计算机科学 ›› 2018, Vol. 45 ›› Issue (6A): 143-145.

• 模式识别与图像处理 • 上一篇    下一篇

基于GPU的图像监督分类算法的研究

李思瑶,周海芳,方民权   

  1. 国防科技大学计算机学院 长沙410073
  • 出版日期:2018-06-20 发布日期:2018-08-03
  • 作者简介:李思瑶 硕士,主要研究方向为高性能图像处理;周海芳 教授,主要研究方向为高性能图像处理、并行算法;方民权 博士,主要研究方向为高性能图像处理。
  • 基金资助:
    国家自然科学基金:CPU/GPU异构系统下高光谱遥感影像降维多级协同并行计算方法及优化策略(61272146)资助

Research of Image Classification Algorithm Based on GPU

LI Si-yao, ZHOU Hai-fang,FANG Min-quan   

  1. School of Computer Science,National University of Defense Technology,Changsha 410073,China
  • Online:2018-06-20 Published:2018-08-03

摘要: 文中介绍了3种经典的图像分类算法在GPU上的实现,分别是简单贝叶斯分类、KNN、SNN分类。GPU与CPU协同处理是目前使用得较多的结构模式。一般在GPU上执行计算量比较大的程序,CPU负责指挥协调。文中对这3种算法进行了测试,通过实验分析,3种算法的GPU并行程序分别获得了平均72.472,149.536,125.39倍的加速效果。使用的GPU架构是Tesla k20c。贝叶斯、KNN和SNN算法是监督分类算法。实验给出了3种算法图像处理的结果和时间,其均符合要求。

关键词: KNN, SNN, 监督学习算法, 简单贝叶斯算法

Abstract: This paper introduced three classical image classification algorithms based on GPU,which are Bayes,KNN and SNN. Coprocessing of GPU and CPU is a structure pattern used frequently. Programs with large amount of calculation are working on the GPU,and CPU is used to control.This paper tested the programs.Through testing,the times of the acceleration effect of working on the GPU are 72.472,149.536,125.39.The used framework is Tesla k20c.Bayesian,KNN and SNN algorithms are based on supervised classification.The experiment shows the image process results and the times,which meet the requirements.

Key words: KNN, Simple Bayesian algorithm, SNN, Supervised learning algorithm

中图分类号: 

