Detecting faces under occlusion remains a significant challenge in computer vision due to variations caused by masks,sunglasses,and other obstructions.Addressing this issue is crucial for applications such as surveill...Detecting faces under occlusion remains a significant challenge in computer vision due to variations caused by masks,sunglasses,and other obstructions.Addressing this issue is crucial for applications such as surveillance,biometric authentication,and human-computer interaction.This paper provides a comprehensive review of face detection techniques developed to handle occluded faces.Studies are categorized into four main approaches:feature-based,machine learning-based,deep learning-based,and hybrid methods.We analyzed state-of-the-art studies within each category,examining their methodologies,strengths,and limitations based on widely used benchmark datasets,highlighting their adaptability to partial and severe occlusions.The review also identifies key challenges,including dataset diversity,model generalization,and computational efficiency.Our findings reveal that deep learning methods dominate recent studies,benefiting from their ability to extract hierarchical features and handle complex occlusion patterns.More recently,researchers have increasingly explored Transformer-based architectures,such as Vision Transformer(ViT)and Swin Transformer,to further improve detection robustness under challenging occlusion scenarios.In addition,hybrid approaches,which aim to combine traditional andmodern techniques,are emerging as a promising direction for improving robustness.This review provides valuable insights for researchers aiming to develop more robust face detection systems and for practitioners seeking to deploy reliable solutions in real-world,occlusionprone environments.Further improvements and the proposal of broader datasets are required to developmore scalable,robust,and efficient models that can handle complex occlusions in real-world scenarios.展开更多
Face detection is a critical component inmodern security,surveillance,and human-computer interaction systems,with widespread applications in smartphones,biometric access control,and public monitoring.However,detecting...Face detection is a critical component inmodern security,surveillance,and human-computer interaction systems,with widespread applications in smartphones,biometric access control,and public monitoring.However,detecting faces with high levels of occlusion,such as those covered by masks,veils,or scarves,remains a significant challenge,as traditional models often fail to generalize under such conditions.This paper presents a hybrid approach that combines traditional handcrafted feature extraction technique called Histogram of Oriented Gradients(HOG)and Canny edge detection with modern deep learning models.The goal is to improve face detection accuracy under occlusions.The proposed method leverages the structural strengths of HOG and edge-based object proposals while exploiting the feature extraction capabilities of Convolutional Neural Networks(CNNs).The effectiveness of the proposed model is assessed using a custom dataset containing 10,000 heavily occluded face images and a subset of the Common Objects in Context(COCO)dataset for non-face samples.The COCO dataset was selected for its variety and realism in background contexts.Experimental evaluations demonstrate significant performance improvements compared to baseline CNN models.Results indicate that DenseNet121 combined with HOG outperforms other counterparts in classification metrics with an F1-score of 87.96%and precision of 88.02%.Enhanced performance is achieved through reduced false positives and improved localization accuracy with the integration of object proposals based on Canny and contour detection.While the proposed method increases inference time from 33.52 to 97.80 ms,it achieves a notable improvement in precision from 80.85% to 88.02% when comparing the baseline DenseNet121 model to its hybrid counterpart.Limitations of the method include higher computational cost and the need for careful tuning of parameters across the edge detection,handcrafted features,and CNN components.These findings highlight the potential of combining handcrafted and learned features for occluded face detection tasks.展开更多
The lack of facial features caused by wearing masks degrades the performance of facial recognition systems.Traditional occluded face recognition methods cannot integrate the computational resources of the edge layer a...The lack of facial features caused by wearing masks degrades the performance of facial recognition systems.Traditional occluded face recognition methods cannot integrate the computational resources of the edge layer and the device layer.Besides,previous research fails to consider the facial characteristics including occluded and unoccluded parts.To solve the above problems,we put forward a device-edge collaborative occluded face recognition method based on cross-domain feature fusion.Specifically,the device-edge collaborative face recognition architecture gets the utmost out of maximizes device and edge resources for real-time occluded face recognition.Then,a cross-domain facial feature fusion method is presented which combines both the explicit domain and the implicit domain facial.Furthermore,a delay-optimized edge recognition task scheduling method is developed that comprehensively considers the task load,computational power,bandwidth,and delay tolerance constraints of the edge.This method can dynamically schedule face recognition tasks and minimize recognition delay while ensuring recognition accuracy.The experimental results show that the proposed method achieves an average gain of about 21%in recognition latency,while the accuracy of the face recognition task is basically the same compared to the baseline method.展开更多
The enrichment of chloride anion within the occluded cell （OC） for Type 304 austenitic stainless steel in low chloride concentration solution has been investigated by means of a simulated OC. The influence of the en...The enrichment of chloride anion within the occluded cell （OC） for Type 304 austenitic stainless steel in low chloride concentration solution has been investigated by means of a simulated OC. The influence of the enrichment of chloride anion on stress corrosion crack （SCC） of Type 304 stainless steel has been studied. It was observed that the amount of chloride anion migration was proportional to the charge flowing through the anode. Owning to the effects of enrichment of chloride anion, low chloride concentration solution could induce SCC for Type 304 stainless steel.展开更多
