CAJ | 학술논문

针对机场目标检测时目标特征提取的单一性问题，提出一种基于整体特征和局部特征互补的多光谱图像的机场目标检测方法。提取多光谱图像的整体轮廓特征，依据轮廓面积的大小得到机场潜在区域；提取潜在区域图像的纹理特征和颜色特征；将这些整体特征和局部特征输入SVM分类器，判断该图像是否包含机场。实验结果表明，该方法能够有效检测出机场目标。
침대궤장목표검측시목표특정제취적단일성문제，제출일충기우정체특정화국부특정호보적다광보도상적궤장목표검측방법。제취다광보도상적정체륜곽특정，의거륜곽면적적대소득도궤장잠재구역；제취잠재구역도상적문리특정화안색특정；장저사정체특정화국부특정수입SVM분류기，판단해도상시부포함궤장。실험결과표명，해방법능구유효검측출궤장목표。
For the incompletion of feature extraction in airport target detection ,a method combining complementary features of the global and local features was presented .The global contour features of the multispectral images were extracted and the air‐port potential areas were got according to areas of the contours .The local texture and color features of potential areas were ex‐tracted .All global and local features mentioned above were input to the SVM classifier so as to determine whether the image con‐tained the airport .Experimental results show the method can effectively detect the airport .