Fast 3D Visualization of Massive Seismic Data in Deep Clustering Index

doi:10.19678/j.issn.1000-3428.0063095

Abstract

Abstract: The 3D visualization technology of seismic data can intuitively analyze geological structure information, which can provide data support for geological exploration and other research.To solve the problems of display delay and image jump lag when loading large scale data in traditional volume-rendering algorithms, this study proposes a fast 3D visualization algorithm of massive seismic data based on deep clustering index.Firstly, the spatial feature representation of data is learned by variational autoencoding and deep clustering;then, the clustering performance is improved by iterative optimization of objective function.Thus, the problem of overlapping nodes caused by uneven spatial data distribution is solved, and an efficient index structure is established to improve the efficiency of real-time data reading.Additionally, a temporal convolutional network is used to predict the position of the next viewpoint, and the potential data is loaded into memory in advance to avoid the problem of picture-stuck jump caused by centralized loading when the data amount is too large.Finally, combined with the viewpoint dynamic-division scheduling model based on double-layer visual cone, redundant nodes are removed to reduce the system load, and improve the data-rendering speed and fluency.The experimental results show that, compared with the Hilbert R-Tree (HRT), the time of querying data sub-blocks on the index structure of algorithm in this paper is reduced by 64.14%~66.37%, and the accuracy of predicting viewpoints is improved by 12.08%~22.70%, compared with the Lagrange interpolation algorithm.The real-time frame rate can also be relatively stable and smooth on a large subset.The rendering performance of the whole system achieves the desired effect under the premise of ensuring the image quality.

Key words: 3D visualization, deep learning, deep clustering, Hilbert R-Tree(HRT), sequence trajectory prediction forecasting, view frustum culling

摘要： 地震数据的三维可视化能够直观反映地质的相关结构信息，为地质勘探等研究提供数据支持。针对传统体绘制算法在集中载入海量数据时存在显示延迟、画面跳跃、卡顿等问题，提出一种快速三维可视化算法。使用变分自编码器和深度聚类学习数据的空间特征表示，通过迭代优化目标函数提高聚类性能，以解决因空间数据分布不均造成的节点重叠问题。建立高效的索引结构，提高数据实时读取的效率，通过时序卷积网络预测下一个视点位置，提前将潜在数据载入内存，避免因海量数据集中加载导致画面卡顿、跳跃。使用基于双层视锥体的视点动态划分调度模型，剔除不必要的绘制节点及减轻系统负荷，从而提高数据渲染速度和流畅度。实验结果表明，该算法在索引结构上查询数据块的时间相比希尔伯特R树算法减少了64.14%~66.37%，预测视点的正确率相比拉格朗日插值算法提高了12.08%~22.70%，实时帧率在较大规模的子集上也能够相对稳定平滑，在保证图像质量的前提下整体系统的渲染性能达到预期效果。

关键词: 三维可视化, 深度学习, 深度聚类, 希尔伯特R树, 时序轨迹预测, 视锥体裁剪

CLC Number:

TP391

TANG Wenlin, XIE Kai, WEN Chang, HE Jianbiao. Fast 3D Visualization of Massive Seismic Data in Deep Clustering Index[J]. Computer Engineering, 2022, 48(11): 275-283.

汤文琳, 谢凯, 文畅, 贺建飚. 深度聚类索引下的海量地震数据快速三维可视化[J]. 计算机工程, 2022, 48(11): 275-283.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0063095

