An Image-based Virtual Presentation And Aided Design System For Textile Products

With the developments in computer graphics, artificial intelligence, virtual reality and human-machine interactive, especially in the explosions of internet and multimedia technology, a lot of competitive applications come into being in one night. Information society also put great challenges before some traditional fields like textile products industry which is also offered good opportunities in technique renovation and productivity improvement. Textile is born with plenty of graphics characteristics just like color and texture, which makes it very suitable to computer image processing.Virtual presentation and computer aided design have found lots of applications in E-Commerce in recent years. Realistic effects and convenient interaction still remain great challenges. We design and implement an image-based presentation system for textile exhibition (named Easyshow). Very realistic effects are achieved by using novel image processing algorithms. We also provide powerful management functions and various aided design tools.In this paper, we first discuss several key problems in product virtual presentation. With the comparison between two different presentation methods, we come to the conclusion that it's the best way to design a textile product presentation system with image processing techniques.Then we put forward a complete and productive system framework. Easyshow is a typical system developed under Win32 environment, with database supported in back end, and doing images processing and rendering works artificially in front. Generally speaking, there are four components in the system. They are database sub-component, image processing sub-component, aided design sub-component and security sub-component.Grounded on the analysis of textile products attributes and cloth business characteristics, some efforts have been performed on these issues in this dissertation. The main contributions of this dissertation include: a sub-area directions based texturemapping algorithm, a Lab color space base pixel blending algorithm, Knockout matting based boundary auto detection, TIP depth image based mobile lighting simulation, fragment-based simple object completion in images and an attempt in grain concealment with Poisson image editing method.With the cooperation with Shanghai Textile Time Ltd., we expand the system database support capacity and put it into market successfully. Considering the multi-platform serviceability, we made a PC to embedded mobile device migration experiment.