融合多层语义特征的测井电成像空白条带填充深度神经网络方法A Deep Neural Network Method for Filling Blank Strips in Electrical Logging Images Based on Fusion of Multi-Level Semantic Features
袁晓涛,马旭成,肖仕军,邓兆元,穆拉帝力·穆塔力夫,金海峰
摘要(Abstract):
针对常规的编码器-解码器网络模型和U-Net网络模型进行复杂岩性(如砂砾岩)测井电成像空白条带充填时效果较差的问题,提出一种融合多层语义特征的深度神经网络方法。在编码部分,通过带有空洞卷积的骨干网络和空洞空间特征提取模块得到图像的高层语义特征,进行上采样后,与骨干网络提取的低层语义特征融合,最后经过多层上采样解码,得到输出图像。在砂砾岩电成像图像的实际处理中,本文模型的图像填充连续性更好,砾岩颗粒边缘的清晰度和准确性得到有效提升,为电成像图像的后续处理提供了准确的图像信息。
关键词(KeyWords): 电成像测井;深度神经网络;语义特征融合
基金项目(Foundation): 国家科技重大专项“大型油气田及煤层气开发”资助课题“苏丹3/7区高凝油油藏高效开发技术”(2011ZX05032-002)
作者(Author): 袁晓涛,马旭成,肖仕军,邓兆元,穆拉帝力·穆塔力夫,金海峰
参考文献(References):
- [1] 杜闪闪,韩超.基于总变分模型的改进图像修复算法[J].激光与光电子学进展,2018,55(7):169-174.DU Shanshan,HAN Chao.An improved image inpainting algorithm based on total variation model[J].Laser & Optoelectronics Progress,2018,55(7):169-174.
- [2] 何凯,牛俊慧,沈成南,等.基于SSIM的自适应样本块图像修复算法[J].天津大学学报(自然科学与工程技术版),2018,51(7):763-767.HE Kai,NIU Junhui,SHEN Chengnan,et.al.Image inpainting algorithm with adaptive patch using SSIM[J].Journal of Tianjin University(Science and Technology),2018,51(7):763-767.
- [3] 高成英,徐仙儿,罗燕媚,等.基于稀疏表示的物体图像修复[J].计算机学报,2019,42(9):1953-1965.GAO Chengying,XU Xianer,LUO Yanmei,et.al.Object image inpainting based on sparse representation[J].Chinese Journal of Computers,2019,42(9):1953-1965.
- [4] HURLEY N F,ZHANG T F.Method to generate full-bore images using borehole images and multipoint statistics[J].SPE Reservoir Evaluation and Engineering,2011,14(2):204-214.
- [5] 孙健孟,赵建鹏,赖富强,等.电测井图像空白条带填充方法[J].测井技术,2011,35(6):532-537.SUN Jianmeng,ZHAO Jianpeng,LAI Fuqiang.et al.Methods to fill in the gaps between pads of electrical logging images[J].Well Logging Technology,2011,35(6):532-537.
- [6] YAMADA T,QUESADA D,ETCHECOPAR A,et al.Revisiting porosity analysis from electrical borehole images:integration of advanced texture and porosity analysis[C].54th SPWLA Annual Logging Symposium,Society of Petrophysicists and Well-Log Analysts,New Orleans,2013,paper E.
- [7] 傅少庆,伍东,李斐,等.电成像空白条带自适应修复方法[C].第十八届测井年会论文集,2013:391-398.
- [8] 王俊华.基于电成像测井图像处理关键技术研究[D].武汉:长江大学,2015.
- [9] 邢强,张晋言.应用形态分量分析算法研究电成像测井空白条带填充[J].测井技术,2017,41(5):555-559.XING Qiang,ZHANG Jinyan.Filling in gaps between pads of electrical logging images with MCA method[J].Well Logging Technology,2017,41(5):555-559.
- [10] ULYANOV D,VEDALDI A,LEMPITSKY V.Deep image prior[C].IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE,Salt Lake City,2018:9446-9454.
- [11] 王哲峰,高娜,曾蕊,等.基于深度学习模型的测井电成像空白条带充填方法[J].测井技术,2019,43(6):578-582.WANG Zhefeng,GAO Na,ZENG Rui,et al.A gaps filling method for electrical logging images based on a deep learning model[J].Well Logging Technology,2019,43(6):578-582.
- [12] 陈建华,杨丽丽,赵延静,等.深度神经网络模型在测井电成像图像处理中的应用[J].电子测量技术,2021,44(4):138-143.CHEN Jianhua,YANG Lili,ZHAO Yanjing,et al.Application of deep neural network model in processing of electrical logging images[J].Electronic Measurement Technology,2021,44(4):138-143.
- [13] 徐胜军,欧阳朴衍,郭学源,等.多尺度特征融合空洞卷积ResNet遥感图像建筑物分割[J].光学精密工程,2020,28(7):1588-1599.XU Shengjun,OUYANG Puyan,GUO Xueyuan,et al.Building segmentation in remote sensing image based on multiscale-feature fusion dilated convolution resnet[J].Optics and Precision Engineering,2020,28(7):1588-1599.