一种BPNNs识别算法的医学检测泛实时性问题研究

doi:10.11896/j.issn.1002-137X.2018.06.053

摘要/Abstract

摘要： 尿沉渣空间环境的复杂性,导致采集的有形成分图像存在较多冗余信息,提取有效的图像信息变得较为困难,进而使得识别系统需要处理的数据量十分巨大。虽然BP神经网络算法的串行版本DJ8000系统平台解决了细胞等有形成分的识别准确率问题,但其不能满足尿沉渣图像医学检验的实时性要求。为此,提出了基于BP神经网络算法优化的并行处理GPU框架的系统平台。它采用并行优化框架,同步高效地对数据进行加速处理;同时,以GPU 计算和测试平台为硬件系统支持,无论是在硬件指标、数据传输及总线技术还是软硬件的兼容性方面,都有助于解决算法中时常出现的负载不均衡的问题。实验数据表明,BP神经网络尿沉渣识别算法在优化并行框架的GPU 系统处理平台上显示的加速比、时效比和运行时间等相关性能参数值都有所提升。相比于DJ8000系统平台,优化的AMD HD7970 和 NVIDIAGTX680 两个并行处理GPU框架系统平台相应的加速比参数值分别是前者的10.82~21.35个和7.63~15.28个标准当量。实验数据充分说明,优化并行框架的GPU处理系统中相关的逻辑数据、地址数据和线性寻程的函数映射关系均能相互动态分配对接并优化算法架构,实现软件到硬件系统的最优比映射,最终解决由于线程间负载不均衡导致的性能瓶颈问题,从而有效地化解了医学领域实时检测中的时效性这一难题。

关键词: BP神经网络, GPU平台, 并行优化, 负载不均衡, 线程协调

Abstract: Due to the complexity of the urine sediment space environment,there is much redundant information of the collected tangible component image,and it also becomes difficult to extract effective image information.Therefore,the amount of data that need to be deal with is huge.Although the serial version DJ8000 system platform of BP neural network algorithm solves the problem of recognition accuracy of tangible components such as cells,it can’t meet the real-time requirement of urine sediment image medical examination.To solve this problem,this paper presented a system platform of parallel processing GPU framework based on BP neural network algorithm optimization.It uses parallel optimization framework to synchronize and accelerate processing of data efficiently.At the same time,it supports the hardware platform based on GPU computing and test platform.Whether from the hardware indicators,data transmission and bus technology or hardware and software compatibility,it will help solve the problems,which often occur in the uneven load irregularities.Experimental data show that BP neural network algorithm for urinary sediment identification improve the performance parameters such as speedup,aging ratio and running time on GPU platform processing platform.Compared with DJ8000 system platform,the parallel processing GPU framework system platforms of AMD HD7970 and NVIDAGTX680 are optimized,and their corresponding acceleration ratio parameter values are 10.82~21.35 and 7.63~15.28 standard equivalents respectively.The data show that optimizing the mapping relationship between logical data,address data and linear seeking function in the GPU processing system of parallel frame can dynamically allocate and optimize the algorithm structure and optimize the mapping between software and hardware system.Finally,it solves the problem of performance bottlenecks caused by load imbalance between threads.Thus,it effectively resolves the problem of real-time detection in urinary sediment environment.

Key words: BP neural networks, GPU platform, Load imbalance, Parallel optimization, Thread coordination

中图分类号:

TP391

刘玉成, 理查德·丁, 张颖超. 一种BPNNs识别算法的医学检测泛实时性问题研究[J]. 计算机科学, 2018, 45(6): 301-307. https://doi.org/10.11896/j.issn.1002-137X.2018.06.053

LIU Yu-cheng, Richard·DING, ZHANG Ying-chao. Research on Pan-real-time Problem of Medical Detection Based on BPNNs Recognition Algorithm[J]. Computer Science, 2018, 45(6): 301-307. https://doi.org/10.11896/j.issn.1002-137X.2018.06.053

