Showing papers in "CAAI Transactions on Intelligence Technology in 2022"

TL;DR: In this article , a counting-based secret sharing method was proposed to enhance image authentication by adjusting embedding the watermarking data over the images by innovative redistribution of shares to be embedded spread over all the images.

TL;DR: In this article , a method called CBIR-similarity measure via artificial neural network interpolation (CBIR-SMANN) has been presented, which is based on Skewness, mean, kurtosis and standard deviation features were extracted then given to ANN for interpolation.

TL;DR: A comprehensive overview of the forecasting models based on deep learning in the field of wind energy can be found in this paper , which includes time-series-based recurrent neural networks, restricted Boltzmann machines, convolutional neural networks as well as auto-encoder-based approaches.

TL;DR: In this paper , a robust deformed denoising CNN (RDDCNN) is proposed, which contains three blocks: a deformable block (DB), an enhanced block (EB), and a residual block (RB).

TL;DR: In this article , a new prediction learning model is proposed to improve the dynamic performance of the alpha-beta filter algorithm, which has been widely researched for various applications, for example, navigation and target tracking systems.

TL;DR: In this paper , the authors compared and contrasted the aims of economic dispatch, emission dispatch, fractional programing based combined economic emission dispatch and environmental restricted economic dispatch (ECED) for a lowvoltage microgrid system.

TL;DR: A systematic literature review as discussed by the authors summarizes and synthesises the major themes of CQA research related to questions, answers, and users, and concludes that the use of deep learning in CQAs is on an upward trajectory.

TL;DR: This study discusses current wireless network research, brief discussions on ML methods that can be effectively applied to the wireless networking domain, some tools available to support and customise efficient mobile system design, and some unresolved issues for future research directions.

TL;DR: In this article , a local binary pattern-based reversible data hiding (LBP-RDH) technique has been proposed to maintain a fair symmetry between the perceptual transparency and hiding capacity.

TL;DR: In this article , a parameter optimization strategy of control systems for uncertain closed-loop wireless power transfer (WPT) systems using the modified genetic algorithm (MGA) is proposed, and three cost functions are designed according to the characteristics of the three stages, and then the optimal controller parameters of each stage are obtained by using MGA.

TL;DR: In this article , the authors proposed a model using one-dimensional CNN architecture for anomaly detection in cyber-security, which divides network traffic data into transmission control protocol (TCP), user datagram protocol (UDP), and other protocol categories in the first phase, then each category is treated independently.

TL;DR: In this article , the authors systematically and precisely outline the most prominent and up-to-date research of automatic depression recognition by intelligent speech signal processing so far, including methods for acoustic feature extraction, algorithms for classification and regression, as well as end to end deep models.

TL;DR: In this article , a light-weight scale-adaptive fitness evaluation-based particle swarm optimisation (SAFE-PSO) approach is proposed to solve the hyperparameter and architecture optimization problem.

TL;DR: In this article , the authors proposed a robust control and tracking for Hamiltonian systems with unknown perturbations by using the operator-based robust right coprime factorization method.

TL;DR: In this article , a Python-based algorithm is proposed to find the parameters of each camera, rectify the images, create the disparity maps and finally use these maps for distance measurements.

TL;DR: In this article , a humanoid sliding mode neural network controller based on the human gait is proposed to improve the rehabilitation training effect of lower limb exoskeleton robots for patients with lower extremity dysfunction.

TL;DR: Li et al. as mentioned in this paper proposed a wafer data augmentation method based on a convolutional autoencoder by adding random Gaussian noise to the hidden layer, which has an average accuracy of 95.4% on the WM-811k wafer dataset with only 173.643 KB of the parameters and 316.194 M of FLOPs, and takes only 22.99 s to detect 1000 wafer pictures.

TL;DR: Wang et al. as mentioned in this paper proposed a recursive recurrent network model to solve the kinematic control issue for manipulators with different levels of physical constraints, which can be formulated as a new manifold system to ensure control solution within all of the joint constraints in different orders.

TL;DR: In this paper , the importance of certain features varies depending on specific tasks, that is, specific tasks exhibit feature bias, and they designed two classification tasks based on human intuition to train deep neural models to identify the anticipated biases.

TL;DR: A comprehensive analysis of machine learning approaches in the field of diagnosing COVID-19 has been conducted, and for the detection of chronic diseases in patients, to identify symptoms of CO VID-19 virus infection in advance, and control the situation a healthcare system has been proposed.

TL;DR: In this paper , an early Japanese book reorganization system was proposed by combining image processing and deep learning techniques, which achieved an accuracy of 79.8, 80.3, and 80.0%.

TL;DR: In this article , a memory-augmented autoencoder for hyperspectral anomaly detection (MAENet) is proposed to address the problem of the abnormal samples are usually reconstructed well along with the normal background samples.

TL;DR: In this paper , a new combination of multi-gradient-direction and deep convolutional neural networks for arecanut disease identification was presented, namely, rot, split and rot-split.

TL;DR: Based on the predigestion of the dynamic model of the intelligent firefighting vehicle, a linear 2-DOF lateral dynamic model and a preview error model are established in this paper to solve the problems of a highly nonlinear vehicle model, time-varying parameters, output chattering, and poor robustness.

TL;DR: In this paper , the authors proposed Histogram Augmentation Technique (HAT) to generate a dataset whose distribution is similar to that of the original dataset, where separate algorithms are designed for continuous and discrete columns.

TL;DR: Wang et al. as mentioned in this paper designed a fully automatic fracture detection model based on a deep convolution neural network (CNN), which used cascade R-CNN, attention mechanism, and atrous convolution to optimise the detection of small fractures in a large X-ray image with big local variations.

TL;DR: Wang et al. as mentioned in this paper proposed SpikeGoogle, which is implemented with GoogLeNet-like inception module, different convolution kernels and max-pooling layer are included to capture deep features across diverse scales.

TL;DR: Wang et al. as discussed by the authors proposed a medical data publishing method based on sensitivity determination to seek the trade-off between information utility and privacy security, which focusses on the protection of sensitive values with high sensitivity and assigns the highly sensitive disease information to groups as evenly as possible.

TL;DR: Zhang et al. as discussed by the authors proposed a multiscale attention encoder to capture the multi-scale contextual information of the aerial/street-view images to bridge the domain gap between these two view images.

TL;DR: Zhang et al. as discussed by the authors proposed an attentive pyramid spatial attention (APSA) module, which can increase the receptive field and enhance the information, and added the context fusion prediction branch that fuses high-semantic and lowsemantic prediction results, which effectively improves the segmentation accuracy of small data sets.