CAJ | 학술논문

针对机场跑道深度图像，研究并提出了基于深度特征和Adaboost的机场跑道异物识别算法。文中采用分水岭分割方法将可疑的异物区域分割出来。然后在提取深度特征的基础上，采用Adaboost分类器实现异物识别。实验结果表明该算法能够准确地完成机场跑道的异物识别。
침대궤장포도심도도상，연구병제출료기우심도특정화Adaboost적궤장포도이물식별산법。문중채용분수령분할방법장가의적이물구역분할출래。연후재제취심도특정적기출상，채용Adaboost분류기실현이물식별。실험결과표명해산법능구준학지완성궤장포도적이물식별。
A recognition algorithm based on the depth feature and Adaboost is presented to distinguish FOD (foreign object damage)from airport runway depth images. Firstly, the target region of the image including FOD is picked up by the watershed segmentation method. Secondly, different kinds of depth features are extracted from the target region, the depth features, Histogram mean and Histogram variance. At last, Adaboost classifier is used to distinguish FOD automatically. Experimental results indicate that the proposed method can obviously improve the accuracy of FOD's recognition.