首页|基于偏振成像的复杂反射率冰形三维测量

基于偏振成像的复杂反射率冰形三维测量

扫码查看
飞机机翼结冰是飞行安全的重要隐患,防冰除冰研究对于飞行器动力性能评估、安全评估等有着非常重要的意义,而准确地测量结冰外形是研究机翼结冰的基础.不同于普通物体,冰体具有复杂的反射率,冰面杂散光较为严重,传统方法难以精确提取冰面的中心线.为此,提出了一种基于偏振成像技术的复杂反射率冰形测量方法,利用偏振特性解决冰面杂散光的问题.将偏振相机捕获的原始图像分离并插值,逐像素筛选最大特征值及其特征向量,融合为最优的偏振信息,并基于Steger算法提取激光中心线,实现冰形三维重建.实验结果表明,对平板形冰面测量误差的标准差为0.123 7mm,所提方法能够有效地重建出具有复杂反射率的冰形三维形貌.
3D Measurement of Complex Reflectance Ice Shapes Based on Polarization Imaging
Icing on the wings is an important potential danger for flight safety,the study of anti-de-icing is of great significance for the evaluation of aircraft power performance,safety assessment,etc.,thus the accurate measurement of the icing profile is the basis for the study of icing on the wings.Different from ordinary objects,the ice has more complex reflectivity,and the stray light of ice surface is more serious.The traditional method is difficult to accurately extract the centerline of the ice surface.Therefore,to solve the problem of stray light on the ice surface,a high-accuracy measurement method of ice shape with complex reflectance based on polarization imaging technique by using the polarization property is proposed.The original images captured by the polarization camera are separated and interpolated,and the maximal eigenvalues and their eigenvectors are selected pixel by pixel,fused into the optimal polarization information.Then the laser centerline based on Steger's algorithm to realize the 3D reconstruction of the ice shape is extracted.The experimental results show that the standard deviation of the measurement error is 0.123 7 mm for the flat plate-shaped ice surface,and the proposed method is able to effectively reconstruct the three-dimensional shape of the ice shape with complex reflectivity.

ice shape 3D measurementstray lightanti-de-icingpolarization imagingSteger algorithm

魏振东、王杰、王启江、朱思洁、王亚军

展开 >

四川大学电子信息学院,四川成都 610065

冰形三维测量 杂散光 防除冰 偏振成像 Steger算法

四川省科技计划国家自然科学基金结冰与防除冰重点实验室开放基金

2023NSFSC049662075143IADL20200308

2024

光学与光电技术
华中光电技术研究所 武汉光电国家实验室 湖北省光学学会

光学与光电技术

CSTPCD
影响因子:0.351
ISSN:1672-3392
年,卷(期):2024.22(2)
  • 26