A study of the photometric accuracy of deconvolved astronomical images was undertaken by processing two simulated images with several well-known algorithms. The simulations consisted of an image of a simple star field and an image of an active galactic nucleus (AGN). Each image was mathematically convolved with a known PSF and then deconvolved to recover the original image. Measurements and comparisons were made before and after the deconvolution to check for photometric accuracy. The best result for the star field was obtained with the multiscale Richardson-Lucy algorithm. When measuring the differential photometry of the nucleus of the AGN between different frames, leaving the image in its convolved state gave a closer variability result to the original than any deconvolution method.