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.

Associative Memory Optimized Method on Deep Neural Networks for Image Classification.

Visual Image Classification Based on Auto-Associative Memory

Biological lateral inhibition and digital cellular neural network applied on associations memory

Memory transformation networks for weakly supervised visual classification

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

A Memory Model for Visual Image Classification Based on Residual Neural Network and Bayesian Decision

A New Pattern Associative Memory Model for Image Recognition Based on Hebb Rules and Dot Product

Research on Optimization of Image Recognition Algorithm Based on Deep Learning

Deep Associative Neural Network for Associative Memory Based on Unsupervised Representation Learning

A Visual Memory Information Interaction Mechanism Inspired Graph Neural Network for Highly Imbalanced Image Classification

A memory model for image recognition and classification based on convolutional neural network and Bayesian decision

Optimized Visual Recognition Through a Deep Convolutional Neural Network With Hierarchical Modular Organization

An Image Classification Method Based on Adaptive Attention Mechanism and Feature Extraction Network

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

Optimizing Image Classification: Automated Deep Learning Architecture Crafting with Network and Learning Hyperparameter Tuning

Deep Learning Model of Image Classification Using Machine Learning

Semantic Memory Neural Network Model Based on Two-Dimension Coding Method

Attention-Augmented Memory Network for Image Multi-Label Classification

Memory Matching Networks for One-Shot Image Recognition

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

Improving Image Similarity Learning by Adding External Memory