基于阵列相机的多光谱成像系统光谱重建算法

Multi-spectral Imaging System Spectral Reconstruction Algorithm Based on Group of Camera

针对传统三通道RGB相机在光源光谱已知条件下不能完全恢复物体表面光谱反射率的缺点,本文构造一套多光谱成像阵列相机系统。该阵列相机采用12个大恒DH-HV1300FM型相机,且11个镜头装有波长不同的滤光片。本文结合阵列相机多通道数的优势,提出一种MSIS-GOC(Multi-spectral Imaging System based on Group of Camera)算法,能够可靠并有效地重建场景的光谱反射率。仿真实验结果分析验证了该系统的有效性。

Recovery of surface spectral reflectance is impossible exclusively with conventional RGB cameras. The appearance of a captured image depends on both the illumination spectrum and the spectral reflectances of objects in the scene. Even if the illumination spectrum is known,an RGB camera provides only 3 measurements which is insufficient to recover the spectral reflectance This paper constructs a set of multi-spectral imaging camera system,which consists of 12 DH-HV1300FM cameras. And eleven len are installed with filters with different wavelengths. The paper proposes an algorithm of multi-spectral imaging system based on group of camera to recover the spectral reflectance of the scene efficiently and robustly,that combines with the advantages of using multiple spectral sources and the groups of multispectral camera. Finally,simulation verifies the validity of this system.