On the Key Technology of Target Detection Based on Compressive Ghost Imaging
L-H. Zhang, Y. Kang, B. Li, D. Liang, Z-L. Pan, D-W. Zhang and X-H. Ma
Efficient target detection is an important technology for many vision applications, in this paper, we present a method based on compressive ghost imaging programs to achieve target detection by using background subtraction. Ghost imaging offers great potentiality, with respect to standard imaging, to obtain the imaging of objects located in optically harsh or noisy environment. In our method, the target object is irradiated by laser firstly and the object information is captured by ghost imaging, the measured value of background subtraction image obtained by background subtraction is sparse and image is reconstructed by compressive sensing directly. The experiments suggest that our method can not only obtain high quality reconstructed image with lower sampling rate, but also extract the target image in different illumination conditions accurately. This technique can apply to target detection and obtain high quality reconstructed target image.
Keywords: Laser irradiation, target detection, ghost imaging, compressive sensing, background subtraction, sparsity