Abstract:The deep convolutional neural networks (DCNN) require large number of training data to avoid overfitting, which makes it unsuitable for processing small-scale image datasets. The transfer learning using DCNN (TCNN) reuses pre-trained layers to generate a mid-level image representation so that the optimization of more than millions CNN parameters can be avoided. By this way, overfitting problem in small-scale data can be alleviated. However, although now many public DCNNs have been trained and can be reused, the existing TCNNs are formed by only a single pre-trained DCNN structure and cannot make full use of multiple structures of pre-trained DCNNs. At the same time, the existing ensemble CNNs have not enough good representation ability. To address this problem, we combine the conventional ideas of ensemble CNNs and propose three ensemble TCNNs (TECNN). They are the voting method based on the combination of all TCNNs, the PickOver method by finding the optimal combination, and weighted method by finding weighted combination. Different from the existing ensemble CNNs, the proposed methods do not need to retrain the component CNNs and generate ensemble transferring representations by transferring the pre-trained mid-level parameters. The mathematical models of those three methods are also provided. Their versions of using fine-tuning are also compared in the experiments. In addition, we replace the Softmax classifier with ensemble linear classifiers in the full-connection layer. They outperform the current state of the art algorithms on Caltech ImageNet and some internet image data. All this research has released as an open source library called Transferring Image Ensemble Representations using Deep Convolutional Neural Networks (TECNN). The source codes and relevant datasets in different versions are available from: http://www.cquptshuyinxia.com/TECNN.html.

Image Classification Method Based on Improved DCNN Combined Transfer Learning

Research on Image Classification Method of Features of Combinatorial Convolution

Noisy Softmax: Improving the Generalization Ability of DCNN Via Postponing the Early Softmax Saturation

Training DCNN by Combining Max-Margin, Max-Correlation Objectives, and Correntropy Loss for Multilabel Image Classification.

Fine-Grained Image Classification Using Modified DCNNs Trained by Cascaded Softmax and Generalized Large-Margin Losses

An Improved Remote Sensing Image Classification Method Based on DCNN

Applying Improved Convolutional Neural Network in Image Classification

Fast image classification algorithm based on deep convolutional neural network

CDTNet: Improved Image Classification Method Using Standard, Dilated and Transposed Convolutions

Deep convolution neural network learning based on deconvolution feature extraction

An Improved Convolutional Neural Network Algorithm and Its Application in Multilabel Image Labeling

Natural Image Classification Based on Multi-Parameter Optimization in Deep Convolutional Neural Network

Transferring Ensemble Representations Using Deep Convolutional Neural Networks for Small-Scale Image Classification.

Midcn: A Multiple Instance Deep Convolutional Network For Image Classification

A new image classification method using CNN transfer learning and web data augmentation.

A Study on CNN Transfer Learning for Image Classification

Hierarchical Transfer Convolutional Neural Networks for Image Classification

Coarse-to-fine Trained Multi-Scale Convolutional Neural Networks for Image Classification.

3D Face Recognition Method Based on Deep Convolutional Neural Network

Improvement of the AlexNet Networks for Large-Scale Recognition Applications

Transductive Semi-Supervised Deep Learning Using Min-Max Features