Mingjie Wang
ABSTRACT
With the advent of the era of big data, various types of image data have exploded. Traditional image classification algorithms have the problems of low efficiency and low accuracy, and cannot meet the processing requirements of massive image data. Aiming at the above problems, this paper proposes a massive image classification method based on parallel hybrid classifier algorithm. This algorithm combines the Adaboost algorithm with the RBF algorithm, combines multiple RBF classifiers into a strong classifier, and uses the MapReduce parallel programming model to parallelize the Adaboost-RBF algorithm. The performance test of the algorithm was performed. using the Caltech256 data set. The test results show that compared with the ordinary Adaboost-RBF algorithm, the parallel hybrid classifier algorithm takes less time to run and the average classification accuracy is increased by 26%. This algorithm can meet the needs of automatic classification of massive images.
- Fernández-Delgado M, Cernadas E, Barro S, et al. Do we need hundreds of classifiers to solve real world classification problems [J]. Journal of Machine Learning Research (S1532-4435), 2014, 15 (1): 3133--3181.Google Scholar
- Wang Wei, Wei Ke. Research on massive image classification algorithm based on cloud platform [J]. Modern Electronic Technology, 2019, 42 (19): 46--49.Google Scholar
- Li Xiangyu, Sun Xiaoqing. Research on image classification analysis based on deep learning [J]. China Informatization, 2019 (04): 48--50.Google Scholar
- Zhang Xu, Jiang Jianguo, Hong Richang, et al. Fast image classification algorithm based on Naive Bayes K-nearest neighbor [J]. Journal of Beijing University of Aeronautics and Astronautics, 2015, 41 (2): 302--310.Google Scholar
- Tian Dongxue. Research on massive remote sensing image classification method based on distributed storage [D]. Northeast Forestry University, 2019.Google Scholar
- Gu Nannan, Yao Peiyang, Jiao Zhiqiang. Parallel guided mutation evolution algorithm for large-scale Web service composition problems on the MapReduce platform [J / OL]. Computer Application Research: 1--6 [2020-05-11].Google Scholar
- Lin Dannan, Huang Rui. Research on MapReduce Parallel Clustering Optimization Algorithm in Big Data Mining [J]. Journal of Taiyuan Teachers College (Natural Science Edition), 2019, 18 (04): 49--53.Google Scholar
- Ge Maosong, Wang Yongli, Zhang Liming, Zhao Jiabin, Yu Zhanlong, Zhang Guozhong. MapReduce-based parallel data flow scheduling strategy [J]. Computer Knowledge and Technology, 2019, 15 (26): 11--12.Google Scholar
- Xu Xiaobin, Duan Haribin, Guo Yanjie, Deng Yimin. A Cascade Adaboost and CNN Algorithm for Drogue Detection in UAV Autonomous Aerial Refueling [J]. Neurocomputing, 2020 (prepublish).Google Scholar
- Robert E S.1999. Theoretical views of boosting. Proceedings of European Conference on Computational Learning Theory. Nordkirchen, Germany, Springer-Verlag.Google Scholar
- Cao Jianfang, Chen Junjie, Li Haifang.2013. Research on image emotion classification method based on Adaboost-BP neural network. Journal of Shanxi University (Natural Science Edition), 36 (3): 331--337.Google Scholar
Index Terms
- Research on Massive Image Classification Method Based on Parallel Hybrid Classifier Algorithm
Recommendations
The Naïve Bayesian Classifier Learning Algorithm Based on Adaboost and Parameter Expectations
CSO '10: Proceedings of the 2010 Third International Joint Conference on Computational Science and Optimization - Volume 02Naïve Bayesian classifier is a simple classification method based on Bayes statistics, which is one of the most popular classifiers and has been successfully applied to many fields. To improve the generalization ability of the naïve Bayesian classifier, ...
Adaboost with SVM-based classifier for the classification of brain motor imagery tasks
UAHCI'11: Proceedings of the 6th international conference on Universal access in human-computer interaction: users diversity - Volume Part IIThe Adaboost with SVM-based component classifier is generally considered to break the Boosting principle for the difficulty in training of SVM and have imbalance between the diversity and accuracy over basic SVM classifiers. The Adaboost classifier in ...
AdaBoost classifiers for pecan defect classification
Highlights The performance of AdaBoost algorithms were compared with support vector machine and Bayesian classifiers for pecan defect classification. AdaBoost classifiers took least time and gave best classification accuracy. AdaBoost classifiers ...
Comments