Search

COVID 19 pandemic is causing a global health epidemic. The most powerful safety tool is wearing a face mask in public places and everywhere else. The COVID 19 outbreak forced governments around the world to implement lockdowns to deter virus transmission. According to survey reports, wearing a face mask at public places reduces the risk of transmission significantly. In this paper, an IoT-enabled smart door that uses a machine learning model for monitoring body temperature and face mask detection. The proposed model can be used for any shopping mall, hotel, apartment entrance, etc. As an outcome a cost-effective and reliable method of using AI and sensors to build a healthy environment. Evaluation of the proposed framework is done by the Face Mask Detection algorithm using the TensorFlow software library. Besides, the body temperature of the individual is monitored using a non-contact temperature sensor. This proposed system can detect the users from COVID 19 by enabling the Internet of Things (IoT) technology.

The work concerns the monitoring of the edge condition based on acoustic emission (AE) signals. The tool edge condition was determined by the wear width on the flank face. The processed material was an aluminum-ceramic composite containing 10% SiC. A carbide milling cutter with a diamond coating was used as the tool. Based on the AE signals, appropriate measures were developed that were correlated with the edge condition. Machine learning methods were used to assess the milling cutter's degree of wear based on AE signals. The applied approach using a decision tree allowed the prediction error of the tool condition class with a value below 6%. The method was also compared with other machine learning methods such as neural networks and the k-nearest neighbor algorithm.

The intersection line information of the point cloud between the coal wall and the roof can not only accurately reflect the direction information of the scraper conveyor but also provide a preliminary basis for realizing the intelligent coal mine. However, the indirect method of using deep learning to segment the point cloud of coal mine working face cannot make full use of the rich information provided by the point cloud data. The direct method of using deep learning to segment the point cloud ignores the local feature relationship between points. Therefore, we propose to use dynamic graph convolution neural networks (DGCNNs) to segment the point cloud of the coal wall and roof so as to obtain the intersection line between them. First, in view of the characteristics of heavy dust and strong electromagnetic interference in the environment of the coal mine working face, we have installed an underground inspection robot so that we use light detection and ranging to obtain the point cloud of the coal mine working face. At the same time, we put forward a fast labeling method of the point cloud of the coal mine working face and an efficient training method of the depth neural network. Second, on the basis of edge convolution, being the greatest innovation of DGCNNs, we analyze the influence of the number of layers, K value, and output feature dimension of edge convolution on the effect of DGCNNs segmenting the point cloud of the coal mine working face and obtaining the intersection line of the coal wall and roof. Finally, we compare DGCNNs with PointNet and PointNet++. The results show that the DGCNN exhibits the best performance. What is more, the results provide a research foundation for the application of DGCNNs in the field of energy. Last but not least, the research results provide a direct and key basis for the adjustment of the scraper conveyor, which is of great significance for an intelligent coal mine working face and accurate construction of a geological information model.

The zero-shot semantic segmentation requires models with a strong image understanding ability. The majority of current solutions are based on direct mapping or generation. These schemes are effective in dealing with the zero-shot recognition, but they cannot fully transfer the visual dependence between objects in more complex scenarios of semantic segmentation. More importantly, the predicted results become seriously biased to the seen-category in the training set, which makes it difficult to accurately recognize the unseen-category. In view of the above two problems, we propose a novel zero-shot semantic segmentation model based on meta-learning. It is observed that the pure semantic space expression has certain limitations for the zero-shot learning. Therefore, based on the original semantic migration, we first migrate the shared information in the visual space by adding a context-module, and then migrate it in the visual and semantic dual space. At the same time, in order to solve the problem of biasness, we improve the adaptability of the model parameters by adjusting the parameters of the dual-space through the meta-learning, so that it can successfully complete the segmentation even in the face of new categories without reference samples. Experiments show that our algorithm outperforms the existing best methods in the zero-shot segmentation on three datasets of Pascal-VOC 2012, Pascal-Context and Coco-stuff.

Over the past decades, three-dimensional(3D) face recognition has developed rapidly due to its intrinsic invariance to pose and illumination changes. Yet despite this, deep learning is rarely used in 3D face recognition due to lack of training dat a. Noting that the majority of the recognition processes are based on facial depth map instead of real 3D data, we focus on the generation of facial depth map to surmount the shortage of training data. In this paper, we present a novel paradigm, Facial Depth Descend, which generates facial depth map from existing 3D face data. The key to this generation paradigm is to recombine facial components from existing faces and thereafter generate brand new faces based on them. We then propose a learning framework RPC to generate recognition-friendly faces. First, it extracts three facial components (eyes, nose and mouth) from the 3D data of real faces. Then, with these components as input, a Relative Location Estimator (RLE) is used to predict the relative location among them so that they can be composed in a reasonable manner. Finally, the method feeds a Pix2Pix network with these composed facial components to extrapolate areas surrounding them and output a full facial depth map. We here enforce an identity preserving loss on the generation network to make the facial depth map more discriminative and favorable for recognition tasks. Since we can choose different identities’ different components as input, RPC is theoretically able to generate a vast number of novel 2.5D faces. More specifically, RPC can generate an enlarged face dataset as large as N 3 identities, where N is the number of identities in the original one. Experiments show that, adding the generated depth maps to the training dataset can help improve the recognition rate in 2.5D face recognition.

