The Research Of Volume Rendering Algorithm With Mass Data Based On Vector Quantization Compressed

Volume Rendering is a technology which generate a two-dimensional image from three-dimensional data field directly. With the development of graphics hardware,based on GPU accelerated volume rendering algorithms become mainstream institutional fields. Compression Volume Rendering is a new volume rendering method which is also involved volume compression and volume rendering. Compression Volume Rendering drawn mainly to solve the memory capacity constraints when dealing with massive data volume rendering. The existing compression volume rendering methods in dealing with massive seismic data lost the important information severely, and can not balance the compression ratio and distortion of the issue. This paper mainly address this issue, and combine seismic data processing to research the compression algorithm. The main contribution of this work is as follows:(1) Proposed a region of interest(ROI) based vector quantization compression volume rendering method: In seismic data processing, geological experts are really interested in some certain region of the entire data. We can get the overall compression ratio by concentrate its limited resources to the region and increase the distortions outside the area. This is to keep the overall compression ratio and maintain the accuracy of the region. This paper Base on this analysis, combined with image processing thoughts on the region of interest(Region of Interest ROI) proposed ROI-based vector quantization compression volume rendering. The basic idea is: based on the ROI’s description, stripped from the body of important regional data and improve the current vector quantization algorithm, through calibration region to influence the trend of compression, ensuring distortion within ROI. In this paper, based on the geological horizon feature in seismic data processing to calibrate ROI, and do the simulation. Experimental results show that this method can satisfy the current compression ratio, ensure the accuracy of ROI, and we can see the overall picture of the whole data.(2) Proposed a target confidence based vector quantization compression volume rendering method: In seismic data, some meaningful geological structure are usually reflected in its value. Users can not describe this structure by way of the calibration area, but the user can give the appropriate judgment based on experience. Based on this analysis, this paper propose target vector confidence compression volume rendering methods. The basic idea is: analysis the target after modeling, and give the CF of the data in volume by the user’s experience, and introduce the level to the compression algorithms, Finally compress the volume by the CF division. Firstly, we input the theoretical data to the algorithm From the two angle of Numerical analysis and image effects to verify the correctness of the method. The algorithm is then applied to the actual seismic data processing, to obtain a certain effect.(3) design and implementation of the overall structure of the three-dimensional visualization software system, and design, development, maintenance the underlying seismic data management module completely to provide a stable fast data access interface for the upper application.