自适应不规则纹理的胶囊内镜图像无损压缩

doi:10.19678/j.issn.1000-3428.0063208

计算机工程 ›› 2022, Vol. 48 ›› Issue (10): 21-27,36. doi: 10.19678/j.issn.1000-3428.0063208

自适应不规则纹理的胶囊内镜图像无损压缩

黄胜^1,3, 向思皓^1,3, 胡峰², 马婷², 卢冰^1,3

1. 重庆邮电大学通信与信息工程学院, 重庆 400065;
2. 江苏势通生物科技有限公司, 江苏南通 226399;
3. 重庆邮电大学光通信与网络重点实验室, 重庆 400065

收稿日期:2021-11-11 修回日期:2021-12-26 发布日期:2022-01-05
作者简介:黄胜(1974—),男,教授、博士,主研方向为视频图像编码、人工智能、信道编码;向思皓,硕士研究生;胡峰、马婷,工程师;卢冰,讲师、博士。
基金资助:
重庆市教育委员会科学技术研究计划项目（KJQN201900635）；江苏势通生物科技有限公司项目“基于医疗影像大数据中心的异常图像标注及机器学习系统”（E2019-78）。

Adaptive Lossless Compression of Capsule Endoscopy Image with Irregular Texture

HUANG Sheng^1,3, XIANG Sihao^1,3, HU Feng², MA Ting², LU Bing^1,3

1. School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
2. Jiangsu Citron Biotechnology Co., Ltd., Nantong, Jiangsu 226399, China;
3. Key Laboratory of Optical Communication and Network, Chongqing University of Posts and Telecommunications, Chongqing 400065, China

Received:2021-11-11 Revised:2021-12-26 Published:2022-01-05

摘要/Abstract

摘要： 为缓解无线胶囊内镜图像在电子设备以及服务器中的存储压力，提出一种自适应不规则纹理的无损压缩算法。在图像块内，利用扩展角度预测模式寻找与待预测像素最邻近的5个参考像素，并给其中3个参考像素分配不同权重，同时根据邻近像素值梯度变化规律，扩大待预测像素在不规则纹理方向上的预测值选择范围，基于图像块的最小信息熵选择最优的预测值，将真实值与预测值作差获得预测残差，以适应不规则纹理图像。利用跨分量预测模式选择最优的预测系数，构建符合图像块内预测残差分布规律的线性关系，从而消除当前编码像素中3个分量的冗余数据。结合Deflate算法对经多角度预测模式与跨分量预测模式预测后的剩余残差进行熵编码。实验结果表明，该算法在Kvasir-Capsule数据集上的无损压缩比平均为5.81，相比WebP、SAP、MDIP等算法，具有较优的压缩性能，能够有效提高图像的冗余消除率，其中相较WebP算法的冗余消除率提高约1.9%。

关键词: 胶囊内镜图像, 无损压缩, 预测编码, 扩展角度模式, 不规则纹理

Abstract: This study proposes an adaptive lossless compression algorithm with irregular texture to alleviate the storage pressure of wireless capsule endoscope images in electronic devices and servers.The extended angle prediction mode is used to obtain five reference pixels closest to the pixel to be predicted in the image block and assign different weights to three of them.Based on the gradient variation in the adjacent pixel values, the selection range of the predicted value of the pixel in the irregular texture direction is expanded.Moreover, the optimal predicted value is selected based on the minimum information entropy of the image block.The prediction residual is obtained by comparing the actual and predicted values to adapt to the irregular texture image.The cross-component prediction mode is used to select the optimal prediction coefficient and establish a linear relationship conforming to the residual distribution in the image block; this is done to eliminate redundant data among the three component prediction residuals in the current coded pixel.Entropy coding of the residual residuals predicted by the multi-angle and cross-component prediction modes in combination with the Deflate algorithm.The experimental results show that the lossless compression ratio of this algorithm on the Kvasir-Capsule dataset reaches 5.81 on average.The proposed algorithm has better compression performance than WebP, SAP, MDIP, and other algorithms.It shows an improved redundancy elimination ratio of images, which increases by approximately 1.9% compared to that of WebP.

Key words: capsule endoscopy image, lossless compression, predictive coding, extended angle mode, irregular texture

中图分类号:

TN919.8

黄胜, 向思皓, 胡峰, 马婷, 卢冰. 自适应不规则纹理的胶囊内镜图像无损压缩[J]. 计算机工程, 2022, 48(10): 21-27,36.

