Ion-beam radiography (iRad) could potentially improve the quality control of ion-beam therapy. The main advantage of iRad is the possibility to directly measure the integrated stopping power. Until now there is no clinical implementation of iRad. Topics of ongoing research include developing dedicated detection systems to achieve the desired spatial resolution (SR) and investigating different ion types as imaging radiation. This work focuses on the theoretical and experimental comparison of proton (pRad) and helium-beam radiography (αRad). The experimental comparison was performed with an in-house developed detection system consisting of silicon pixel detectors. This system enables the measurement of energy deposition of single ions, their tracking, and the identification of the ion type, which is important for αRad due to secondary fragments. A 161 mm-thick PMMA phantom with an air gap of 1 mm placed at different depths was imaged with a 168 MeV u-1 proton/helium-ion beam at the Heidelberg ion-beam therapy center. The image quality in terms of SR and contrast-to-noise ratio (CNR) was evaluated. After validating MC simulations against experiments, pRad and αRad were compared to carbon-beam radiography (cRad) in simulations. The theoretical prediction that the CNR of pRad and αRad is equal at similar imaging doses was experimentally confirmed. The measured SR of αRad was 55% better compared to pRad. The simulated cRads showed the expected improvement in SR and the decreased CNR at the same dose compared to the αRads, however only at dose levels exceeding typical doses of diagnostic x-ray projections. For clinically applicable dose levels, the cRads suffered from an insufficient number of carbon ions per pixel (220 μm × 220 μm). In conclusion, it was theoretically and experimentally shown that αRad provides a better SR than pRad without any disadvantages concerning the CNR. Using carbon ions instead of helium ions leads to a better SR at the cost of higher doses.