摘要
针对传统基于极端学习机(ELM)的纹理分类方法容易出现输出不稳定的缺点,本文将线性和非线性模型进行融合,改进了传统动力学模型。利用ELM能够快速学习的特点,将其作为基分类器,同时利用线性吸引子和局部吸引子的迭代,实现多个ELM分类器的最佳融合,以提升学习模型的泛化能力和稳定性。利用动力模型实现多分类器的融合有助于寻求多个基分类器之间的一致性,摒弃了基分类器中判别错误的样本输出。通过对CUReT纹理数据库的分类结果,与传统纹理分类方法相比,本文算法的稳定性和分类准确率都得到了一定程度的提升,取得了理想的分类效果。
For the unstable output of the traditional texture classification methods based on extreme learning machine (ELM) ,this paper presents a new approach for the automatic classification of texture images. In order to improve the generalization ability and the robustness of the learning model, this paper improves the traditional dynamical model by fusing the linear and nonlinear models together. Due to the faster learning speed,ELM is used as the basic classifier in this paper. Moreover,a proposed dynamical model is utilized to realize the optimal fusion of multiple ELMs with the iteration of linear and local at- tractor. By combining multiple classifiers with the improved dynamical model, the corrupted classifier outputs are discarded according to the classifier agreement. The experimental results on CURET texture database demonstrate that the proposed approach can improve the stability and classification accuracy significantly,and achieve a more ideal texture classification compared with the traditional methods.
出处
《光电子.激光》
EI
CAS
CSCD
北大核心
2015年第4期752-757,共6页
Journal of Optoelectronics·Laser
基金
国家自然科学基金(61271326)资助项目