HUANG Sheng, XIANG Sihao, HU Feng, MA Ting, LU Bing. Adaptive Lossless Compression of Capsule Endoscopy Image with Irregular Texture[J]. Computer Engineering, 2022, 48(10): 21-27,36.

https://www.ecice06.com/CN/Y2022/V48/I10/21

图/表 10

参考文献

[1] MALATHKAR N V, SONI S K.Low complexity image compression algorithm based on hybrid DPCM for wireless capsule endoscopy[J].Biomedical Signal Processing and Control, 2019, 48:197-204.
[2] BABU C R, CHANDY D A.DPCM based compressor for capsule endoscopic videos[C]//Proceedings of International Conference on Signal Processing and Communication.Washington D.C., USA:IEEE Press, 2017:89-93.
[3] YANG H L, WANG H K, CHEN Y M, et al.Multi-weights intra prediction with double reference lines[C]//Proceedings of IEEE International Conference on Image Processing.Washington D.C., USA:IEEE Press, 2019:3562-3566.
[4] LUCAS L F R, RODRIGUES N M M, CRUZ L A, et al.Lossless compression of medical images using 3D predictors[J].IEEE Transactions on Medical Imaging, 2017, 36(11):2250-2260.
[5] LEI M, LUO F L, ZHANG X F, et al.Two-step progressive intra prediction for versatile video coding[C]//Proceedings of IEEE International Conference on Image Processing.Washington D.C., USA:IEEE Press, 2020:1137-1141.
[6] MÖLLER P, STRUTZ T.Improved lossless image compression using adaptive image rotation[C]//Proceedings of the 27th European Signal Processing Conference.Washington D.C., USA:IEEE Press, 2019:1-4.
[7] KOC B, ARNAVUT Z, SARKAR D, et al.A new technique for lossless compression of color images based on hierarchical prediction, inversion and context adaptive coding[C]//Proceedings of Data Compression Conference.Washington D.C., USA:IEEE Press, 2019:584.
[8] XU W Z, FU F F, WANG Y, et al.Discrete wavelet transform-based fast and high-efficient lossless intraframe compression algorithm for high-efficiency video coding[J].Journal of Electronic Imaging, 2019, 28(1):1-17.
[9] 李庆忠, 谷娜娜, 刘晓丽.基于小波域降采样的低比特率图像压缩算法[J].计算机工程, 2013, 39(4):272-275, 279. LI Q Z, GU N N, LIU X L.Low bit-rate image compression algorithm based on down-sampling in wavelet domain[J].Computer Engineering, 2013, 39(4):272-275, 279.(in Chinese)
[10] CHANG Y J, JHU H J, JIANG H Y, et al.Multiple reference line coding for most probable modes in intra prediction[C]//Proceedings of Data Compression Conference.Washington D.C., USA:IEEE Press, 2019:559.
[11] LI J H, LI B, XU J Z, et al.Efficient multiple-line-based intra prediction for HEVC[J].IEEE Transactions on Circuits and Systems for Video Technology, 2018, 28(4):947-957.
[12] KO H, LEE J, KANG J, et al.Edge-preserving reference sample filtering and mode-dependent interpolation for intra-prediction[J].IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(1):205-216.
[13] YEE D, SOLTANINEJAD S, HAZARIKA D, et al.Medical image compression based on region of interest using Better Portable Graphics (BPG)[C]//Proceedings of International Conference on Systems, Man, and Cybernetics.Washington D.C., USA:IEEE Press, 2017:216-221.
[14] FAN B.Selective compression of medical images via intelligent segmentation and 3D-SPIHT coding[D].Milwaukee:University of Wisconsin-Milwaukee, 2018.
[15] MENTZER F, AGUSTSSON E, TSCHANNEN M, et al.Practical full resolution learned lossless image compression[C]//Proceedings of Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2019:10621-10630.
[16] VAN O A, KALCHBRENNER N, KAVUKCUOGLU K.Pixel recurrent neural networks[C]//Proceedings of the 33rd International Conference on Machine Learning.New York, USA:ACM Press, 2016:1747-1756.
[17] RHEE H, JANG Y I, KIM S, et al.Channel-wise progressive learning for lossless image compression[C]//Proceedings of IEEE International Conference on Image Processing.Washington D.C., USA:IEEE Press, 2020:1113-1117.
[18] ZHOU M H, GAO W, JIANG M Q, et al.HEVC lossless coding and improvements[J].IEEE Transactions on Circuits and Systems for Video Technology, 2012, 22(12):1839-1843.
[19] LI J R, WANG M, ZHANG L, et al.Unified intra mode coding based on short and long range correlations[J].IEEE Transactions on Image Processing, 2020, 29:7245-7260.
[20] KIM W S, PU W, KHAIRAT A, et al.Cross-component prediction in HEVC[J].IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(6):1699-1708.
[21] GOOGLE.WebP:a new image format for the Web[EB/OL].[2021-10-08].https://github.co-m/webmproject/releases/tag/v1.2.0.
[22] THAMBAWITA V, RIEGLER M, HICKS S, et al.The Kvasir-Capsule dataset[EB/OL].[2021-10-08].https://osf.io/dv2ag/.
[23] ZHANG L, WU X L, BUADES A, et al.McMaster dataset[EB/OL].[2021-10-08].https://www4.comp.polyu.edu.hk/~cslzhang/CDM_Dataset.htm.
[24] KRASIN I, DUERIG T, ALLDRIN N, et al., Open images dataset[EB/OL].[2021-10-08].https://storage.googleapis.com/openimages/web/index.html.
[25] FRANZEN R.Kodak lossless true color image suite[EB/OL].[2021-10-08].http://r0k.us/graphics/kodak/.
[26] KHAN T H, WAHID K A.Design of a lossless image compression system for video capsule endoscopy and its performance in in-vivo trials[J].Sensors, 2014, 14(11):20779-20799.