Face images, as an information carrier, are rich in sensitive information. Direct publication of these images would cause privacy leak, due to their natural weak privacy. Most of the existing privacy protection methods for face images adopt data publication under a non-interactive framework. However, the E-effect under this framework covers the entire image, such that the noise influence is uniform across the image. To solve the problem, this paper proposes region growing publication (RGP), an algorithm for the interactive publication of face images under differential privacy. This innovative algorithm combines the region growing technique with differential privacy technique. The privacy budget E is dynamically allocated, and the Laplace noise is added, according to the similarity between adjacent sub-images. To measure this similarity more effectively, the fusion similarity measurement mechanism (FSMM) was designed, which better adapts to the intrinsic attributes of images. Different from traditional region growing rules, the FSMM fully considers various attributes of images, including brightness, contrast, structure, color, texture, and spatial distribution. To further enhance algorithm feasibility, RGP was extended to atypical region growing publication (ARGP). While RGP limits the region growing direction between adjacent sub-images, ARGP searches for the qualified sub-images across the image, with the aid of the exponential mechanism, thereby expanding the region merging scope of the seed point. The results show that our algorithm can satisfy E-differential privacy, and the denoised image still have a high availability.

Our country has a vast territory and rich resources, but it is a country with more coal, less oil, and poor gas. With the increase of our population, the development of society, and the more severe international situation, coal has become more important for our country’s economic development and energy. Security plays an irreplaceable role. Based on the neural network, this paper studies and controls the underground pressure law of the coal mine’s soft rock heading face, aiming at the safe and efficient mining of the first face and providing an experience for the next face. This paper mainly uses BP neural network learning algorithm and support pressure algorithm to measure and study the ground pressure law of coal mine soft rock heading face and establishes the ground pressure online monitoring system, which is used to analyze and summarize the ground pressure abnormal area during the mining of the working face, so as to provide the basis for safe mining of the working face. Through the field measured data, the initial pressure step and periodic pressure step at the upper, middle, and lower parts of the working face, the average working resistance of the support at the working face during pressure, and the dynamic load coefficient of the support are obtained. It is analyzed that the support in the middle of the working face has a large load and the pressure is obvious. The experimental results show that the initial support force of the whole working face is approximately normally distributed, the proportion of the initial support force in the range of 10–30 MPa accounts for more than 85% of the total statistics, and the frequency of the initial support force in the upper, middle, and lower stations at 10–25 MPa is 55%–65%.

Clustering of face images is a well-known clustering problem with several applications as for example the automated face tagging in photo albums. The performance of face clustering algorithms largely depends on the face representations that are used. Therefore, the exploitation of the internal face representation (embedding) obtained by feeding the face image to a deep face network has led to increased clustering performance. In this work, we perform face clustering by combining multiple representations of each image obtained from several deep face networks. The multiple deep representations are not exploited in a straightforward manner (simple vector concatenation). Instead a weight is assigned to each representation to reflect its importance on the clustering result and a weighted clustering algorithm is employed to automatically adjust the weights and, consequently, the contribution of each representation on the clustering solution. It should also be noted that in the proposed approach, the number of clusters (number of faces) is not given as input, but it is automatically estimated using the silhouette criterion. The conducted experiments have shown that the weighted combination of deep face representations leads to improved clustering performance compared to using single representations.

The spur face gear is a novel drive that is widely applied in national defense. To improve the production efficiency and meshing properties of spur face gears, a hobbing method for spur face gears with a bidirectional modification is proposed. To obtain the tooth surface of the gear by hobbing, the basic worm of the spherical hob and the tooth surface equation by hobbing were established. To obtain a convex surface, modification along the contact trace and contact line was proposed. Then, a hobbing method for face gears with bidirectional modification was studied based on current machine tools. Therefore, a prototype was produced, which illustrates the feasibility of manufacturing the design. Detection methods revealed that the maximum error of an unmodified tooth surface is −11.2 μm, and the maximum deviation errors along the contact trace and contact line are −2.7 μm and −2.1 μm, respectively. The results show that hobbing is an effective method for bidirectional modification of gear teeth.

Facial Expression Recognition (FER) is a task usually framed as predicting an emotional state given a facial image. FER has received numerous attentions from both business and academia as it could serve as a crucial component in various applications, e.g., automatic evaluation of customer satisfaction, sign language recognition, and human–computer interaction. Despite active research on the subject, facial expression recognition remains greatly challenging due to the diversity of individual expressions, shapes, and sizes of face, eyes, mouth, and other facial features, as well as orientation, alignment, and lighting. In this paper, we aim to improve the fast and robust online-learning FER with unseen data identification. We compare two widely used feature extraction methods for FER, namely Curvelet Transform (CT) and Local Curvelet Transform (LCT). Furthermore, we explore factors underlying several online extreme learning approaches for unseen data identification. Our experimental results demonstrate that (1) CT is suitable for a cleaner and well-prepared dataset, while LCT seems to work well on a dataset with diverse quality and on level of consistency. (2) The Identity Structural Tolerance Sequential Circular Extreme Learning Machine outperforms other Extreme Learning algorithms employed in FER. (3) LC can provide unseen identification capability to Extreme Learning algorithms. These findings emphasize the common underlying foundation between the extreme learning approach and other traditional learning approaches.

Recent works based on deep learning and facial priors have performed well in superresolving severely degraded facial images. However, due to the limitation of illumination, pixels of the monitoring probe itself, focusing area, and human motion, the face image is usually blurred or even deformed. To address this problem, we properly propose Face Restoration Generative Adversarial Networks to improve the resolution and restore the details of the blurred face. They include the Head Pose Estimation Network, Postural Transformer Network, and Face Generative Adversarial Networks. In this paper, we employ the following: (i) Swish-B activation function that is used in Face Generative Adversarial Networks to accelerate the convergence speed of the cross-entropy cost function, (ii) a special prejudgment monitor that is added to improve the accuracy of the discriminant, and (iii) the modified Postural Transformer Network that is used with 3D face reconstruction network to correct faces at different expression pose angles. Our method improves the resolution of face image and performs well in image restoration. We demonstrate how our method can produce high-quality faces, and it is superior to the most advanced methods on the reconstruction task of blind faces for in-the-wild images; especially, our 8 × SR SSIM and PSNR are, respectively, 0.078 and 1.16 higher than FSRNet in AFLW.

