【摘要】 为了验证一维距离像用于光学目标分类识别的可行性,解决复杂环境和远距离条件下光学目标的探测和识别难题,首次尝试利用激光一维距离像识别具有不同光学参数的猫眼目标,并提出基于双谱的激光一维距离像的稳定特征提取方法。该方法通过信号积累和滤波处理来提升复杂背景下的作用距离和回波信噪比增益,利用双谱计算同时实现远距离目标探测和猫眼目标的稳定特征提取;并通过奇异值分解和主分量计算降低双谱特征提取的运算量;通过处理不同距离、类型及入射角的真实数据样本,验证双谱特征的稳定性和有效性。该方法可以识别具有猫眼效应的光学目标和角反射器,与现有方法比较,具有远距离、低信噪比探测和识别的优点。更多还原

【Abstract】 To verify the feasibility of optical target recognition from laser one-dimensional range profile,laser one-dimensional range profile was used to detect and recognize"cat-eye"targets with different parameters,and a feature extracting method of laser one-dimensional range profile based on bi-spectrum is proposed. Operating range and the SNR gain of laser returns under complicated backgrounds can be improved by signal accumulation and filtering process. And,the remote target detection and robust feature extraction of"cat-eye"targets were realized by using bi-spectrum calculation. By using singular value decomposition and principal component analysis,the computation of the bi-spectrum can be reduced. Lastly,the robustness and effectiveness of the eigenvector were verified by analyzing the real data samples of different distances,types and incidence angles. Compared with other methods,the method can recognize "cat-eye"target and corner reflector,and have the advantages of far operation range for low SNR target.更多还原