基于残差补偿极限学习机的图像分类算法

Image Classification Based on Error Compensation Extreme Learning Machine

图像分类是人工智能领域的基本研究内容之一。在图像分类任务中,特征的提取和分类器的选择是影响分类正确率的重要因素。本文提出一种基于残差补偿极限学习机的图像分类算法,该方法对原有极限学习机进行改进,在网络训练过程中通过对误差的不断矫正和补偿,获得了性能良好的分类器。并通过特征融合的方法实现不同特征的串联拼接,提高了提取特征的质量,进一步提高了图像分类的准确性。实验结果表明,本文提出的图像分类算法有较好的性能,在多个数据集下均有良好的性能表现。

Image classification is a vital research area in machine learning.In terms of image classification,how to extract features an choose the classifier have great influence on classification accuracy.In this paper,an image classification algorithm which is based on error compensation extreme learning machine is proposed.This algorithm In the process of network training,a classifier with good performance is obtained by continuously correcting and compensating errors.Besides,we use the method of feature fusion to realize the cascade stitching of different features,which improves the quality of feature extraction and further improves the accuracy of image classification.Experimental results show that the algorithm proposed in this paper has better performance on several benchmark datasets.