Nissl histology underpins our understanding of brain anatomy and architecture. Despite its importance, no high-resolution datasets are currently available in the literature for 14-day-old rats. To remedy this issue and demonstrate the utility of such a dataset, we have acquired over 2000 high-resolution images (0.346 μm per pixel) from eight juvenile rat brains stained with cresyl violet. To analyze this dataset, we developed a semi-automated pipeline using open-source software to perform cell density quantification in the primary somatosensory hindlimb (S1HL) cortical column. In addition, we performed cortical layer annotations both manually and using a machine learning model to expand the number of annotated samples. After training the model, we applied it to 262 images of the S1HL, retroactively assigning segmented cells to specific cortical layers, enabling cell density quantification per layer rather than just for entire brain regions. The pipeline improved the efficiency and reliability of cell density quantification while accurately assigning cortical layer boundaries. Furthermore, the method is adaptable to different brain regions and cell morphologies. The full dataset, annotations, and analysis tools are made publicly available for further research and applications.