Digital autoradiography (DAR) is a powerful tool to quantitatively determine the distribution of a radiopharmaceutical within a tissue section and is widely used in drug discovery and development. However, the low image resolution and significant background noise can result in poor correlation, even errors, between radiotracer distribution, anatomical structure, and molecular expression profiles. Differing from conventional optical systems, the point spread function (PSF) in DAR is determined by properties of radioisotope decay, phosphor and digitizer. Calibration of an experimental PSF a priori is difficult, prone to error, and impractical. We have developed a content-adaptive restoration algorithm to address these problems. Methods: We model the DAR imaging process using a mixed Poisson-Gaussian model, and blindly restore the image by a Penalized Maximum-Likelihood Expectation-Maximization algorithm (PG- PEM). PG-PEM implements a patch-based estimation algorithm with "Density-Based Spatial Clus- tering of Applications with Noise" to estimate noise parameters, and utilizes L2 and Hessian Frobenius (HF) norms as regularization functions to improve performance. Results: First, PG-PEM outperformed other restoration algorithms at the denoising task (p<0.01). Next, we implemented PG-PEM on pre-clinical DAR images (18F-FDG treated mice tumor and heart, 18F-NaF treated mice femur) and clinical DAR images (bone biopsy sections from 223RaCl2 treated castrate resistant prostate cancer patients). DAR images restored by PG-PEM of all samples achieved significantly higher effective resolution, contrast to noise ratio (CNR), and a lower standard deviation of background (STDB) (p<0.0001). Additionally, by comparing the registration results between the clinical DAR images and the segmented bone masks from the corresponding histological images, the radiopharmaceutical distribution was significantly improved (p<0.0001). Conclusion: PG-PEM is able to increase resolution and contrast while robustly accounting for DAR noise, and demonstrates the capacity to be widely implemented to improve pre- and clinical DAR imaging of radiopharmaceutical distribution.