Rivet Flushness Measurement of Aircraft Skin Based on Density-Geometry Collaborative Constraint

The assembly quality of aircraft skin rivets directly affects aerodynamic performance and structural integrity, particularly in high-speed aircraft and stealth fighters. Consequently, rivet flushness measurement is a critical aspect of aircraft assembly inspection. However, existing methods often suffer from high computational complexity, poor noise resistance, and redundant feature extraction, which limits measurement accuracy and efficiency. To address these limitations, this paper proposes a density-geometry collaborative constraint method that integrates geometric feature analysis with an optimization algorithm. Specifically, in the data preprocessing stage, a lightweight contour extraction method is developed to reduce the processed data volume to 3.5–6.8% of the original scale, thereby mitigating the computational burden and eliminating feature redundancy. In the rivet feature screening stage, a density-geometry dual screening strategy is proposed to precisely eliminate abnormal clusters via a dual-threshold filtering mechanism, enhancing antinoise capability. In the quantitative evaluation stage, a circular model reconstruction method combining probability sampling and dynamic weight optimization is developed to accurately identify and locate the circular contour of the rivet head. The proposed method enables rapid and accurate rivet feature identification and flushness measurement.