Intelligent Dimension Annotation in Engineering Drawings: A Case-Based Reasoning and MKD-ICP Algorithm Approach

To address the demands for accuracy and completeness in engineering drawing dimension annotation, this paper presents an intelligent dimensioning method that integrates Case-Based Reasoning (CBR), K-Dimensional Tree (KD-Tree), and an enhanced Iterative Closest Point (ICP) algorithm. The proposed approach leverages a historical case database to extract key features from similar cases, providing high-quality initial references for the ICP algorithm. By combining KD-Tree’s efficient spatial search capabilities with ICP’s precise point cloud alignment, the method achieves both efficient mapping and accurate alignment of dimension information. Applied to creating engineering drawings of refrigerated van design as a case study, the results demonstrate that this method significantly enhances the efficiency and precision of dimension annotation, minimizes manual intervention and error rates, and showcases broad application potential in complex engineering design scenarios. The contributions include an innovative intelligent dimensioning method, the MKD-ICP algorithm for dimension mapping and alignment, and empirical validation of the approach’s effectiveness.