Currently, methods for recognizing objects in images work poorly and use intellectually unsatisfactory methods. The existing identification systems and methods do not completely solve the problem of identification, namely, identification in difficult conditions: interference, lighting, various changes on the face, etc. To solve these problems, a local detector for a reprint model of an object in an image was developed and described. A transforming autocoder (TA), a model of a neural network, was developed for the local detector. This neural network model is a subspecies of the general class of neural networks of reduced dimension. The local detector is able, in addition to determining the modified object, to determine the original shape of the object as well. A special feature of TA is the representation of image sections in a compact form and the evaluation of the parameters of the affine transformation. The transforming autocoder is a heterogeneous network (HS) consisting of a set of networks of smaller dimension. These networks are called capsules. Artificial neural networks should use local capsules that perform some rather complex internal calculations on their inputs, and then encapsulate the results of these calculations in a small vector of highly informative outputs. Each capsule learns to recognize an implicitly defined visual object in a limited area of viewing conditions and deformations. It outputs both the probability that the object is present in its limited area and a set of “instance parameters” that can include the exact pose, lighting, and deformation of the visual object relative to an implicitly defined canonical version of this object. The main advantage of capsules that output instance parameters is a simple way to recognize entire objects by recognizing their parts. The capsule can learn to display the pose of its visual object in a vector that is linearly related to the “natural” representations of the pose that are used in computer graphics. There is a simple and highly selective test for whether visual objects represented by two active capsules A and B have the correct spatial relationships for activating a higher-level capsule C. The transforming autoencoder solves the problem of identifying facial images in conditions of interference (noise), changes in illumination and angle.

O uso de sistemas biométricos de reconhecimento facial automatizado na Segurança Pública requer grande cuidado, considerando que estamos tratando da possibilidade de privação de liberdade. Dessa forma, é de extrema importância que o processo de identificação seja baseado em métodos técnicos científicos, respeitando os princípios de unicidade, imutabilidade, perenidade, praticabilidade e classificabilidade. O objetivo do presente estudo é revisar a literatura sobre a aplicação dos sistemas biométricos de reconhecimento facial no processo de identificação humana por meio de imagem e sua aplicação na Segurança Pública. Foi feita uma revisão de literatura narrativa, bem como a descrição e uma demonstração da técnica do exame pericial de comparação facial visando identificação humana. Os estudos revelaram que, por mais que se tenha avançado na tecnologia de reconhecimento facial automatizado, diversos fatores são sensíveis e influenciam a acuracidade dos seus resultados em cenários práticos, sendo indicada a verificação humana. Em relação ao seu uso na segurança pública, são relatadas experiências de sucesso e de equívocos, tendo-se observado o seu banimento em vários locais em função da falta de acurácia, intromissão de privacidade e tendências discriminatórias. Deste modo, a melhor solução para o uso em Segurança Pública, dado o estado da arte da tecnologia de reconhecimento facial utilizada, aponta para uma convergência de dois elementos fundamentais nesse processo: máquina e humano. Por esta razão, este estudo indica o uso dos sistemas biométricos de reconhecimento facial como etapa prévia à identificação facial por imagem realizada por meio de técnica pericial.

Unconstrained face alignment is still a challenging problem due to the large poses, partial occlusions and complicated illuminations. To address these issues, in this paper, we propose a mixed attention hourglass network (MAttHG) to learn more discriminative representations by modeling the correlated relationships between features. Specifically, by integrating the attention module from features of different levels in the stacked hourglass networks, MAttHG can capture rich contextual correlations, which can be further used to combine local features to better model the spatial position relationship of facial landmarks. Furthermore, by combining the hourglass network and the attention module, MAttHG can effectively model the global attention and local attention to enhance the facial shape constraints for robust face alignment. Moreover, a head pose prediction module is designed to adaptively adjust the weight of each sample in the training set and redefine the loss function for addressing the problem of data imbalance. Experimental results on challenging benchmark datasets demonstrate the superiority of our MAttHG over state-of-the-art face alignment methods.

Hyperspectral face recognition plays an important role in remote sensing. However, it faces many challenges such as difficulty in data acquisition, low signal to noise ratio (SNR), and high dimensionality. In this paper, we develop a novel method for hyperspectral face recognition by extracting histogram of oriented features (HOG) and using collaborate representation-based classifier (CRC) to classify unknown face data cubes. To improve overall classification rates, we also implement noise reduction in hyperspectral face data cubes. We also crop the face images by a bounding box and use this bounding box image to classify the testing faces. Experiments show that our new method outperforms several existing methods for both the PolyU-HSFD dataset and the CMU-HSFD dataset for hyperspectral face recognition. The contribution of this paper is the following: We introduce the MNF-based denoising method in this paper, which is new to the best of our knowledge. We also combine it with HOG features and CRC classifier so that better recognition rate can be achieved.