In patients with a malignant biliary obstruction who require biliary drainage,a self-expandable metallic stent(SEMS) provides longer patency duration than a plastic stent(PS).Nevertheless,a stent occlusion by tumor in...In patients with a malignant biliary obstruction who require biliary drainage,a self-expandable metallic stent(SEMS) provides longer patency duration than a plastic stent(PS).Nevertheless,a stent occlusion by tumor ingrowth,tumor overgrowth and biliary sludge may develop.There are several methods to manage occluded SEMS.Endoscopic management is the preferred treatment,whereas percutaneous intervention is an alternative approach.Endoscopic treatment involves mechanical cleaning with a balloon and a second stent insertion as stent-in-stent with either PS or SEMS.Technical feasibility,patient survival and cost-effectiveness are important factors that determine the method of re-drainage and stent selection.展开更多
Oil-cracked gas,as the main type of high-over mature marine natural gas in China,is mainly derived from occluded hydrocarbon.So it is significant to carry out quantitative study on occluded hydrocarbon.In this paper,t...Oil-cracked gas,as the main type of high-over mature marine natural gas in China,is mainly derived from occluded hydrocarbon.So it is significant to carry out quantitative study on occluded hydrocarbon.In this paper,the occluded hydrocarbon volume of the main basins in China was calculated depending on their types,abundances and evolution stages by means of the forward method(experimental simulation)and the inversion method(geologic profile dissection).And then,occluded hydrocarbon evolution models were established for five types of source rocks(sapropelic,sapropelic prone hybrid,humic prone hybrid,humic and coal).It is shown that the hydrocarbon expulsion efficiency of sapropelic and sapropelic prone hybrid excellent source rocks is lower than 30%at the low-maturity stage,30%e60%at the principal oil generation stage,and 50%e80%at the high-maturity stage,which are all about 10%higher than that of humic prone hybrid and humic source rocks at the corresponding stages.The resource distribution and cracked gas expulsion of occluded hydrocarbon since the high-maturity stage of marine source rocks in the Sichuan Basin were preliminarily calculated on the basis of the evolution models.The cracked gas expulsion is 230.4×10^(12) m^(3) at the high evolution stage of occluded hydrocarbon of the Lower Cambrian Qiongzhusi Fm in this basin,and 12.3×10^(12) m^(3) from the source rocks of Sinian Doushantuo Fm,indicating good potential for natural gas resources.It is indicated that the favorable areas of occluded hydrocarbon cracked gas in the Qiongzhusi Fm source rocks in the Sichuan Basin include GaoshitieMoxi,Ziyang and Weiyuan,covering a favorable area of 4.3×10^(4) km^(2).展开更多
An artificial localized corrosion system is assembled and some parameters related to the localized corrosion in active dissolution state (i.e., non-passive state) have been studied. The results showed that the develop...An artificial localized corrosion system is assembled and some parameters related to the localized corrosion in active dissolution state (i.e., non-passive state) have been studied. The results showed that the developed electrochemical system can satisfactorily imitate a naturally formed localized corrosion and the coupling current can indicate the maximum localized propagating rate. In this artificial system, the anodic dissolution reaction followed the auto-catalytic mechanism. The localized corrosion current density was dependent on the area ratio R of the cathode to the occluded anode. While R was equal to or more than 6, the coupling current reached at a maximum value and did not alter with the increase in R-value. Therefore, R=7 is chosen as one of these optimum parameters used in constructing the system, with which the biggest galvanic current might be obtained. In contrast, the thickness of the polymer filler separating the occluded anode area from the bulk electrolyte solution and the volume of the occluded anode area did not affect the corrosion current obviously. They might affect the response time to approach a steady state.展开更多
Three kerogen samples(JJZG-1,JJZG-2 and JJZG-3)isolated from the Permian Lucaogou shales of varying biodegradation levels(BLs≈0,3 and 7,respectively)were subjected to sequential stepwise pyrolysis combined with on-li...Three kerogen samples(JJZG-1,JJZG-2 and JJZG-3)isolated from the Permian Lucaogou shales of varying biodegradation levels(BLs≈0,3 and 7,respectively)were subjected to sequential stepwise pyrolysis combined with on-line detection of gas chromatography-mass spectrometry(GC-MS).Occluded fractions(bitumenⅡ)released at low-temperature steps(≥410℃)show consistent biodegradative signatures with that reported for solvent-extracted fractions(bitumenⅠ)of the original shales,e.g.,broad range of abundant n-alkanes,isoprenoids and regular hopanes for the non-biodegraded JJZG-1;trace n-alkanes and abundant hopanes for the moderately biodegraded JJZG-2;and no n-alkanes but still prominent hopanes including the microbially produced 25-nohopanes for the severely biodegraded JJZG-3.This consistency between bitumenⅡand bitumenⅠfractions indicates the biodegradability of the kerogenoccluded bitumenⅡwith limited protection from host kerogen.A minor level of protection was suggested by the trace distribution of n-alkanes in the bitumenⅡof JJZG-2,whereas the bitumenⅠhad no nalkanes.The kerogen itself was more resistant to biodegradation as reflected by the persistence of high abundances of both n-alkanes and hopanes in the high temperature(≥460℃)products of all three kerogen samples.However,the relative abundances of these product groups did show some evidence of biodegradation alteration,e.g.,ratios of n-C_(15)alkene/C_(27)hop-17(21)-ene at 510℃pyrolysis decreased by order of magnitude from the non-biodegraded(JJZG-1=27.4)to highly biodegraded(0.3 for JJZG-3)samples.The reduced biodegradation impact on the kerogen fraction(Cf.bitumen fractions)was also evident by the absence of 25-norhopanes in the high-temperature analysis of the JJZG-3 kerogen.展开更多
Autonomous driving is a promising way to future safe,efficient,and low-carbon transportation.Real-time ac-curate target detection is an essential precondition for the generation of proper following decision and contro...Autonomous driving is a promising way to future safe,efficient,and low-carbon transportation.Real-time ac-curate target detection is an essential precondition for the generation of proper following decision and control signals.However,considering the complex practical scenarios,accurate recognition of occluded targets is a major challenge of target detection for autonomous driving with limited computational capability.To reveal the overlap and difference between various occluded object detection by sharing the same available sensors,this paper presents a review of detection methods for occluded objects in complex real-driving scenarios.Considering the rapid development of autonomous driving technologies,the research analyzed in this study is limited to the recent five years.The study of occluded object detection is divided into three parts,namely occluded vehicles,pedes-trians and traffic signs.This paper provided a detailed summary of the target detection methods used in these three parts according to the differences in detection methods and ideas,which is followed by the comparison of advantages and disadvantages of different detection methods for the same object.Finally,the shortcomings and limitations of the existing detection methods are summarized,and the challenges and future development prospects in this field are discussed.展开更多
