研究目的
To propose a new method for generating multi-view-angle range images for complicated polyhedrons in 3D space, improving the representation of structure features compared to single view angle methods.
研究成果
The proposed multi-view-angle range images generation method effectively represents the structure features of complicated polyhedrons in 3D space, outperforming single view angle methods in terms of consistency with reality and quality of generated images. The method's validation through experiments demonstrates its potential for high-precision structure measurement applications.
研究不足
The method's effectiveness is dependent on the selection of appropriate view angles and the scale of the structure element used in mathematical morphology operations. The computation time may increase with the complexity of the polyhedron and the resolution of the range images.
1：Experimental Design and Method Selection:
The method introduces the concept of three-view drawing from mechanical cartography into range image generation, selecting six view angles (negative and positive directions of ??-, ??-, and ??-axes) for comprehensive coverage. A novel iterative mathematical morphology operation is proposed for generating high-quality range images.
2：Sample Selection and Data Sources:
Point clouds from two models, 'fandisk' and 'oil pump', were used, obtained via laser scanning technology.
3：List of Experimental Equipment and Materials:
Laser scanning instruments for acquiring point cloud data, MATLAB software for processing.
4：Experimental Procedures and Operational Workflow:
The process involves selecting view angles, gridding operation to convert 3D point cloud data into 2D images, and applying mathematical morphology operations to generate range images.
5：Data Analysis Methods:
The quality of generated range images was evaluated by comparing mean values of gray level deviations and Hausdorff distances against standard images extracted from 2D views of models.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
