Nano-imaging has played a vital role in biology, chemistry and physics as it enables interrogation of material with sub-nanometer resolution. Primary means of achieving such atomic scale interrogation of matter are based on the principles of atomic force microscopy (AFM) and scanning tunneling microscopy (STM). However, current nano-imaging techniques are too slow to be useful in the high speed applications of interest such as studying the evolution of certain biological processes over time that involve very small time scales. In this work, we present a high speed one-bit imaging technique using dynamic mode AFM with a high quality factor cantilever. We model the high quality factor cantilever system using a Markovian model which incorporates the inherent system memory due to the inter-symbol interference and the cantilever state. Next, we pose the imaging problem as one of finding the maximum a posteriori (MAP) symbol detector for this model. This is solved by adapting the BCJR algorithm for our channel model. Furthermore, we propose an improved MAP symbol detector that incorporates a learned prior from the previous scan line while detecting the features on the current scan line. Experimental results demonstrate that our proposed algorithm provides significantly better image resolution compared to current nano-imaging techniques at high scanning speed.