Geo-spatial object detection in high-resolution satellite images has many applications in urban planning, military applications, maritime surveillance, environment control and management. Despite the success of convolutional neural networks in object detection tasks in natural images, the current deep learning models face challenges in geo-spatial object detection in satellite images due to complex background, arbitrary views and large variations in object sizes. In this paper, we propose a framework that tackles these problems in efficient and effective way. The framework consists of two stages. The first stage generates multi-scale object proposals and the second stage classifies each proposal into different classes. The first stage utilizes feature pyramid network to obtain multi-scale feature maps and then convert each level of the pyramid into an independent multi-scale proposal generator by appending multiple region proposal networks (RPNs). We define scale range for each RPN to capture different scales of the target. The multi-scale object proposals are provided as input to the detection sub-network. We evaluate proposed framework on publicly available benchmark dataset, and from the experiment results, we demonstrate that proposed framework outperformed other reference methods

Recently, deep convolution neural networks (DCNNs) have obtained remarkable performance in exploring single image super-resolution (SISR). However, most of the existing CNN-based SISR methods only focus on increasing the width and depth of the network to improve SR performance, which makes them face a heavy computing burden. In this paper, we propose a lightweight dynamic dual attention iteration network (DDAIN) for SISR. Specifically, to better realize the attention of the channel and the convolution kernel, we design a dynamic convolution unit (DYCU) at the head of the network. It improves the SR performance by enhancing the complexity of the model without increasing the width and depth of the network. Compared with the traditional static convolution, it can extract more abundant high and low-frequency image features according to different input images. Moreover, to recover the high-frequency detail features of images with different resolutions as much as possible, we embed multiple dual residual attention (DRA) in the feature refinement unit (FRU). Finally, to alleviate the height discomfort caused by SR, we introduce iterative loss Liter to optimize the training process further. Extensive experimental results on benchmark show that the performance of the DDAIN in different degradation models exceeds some existing classical methods.

Landry, Minsky and Taylor (LMT) introduced two polynomial invariants of veering triangulations—the taut polynomial and the veering polynomial. Here, we consider a pair of taut polynomials associated to one veering triangulation, the upper and the lower one, and analogously the upper and lower veering polynomials. We prove that the upper and lower taut polynomials are equal. In contrast, the upper and lower veering polynomials of the same veering triangulation may differ by more than a unit. We give algorithms to compute all these invariants. LMT related the Teichmüller polynomial of a fibered face of the Thurston norm ball with the taut polynomial of the associated layered veering triangulation. We use this result to give an algorithm to compute the Teichmüller polynomial of any fibered face of the Thurston norm ball.\ud \ud

Law offenders take advantage of face masks to conceal their identities and in the present time of the COVID-19 pandemic wearing face masks is a new norm which makes it a daunting task for the investigation agencies to identify the offenders. To address the issue of detection of people wearing face masks using surveillance cameras, we propose a novel face mask vision system that is based on an improved tiny YOLO v4 object detector. The face masks detection network of the proposed vision system is developed by integrating tiny YOLO v4 with spatial pyramid pooling (SPP) module and additional YOLO detection layer and tested and validated on a self-created face masks detection dataset consisting of more than 50,000 images. The proposed tiny YOLO v4-SPP network achieved a mAP (mean average precision) value of 64.31% on the employed dataset which was 6.6% higher than tiny YOLO v4. Specifically, for detection of the presence of a small object like a face mask on the face region, the proposed tiny YOLO v4-SPP based vision system achieved an AP (average precision) of 84.42% which was 14.05% higher than the original tiny YOLO v4 thus, ensuring that the proposed network is capable of accurate detection of a mask on the face region in real-time surveillance applications where visibility of complete face area is a guideline.

Low-resolution face images can be found in many practical applications. For example, faces captured from surveillance videos are typically in small sizes. Existing face recognition deep networks, trained on high-resolution images, perform poorly in recognizing low-resolution faces. In this work, an improved multi-branch network is proposed by combining ResNet and feature super-resolution modules. ResNet is for recognizing high-resolution facial images and extracting features from both high- and low-resolution images. Feature super-resolution modules are inserted before the classifier of ResNet for low-resolution facial images. They are used to increase feature resolution. The proposed method is effective and simple. Experimental results show that the recognition accuracy for high-resolution face images is high, and the recognition accuracy for low-resolution face images is improved.

Most traffic accidents are caused by human error, i.e. drowsiness. A drowsiness detection system is then developed to respond to this situation. In this work, the drowsiness detection system is built through the OpenCV library by combining the Haar Cascade Classifier algorithm with Blur, Canny, and Contour function. Haar Cascade Classifier was used to detect areas of face and eyes whereas the combination of Blur, Canny, and Contour functions are used to detect the driver's eyes and analyze the opening or closing of the driver's eyes. The performance of the drowsiness detection system was tested through four variables; kernel size, threshold value, lighting condition (morning, noon, afternoon, and night), and eye's characteristic (eyeglasses or not). Based on the experiments, the best kernel size to detect the driver's eyes is 4,4. Then, the best lower threshold and upper thresholds are 70–110 and 210–240. Subsequently, the light conditions have a 20% error rate to the system. The eye's characteristic has a 16,7% error rate to the system.

The use of face masks in the current era is one of the special regulations in many countries including Indonesia to prevent the spread of coronavirus. However, not all people strongly agree to wear masks because they feel uncomfortable to wear even in crowded places require the use of masks such as shopping malls, hospitals, factories, stations and others by checking manually. Therefore, in the study proposed automatic detection of masks with YOLOv4 with the stage of data collection recording community activities in crowded places, labeling images of masks and non masks. The labelling results were conducted in training that resulted in 90.3% accuracy in the 2000 ierasi, the last of which was video testing in three different crowd locations: taxes, city parks and highways. Based on the test results, YOLOv4 can detect masks and non masks on videos with different obstruction conditions such as people wearing helmets, hand obstacles. However, for the detection of people with tissue obstruction conditions and improper position of wearing masks has not resulted in good detection.