Compared with traditional nickel-titanium alloy patent foramen ovale occluders,which are widely used in clinical practice,biodegradable patent foramen ovale occluders have obvious differences in material characteristi...Compared with traditional nickel-titanium alloy patent foramen ovale occluders,which are widely used in clinical practice,biodegradable patent foramen ovale occluders have obvious differences in material characteristics,interventional operation mode and postoperative management strategy.This article gives expert consensus on the selection of clinical indications and standardized operating procedures,so as to standardize the clinical application of biodegradable patent foramen ovale occluders.展开更多
Dissimilar metal joining between NiTi shape memory alloy（SMA） and stainless steel was conducted.A cluster of NiTi SMA wires were first joined with tungsten inert gas（TIG） welding process,then the NiTi SMA TIG weld...Dissimilar metal joining between NiTi shape memory alloy（SMA） and stainless steel was conducted.A cluster of NiTi SMA wires were first joined with tungsten inert gas（TIG） welding process,then the NiTi SMA TIG weld was welded to a stainless steel pipe with laser spot welding process.The microstructure of the welds was examined with an optical microscope and the elemental distribution in the welds was measured by electron probe microanalysis（EPMA）.The results show that TiC compounds dispersively distribute in the NiTi SMA TIG weld.However,the amount of TiC compounds greatly decreases around the fusion boundary of the laser spot weld between the NiTi SMA and stainless steel.Mutual diffusion between NiTi shape memory alloy and stainless steel happen within a short distance near the fusion boundary,and intermetallic compounds such as Ni3Ti＋（Fe,Ni）Ti appear around the fusion boundary.展开更多
Video surveillance is an active research topic in computer vision.In this paper,humans and cars identifcation technique suitable for real time video surveillance systems is presented.The technique we proposed includes...Video surveillance is an active research topic in computer vision.In this paper,humans and cars identifcation technique suitable for real time video surveillance systems is presented.The technique we proposed includes background subtraction,foreground segmentation,shadow removal,feature extraction and classifcation.The feature extraction of the extracted foreground objects is done via a new set of afne moment invariants based on statistics method and these were used to identify human or car.When the partial occlusion occurs,although features of full body cannot be extracted,our proposed technique extracts the features of head shoulder.Our proposed technique can identify human by extracting the human head-shoulder up to 60%–70%occlusion.Thus,it has a better classifcation to solve the issue of the loss of property arising from human occluded easily in practical applications.The whole system works at approximately 16 29 fps and thus it is suitable for real-time applications.The accuracy for our proposed technique in identifying human is very good,which is 98.33%,while for cars identifcation,the accuracy is also good,which is 94.41%.The overall accuracy for our proposed technique in identifying human and car is at 98.04%.The experiment results show that this method is efective and has strong robustness.展开更多
In order to provide scientific basis for advanced applications of near eutectic Al-Si-Mg alloys as architectural profiles, a comparative study on the corrosion resistance of an as-extruded near eutectic AI-Si-Mg alloy...In order to provide scientific basis for advanced applications of near eutectic Al-Si-Mg alloys as architectural profiles, a comparative study on the corrosion resistance of an as-extruded near eutectic AI-Si-Mg alloy and AA6063 aluminium alloy was carried out by means of neutral salt spray test. The corroded surfaces of the alloys were examined with optical microscopy and scanning electron microscope （SEM）. Results show that the corrosion type of these two alloys is pitting corrosion. The number of corrosion pits in the AA6063 aluminium alloy is more than that in the near eutectic AI-Si-Mg alloy, but the pits in the latter alloy are much larger and deeper. Because the relatively low polarization resistance of the near eutectic alloy leads to poorer repassivation ability, autocatalytic acidification occurs once a pit forms. Thus, occluded corrosion cells are developed in this alloy.展开更多
The difficulty of multiple targets tracking is how to quickly fulfill the target matching from one flame image to another and fix the position of the target. In order to accurately choose target feature information fo...The difficulty of multiple targets tracking is how to quickly fulfill the target matching from one flame image to another and fix the position of the target. In order to accurately choose target feature information for reliable matching, simplify operations under the reliable precondition, and realize precise moving objects tracking, an approach based on Kalman prediction and feature matching was proposed. The position of the target in next frame image was predicted by Kalman, and then the moving objects of two adjacent frames were matched by the centroid and area methods. When occlusion occurs, the best matching result was found to realize tracking by matching matrix algorithm. The simulation results show that the proposed method can achieve multiple targets tracking accurately and in real-time under complicated motion movements.展开更多