选择文件类型/文献管理软件名称

选择包含的内容

自适应不规则纹理的胶囊内镜图像无损压缩

Adaptive Lossless Compression of Capsule Endoscopy Image with Irregular Texture

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献

相关文章 12

编辑推荐

Metrics

本文评价

[1]	符世园, 张敏行, 高宇, 汪璐, 程耀东. 面向同步辐射光源图像的可并行智能压缩方法[J]. 计算机工程, 2023, 49(1): 234-241,249.
[2]	安晨, 汪成亮, 廖超, 肖诗童. 基于注意力关系网络的无线胶囊内镜图像分类方法[J]. 计算机工程, 2021, 47(10): 252-259,268.
[3]	王豫新, 高美凤. 一种改进的固件增量更新算法[J]. 计算机工程, 2020, 46(10): 210-215.
[4]	鄢海舟,胥布工,石东江,郑伟德. 无损压缩算法LZW前缀编码优化及应用[J]. 计算机工程, 2017, 43(3): 299-303.
[5]	解瑞云,海本斋. 基于自适应霍夫曼和Golomb-Rice混合编码的WSN无损压缩算法[J]. 计算机工程, 2016, 42(7): 86-93.
[6]	张萍, 孔月萍, 江永林. 基于矩形编码的抖动半调图像无损压缩算法[J]. 计算机工程, 2012, 38(20): 188-190.
[7]	孔月萍, 曹慧敏, 耿烨. 基于分形特征的半调图像无损压缩算法[J]. 计算机工程, 2012, 38(19): 222-225.
[8]	陈钢, 张茹, 钮心忻, 杨义先. 大容量矢量地图可逆水印算法[J]. 计算机工程, 2010, 36(21): 129-131.
[9]	姜磊;黄广君. 自适应的无损数据压缩算法[J]. 计算机工程, 2008, 34(1): 102-104.
[10]	邓文平;朱培栋;卢锡城. 面向相似数据集的关系数据库压缩[J]. 计算机工程, 2007, 33(14): 69-70,8.
[11]	吴晶;陈淑珍;曾立波;隗芬芬. 连续帧医学图像无损渐进传输的研究与实现[J]. 计算机工程, 2006, 32(23): 259-260,.
[12]	朱曼洁;赵岭. 基于可逆整数DCT的遥感图像的无损压缩[J]. 计算机工程, 2006, 32(21): 273-274,.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

自适应不规则纹理的胶囊内镜图像无损压缩

Adaptive Lossless Compression of Capsule Endoscopy Image with Irregular Texture

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献

相关文章 12

编辑推荐

Metrics

本文评价