The development of web cameras and smart phones is mature, and more and more facial recognition-related applications are implemented on embedded systems. The demand for real-time face recognition on embedded systems is also increasing. In order to improve the accuracy of face recognition, most of the modern face recognition systems consist of multiple deep neural network models for recognition. However, in an embedded system, integrating these complex neural network models and execute simultaneously is not easy to achieve the goal of real-time recognition of human faces and their identities. In view of this, this study proposes a new frame analysis mechanism, continuous frames skipping mechanism (CFSM), which can analyze the frame in real time to determine whether it is necessary to perform face recognition on the current frame. Through the analysis of CFSM, the frames that do not need to be re-recognized for face are omitted. In this way, the workload of the face recognition system will be greatly reduced to achieve the goal of real-time face recognition in the embedded system. The experimental results show that the proposed CFSM mechanism can greatly increase the speed of face recognition in the video on the embedded system, achieving the goal of real-time face recognition.

The face gear power-split system has huge superiorities over the traditional transmission form in the application of modern rotorcraft, and it has become the research trend of the industry in recent years. Thus this paper took the double input face gear split-parallel transmission system used in the rotorcraft as the research target, and established its dynamics model through the lumped parameter theory. Based on the Newtonian second law, the dynamics equations were built and solved to gain the meshing forces and load sharing coefficients of the transmission system. Simultaneously, the impacts of the eccentric errors, support stiffness, and torsional stiffness on the load sharing characteristics were studied. The results show that the meshing forces and load sharing coefficients of each gear pair have periodic changes; the eccentric errors of each drive stage gear have only a significant effect on the corresponding drive stage. Moreover, the changes in the support stiffness of the split-torque shafts and double gear shafts mainly affect the load distribution of the parallel stage, and the shaft torsional stiffness is less sensitively to maintain load balance. In addition, the increment of the shaft stiffness increases the load sharing coefficients of the corresponding gear pairs.

O presente trabalho trata-se de uma revisão narrativa de literatura com o objetivo descrever as principais indicações e técnicas utilizadas na cirurgia de reconstrução da face em tumores avançados. As neoplasias de cabeça e pescoço como, por exemplo, os variados tipos de câncer de pele, afetam uma região de grande complexidade do organismo e as cirurgias têm como principal intuito a cura da doença. No entanto, procedimentos de reconstrução facial são de suma importância pois promovem a manutenção da fisiologia e estética do indivíduo, já que a modificação facial pode causar repercussão psicológica por conta da perda de identidade do indivíduo. As técnicas de enxerto e de retalho cirúrgico se mostraram as mais eficazes para reconstrução da face em tumores avançados e suas indicações são determinadas pela necessidade tecidual e óssea, devendo, cada caso, ser avaliado individualmente. Assim, acredita-se que as cirurgias para reconstrução facial estão em constante evolução, visando resultados cada vez melhores incluindo a qualidade de vida dos pacientes.

In view of the randomness of rockfalls shape and irregularity of the bottom floor of working face in steeply dipping coal seams (SDCS), it is difficult to accurately simulate rockfall movement, and it is consequently unable to effectively protect against multirockfalls. Therefore, a method for generating random shape rockfalls based on ellipsoid equation is proposed, and a 3D grid model of real bottom floor of working face is established based on the geographic information system data. In order to verify the accuracy and feasibility of the method and 3D model, the trajectory simulated by Rockyfor3D software is compared, and the proposed method and 3D model prove to be effective in simulating rockfall movement more accurately. Then the proposed method and 3D grid model are applied to solve the problem of multirockfalls protection in numerical simulation, and the main factors affecting the structural stress response of protective netting are analyzed by taking three incident modes of parallel heights, ladder parallel, and the same trajectory. In the simulation, it is found out that the trajectory of irregular rockfalls is greatly affected by the shape of rockfall and working face floor; during the process of multiple rockfalls colliding with the protective netting, the peak stress on the protective netting is inversely proportional to both the time interval between each rockfall and the distance between each rockfall. The findings presented in this research contribute to rockfall prediction and protection against rockfall hazards.

Emotion recognition from facial images is considered as a challenging task due to the varying nature of facial expressions. The prior studies on emotion classification from facial images using deep learning models have focused on emotion recognition from facial images but face the issue of performance degradation due to poor selection of layers in the convolutional neural network model.To address this issue, we propose an efficient deep learning technique using a convolutional neural network model for classifying emotions from facial images and detecting age and gender from the facial expressions efficiently. Experimental results show that the proposed model outperformed baseline works by achieving an accuracy of 95.65% for emotion recognition, 98.5% for age recognition, and 99.14% for gender recognition.

We explored the neural signatures of face familiarity using cross-participant and cross-experiment decoding of event-related potentials, evoked by unknown and experimentally familiarized faces from a set of experiments with different participants, stimuli, and familiarization-types. Participants were either familiarized perceptually, via media exposure, or by personal interaction. We observed significant cross-experiment familiarity decoding involving all three experiments, predominantly over posterior and central regions of the right hemisphere in the 270 – 630 ms time window. This shared face familiarity effect was most prominent between the Media and Personal, as well as between the Perceptual and Personal experiments. Cross-experiment decodability makes this signal a strong candidate for a general neural indicator of face familiarity, independent of familiarization methods and stimuli. Furthermore, the sustained pattern of temporal generalization suggests that it reflects a single automatic processing cascade that is maintained over time.HighlightsWe investigated if a general neural signature of face familiarity existA cross-experiment decoding analysis of EEG data was usedThe analysis involved perceptual, media and personal familiarization methodsWe found a preserved pattern of familiarity decoding across experiments between 270 and 630 ms post-stimulusThis signature is consistent with previous reports on face familiarity effects

