Computational ghost imaging by using complementary illumination patterns

Computational ghost imaging (CGI) is a single-pixel imaging technique by illuminating the object with structured-light. The image of the object can be retrieved by the correlation of numerous power measurements and the corresponding illumination patterns. Although CGI owns many unique merits, its shortcoming is apparent. The demand of numerous measurements is time-consuming, and the reconstructed image always suffers from speckle-like noise. In this paper we proposed to use complementary illumination patterns to perform CGI. In addition, we applied Gerchberg-Saxton-like algorithm to optimize the reconstructed image. By this way, the time of reconstruction is reduced. In addition, the signalto-noise ratio (SNR) significantly increases in comparison with that by using complete random illumination patterns.