AdaBoost algorithm based on Haar features and Local Binary Patterns (LBP)features are widely used in machine learning method, and deep learning algorithm is a hot research field in recent years. Remote sensing image recognition meets the needs of human beings in the social development of Information Technology Intelligent. Combined machine learning method with the three features, a certain number of buildings in the remote sensing image were trained and performance test, and then the test results were evaluated, respectively. The results show that the features performance of the deep learning algorithm is better than the other two kinds of features, and the accuracy can meet the requirements of the application of remote sensing image interpretation.