The Target Classification of Optical Remote Sensing Image Based on Hierarchical Features and AdaBoost Algorithm

Xiaofei Ji, Ningli Qin, Yangyang Wang
2017 International Journal of Signal Processing, Image Processing and Pattern Recognition  
In order to accurately classify and recognize specific targets in remote sensing images, a novel recognition method based on hierarchical features and AdaBoost algorithm is proposed. Hierarchical features can represent the distribution characteristics of targets effectively by applying the image pyramid idea to the BoF-SIFT feature extraction. The AdaBoost algorithm with support vector machine(SVM) as the weak classifiers is used for targets classification. The proposed classification and
more » ... ition method is tested on our remote sensing image dataset. The results verified the method can achieve higher recognition accuracy.
doi:10.14257/ijsip.2017.10.1.03 fatcat:uqntimy5n5gqdj7giipad6hkky