The primary aim of this research is to tackle the issue of inaccurate and inconsistent ear disease detection, particularly in remote and under-resourced areas. Traditional diagnostic methods employed by general practitioners and otolaryngologists have shown limitations, underscoring the need for more reliable approaches in these challenging healthcare environments. In response to this issue, a digital otoscope powered by the Rockchip RK3566 processor and enhanced with machine learning capabilities has been developed. This device, using a novel camera system, captures high-resolution images of the patient's ear and applies image classification algorithms to identify and categorize various ear conditions in real time. The design of OtoVision emphasizes ease of use, affordability, and adaptability to different healthcare settings, aiming to make advanced diagnostic tools more accessible to underserved populations. Our testing and analysis reveal that OtoVision significantly enhances the accuracy of ear disease diagnosis. In controlled settings, the device achieved an accuracy rate of approximately 87% when connected to a desktop computer and 82% when operating on a standalone Rockchip single-board computer. These results indicate a substantial improvement over traditional diagnostic methods and demonstrate the potential of integrating machine learning technologies into medical diagnostics. OtoVision represents a step forward in the field of medical diagnostics, particularly for the detection of ear diseases in areas where specialist care is scarce. By leveraging machine learning and digital imaging, OtoVision offers a more accurate, accessible, and cost-effective solution compared to conventional methods. Ongoing development will focus on conducting extensive field testing.