Research On The Algorithms And Experiment For Image Based Modelling

Traditionally, the methods for data collection in Reverse Engineering are mostly touch-measurement. Although these methods are good for high precision, the shortcomings of them have limited the application scopes. The non-touch-measurement based on optics principle , in the other side, can avoid the limitation of touch-measurement, and with the development of technology, the measurement precision have been becoming higher and higher. Based on the stereo-vision theory in the field of computer vision, this paper presented a method for product profile reverse designing according to the theory Image Based 3D Modeling.Firstly, the advantages and the application scopes of existed 3D modeling methods were analyzed and Image Based 3D Modeling method in computer vision field was selected based on the requirements for product profile reverse designing; Then several Corner Detected Arithmetic Operators introduced in the existing papers were compared and Harris Operator was selected for corner detected due to its better stability. Furthermore, according to the analysis of advantages and disadvantages of Camera Calibration Algorithm, Matlab program was written to realize linear camera calibration based on coplanar points, which considered the aberrance of camera lens and avoided nonlinear optimize, and at the same time calculated all interior and exterior parameters of the camera; After the camera parameters obtained, the corner points'coordinates in 3D space were calculated based on the corresponding image points'coordinates in two or more images. Therefore, the deepness information of the object can be obtained from several images; Finally, 3D model of the object was constructed in Rhino by using the discrete and sparse points cloud of the corner points.The 3D modeling method presented in this paper was not limited by the sizes and the location of the objects. Only one digital camera is required in the whole process. Comparing with the traditional modeling system, the requirements for equipment was rapidly reduced At the same time, the model from Rhino could be output as a STL file which could be put into Rapid Prototype Machine directly for manufacturing. That means Image Based Modeling and RP manufacturing can be integrated seamlessly. By introducing the theory and its algorithm in the field of computer vision, the theory and methods in Reverse Engineering was enriched. It would be valuable in theory and have broad application prospects in practice.