Abstract:Deep Neural Networks (DNNs) are commonly used methods in computational intelligence. Most prevalent DNN-based image classification methods are dedicated to promoting the performance by designing complicated network architectures and requiring large amounts of model parameters. These large-scale DNN-based models are performed on all images consistently. However, since there are meaningful differences between images, it is difficult to accurately classify all images by a consistent network architecture. For example, a deeper network is fit for the images that are difficult to be distinguished, but may lead to model overfitting for simple images. Therefore, we should selectively use different models to deal with different images, which is similar to the human cognition mechanism, in which different levels of neurons are activated according to the difficulty of object recognition. To this end, we propose a Hierarchical Convolutional Neural Network (HCNN) for image classification in this paper. HCNNs comprise multiple sub-networks, which can be viewed as different levels of neurons in humans, and these sub-networks are used to classify the images progressively. Specifically, we first initialize the weight of each image and each image category, and these images and initial weights are used for training the first sub-network. Then, according to the predicted results of the first sub-network, the weights of misclassified images are increased, while the weights of correctly classified images are decreased. Furthermore, the images with the updated weights are used for training the next sub-networks. Similar operations are performed on all sub-networks. In the test stage, each image passes through the sub-networks in turn. If the prediction confidences in a sub-network are higher than a given threshold, then the results are output directly. Otherwise, deeper visual features need to be learned successively by the subsequent sub-networks until a reliable image classification result is obtained or the last sub-network is reached. Experimental results show that HCNNs can obtain better results than classical CNNs and the existing models based on ensemble learning. HCNNs have 2.68% higher accuracy than Residual Network 50 (Resnet50) on the ultrasonic image dataset, 1.19% than Resnet50 on the chimpanzee facial image dataset, and 10.86% than Adaboost-CNN on the CIFAR-10 dataset. Furthermore, the HCNN is extensible, since the types of sub-networks and their combinations can be dynamically adjusted.

深度学习经典论文翻译专栏收录该内容

ImageNet classification with deep convolutional neural networks

Deep Learning Model of Image Classification Using Machine Learning

A Novel Deep Learning Network Architecture with Cross-Layer Neurons

Research on image classification leveraging deep convolutional neural networks and visual cognition

Going deeper with convolutions

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

Deep Residual Learning 3 . 1 . Residual Learning

Bio-Inspired Deep Spiking Neural Network For Image Classification

Performance Evaluation of Deep Learning Classification Network for Image Features

Very Deep Convolutional Networks for Large-Scale Image Recognition

Multi-column Deep Neural Networks for Image Classification

DenseUNet: Improved Image Classification Method Using Standard Convolution and Dense Transposed Convolution

Image retrieval and classification on deep convolutional SparkNet

Hyper-sausage coverage function neuron model and learning algorithm for image classification

Deep Learning Using Isotroping, Laplacing, Eigenvalues Interpolative Binding, and Convolved Determinants with Normed Mapping for Large-Scale Image Retrieval

Error-Driven Incremental Learning in Deep Convolutional Neural Network for Large-Scale Image Classification

Two-Stage Model Compression and Acceleration: Optimal Student Network for Better Performance

A Novel Progressive Image Classification Method Based on Hierarchical Convolutional Neural Networks

Sparsing Deep Neural Network Using Semi-Discrete Matrix Decomposition

Return of the Devil in the Details: Delving Deep into Convolutional Nets

深度学习经典论文翻译 专栏收录该内容