BISHOP’s effective stress or two state stress variables are unsatisfactory for unsaturated soils where one of fluid phases is discontinuous, so new expressions of effective stress should be founded. The approach for ...BISHOP’s effective stress or two state stress variables are unsatisfactory for unsaturated soils where one of fluid phases is discontinuous, so new expressions of effective stress should be founded. The approach for derivation was according to the principle of equilibrium of forces (i.e., the stress-sharing principle), and it was firstly validated by demonstrating TERZAGHI’s principle of effective stress. And then, the derivations were subdivided into four parts according to different pore air states: 1) air bubbles were spherical and suspended in pore water; 2) air bubbles were bound on soil skeleton; 3) air bubbles held almost the single section of pore; 4) air phase was continuous. The different formulae of effective stress were presented. Conclusions are drawn as follows: 1) For nearly-saturated soils, the "real" effective stress would be a little smaller than TERZAGHI’s effective stress; 2) For soils in which air phase is discontinuous in the form of bubbles, a new concept of pore air elastic pressure is put forward, and the total stress can be constituted by effective stress, pore water pressure and pore air elastic pressure; 3) For soils in which air phase is continuous, effective stress is equal to the value of the total stress plus suction; 4) Suction can be divided into two parts: one is the effect caused by additional pressure, and the other is the contract action by the "skin".展开更多
Chronic thromboembolic pulmonary hypertension (CTEPH) comprises organizing thrombotic obstructions in the pulmonary arteries by nonresolving thromboemboli, formation of fibrosis and remodeling of pulmonary blood vesse...Chronic thromboembolic pulmonary hypertension (CTEPH) comprises organizing thrombotic obstructions in the pulmonary arteries by nonresolving thromboemboli, formation of fibrosis and remodeling of pulmonary blood vessels. Surgical pulmonary endarterectomy (PEA) is the therapy of choice for patients with surgically accessible CTEPH, which leads to a profound improvement in hemodynamics, functional class and survival. Select- ing the candidates that will benefit from surgery is still a challenging task. Criteria for surgical suitability have been described but the decision-making for or against surgical intervention remains still subjective. The optimal characterization of the reciprocal contribution of large vessel and small vessel disease in the elevation of pulmonary vascular resistance is crucial for the indication and outcome of PEA. Recently, Toshner et al intended to validate the partition resistance into small and large vessels compartments (upstream resistance:Rup) by the occlusion technique in the preoperative assessment of PEA. We discuss the advantages and disadvantages of Rup and compare it with other hemodynamic predictor to evaluate operative risk in CTEPH patients.展开更多
Biometric recognition refers to the process of recognizing a person’s identity using physiological or behavioral modalities,such as face,voice,fingerprint,gait,etc.Such biometric modalities are mostly used in recogni...Biometric recognition refers to the process of recognizing a person’s identity using physiological or behavioral modalities,such as face,voice,fingerprint,gait,etc.Such biometric modalities are mostly used in recognition tasks separately as in unimodal systems,or jointly with two or more as in multimodal systems.However,multimodal systems can usually enhance the recognition performance over unimodal systems by integrating the biometric data of multiple modalities at different fusion levels.Despite this enhancement,in real-life applications some factors degrade multimodal systems’performance,such as occlusion,face poses,and noise in voice data.In this paper,we propose two algorithms that effectively apply dynamic fusion at feature level based on the data quality of multimodal biometrics.The proposed algorithms attempt to minimize the negative influence of confusing and low-quality features by either exclusion or weight reduction to achieve better recognition performance.The proposed dynamic fusion was achieved using face and voice biometrics,where face features were extracted using principal component analysis(PCA),and Gabor filters separately,whilst voice features were extracted using Mel-Frequency Cepstral Coefficients(MFCCs).Here,the facial data quality assessment of face images is mainly based on the existence of occlusion,whereas the assessment of voice data quality is substantially based on the calculation of signal to noise ratio(SNR)as per the existence of noise.To evaluate the performance of the proposed algorithms,several experiments were conducted using two combinations of three different databases,AR database,and the extended Yale Face Database B for face images,in addition to VOiCES database for voice data.The obtained results show that both proposed dynamic fusion algorithms attain improved performance and offer more advantages in identification and verification over not only the standard unimodal algorithms but also the multimodal algorithms using standard fusion methods.展开更多
Person search mainly consists of two submissions,namely Person Detection and Person Re-identification(reID).Existing approaches are primarily based on Faster R-CNN and Convolutional Neural Network(CNN)(e.g.,ResNet).Wh...Person search mainly consists of two submissions,namely Person Detection and Person Re-identification(reID).Existing approaches are primarily based on Faster R-CNN and Convolutional Neural Network(CNN)(e.g.,ResNet).While these structures may detect high-quality bounding boxes,they seem to degrade the performance of re-ID.To address this issue,this paper proposes a Dual-Transformer Head Network(DTHN)for end-to-end person search,which contains two independent Transformer heads,a box head for detecting the bounding box and extracting efficient bounding box feature,and a re-ID head for capturing high-quality re-ID features for the re-ID task.Specifically,after the image goes through the ResNet backbone network to extract features,the Region Proposal Network(RPN)proposes possible bounding boxes.The box head then extracts more efficient features within these bounding boxes for detection.Following this,the re-ID head computes the occluded attention of the features in these bounding boxes and distinguishes them from other persons or backgrounds.Extensive experiments on two widely used benchmark datasets,CUHK-SYSU and PRW,achieve state-of-the-art performance levels,94.9 mAP and 95.3 top-1 scores on the CUHK-SYSU dataset,and 51.6 mAP and 87.6 top-1 scores on the PRW dataset,which demonstrates the advantages of this paper’s approach.The efficiency comparison also shows our method is highly efficient in both time and space.展开更多
The paper deals with the fluid field of web forming in wet-laid non-woven production.The influence of the turbulent flow on blending fiber and occluded fluid produced in pulp flow has been discussed in theory and prac...The paper deals with the fluid field of web forming in wet-laid non-woven production.The influence of the turbulent flow on blending fiber and occluded fluid produced in pulp flow has been discussed in theory and practice.The suitable use of the imported velocity of pulp is very important in producing wet-laid products of good quality.展开更多
基金funded by A’Sharqiyah University,Sultanate of Oman,under Research Project grant number(BFP/RGP/ICT/22/490).