参考文献

[1]JIAO L,YANG S Y,LIU F,et al.Seventy Years Beyond Neural Networks:Retrospect and Prospect[J].Chinese Journal of Computers,2016,39(8):1697-1716.(in Chinese)
焦李成,杨淑媛,刘芳,等.神经网络七十年:回顾与展望[J].计算机学报,2016,39(8):1697-1716.
[2]RAFAEL C G,RICHARD E W,EDDINS S L.数字图像处理的MATLAB实现[M].北京:清华大学出版社,2013.
[3]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Image Net classification with deep convolutional neural networks[C]//Proceedings of the Neural Information Processing Systems.Lake Tahoe,USA,2012:1097-1105
[4]LIU Y C,ZHANG Y C.Image Colorimetry Segmentation Based on White Balance Algorithm[J].Computer Engineering,2012,38(20):195-196.(in Chinese)
刘玉成,张颖超.基于白平衡算法的图像色度学分割.[J].计算机工程,2012,38(20):195-196.
[5]TZENG S,ATNEY A,OWENS J D.Task management for irregular-parallel workloads on the GPU[C]//Proceedings of the Conference on High Performance Graphics.Ville,Switzerland,2010:29-37.
[6]SHARMA B,THOTA R,VYDYANATHAN N,et al.Towards a Robust,Real-time Face Processing System Using CUDA-enabled GPUs[C]//Proceedings of the 2009 IEEE International Conference on High Performance Computing.Kochi,India,2009:368-377.
[7]GHORAYEB H,STEUX B,LAURGEAU C.Boosted Algorithms for Visual Object Detection on Graphics Processing Units[C]//Proceedings of the 7th Asian Conference on Computer Vision.Hyderabad,India,2006:254-263.
[8]MERRILL D,GARIAND M,GRIMSHAW A.Scalable GPU Graph Traversal[C]//Proceedings of the 17th ACM SIGPLAN symposium on Principles and Parallel Pragramming.New York,USA,2012:117-128.
[9]AILA T,LAINE S.Understanding the Efficiency of Ray Traversal on GPUs[C]//Proceedings of the Conference on High Performance Graphics.New York,USA,2009:145-150.
[10]CEDERMAN D,TSIGAS P.On Dynamic Load Balancing on Graphics Processors[C]//Proceedings of the 23rd ACM Symposium on Graphic Hardware.Ville,Switzerland,2008:57-64.
[11]CHATTERJEE S,GROSSMAN M,SBIRLEA A,et al.Dynamic Task Parallelism with a GPU Work-Stealing Runtime System[C]//Proceedings of the 24th International Workshop on Languages and Compilers for Parallel Computing.Fort Collins,USA,2011:203-217.
[12]NVIDIA Corporation.NVIDIA GeForce GTX 750 Ti:Featuring First-Generation Maxwell GPU Technology.Designed for Extreme Performance per Watt[R].IEEE,2014.
[13]AMD Corporation.Accelerated Parallel Processing OpenCLTM[R].IEEE,2014.
[14]CIRESAN D,MEIER U,MASCI J,et al.A committee of neural networks for traffic sign classification[C]//Proceedings of the International Joint Conference on Neural Networks.San Jose,USA,2011:1918-1921.
[15]JIA Y Q,SHELHAMER E,DONAHUE J,et al.Caffe:convolutional architecture for fast feature embedding[C]//Proceedings of the ACM International Conference on Multimedia.Orlando,FL,USA,2014,675-678.
[16]LIU M,QUAN T,LUAN S.An attribute recognition system based on rough-set theory-fuzzy neural network and fuzzy expert system[C]//Fifth World Congress onIntelligent Control and Automation,2004(WCICA 2004).IEEE,2004:2355-2359.
[17]PEI Y K.Study on Urine Sediment Recognition System Based on Neural Network and Fuzzy Reasoning[D].Nanjing:Nanjing University of Science and Technology,2008.(in Chinese)
裴元焜.基于神经网络与模糊推理的尿沉渣识别系统研究[D].南京:南京信息工程大学,2008.
[18]SHEN M L.Study on Urinary Sediments Visible Component Automation Clssfication System[D].Changchun:Changchun University of Science and Technology,2006.(in Chinese)
沈美丽.尿沉渣有形成分自动分类系统研究[D].长春:长春理工大学,2006.
[19]BURTSCHER M,NASRE R,PINGALI K.A Quantitative Study of Irregular Programs on GPUs[C]//Proceedings of the 2012 IEEE International Symposium on Workload Characte-rization.La Jolla,USA,2012:141-151.
[20]NASRE R,BURTSCHER M,PINGALI K.Atomic-free Irregular Computations on GPUs[C]//Proceedings of the 6th Workshop on General Purpose Processor Using Graphics Processing Units.New York,USA,2013:96-107.
[21]NASRE R,BURTSCHER M,PINGALI K.Data-Driven Versus Topology- driven Irregular Computations on GPUs[C]//Proceedings of IEEE 27th International Symposium on Parallel & Distributed Processing.Boston,USA,2013:463-474.
[22]YAN S G,LONG G P,ZHANG Y Q.StreamScan:Fast Scan Algorithms for GPUs Without Global Barrier Synchronization[C]//Proceedings of the 18th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming.Guangzhou,China,2013.

