A method for identifying grid features on wall panel surfaces for laser nondestructive detection systems

Purpose This paper aims to propose a preprocessing scheme for an online laser nondestructive testing system for panel thickness, based on the characteristics of the Stereolithography (STL) file format for storing three-dimensional geometric models, to provide precise data support. Design/methodology/approach This paper presents a surface mesh feature recognition technique based on STL files. The technique combines the midpoint matching method of adjacent edges and the divide-and-conquer approach to generate mesh contours and hole boundaries, achieving hierarchical extraction of complex panel machining surfaces, meshes and edge features. To further enhance data organization and management efficiency, a modeling technique of a progressive data organization model is proposed to quickly determine the affiliation between machining surfaces and meshes. Findings Through the method proposed in this paper, each line segment in the processed STL model file is stored only once, with segments connected end-to-end to form each contour of the model, with a precision of 0.006 mm. The results meet the preprocessing requirements of the laser nondestructive testing system. Originality/value With the widespread application of Computer-Aided Design (CAD) in the engineering field, the demand for extracting key features from CAD models is increasing. The preprocessing method proposed in this paper provides the location coordinates and reference values for inspection points for the laser nondestructive testing system and enables automatic planning of inspection paths. Moreover, this preprocessing method can also be applied to model processing, machining path planning and machining accuracy inspection of panel-type workpieces in the manufacturing industry.