文摘Detecting faces under occlusion remains a significant challenge in computer vision due to variations caused by masks,sunglasses,and other obstructions.Addressing this issue is crucial for applications such as surveillance,biometric authentication,and human-computer interaction.This paper provides a comprehensive review of face detection techniques developed to handle occluded faces.Studies are categorized into four main approaches:feature-based,machine learning-based,deep learning-based,and hybrid methods.We analyzed state-of-the-art studies within each category,examining their methodologies,strengths,and limitations based on widely used benchmark datasets,highlighting their adaptability to partial and severe occlusions.The review also identifies key challenges,including dataset diversity,model generalization,and computational efficiency.Our findings reveal that deep learning methods dominate recent studies,benefiting from their ability to extract hierarchical features and handle complex occlusion patterns.More recently,researchers have increasingly explored Transformer-based architectures,such as Vision Transformer(ViT)and Swin Transformer,to further improve detection robustness under challenging occlusion scenarios.In addition,hybrid approaches,which aim to combine traditional andmodern techniques,are emerging as a promising direction for improving robustness.This review provides valuable insights for researchers aiming to develop more robust face detection systems and for practitioners seeking to deploy reliable solutions in real-world,occlusionprone environments.Further improvements and the proposal of broader datasets are required to developmore scalable,robust,and efficient models that can handle complex occlusions in real-world scenarios.
基金funded by A’Sharqiyah University,Sultanate of Oman,under Research Project Grant Number(BFP/RGP/ICT/22/490).
文摘Face detection is a critical component inmodern security,surveillance,and human-computer interaction systems,with widespread applications in smartphones,biometric access control,and public monitoring.However,detecting faces with high levels of occlusion,such as those covered by masks,veils,or scarves,remains a significant challenge,as traditional models often fail to generalize under such conditions.This paper presents a hybrid approach that combines traditional handcrafted feature extraction technique called Histogram of Oriented Gradients(HOG)and Canny edge detection with modern deep learning models.The goal is to improve face detection accuracy under occlusions.The proposed method leverages the structural strengths of HOG and edge-based object proposals while exploiting the feature extraction capabilities of Convolutional Neural Networks(CNNs).The effectiveness of the proposed model is assessed using a custom dataset containing 10,000 heavily occluded face images and a subset of the Common Objects in Context(COCO)dataset for non-face samples.The COCO dataset was selected for its variety and realism in background contexts.Experimental evaluations demonstrate significant performance improvements compared to baseline CNN models.Results indicate that DenseNet121 combined with HOG outperforms other counterparts in classification metrics with an F1-score of 87.96%and precision of 88.02%.Enhanced performance is achieved through reduced false positives and improved localization accuracy with the integration of object proposals based on Canny and contour detection.While the proposed method increases inference time from 33.52 to 97.80 ms,it achieves a notable improvement in precision from 80.85% to 88.02% when comparing the baseline DenseNet121 model to its hybrid counterpart.Limitations of the method include higher computational cost and the need for careful tuning of parameters across the edge detection,handcrafted features,and CNN components.These findings highlight the potential of combining handcrafted and learned features for occluded face detection tasks.
基金supported by National Natural Science Foundation of China(61901071,61871062,61771082,U20A20157)Science and Natural Science Foundation of Chongqing,China(cstc2020jcyjzdxmX0024)+6 种基金University Innovation Research Group of Chongqing(CXQT20017)Program for Innovation Team Building at Institutions of Higher Education in Chongqing(CXTDX201601020)Natural Science Foundation of Chongqing,China(CSTB2022NSCQ-MSX0600)Youth Innovation Group Support Program of ICE Discipline of CQUPT(SCIE-QN-2022-04)Chongqing Municipal Technology Innovation and Application Development Special Key Project(cstc2020jscxdxwtBX0053)China Postdoctoral Science Foundation Project,China(2022MD723723)Chongqing Postdoctoral Research Project Special Funding,China(2023CQBSHTB3092)。
文摘The lack of facial features caused by wearing masks degrades the performance of facial recognition systems.Traditional occluded face recognition methods cannot integrate the computational resources of the edge layer and the device layer.Besides,previous research fails to consider the facial characteristics including occluded and unoccluded parts.To solve the above problems,we put forward a device-edge collaborative occluded face recognition method based on cross-domain feature fusion.Specifically,the device-edge collaborative face recognition architecture gets the utmost out of maximizes device and edge resources for real-time occluded face recognition.Then,a cross-domain facial feature fusion method is presented which combines both the explicit domain and the implicit domain facial.Furthermore,a delay-optimized edge recognition task scheduling method is developed that comprehensively considers the task load,computational power,bandwidth,and delay tolerance constraints of the edge.This method can dynamically schedule face recognition tasks and minimize recognition delay while ensuring recognition accuracy.The experimental results show that the proposed method achieves an average gain of about 21%in recognition latency,while the accuracy of the face recognition task is basically the same compared to the baseline method.
