We examine the impact of background lumiphore on image quality in luminescence optical tomography. A modification of a previously described [J. Chang, H. L. Graber, and R. L. Barbour, J. Opt. Soc. Am. A 14, 288–299 (1997); J. Chang, H. L. Graber, and R. L. Barbour, IEEE Trans. Biomed. Eng. 44, 810–822 (1997)] that estimates the background luminescence directly from the detector readings is developed. Numerical simulations were performed to calculate the diffusion–regime limiting form of forward–problem solutions for a specific test medium. We performed image reconstructions with and without white noise added to the detector readings, using both the original and improved versions of the algorithm. The results indicate that the original version produces unsatisfactory reconstructions when background lumiphore is present, whereas the improved algorithm yields qualitatively better images, especially for small target–to–background luminescence yield ratios.