【摘要】 研究机场的障碍物有效三维识别问题。机场上的障碍物规则很难形成有效的数学约束,造成三维识别效果不好。为了避免上述缺陷,提出了一种基于PCA和LDA融合算法的机场障碍物识别方法。针对三维地理定位采集的图像,进行二值化处理,从而提高机场障碍物图像的对比度,为机场障碍物识别提供依据。利用PCA和LDA融合算法,提取机场障碍物图像的低维度特征,根据特征对机场障碍物进行有效的识别,从而使飞机在行驶的过程中对障碍物进行规避,保证了飞机的安全。实验结果表明,利用上述算法进行机场障碍物识别,可以有效提高识别的准确性,从而保证飞机在跑道上行驶的过程中能够对障碍物进行准确的规避,取得了令人满意的结果。更多还原

【Abstract】 The effective 3- D recognition problem of airport barrier was studied in this paper. It is difficult for the rule of airport barrier to form an effective mathematical constraint,resulting in bad 3- D recognition effect. In order to avoid the above defects,this paper proposed an airport barrier recognition method based on the fusion algorithm of PCA and LDA. For the collected 3- D geographical location images,the binarization processing was implemented,so as to improve the contrast of the airport barrier image and provide the basis for the airport barrier recognition.PCA and LDA fusion algorithm was employed to extract the low dimension characteristics of airport barrier image to identify the airport barrier effectively according to the characteristics,so that the aircraft can avoid barrier in the process of driving and ensure the safety of the aircraft. The experimental results show that the algorithm presented in this paper can effectively improve the accuracy of recognition,thus ensuring that the aircraft on the run way can accurately avoid the barrier and achieve satisfactory results.更多还原