文摘The enrichment of chloride anion within the occluded cell （OC） for Type 304 austenitic stainless steel in low chloride concentration solution has been investigated by means of a simulated OC. The influence of the enrichment of chloride anion on stress corrosion crack （SCC） of Type 304 stainless steel has been studied. It was observed that the amount of chloride anion migration was proportional to the charge flowing through the anode. Owning to the effects of enrichment of chloride anion, low chloride concentration solution could induce SCC for Type 304 stainless steel.
文摘In patients with a malignant biliary obstruction who require biliary drainage,a self-expandable metallic stent(SEMS) provides longer patency duration than a plastic stent(PS).Nevertheless,a stent occlusion by tumor ingrowth,tumor overgrowth and biliary sludge may develop.There are several methods to manage occluded SEMS.Endoscopic management is the preferred treatment,whereas percutaneous intervention is an alternative approach.Endoscopic treatment involves mechanical cleaning with a balloon and a second stent insertion as stent-in-stent with either PS or SEMS.Technical feasibility,patient survival and cost-effectiveness are important factors that determine the method of re-drainage and stent selection.
基金National Major S&T Project“Formation Conditions,Accumulation Regularity and Target Evaluation of Large Gas Fields”(Grant No.2011ZX05007).
文摘Oil-cracked gas,as the main type of high-over mature marine natural gas in China,is mainly derived from occluded hydrocarbon.So it is significant to carry out quantitative study on occluded hydrocarbon.In this paper,the occluded hydrocarbon volume of the main basins in China was calculated depending on their types,abundances and evolution stages by means of the forward method(experimental simulation)and the inversion method(geologic profile dissection).And then,occluded hydrocarbon evolution models were established for five types of source rocks(sapropelic,sapropelic prone hybrid,humic prone hybrid,humic and coal).It is shown that the hydrocarbon expulsion efficiency of sapropelic and sapropelic prone hybrid excellent source rocks is lower than 30%at the low-maturity stage,30%e60%at the principal oil generation stage,and 50%e80%at the high-maturity stage,which are all about 10%higher than that of humic prone hybrid and humic source rocks at the corresponding stages.The resource distribution and cracked gas expulsion of occluded hydrocarbon since the high-maturity stage of marine source rocks in the Sichuan Basin were preliminarily calculated on the basis of the evolution models.The cracked gas expulsion is 230.4×10^(12) m^(3) at the high evolution stage of occluded hydrocarbon of the Lower Cambrian Qiongzhusi Fm in this basin,and 12.3×10^(12) m^(3) from the source rocks of Sinian Doushantuo Fm,indicating good potential for natural gas resources.It is indicated that the favorable areas of occluded hydrocarbon cracked gas in the Qiongzhusi Fm source rocks in the Sichuan Basin include GaoshitieMoxi,Ziyang and Weiyuan,covering a favorable area of 4.3×10^(4) km^(2).
文摘An artificial localized corrosion system is assembled and some parameters related to the localized corrosion in active dissolution state (i.e., non-passive state) have been studied. The results showed that the developed electrochemical system can satisfactorily imitate a naturally formed localized corrosion and the coupling current can indicate the maximum localized propagating rate. In this artificial system, the anodic dissolution reaction followed the auto-catalytic mechanism. The localized corrosion current density was dependent on the area ratio R of the cathode to the occluded anode. While R was equal to or more than 6, the coupling current reached at a maximum value and did not alter with the increase in R-value. Therefore, R=7 is chosen as one of these optimum parameters used in constructing the system, with which the biggest galvanic current might be obtained. In contrast, the thickness of the polymer filler separating the occluded anode area from the bulk electrolyte solution and the volume of the occluded anode area did not affect the corrosion current obviously. They might affect the response time to approach a steady state.
基金supported by the National Natural Science Foundation of China(grants No.41972163 and 42173055).
文摘Three kerogen samples(JJZG-1,JJZG-2 and JJZG-3)isolated from the Permian Lucaogou shales of varying biodegradation levels(BLs≈0,3 and 7,respectively)were subjected to sequential stepwise pyrolysis combined with on-line detection of gas chromatography-mass spectrometry(GC-MS).Occluded fractions(bitumenⅡ)released at low-temperature steps(≥410℃)show consistent biodegradative signatures with that reported for solvent-extracted fractions(bitumenⅠ)of the original shales,e.g.,broad range of abundant n-alkanes,isoprenoids and regular hopanes for the non-biodegraded JJZG-1;trace n-alkanes and abundant hopanes for the moderately biodegraded JJZG-2;and no n-alkanes but still prominent hopanes including the microbially produced 25-nohopanes for the severely biodegraded JJZG-3.This consistency between bitumenⅡand bitumenⅠfractions indicates the biodegradability of the kerogenoccluded bitumenⅡwith limited protection from host kerogen.A minor level of protection was suggested by the trace distribution of n-alkanes in the bitumenⅡof JJZG-2,whereas the bitumenⅠhad no nalkanes.The kerogen itself was more resistant to biodegradation as reflected by the persistence of high abundances of both n-alkanes and hopanes in the high temperature(≥460℃)products of all three kerogen samples.However,the relative abundances of these product groups did show some evidence of biodegradation alteration,e.g.,ratios of n-C_(15)alkene/C_(27)hop-17(21)-ene at 510℃pyrolysis decreased by order of magnitude from the non-biodegraded(JJZG-1=27.4)to highly biodegraded(0.3 for JJZG-3)samples.The reduced biodegradation impact on the kerogen fraction(Cf.bitumen fractions)was also evident by the absence of 25-norhopanes in the high-temperature analysis of the JJZG-3 kerogen.