  • TP391
[1]沈欣.光学遥感卫星轨道设计若干关键技术研究[D].北京:清华大学,2012.
[2]林前.图像分割质量的协同评价框架设计与实现[D].成都:西南交通大学,2011.
[3]严蔚敏,吴伟民.数据结构(C语言版)[M].北京:清华大学出版社,2011.
[4]史春.基于遥感处理的交通信息提取研究和实现[D].北京:清华大学,2008.
[5]陈锦伟.遥感图像配准和拼接技术研究[D].杭州:浙江大学,2004.
[6]陈忠.高分辨率遥感图像分类技术研究[D].北京:中国科学院遥感应用研究所,2006.
[7]赵春晖,刘春红.超谱遥感图像降维方法研究现状与分析[J].中国空间科学技术,2004(5):28-36.
[8]周海芳,赵进.基于GPU的遥感图像配准并行程序设计与存储优化[J].计算机研究和发展,2012,49(2):281-286.
[9]陈志远,黄少滨,韩丽丽.现代模态逻辑在计算机科学中的应用研究[J].计算机科学,2013,40(S1):70-76.
[1] 张仁杰, 陈伟, 杭梦鑫, 吴礼发.
基于变分自编码器的不平衡样本异常流量检测
Detection of Abnormal Flow of Imbalanced Samples Based on Variational Autoencoder
计算机科学, 2021, 48(7): 62-69. https://doi.org/10.11896/jsjkx.200600022
[2] 赵志强, 易秀双, 李婕, 王兴伟.
基于GR-AD-KNN算法的IPv6网络DoS入侵检测技术研究
Research on DoS Intrusion Detection Technology of IPv6 Network Based on GR-AD-KNN Algorithm
计算机科学, 2021, 48(6A): 524-528. https://doi.org/10.11896/jsjkx.200500001
[3] 黄铭, 孙林夫, 任春华, 吴奇石.
改进KNN的时间序列分析方法
Improved KNN Time Series Analysis Method
计算机科学, 2021, 48(6): 71-78. https://doi.org/10.11896/jsjkx.200500044
[4] 贺苗苗, 郭卫斌.
基于KNN与矩阵变换的图节点嵌入归纳式学习算法
Inductive Learning Algorithm of Graph Node Embedding Based on KNN and Matrix Transform
计算机科学, 2021, 48(3): 201-205. https://doi.org/10.11896/jsjkx.191200156
[5] 包宗铭, 龚声蓉, 钟珊, 燕然, 戴兴华.
基于双向KNN排序优化的行人再识别算法
Person Re-identification Algorithm Based on Bidirectional KNN Ranking Optimization
计算机科学, 2019, 46(11): 267-271. https://doi.org/10.11896/jsjkx.181001861
[6] 何佶星,陈汶滨,牟斌皓.
流行度划分结合平均偏好权重的协同过滤个性化推荐算法
Coordination Filtering Personalized Recommendation Algorithm Considering Average
Preference Weight and Popularity Division
计算机科学, 2018, 45(6A): 493-496.
[7] 陈静杰,车洁.
基于标准欧氏距离的燃油流量缺失数据填补算法
Fuel Flow Missing-value Imputation Method Based on Standardized Euclidean Distance
计算机科学, 2017, 44(Z6): 109-111. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.023
[8] 冯飞,姜宝华,刘培学,陈玉杰.
改进2DPCA算法在人脸识别中的应用
Application of Improved 2DPCA Algorithm in Face Recognition
计算机科学, 2017, 44(Z11): 267-268. https://doi.org/10.11896/j.issn.1002-137X.2017.11A.056
[9] 薛忠斌,白利光,何宁,周烜,周歆,王珊.
路网中高吞吐量移动对象实时查询算法
Throughput Oriented Real-time Query Processing Algorithm for Moving Objects in Road Network
计算机科学, 2017, 44(3): 16-19. https://doi.org/10.11896/j.issn.1002-137X.2017.03.004
[10] 王立,王欣,马朝东.
一种基于本体KNN的分布式缓存数据交换策略
Distributed Caching Strategy for Data Exchange Program Based on Ontology and KNN Algorithm
计算机科学, 2016, 43(Z11): 316-319. https://doi.org/10.11896/j.issn.1002-137X.2016.11A.074
[11] 张丽平,经海东,李松,崔环宇.
障碍空间中基于Voronoi图的k最近邻查询
k Nearest Neighbor Query Based on Voronoi Diagram for Obstructed Spaces
计算机科学, 2016, 43(5): 174-178. https://doi.org/10.11896/j.issn.1002-137X.2016.05.032
[12] 华辉有,陈启买,刘海,张阳,袁沛权.
一种融合Kmeans和KNN的网络入侵检测算法
Hybrid Kmeans with KNN for Network Intrusion Detection Algorithm
计算机科学, 2016, 43(3): 158-162. https://doi.org/10.11896/j.issn.1002-137X.2016.03.030
[13] 王培重,郑南山,张言哲.
基于动态K值及AP MAC地址筛选的室内定位算法
Indoor Positioning Algorithm Based on Dynamic K Value and AP MAC Address Match
计算机科学, 2016, 43(1): 163-165. https://doi.org/10.11896/j.issn.1002-137X.2016.01.037
[14] 徐晓丹,姚明海,刘华文,郑忠龙.
基于kNN的多标签分类预处理方法
Pre-processing Method of Multi-label Classification Based on kNN
计算机科学, 2015, 42(5): 106-108. https://doi.org/10.11896/j.issn.1002-137X.2015.05.021
[15] 钱燕燕,李永忠,余西亚.
基于多标记与半监督学习的入侵检测方法研究
Intrusion Detection Method Based on Multi-label and Semi-supervised Learning
计算机科学, 2015, 42(2): 134-136. https://doi.org/10.11896/j.issn.1002-137X.2015.02.029
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!