The ensemble deep learning model for novel COVID-19 on CT images.

Novel memristive hyperchaotic system designs and their engineering applications have received considerable critical attention. In this paper, a novel multistable 5D memristive hyperchaotic system and its application are introduced. The interesting aspect of this chaotic system is that it has different types of coexisting attractors, chaos, hyperchaos, periods, and limit cycles. First, a novel 5D memristive hyperchaotic system is proposed by introducing a flux-controlled memristor with quadratic nonlinearity into an existing 4D four-wing chaotic system as a feedback term. Then, the phase portraits, Lyapunov exponential spectrum, bifurcation diagram, and spectral entropy are used to analyze the basic dynamics of the 5D memristive hyperchaotic system. For a specific set of parameters, we find an unusual metastability, which shows the transition from chaotic to periodic (period-2 and period-3) dynamics. Moreover, its circuit implementation is also proposed. By using the chaoticity of the novel hyperchaotic system, we have developed a random number generator (RNG) for practical image encryption applications. Furthermore, security analyses are carried out with the RNG and image encryption designs.

/pdf/chaos-based-application-of-a-novel-multistable-5d-memristive-3hjux0wowk.pdf

Chaos-Based Application of a Novel Multistable 5D Memristive Hyperchaotic System with Coexisting Multiple Attractors

Colonic polyps are more likely to be cancerous, especially those with large diameter, large number and atypical hyperplasia. If colonic polyps cannot be treated in early stage, they are likely to develop into colon cancer. Colonoscopy is easily limited by the operator’s experience, and factors such as inexperience and visual fatigue will directly affect the accuracy of diagnosis. Cooperating with Hunan children’s hospital, we proposed and improved a deep learning approach with global average pooling (GAP) in colonoscopy for assisted diagnosis. Our approach for assisted diagnosis in colonoscopy can prompt endoscopists to pay attention to polyps that may be ignored in real time, improve the detection rate, reduce missed diagnosis, and improve the efficiency of medical diagnosis. We selected colonoscopy images from the gastrointestinal endoscopy room of Hunan children’s hospital to form the colonic polyp datasets. And we applied the image classification method based on Deep Learning to the classification of Colonic Polyps. The classic networks we used are VGGNets and ResNets. By using global average pooling, we proposed the improved approaches: VGGNets-GAP and ResNets-GAP. The accuracies of all models in datasets exceed 98%. The TPR and TNR are above 96 and 98% respectively. In addition, VGGNets-GAP networks not only have high classification accuracies, but also have much fewer parameters than those of VGGNets. The experimental results show that the proposed approach has good effect on the automatic detection of colonic polyps. The innovations of our method are in two aspects: (1) the detection accuracy of colonic polyps has been improved. (2) our approach reduces the memory consumption and makes the model lightweight. Compared with the original VGG networks, the parameters of our VGG19-GAP networks are greatly reduced.

/pdf/an-improved-deep-learning-approach-and-its-applications-on-3htf6m9082.pdf

An improved deep learning approach and its applications on colonic polyp images detection

In order to complete the task of the woodland census in Chenzhou, China, this paper carries out a remote sensing survey on the terrain of this area to produce a data set, and used deep learning methods to label the woodland. There are two main improvements in our paper: Firstly, this paper comparatively analyzes the semantic segmentation effects of different deep learning models on remote sensing image datasets in Chenzhou. Secondly, this paper proposed a dense fully convolutional network (DFCN) which combines dense network with FCNmodel and achieves good semantic segmentation effect. DFCNmethod is used to label the woodland in Gaofen-2 (GF-2) remote sensing images in Chenzhou. Under the same experimental conditions, the labeling results are compared with the original FCN, SegNet, dilated convolutional network, and so on. In these experiments, the global pixel accuracy of DFCN is 91.5%, and the prediction accuracy of the “woodland” class is 93%, both of them perform better than that of the other methods. In other indicators, our method also has better performance. Using the method of this paper, we have completed the land feature labeling of Chenzhou area and provided it to customers.

/pdf/woodland-labeling-in-chenzhou-china-via-deep-learning-4c0hs054cm.pdf

Woodland Labeling in Chenzhou, China, via Deep Learning Approach

Because deep neural networks (DNNs) are both memory-intensive and computation-intensive, they are difficult to apply to embedded systems with limited hardware resources. Therefore, DNN models need to be compressed and accelerated. By applying depthwise separable convolutions, MobileNet can decrease the number of parameters and computational complexity with less loss of classification precision. Based on MobileNet, 3 improved MobileNet models with local receptive field expansion in shallow layers, also called Dilated-MobileNet (Dilated Convolution MobileNet) models, are proposed, in which dilated convolutions are introduced into a specific convolutional layer of the MobileNet model. Without increasing the number of parameters, dilated convolutions are used to increase the receptive field of the convolution filters to obtain better classification accuracy. The experiments were performed on the Caltech-101, Caltech-256, and Tubingen animals with attribute datasets, respectively. The results show that Dilated-MobileNets can obtain up to 2% higher classification accuracy than MobileNet.

/pdf/a-new-image-classification-approach-via-improved-mobilenet-2swb4nxjnj.pdf

A New Image Classification Approach via Improved MobileNet Models with Local Receptive Field Expansion in Shallow Layers

As a lightweight deep neural network, MobileNet has fewer parameters and higher classification accuracy. In order to further reduce the number of network parameters and improve the classification accuracy, dense blocks that are proposed in DenseNets are introduced into MobileNet. In Dense-MobileNet models, convolution layers with the same size of input feature maps in MobileNet models are taken as dense blocks, and dense connections are carried out within the dense blocks. The new network structure can make full use of the output feature maps generated by the previous convolution layers in dense blocks, so as to generate a large number of feature maps with fewer convolution cores and repeatedly use the features. By setting a small growth rate, the network further reduces the parameters and the computation cost. Two Dense-MobileNet models, Dense1-MobileNet and Dense2-MobileNet, are designed. Experiments show that Dense2-MobileNet can achieve higher recognition accuracy than MobileNet, while only with fewer parameters and computation cost.

/pdf/a-novel-image-classification-approach-via-dense-mobilenet-2089bar60z.pdf

A Novel Image Classification Approach via Dense-MobileNet Models

With the presentation of super-resolution convolutional neural network, deep learning approach was applied to image super-resolution reconstruction for the first time. By using convolutional neural network, the deep learning approaches can directly learn the mapping relationship between the low-resolution image and high-resolution image, and have achieved better reconstruction effects than the traditional image super-resolution reconstruction methods. Subsequently, a series of improved deep learning approaches have been proposed, and the reconstruction effects have been improved continuously. This paper systematically summa rizes the image super-resolution reconstruction approaches based on deep learning, analyzes the characteristics of different models, and compares the main deep learning models based on the experiments. Furthermore, based on deep learning model, the future research directions of the image super-resolution reconstruction methods based on deep learning models are reasonably predicted.

Yanhong Luo

Papers

A Novel Image Classification Approach via Dense-MobileNet Models

An improved deep learning approach and its applications on colonic polyp images detection

Brief Survey of Single Image Super-Resolution Reconstruction Based on Deep Learning Approaches