Extreme learning machine with coefficient weighting and trained local receptive fields for image classification
- PDF / 1,718,014 Bytes
- 22 Pages / 439.37 x 666.142 pts Page_size
- 76 Downloads / 185 Views
Extreme learning machine with coefficient weighting and trained local receptive fields for image classification Chao Wu 1 & Yaqian Li 2
1
2
& Yaru Zhang & Jing Liu & Bin Liu
1
Received: 23 July 2019 / Revised: 22 June 2020 / Accepted: 29 June 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Local receptive fields based extreme learning machine (ELM-LRF) is widely used to solve image classification problems. However, the performance of ELM-LRF is limited by the single generation method of local receptive fields and the simple network structure. In order to solve these problems and make full use of image information to improve classification accuracy, extreme learning machine with coefficient weighting and trained local receptive fields (ELM-WLRF) is proposed based on ELM-LRF. The structure is mainly composed of convolution blocks, weighting blocks, dimensionality reduction and classification layers. In the convolution block, the principle of the ELM and the method of grouping calculation are used to train the local receptive fields of the two convolutional layers. The trained local receptive fields are used to extract identifiable feature information in the image more stably and adequately. In the weighting block, the principles of ELM and ELM autoencoder (ELM-AE) are used to train channel and spatial weighting coefficients to improve the recognizability of features. In the dimensionality reduction and classification layers, the approximate empirical kernel map (EKM) is used to train the connection weight matrix between each layer to further improve the network training speed and classification accuracy. To demonstrate the effectiveness of the proposed method, ELM-WLRF is tested on the MNIST, NORB and CIFAR-10 databases. The experimental results show that ELM-WLRF achieves superior classification accuracy, i.e. 99.27%, 98.03% and 60.14% respectively, and requires shorter training time compared with other state-of-the-art ELM-LRF-based algorithms. Keywords Extreme learning machine . Local receptive fields . Coefficient weighting . Image classification
* Yaqian Li [email protected]
1
School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China
2
School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China
Multimedia Tools and Applications
1 Introduction Currently, deep neural network (DNN) algorithms [23, 36, 37] based on backpropagation (BP) are widely used to deal with image classification problem. The iterative training process of these algorithms involves a large number of gradient descent search steps, and encounters difficulties including slower convergence speed, local minimum, and frequent human intervention. In order to avoid these problems and complete training of network quickly and efficiently without iteration, Huang et al. [11, 12] proposed a simple and effective Extreme Learning Machine (ELM). However, many improved algorithms [2, 18, 33, 44] based on ELM can only use the extracted features or vector gener
Data Loading...
