一种快速有效特殊天气泊位飞机机型识别方法

Quick valid method for docking aircraft type recognition in bad weather

自动泊位引导系统的关键是飞机定位是否准确,而特殊天气下飞机特征选择和提取是定位准确与否的关键因素。提出了基于自适应加权形态学的高阶神经网络识别算法。首先利用自适应加权形态学提取边缘从中选择特征同时进行预处理,然后利用高阶反馈神经网络进行机型和子型识别。实验证明:该方法具有特征简单、识别率高的特点,能准确识别出飞机型号,得到比较精确的飞机外形描述,可有效对抗噪声、遮挡、阴影等干扰,具有很好的稳健性。该方法满足自动化泊位系统对目标识别模块的稳定可靠、快速准确的要求。

Aircraft type identification is a key factor of aircraft safe docking in airport auto-docking guide systems in bad weather. Based on adaptive weighted morphology, a high neural network method was developed to solve the key issues in docking guide. Adaptive weighted morphology was used to extract the features and preprocess with high neural network then used to classify aircraft types. The experimental results show the effectiveness of the improved identification system. The system provides a good performance in aircraft identification and offers better robustness against noise and poor image quality, which can satisfy the auto docking system requirements with high precision, rapid speed and stabilization.