基金supported by the National Key Research and Devel-opment Program of China under Grant No.2022YFE0102700Dr Yuhan Huang is a recipient of the ARC Discovery Early Career Research Award(DE220100552).
文摘Autonomous driving is a promising way to future safe,efficient,and low-carbon transportation.Real-time ac-curate target detection is an essential precondition for the generation of proper following decision and control signals.However,considering the complex practical scenarios,accurate recognition of occluded targets is a major challenge of target detection for autonomous driving with limited computational capability.To reveal the overlap and difference between various occluded object detection by sharing the same available sensors,this paper presents a review of detection methods for occluded objects in complex real-driving scenarios.Considering the rapid development of autonomous driving technologies,the research analyzed in this study is limited to the recent five years.The study of occluded object detection is divided into three parts,namely occluded vehicles,pedes-trians and traffic signs.This paper provided a detailed summary of the target detection methods used in these three parts according to the differences in detection methods and ideas,which is followed by the comparison of advantages and disadvantages of different detection methods for the same object.Finally,the shortcomings and limitations of the existing detection methods are summarized,and the challenges and future development prospects in this field are discussed.
基金Chinese Academy of Medical Sciences Fuwai Hospital High Level Hospital clinical research fund(2022-GSPGG-18).
文摘Compared with traditional nickel-titanium alloy patent foramen ovale occluders,which are widely used in clinical practice,biodegradable patent foramen ovale occluders have obvious differences in material characteristics,interventional operation mode and postoperative management strategy.This article gives expert consensus on the selection of clinical indications and standardized operating procedures,so as to standardize the clinical application of biodegradable patent foramen ovale occluders.
基金Project(50974046/E041607) supported by the National Natural Science Foundation of China
文摘Dissimilar metal joining between NiTi shape memory alloy（SMA） and stainless steel was conducted.A cluster of NiTi SMA wires were first joined with tungsten inert gas（TIG） welding process,then the NiTi SMA TIG weld was welded to a stainless steel pipe with laser spot welding process.The microstructure of the welds was examined with an optical microscope and the elemental distribution in the welds was measured by electron probe microanalysis（EPMA）.The results show that TiC compounds dispersively distribute in the NiTi SMA TIG weld.However,the amount of TiC compounds greatly decreases around the fusion boundary of the laser spot weld between the NiTi SMA and stainless steel.Mutual diffusion between NiTi shape memory alloy and stainless steel happen within a short distance near the fusion boundary,and intermetallic compounds such as Ni3Ti＋（Fe,Ni）Ti appear around the fusion boundary.
文摘Video surveillance is an active research topic in computer vision.In this paper,humans and cars identifcation technique suitable for real time video surveillance systems is presented.The technique we proposed includes background subtraction,foreground segmentation,shadow removal,feature extraction and classifcation.The feature extraction of the extracted foreground objects is done via a new set of afne moment invariants based on statistics method and these were used to identify human or car.When the partial occlusion occurs,although features of full body cannot be extracted,our proposed technique extracts the features of head shoulder.Our proposed technique can identify human by extracting the human head-shoulder up to 60%–70%occlusion.Thus,it has a better classifcation to solve the issue of the loss of property arising from human occluded easily in practical applications.The whole system works at approximately 16 29 fps and thus it is suitable for real-time applications.The accuracy for our proposed technique in identifying human is very good,which is 98.33%,while for cars identifcation,the accuracy is also good,which is 94.41%.The overall accuracy for our proposed technique in identifying human and car is at 98.04%.The experiment results show that this method is efective and has strong robustness.
基金financially supported by Jiangsu Graduate Student Innovation Program of China(No.CXZZ-0146)the Scientif ic Research Foundation of Graduate School of Southeast University(No.YBJJ1235)
文摘In order to provide scientific basis for advanced applications of near eutectic Al-Si-Mg alloys as architectural profiles, a comparative study on the corrosion resistance of an as-extruded near eutectic AI-Si-Mg alloy and AA6063 aluminium alloy was carried out by means of neutral salt spray test. The corroded surfaces of the alloys were examined with optical microscopy and scanning electron microscope （SEM）. Results show that the corrosion type of these two alloys is pitting corrosion. The number of corrosion pits in the AA6063 aluminium alloy is more than that in the near eutectic AI-Si-Mg alloy, but the pits in the latter alloy are much larger and deeper. Because the relatively low polarization resistance of the near eutectic alloy leads to poorer repassivation ability, autocatalytic acidification occurs once a pit forms. Thus, occluded corrosion cells are developed in this alloy.
基金Project(61172089) supported by the National Natural Science Foundation of China
文摘The difficulty of multiple targets tracking is how to quickly fulfill the target matching from one flame image to another and fix the position of the target. In order to accurately choose target feature information for reliable matching, simplify operations under the reliable precondition, and realize precise moving objects tracking, an approach based on Kalman prediction and feature matching was proposed. The position of the target in next frame image was predicted by Kalman, and then the moving objects of two adjacent frames were matched by the centroid and area methods. When occlusion occurs, the best matching result was found to realize tracking by matching matrix algorithm. The simulation results show that the proposed method can achieve multiple targets tracking accurately and in real-time under complicated motion movements.
