In this work, we reconstruct discrete image functions by means Bessel-Fourier polynomials. To measure the image reconstruction we use the Normalized image reconstruction error between the input and reconstructed images. We show that, a good reconstruction performance is found to be available for gray-level images. The reconstruction algorithm is implemented using the first forty zeros of the Bessel functions of the first kind. Experimental results are presented.© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.