One of the usual problems during the construction of underground structures has been the water seepage into these structures. Water seepage into the tunnel, especially during excavation, could reduce the excavation procedure’s rapidity and make the walls unstable. Furthermore, to embed the required pumps for draining leaked water into the tunnel, it is necessary to estimate the tunnel's approximate inlet flow rate. The purpose of this study is to estimate the flow rate in the width of the tunnel inlet portal of the second part of Isfahan metro line 2, which has been done using numerical modeling and matching the results with analytical models. Observation of the sensitivity analysis results showed that the two parameters significantly affect the inlet flow rate of seepage into the tunnel: (1) Soil permeability coefficient, and (2) Groundwater level. Finally, the seepage flow rate was estimated in high water and water scarcity seasons, compared with the analytical results. Comparative parameters of NSE and RMSE were used to estimate the differences between analytical and numeral methods. This estimation showed that Lei (Groundwater 37: 23, 1999), and Lombardi (2002) could estimate the tunnel portal’s flow rate.

Thermal energy transfer (TET) is the main performance of contact interfaces which has been studied at a molecular level. Several investigations on TET were accomplished, however, the influences of liquid film thickness on TET have not been sufficiently examined. Thus, this paper analyses the influences of liquid film thickness on TET across solid–liquid (S-L) interfaces. Two liquid film thicknesses (Lz) of 30 Å and 60 Å have been evaluated, and two shear directions (x- and y-directions) have been tested in the simulation system. It has been found that there is no significant difference in the density distribution of liquid regardless of the shear directions for the same Lz. However, there are differences in the density distribution of liquid between Lz of 30 Å and 60 Å. Based on the results its suggests that, the cut-off of the temperature and velocity at the contact interfaces of solid and liquid is substantially influences by the liquid thickness of the simulation system. It is found that, there are a significant different in the thermal boundary resistance (TBR) for Lz of 30 Å and 60 Å for cases liquid sheared in the x-direction. Whereas TBR for Lz of 30 Å and 60 Å sheared in the y-direction have no significant difference. In conclusion, the TET is affected by the velocity cut-off at the contact interfaces of solid and liquid where larger velocity discontinuity exhibits higher TBR.

The roof usually collapses when exposed to the repeated underground mining; this will potentially lead to the instability of coal wall and even cause serious mining disasters. Based on this specific mining hazard, this work investigates the roof leak characteristics and corresponding control technology based on a coal mine in Guizhou province, China. The theoretical analysis, numerical simulation, and physical testing were carried out. By establishing a physical model, the roof structure and roof “pressure” law under repeated mining are studied, and the instable conditions of face-end roof are analyzed, so as to determine the influencing factors of face-end roof instability. By analyzing different roof structures, the working resistance of hydraulic support with stable face-end roof under repeated mining is determined. The leakage of face-end roof under different support working resistance, tip-to-face distance, and surrounding rock strength is simulated, and different methods are proposed to control face-end roof leaks. The results show that in the process of 17# coal seam mining, the roof collapse is serious, the roof pressure is frequent, and the broken roof instability would form the “granular arch” structure. Through the analysis of the roof structure, the tip-to-face distance, the strength of granular arch structure, and the working resistance of the hydraulic support affect the stability of the face-end roof. Using UDEC simulation software, it is obtained that the higher the strength of roof and surrounding rock, the smaller the tip-to-face distance, the more stable the face-end roof is. The prevention measures such as strengthening the broken face-end roof under repeated mining, improving the support strength of the working face hydraulic support and controlling the reasonable tip-to-face distance are put forward. Through field application, the face-end roof leaks have been effectively controlled, which has played a good reference for the prevention and control of face-end roof leaks under repeated mining.

The correct classification of cancer subtypes is of great significance for the in-depth study of cancer pathogenesis and the realization of accurate treatment for cancer patients. In recent years, the classification of cancer subtypes using deep neural networks and gene expression data has become a hot topic. However, most classifiers may face the challenges of overfitting and low classification accuracy when dealing with small sample size and high-dimensional biological data. In this paper, the Cascade Flexible Neural Forest (CFNForest) Model was proposed to accomplish cancer subtype classification. CFNForest extended the traditional flexible neural tree structure to FNT Group Forest exploiting a bagging ensemble strategy and could automatically generate the model’s structure and parameters. In order to deepen the FNT Group Forest without introducing new hyperparameters, the multilayer cascade framework was exploited to design the FNT Group Forest model, which transformed features between levels and improved the performance of the model. The proposed CFNForest model also improved the operational efficiency and the robustness of the model by sample selection mechanism between layers and setting different weights for the output of each layer. To accomplish cancer subtype classification, FNT Group Forest with different feature sets was used to enrich the structural diversity of the model, which make it more suitable for processing small sample size datasets. The experiments on RNA-seq gene expression data showed that CFNForest effectively improves the accuracy of cancer subtype classification. The classification results have good robustness.

Presently, facial image recognition via a thermal camera is a critical phase in numerous fields. Systems using thermal facial images suffer from numerous problems in face identification. In this paper, a model Edge-Aided Generative Adversarial Network (EA-GAN) is introduced to overcome the difficulties of thermal face identification by synthesizing a visible faces image from the thermal version. To enhance the performance of the Conditional Generative Adversarial Network (CGAN) model for the create realistic face images, the edge information extracted from the thermal image has been used as input, thus lead to improving overall the system's achievement. Moreover, a new model is presented in the present work for face identification by integrating two Convolutional Neural Networks (CNN) to achieve high and rapid accuracy rates. Based on the experiments on the Carl dataset for faces, it is indicated that EA-GAN can synthesize visually comfortable and identity-preserving faces; thus, better performance is achieved in comparison with the state-of-the-art approaches for thermal facial identification.

