光学 精密工程, 2009, 17 (7): 1714, 网络出版: 2009-10-28   

复杂场景下应用成像Ladar的自动目标识别

Implementation of automatic target recognition by imaging Ladar in complex scenes
作者单位
1 电子工程学院 安徽省红外与低温等离子体重点实验室,安徽 合肥 230037
2 武警指挥学院,天津 300350
摘要
针对目标部分被树木遮挡时造成的识别困难,提出了基于成像激光雷达(Ladar)的自动目标识别算法。介绍了采用的数据分类、数据拼接、目标姿态调整和基于高度直方图的目标识别等方法的操作步骤。应用图像处理技术对场景数据进行分类,实现地面、目标和树木的分离;利用图像帧之间内在的相关性,采用“由粗到精”的方法求出数据变换矩阵并进行数据拼接;根据目标位置的法线方向进行目标姿态调整得到目标高度直方图。最后,将采集的目标高度直方图同存储的模型高度直方图进行匹配,识别出目标类型。实验结果表明:目标提取部分占目标总面积的90%以上,对目标的识别率高于99%,基本满足自动目标识别算法的稳定可靠、识别率高、抗干扰能力强等要求。
Abstract
A novel Automatic Target Recognition(ATR) algorithm based on imaging Laser Radar(Ladar) is proposed to solve the target recognition difficulty caused by a tree sheltering,and the algorithms such as data classification,data mosaic,target pose adjustment and target recognition based on a height histogram are studied.The scene data are classified into several parts (ground,targets and trees) by using an image processing technique,and the translation matrixes among different views are calculated from coarse to fine according to their inherent attributions.Then,the data from different views are integrated together.According to its normal directions,each target pose is adjusted to the top view,and its height histogram can be calculated.Finally,the target is recognized by matching the target height histogram with the saved model height histograms.Experimental results indicate that the extracted target part is more than 90% of the whole target area and the recognition ratio is more than 99%,which satisfies the ATR requirements in stabilization,higher recognition ratios and strong anti-jamming.

马超杰, 杨华, 李晓霞, 吴丹, 黄超超. 复杂场景下应用成像Ladar的自动目标识别[J]. 光学 精密工程, 2009, 17(7): 1714. MA Chao-jie, YANG Hua, LI Xiao-xia, WU Dan, HUANG Chao-chao. Implementation of automatic target recognition by imaging Ladar in complex scenes[J]. Optics and Precision Engineering, 2009, 17(7): 1714.

本文已被 4 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!