Research On Ray Casting Algorithm For The Visualization Of Underwater Acoustic Data

In recent years,with the rapid development of marine source exploration,the visualization technology of underwater acoustic data has been widely used.Through the 3D visualization technology,the image of underwater acoustic data is reconstructed,which is beneficial to analyze the data more directly and effectively for researchers.This thesis briefly introduces the research background and the research status of underwater acoustic data’s visualization,summarizes the current research methods of volume data visualization.The performance of several classical volume rendering methods is analyzed and compared.Finally,the ray casting algorithm which has the highest rendered image quality in all volume data visualization methods is used for the visualization of underwater acoustic data.The characteristics of underwater acoustic data are analyzed in detail,the improvement of data classification and rendering of the ray casting algorithm are proposed.In this thesis,the main research achievements are given as follows:For classification of volume data in the ray casting algorithm,the noise data in underwater acoustic data is relatively high and the classification boundary of different substances contained in underwater acoustic data is fuzzy,a novel data classification method using gray scale histogram based kernel fuzzy C means(HKFCM)algorithm and adaptive genetic algorithm is proposed.In the proposed method,firstly,the objective function of HKFCM algorithm is calculated.Secondly,the adaptive genetic algorithm is used to optimize the above objective function;the optimal clustering matrix and the optimal membership matrix are obtained.Finally,the classification of underwater acoustic data is realized by combining the maximum membership criteria.For the rendering part of the ray casting algorithm,the pre-integration based volume rendering integration for ray casting is proposed to improve the continuity of underwater acoustic data.In the proposed method,the first-order linear interpolation reconstruction is used to replace the original zero-order preserving for the reconstruction of underwater acoustic data’s value,then the color value and opacity value of line segment between two adjacent sampling points are calculated.The pre-integration result is used to solve the volume rendering equation,and the rendering process of data visualization is completed.Experimental results show that the classification of underwater acoustic data is realized by the proposed research method,the 3D visualization effect of classified data is good,details of the underwater acoustic data is shown clearly;after the rendering of the pre-integrated classification based volume rendering integration,suspected target area is more clear,the performance of underwater acoustic data is improve and the visualization effect of underwater acoustic data is improved greatly.