Abstract
Handwritten and legacy engineering diagrams are still commonly used in many industries and academic settings, but the lack of digitization limits their utility in modern workflows. While significant effort has been made to digitize handwritten content in other engineering domains, the digitization of handwritten circuit diagrams remains underexplored. Automating this process would enable the development of tools capable of converting handwritten circuit diagrams into machine-readable formats that can be instantly interpreted by circuit simulation software. This will offer significant benefits to both students and industry professionals. However, the lack of publicly available datasets focused on the digitization of handwritten circuit diagrams has slowed progress in this area. To address this gap, we developed the Digitize - Handwritten Circuit Diagram (HCD) Dataset, a comprehensive collection of 1,277 handwritten circuit diagrams contributed by 176 volunteers. The dataset includes detailed annotations for multiple aspects of handwritten circuit diagrams, such as component symbols, text labels, and port locations. It contains 18,602 annotated instances across 17 distinct classes of circuit component symbols and 11,936 annotated text labels associated with these components. For the preparation of ground-truth data for component port localization, we developed an annotation tool, which is publicly available for reuse. The Digitize-HCD dataset has the potential to accelerate research on digitization of handwritten circuit diagrams and contribute to the development of advanced end-to-end tools capable of transforming these diagrams into machine-readable formats.