Although facial beauty prediction (FBP) has achieved high accuracy on images captured in a constrained environment, it is still a challenging task on face images in-the-wild. Moreover, there is no FBP benchmark dataset that includes images with various ethnic, age, gender properties and unrestricted in terms of face expression and pose. In this work, the issue of FBP in real-life scenario is addressed and a multi-ethnic facial beauty dataset, namely MEBeauty, is introduced. All face images are captured in an unconstrained environment and rated by volunteers with various ethnicity, age and gender in order to avoid any cultural and social biases in beauty perception. Different well-known CNNs with layer-wise transfer learning are performed on the dataset. Moreover, the evaluation of knowledge learning from the face recognition task across FBP is conducted. The expected high number of aberrant and outlier faces is considered and the use of various robust loss functions in order to learn deep regression networks for facial beauty prediction is evaluated. Several FBP frameworks are performed on the proposed dataset and widely-used SCUT-FBP 5500 in order to compare their effectiveness on face images in constrained and unconstrained environments.

Wearing masks is an effective and simple method to prevent the spread of the COVID-19 pandemic in public places, such as train stations, classrooms, and streets. It is of positive significance to urge people to wear masks with computer vision technology. However, the existing detection methods are mainly for simple scenes, and facial missing detection is prone to occur in dense crowds with different scales and occlusions. Moreover, the data obtained by surveillance cameras in public places are difficult to be collected for centralized training, due to the privacy of individuals. In order to solve these problems, a cascaded network is proposed: the first level is the Dilation RetinaNet Face Location (DRFL) Network, which contains Enhanced Receptive Field Context (ERFC) module with the dilation convolution, aiming to reduce network parameters and locate faces of different scales. In order to adapt to embedded camera devices, the second level is the SRNet20 network, which is created by Neural Architecture Search (NAS). Due to privacy protection, it is difficult for surveillance video to share in practice, so our SRNet20 network is trained in federated learning. Meanwhile, we have made a masked face dataset containing about 20,000 images. Finally, the experiments highlight that the detection mAP of the face location is 90.6% on the Wider Face dataset, and the classification mAP of the masked face classification is 98.5% on the dataset we made, which means our cascaded network can detect masked faces in dense crowd scenes well.

The article focuses on the gentle introduction of Artificial Intelligence and the concepts of machine learning (ML) and deep learning (DL). The rapid developments made in DL techniques has motivated us to delve into this study. The concept of DL flourishing from basics theoretical concepts to applications. Deep neural networks are now state-of-the-art ML models commonly used in academia and industry in several fields, from image recognition to natural language processing. These advances have an immense potential for medical imaging technology, medical data processing, medical diagnostics and general healthcare. Our aim is two-fold: (1) the survey on DL approaches to medical images (2) the DL-based object detection approaches. The article delivers an effective description of the recent advances, advanced learning technologies and the platforms used for DL approaches. Object detection is the most explored and challenging concept in the field of computer vision systems. This field is receiving greater attention among the researchers since it covers real-time applications such as the face, pedestrian, text etc. The role of object detection is to detect the target objects presented in the image (or) video frames by appropriately classifying into their relevant classes. The review study of object detection begins with the recent works, the datasets used, and the real-time applications are explored from the learning strategies. Finally, the article investigates the challenges of the DL models and discusses promising future directions in both the research areas.

Human emotions, such as smiling or laughing, can be expressed in various forms through the face whenever there are stimuli. These changing faces can reflect the emotional states that are used to identify a normal or an abnormal behaviour. This research aims to study the patterns in human faces and identify the areas of interest (AOI), which is called Facial Landmark Detection (FLD). The investigation of the external elements of eyes is performed, and it consists of the interpalpebral fissure (IPF), the palpebral fissure length (PFL), and the palpebral fissure region (PFR). These elements are applied to classify the emotional differences between neutral and fearful emotions. A method for emotional classification was designed according to the changing values of the IPF, PFL, and PFR. An ID3 algorithm was used to classify the emotions. Three hundred sixty images were derived from horror-thriller-murder movies based on IMDb. This data set was utilized to generate the proposed pattern. This pattern was used to classify the emotions using a decision tree technique that led to the development of an emotional classification model. The accuracy of the emotional classification model between neutral and fearful emotions was 92.50%, thus proving that the proposed model is efficient.

This letter proposes the use of neural networks to realize the passive localization by signal time difference of arrival (TDOA). In the face of multiple complex targets with radiation sources in a specific area, real-time localization is an urgent problem. In this letter, positions of the known targets from the prior data are obtained and their time difference is calculated, which will be connected as data pairs and input to the neural network trained to obtain a corresponding model. Subsequently, unknown targets can be localized by this network. It is verified that the localization accuracy of the algorithm is reliable and its robustness is higher than that of traditional algorithms. The proposed method also shows that great reduction of operation time depending on the previous network training can complete real-time goals.

To solve the problem of data sparsity and cold start, the cross-domain recommendation is a promising research direction in the recommender system. The goal of cross-domain recommendation is to transfer learned knowledge from the source domain to the target domain by different means to improve the performance of the recommendation. But most approaches face the distribution misalignment. In this paper, we propose a joint learning cross-domain recommendation model that can extract domain-specific and common features simultaneously, and only use the implicit feedback data of users without additional auxiliary information. To the best of our knowledge, it is the first attempt to combine the sliced Wasserstein distance and canonical correlation analysis for the cross-domain recommendation scenario. Our one intuition is to reduce the reconstruction error caused by the variational inference based autoencoder model by the optimal transportation theory. Another attempt is to improve the correlation between domains by combining the idea of the canonical correlation analysis. With rigorous experiments, we empirically demonstrated that our model can achieve better performance compared with the state-of-the-art methods.