http://www.ecice06.com/EN/Y2022/V48/I11/275

Figures/Tables 17

References

[1] VISUTSAK P.Marching cubes and histogram pyramids for 3D medical visualization[J].Journal of Imaging, 2020, 6(9):88-98.
[2] LI L F, QI M R.Design of three-dimensional visualization system for medical image based on magnetic resonance angiography data[J].Matrix Science Medica, 2019, 3(2):28-34.
[3] HAN X, LIU J M, TAN B H, et al.Design and implementation of smart ocean visualization system based on extended reality technology[J].Journal of Web Engineering, 2021, 20(2):557-574.
[4] BI M L, XUE X G, LI C H, et al.Discussion on 3D visualization model of geological structure[J].Journal of Physics:Conference Series, 2019, 1345(5):14-21.
[5] 周文辉, 石敏, 朱登明, 等.SEGY格式地震数据的三维可视化[J].计算机应用与软件, 2019, 36(2):78-84, 154. ZHOU W H, SHI M, ZHU D M, et al.Three-dimensional visualization of seismic data in segy format[J].Computer Applications and Software, 2019, 36(2):78-84, 154.(in Chinese)
[6] 王华维, 何柳, 曹轶, 等.大规模科学数据体绘制技术综述[J].国防科技大学学报, 2020, 42(2):1-12. WANG H W, HE L, CAO Y, et al.A survey of the techniques of volume rendering for large-scale scientific data[J].Journal of National University of Defense Technology, 2020, 42(2):1-12.(in Chinese)
[7] AGRAWAL A, KUMAR P.GPU-based focus-driven multi-coordinates viewing system for large volume data visualisation[J].International Journal of Computational Systems Engineering, 2018, 4(2/3):86-92.
[8] XIE K, WU P, YANG S.GPU and CPU cooperation parallel visualisation for large seismic data[J].Electronics Letters, 2010, 46(17):1196-1203.
[9] 李彩林, 陈文贺, 胡善明, 等.采用八叉树和OSG分页结点的海量点云三维可视化[J].计算机工程与应用, 2019, 55(21):234-239. LI C L, CHEN W H, HU S M, et al.Massive point cloud visualization using octree and OSG paging nodes[J].Computer Engineering and Applications, 2019, 55(21):234-239.(in Chinese)
[10] LIMKAR S V, JHA R K.A novel method for parallel indexing of real time geospatial big data generated by IoT devices[J].Future Generation Computer Systems, 2019, 97:433-452.
[11] HAI Y, HAN B, SU W, et al.R-tree index construction of dynamic K-means algorithm[C]//Proceedings of the 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery.Washington D.C., USA:IEEE Press, 2018:831-836.
[12] 侯海耀, 钱育蓉, 英昌甜, 等.基于Hilbert-R树分级索引的时空查询算法[J].计算机应用, 2018, 38(10):2869-2874, 2885. HOU H Y, QIAN Y R, YING C T, et al.Spatio-temporal query algorithm based on Hilbert-R tree hierarchical index[J].Journal of Computer Applications, 2018, 38(10):2869-2874, 2885.(in Chinese)
[13] 王响, 雷小永, 戴树岭.基于视点预测的大规模地形的实时渲染[J].系统仿真学报, 2013, 25(6):1202-1206. WANG X, LEI X Y, DAI S L.Real time rendering of large scale terrain based on viewport prediction[J].Journal of System Simulation, 2013, 25(6):1202-1206.(in Chinese)
[14] CHENG H, XIE K, WEN C, et al.Fast visualization of 3D massive data based on improved Hilbert R-tree and stacked LSTM models[J].IEEE Access, 2021, 9:16266-16278.
[15] SUO Y, CHEN W, CLARAMUNT C, et al.A ship trajectory prediction framework based on a recurrent neural network[J].Sensors(Basel), 2020, 20(18):5133-5143.
[16] MIKHAILOV S, KASHEVNIK A.Car tourist trajectory prediction based on bidirectional LSTM neural network[J].Electronics, 2021, 10(12):1390-1403.
[17] SERPA Y R, RODRIGUES M A F.A draw call-oriented approach for visibility of static and dynamic scenes with large number of triangles[J].The Visual Computer, 2019, 35(4):549-563.
[18] JIANG Y Z, NIU P.View frustum culling algorithm for scene based on adaptive binary tree[C]//Proceedings of the 2nd IEEE Advanced Information Management, Communicates, Electronic and Automation Control Conference.Washington D.C., USA:IEEE Press, 2018:1159-1164.
[19] WEN C, LI L, XIE K.Fast visualisation of massive data based on viewpoint motion model[J].Electronics Letters, 2017, 53(15):1038-1040.
[20] 孙小虎, 韩文军, 潘娟, 等.基于聚合与动态调度的变电站三维可视化研究[J].计算机工程, 2019, 45(12):300-307. SUN X H, HAN W J, PAN J, et al.Research on three-dimensional visualization of substation based on aggregation and dynamic scheduling[J].Computer Engineering, 2019, 45(12):300-307.(in Chinese)
[21] WEI R Q, GARCIA C, EL-SAYED A, et al.Variations in variational autoencoders-a comparative evaluation[J].IEEE Access, 2020, 8:153651-153670.
[22] XIE J Y, GIRSHICK R, FARHADI A.Unsupervised deep embedding for clustering analysis[EB/OL].[2021-10-01].https://arxiv.org/abs/1511.06335.
[23] BAI S J, KOLTER J Z, KOLTUN V.An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[EB/OL].[2021-10-01].https://arxiv.org/abs/1803.01271.
[24] YANN L C, CORINNA C, CHRISTOPHER J C.The mnist database of handwritten digits[EB/OL].[2021-10-01].http://yann.lecun.com/exdb/mnist/.
[25] CAI D, HE X, HAN J.Document clustering using locality preserving indexing[C]//Proceedings of IEEE Transactions on Knowledge and Data Engineering.Washington D.C., USA:IEEE Press, 2005:1624-1637.
[26] KHAN S S, AHMAD A.Cluster center initialization algorithm for K-means clustering[J].Pattern Recognition Letters, 2004, 25(11):1293-1302.

Please choose a citation manager

Content to export