基金Project(50878191) supported by the National Natural Science Foundation of ChinaProject(Y12E090030) supported by Zhejiang Provincial Natural Science Foundation, China
文摘BISHOP’s effective stress or two state stress variables are unsatisfactory for unsaturated soils where one of fluid phases is discontinuous, so new expressions of effective stress should be founded. The approach for derivation was according to the principle of equilibrium of forces (i.e., the stress-sharing principle), and it was firstly validated by demonstrating TERZAGHI’s principle of effective stress. And then, the derivations were subdivided into four parts according to different pore air states: 1) air bubbles were spherical and suspended in pore water; 2) air bubbles were bound on soil skeleton; 3) air bubbles held almost the single section of pore; 4) air phase was continuous. The different formulae of effective stress were presented. Conclusions are drawn as follows: 1) For nearly-saturated soils, the "real" effective stress would be a little smaller than TERZAGHI’s effective stress; 2) For soils in which air phase is discontinuous in the form of bubbles, a new concept of pore air elastic pressure is put forward, and the total stress can be constituted by effective stress, pore water pressure and pore air elastic pressure; 3) For soils in which air phase is continuous, effective stress is equal to the value of the total stress plus suction; 4) Suction can be divided into two parts: one is the effect caused by additional pressure, and the other is the contract action by the "skin".
文摘Chronic thromboembolic pulmonary hypertension (CTEPH) comprises organizing thrombotic obstructions in the pulmonary arteries by nonresolving thromboemboli, formation of fibrosis and remodeling of pulmonary blood vessels. Surgical pulmonary endarterectomy (PEA) is the therapy of choice for patients with surgically accessible CTEPH, which leads to a profound improvement in hemodynamics, functional class and survival. Select- ing the candidates that will benefit from surgery is still a challenging task. Criteria for surgical suitability have been described but the decision-making for or against surgical intervention remains still subjective. The optimal characterization of the reciprocal contribution of large vessel and small vessel disease in the elevation of pulmonary vascular resistance is crucial for the indication and outcome of PEA. Recently, Toshner et al intended to validate the partition resistance into small and large vessels compartments (upstream resistance:Rup) by the occlusion technique in the preoperative assessment of PEA. We discuss the advantages and disadvantages of Rup and compare it with other hemodynamic predictor to evaluate operative risk in CTEPH patients.
文摘Biometric recognition refers to the process of recognizing a person’s identity using physiological or behavioral modalities,such as face,voice,fingerprint,gait,etc.Such biometric modalities are mostly used in recognition tasks separately as in unimodal systems,or jointly with two or more as in multimodal systems.However,multimodal systems can usually enhance the recognition performance over unimodal systems by integrating the biometric data of multiple modalities at different fusion levels.Despite this enhancement,in real-life applications some factors degrade multimodal systems’performance,such as occlusion,face poses,and noise in voice data.In this paper,we propose two algorithms that effectively apply dynamic fusion at feature level based on the data quality of multimodal biometrics.The proposed algorithms attempt to minimize the negative influence of confusing and low-quality features by either exclusion or weight reduction to achieve better recognition performance.The proposed dynamic fusion was achieved using face and voice biometrics,where face features were extracted using principal component analysis(PCA),and Gabor filters separately,whilst voice features were extracted using Mel-Frequency Cepstral Coefficients(MFCCs).Here,the facial data quality assessment of face images is mainly based on the existence of occlusion,whereas the assessment of voice data quality is substantially based on the calculation of signal to noise ratio(SNR)as per the existence of noise.To evaluate the performance of the proposed algorithms,several experiments were conducted using two combinations of three different databases,AR database,and the extended Yale Face Database B for face images,in addition to VOiCES database for voice data.The obtained results show that both proposed dynamic fusion algorithms attain improved performance and offer more advantages in identification and verification over not only the standard unimodal algorithms but also the multimodal algorithms using standard fusion methods.
基金supported by the Natural Science Foundation of Shanghai under Grant 21ZR1426500the National Natural Science Foundation of China under Grant 61873160.
文摘Person search mainly consists of two submissions,namely Person Detection and Person Re-identification(reID).Existing approaches are primarily based on Faster R-CNN and Convolutional Neural Network(CNN)(e.g.,ResNet).While these structures may detect high-quality bounding boxes,they seem to degrade the performance of re-ID.To address this issue,this paper proposes a Dual-Transformer Head Network(DTHN)for end-to-end person search,which contains two independent Transformer heads,a box head for detecting the bounding box and extracting efficient bounding box feature,and a re-ID head for capturing high-quality re-ID features for the re-ID task.Specifically,after the image goes through the ResNet backbone network to extract features,the Region Proposal Network(RPN)proposes possible bounding boxes.The box head then extracts more efficient features within these bounding boxes for detection.Following this,the re-ID head computes the occluded attention of the features in these bounding boxes and distinguishes them from other persons or backgrounds.Extensive experiments on two widely used benchmark datasets,CUHK-SYSU and PRW,achieve state-of-the-art performance levels,94.9 mAP and 95.3 top-1 scores on the CUHK-SYSU dataset,and 51.6 mAP and 87.6 top-1 scores on the PRW dataset,which demonstrates the advantages of this paper’s approach.The efficiency comparison also shows our method is highly efficient in both time and space.
文摘The paper deals with the fluid field of web forming in wet-laid non-woven production.The influence of the turbulent flow on blending fiber and occluded fluid produced in pulp flow has been discussed in theory and practice.The suitable use of the imported velocity of pulp is very important in producing wet-laid products of good quality.