This paper focuses on the following problem: Given a large number of unlabeled face images, group them into individual clusters, and the number of clusters cannot be known in advance. To this end, an Efficient Density-based clustering incorporated with the model of Graph partitioning (EDG) is proposed. 1. Inspired by the progress of graph partitioning clustering, a novel criterion that can be seen as a variant of the Normalized-cut model is employed to measure the similarity between two samples. 2. We only consider the similarities and connections on a subset of all possible pairs, i.e. the top-K nearest neighbors for each sample. Therefore, the computing and storage costs are linear w.r.t. the number of samples. In order to assess the performance of EDG on face images, extensive experiments based on a two-stage framework have been conducted on 19 benchmark datasets (14 middle-scale and 5 large-scale) from the literature. The experimental results have shown the effectiveness and robustness of our model, compared with the state-of-the-art methods. [code]

Concrete face rockfill dams are widely used in hydropower engineering because of their many advantages. As important structures, if the performance of the concrete face slab and the impervious curtain is weakened, leakages and even serious dam failures can occur. Based on an equivalent quasi-continuum medium model, three-dimensional finite-element numerical simulations of the seepage behaviour of Miaojiaba dam, China, are performed. The seepage field and seepage characteristics, including the phreatic line, maximum hydraulic gradient and seepage discharge, are analysed quantitatively and systematically under various abnormal conditions. The results show that cracks in different areas of the face slab and different degrees of degradation in the impervious curtain influence the seepage field in different ways, with cracks in the lower face slab being the most serious. Ensuring that the face slab and impervious curtain are working as intended is very important for the integrity of the dam. Therefore, the design and construction quality of impervious structures should be strictly controlled. Under abnormal conditions, seepage control should be realised to ensure the safety and stability of concrete face rockfill dams.

Face recognition systems are widely used for target recognition and identity authentication, such as automated teller machines, mobile phones, and entrance guard systems. However, face recognition systems are vulnerable to presentation attacks, such as photo, replay, and 3D mask attacks. In particular, silicone mask attacks pose a greater threat to face recognition systems because high-quality silicone masks do living properties. To promote the development of face anti-spoofing detection algorithms for silicone mask attacks, this paper constructs a Silicone Mask Face Motion Video Dataset (SMFMVD) containing 200 real face videos and 200 silicone mask face videos. These videos include different facial motions collected from 20 subjects. Moreover, inspired by the observation that the silicone mask face’s facial movement is not so natural as the real face, we propose a novel silicone mask face anti-spoofing detection method based on visual saliency and facial motion characteristics. Specifically, we compute the visual saliency map of a given face image by simulating two kinds of eye movement behaviors, namely “gaze” and “saccade”. Then, we propose a saliency-weighted histogram of local binary pattern operator to extract facial texture features in spatial domain and a saliency-guided histogram of oriented optical flow operator to extract facial motion features in temporal domain. Finally, the support vector machine is used to fuse two groups of facial features to distinguish real and spoof faces. Extensive experiments on public and self-built datasets show its superiority over the state-of-the-art methods.

Strong rock blocks embedded in a weaker soil matrix are found in many geological units. When tunneling in ground containing cobbles and boulders, extremely challenging conditions can be encountered. Such inconveniences may be avoided by means of appropriate tunneling methods and cutterhead designs, which require the content, frequency, and size of rock blocks to be predicted as accurately as possible. Several approaches have been developed to estimate the block fraction of heterogeneous geomaterials for excavation. However, the estimation of cobble–boulder quantities both all along the tunnel and only partially embedded within the tunnel face remains a critical issue. This study develops a methodology for the estimation of the probability of encountering blocks partially or totally contained within the tunnel excavation area, wherein the area of intersection with the tunnel face is greater than the given critical values. For this purpose, a statistical approach has been implemented in a Matlab routine. The potential of this code is that it provides extremely useful and statistically based information that can be used for making a more rational choice regarding tunneling technique and in terms of designing a suitable cutterhead in order to avoid technical problems during tunnel excavations in heterogeneous ground. The executable code is provided.

In light of the rising demand for biometric-authentication systems, preventing face spoofing attacks is a critical issue for the safe deployment of face recognition systems. Here, we propose an efficient face presentation attack detection (PAD) algorithm that requires minimal hardware and only a small database, making it suitable for resource-constrained devices such as mobile phones. Utilizing one monocular visible-light camera, the proposed algorithm takes two facial photos, taken with and without a flash, respectively. The proposed SpecDiff descriptor is constructed by leveraging two types of reflection: (i) specular reflections from the iris region that have a specific intensity distribution depending on liveness, and (ii) diffuse reflections from the entire face region that represents the 3D structure of a subject’s face. Classifiers trained with the SpecDiff descriptor outperform other flash-based PAD algorithms on both an in-house database and four publicly available databases: NUAA, Replay-Attack, Spoofing in the Wild, and OULU-NPU. Furthermore, the proposed algorithm achieves statistically significantly better accuracy to that of an end-to-end, deep neural network classifier, while being approximately six-times faster execution speed. The limitation of the proposed algorithm is also quantified under various adversarial lighting conditions, to guide users for the safe deployment of the algorithm. The code is publicly available at https://github.com/Akinori-F-Ebihara/SpecDiff-spoofing-detector. Example images of in-house database are also available at https://github.com/Akinori-F-Ebihara/SpecDiff in house database sample.