Automotive Design Renderings, 3d Information Extraction Technology

With the acceleration of the economic globalization process, the world auto industry competition is increasing, and the designing and developing of new models is also increasing. The car body styling is the key and the most difficult part of the car design. Automobile styling can embody exterior design characteristics, and it is the most direct-viewing and the richest part in the automobile appraisal. Good modeling has vital significance to enhance the product competitive power and reflect the corporate culture.At present, reverse engineering based on picture and three-dimensional reconstruction is one of the important ways of enterprise new product development. Using this method may enhance designs and processes efficiency and reduce the manufacturing cycle. In recent years, using the computer vision extract three-dimensional information from the two-dimensional picture to recognition the object has been applied in the automobile styling, overseas research in this aspect's have been quite mature. But this conversion from two-dimensional to three-dimensional is mainly aims at conversion from two-dimensional blueprint. Three-dimensional date extraction and three dimensional restructuring are just at the start stage.In modeling design of automobile product, designers often choose to use hand-drawn sketch or draft of the design to express their creativity in initial period. These paintings on paper now can be converted into 2D digital graphic by scanning. How to extract three-dimensional parameter from this digitized effect chart in order to enhance design efficiency and reduces the mock-up's manufacture is the question which urgently await to be solved in manufactured products design processes such as automobile and so on.Combing perspective principle and image analysis technology and based on deep research on drawing principle of automobile design effect chart, this paper fist convert hand-painted effect chart to 2D digitization image by scanning, and then use digital image processing technology to handle the image including Gray-scale transformation, edge extracting and image segmentation. Finally use data fusion algorithm to fuse and handle the data to complete the extraction of perspective condition and parameter in rendering, and thus to lay foundation for the next three dimensional reconstruction.