相关文章 15

[1]	刘宝宝, 杨菁菁, 陶露, 王贺应. 基于DE-LSTM模型的教育统计数据预测研究 Study on Prediction of Educational Statistical Data Based on DE-LSTM Model 计算机科学, 2022, 49(6A): 261-266. https://doi.org/10.11896/jsjkx.220300120
[2]	徐佳楠, 张天瑞, 赵伟博, 贾泽轩. 面向供应链风险评估的改进BP小波神经网络研究 Study on Improved BP Wavelet Neural Network for Supply Chain Risk Assessment 计算机科学, 2022, 49(6A): 654-660. https://doi.org/10.11896/jsjkx.210800049
[3]	朱旭辉, 沈国娇, 夏平凡, 倪志伟. 基于螺旋进化萤火虫算法和BP神经网络的模型及其在PPP融资风险预测中的应用 Model Based on Spirally Evolution Glowworm Swarm Optimization and Back Propagation Neural Network and Its Application in PPP Financing Risk Prediction 计算机科学, 2022, 49(6A): 667-674. https://doi.org/10.11896/jsjkx.210800088
[4]	夏静, 马中, 戴新发, 胡哲琨. 基于BP神经网络的智能云效能模型 Efficiency Model of Intelligent Cloud Based on BP Neural Network 计算机科学, 2022, 49(2): 353-367. https://doi.org/10.11896/jsjkx.201100140
[5]	郭福民, 张华, 胡瑢华, 宋岩. 一种基于表面肌电信号的腕部肌力估计方法研究 Study on Method for Estimating Wrist Muscle Force Based on Surface EMG Signals 计算机科学, 2021, 48(6A): 317-320. https://doi.org/10.11896/jsjkx.200600021
[6]	程铁军, 王曼. 基于变权组合的突发事件网络舆情趋势预测 Network Public Opinion Trend Prediction of Emergencies Based on Variable Weight Combination 计算机科学, 2021, 48(6A): 190-195. https://doi.org/10.11896/jsjkx.200600094
[7]	石琳姗, 马创, 杨云, 靳敏. 基于SSC-BP神经网络的异常检测算法 Anomaly Detection Algorithm Based on SSC-BP Neural Network 计算机科学, 2021, 48(12): 357-363. https://doi.org/10.11896/jsjkx.201000086
[8]	周俊, 尹悦, 夏斌. 基于LSTM神经网络的声发射信号识别研究 Acoustic Emission Signal Recognition Based on Long Short Time Memory Neural Network 计算机科学, 2021, 48(11A): 319-326. https://doi.org/10.11896/jsjkx.210700034
[9]	焦东来, 王浩翔, 吕海洋, 徐轲. 基于手机传感器轨迹的路面地物检测方法 Road Surface Object Detection from Mobile Phone Based Sensor Trajectories 计算机科学, 2021, 48(11A): 283-289. https://doi.org/10.11896/jsjkx.210200145
[10]	宋岩, 胡瑢华, 郭福民, 袁新亮, 熊睿洋. 基于sEMG的改进SVM+BP肌力预测分层算法 Improved SVM+BP Algorithm for Muscle Force Prediction Based on sEMG 计算机科学, 2020, 47(6A): 75-78. https://doi.org/10.11896/JsJkx.190900143
[11]	诸珺文. 基于改进BP神经网络的SQL注入识别 SQL InJection Recognition Based on Improved BP Neural Network 计算机科学, 2020, 47(6A): 352-359. https://doi.org/10.11896/JsJkx.191200054
[12]	周立鹏, 孟利民, 周磊, 蒋维, 董建平. 基于BP神经网络的摔倒检测算法 Fall Detection Algorithm Based on BP Neural Network 计算机科学, 2020, 47(6A): 242-246. https://doi.org/10.11896/JsJkx.191000077
[13]	吕小敬, 刘钊, 褚学森, 石树鹏, 孟虹松, 黄震春. 面向超大规模并行模拟的LBM计算流体力学软件 Extreme-scale Simulation Based LBM Computing Fluid Dynamics Simulations 计算机科学, 2020, 47(4): 13-17. https://doi.org/10.11896/jsjkx.191000010
[14]	陈燕文,李坤,韩焱,王燕平. 基于MFCC和常数Q变换的乐器音符识别 Musical Note Recognition of Musical Instruments Based on MFCC and Constant Q Transform 计算机科学, 2020, 47(3): 149-155. https://doi.org/10.11896/jsjkx.190100224
[15]	刘晓彤,王伟,李泽禹,沈思婉,姜小明. 基于改进BP神经网络的尿液中红白细胞识别算法 Recognition Algorithm of Red and White Cells in Urine Based on Improved BP Neural Network 计算机科学, 2020, 47(2): 102-105. https://doi.org/10.11896/jsjkx.191100195

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed