Quantum light sources are the core resources for photonics-based quantum information processing.We investigate the spectral engineering of photon triplets generated by third-order spontaneous parametric down-conversio...Quantum light sources are the core resources for photonics-based quantum information processing.We investigate the spectral engineering of photon triplets generated by third-order spontaneous parametric down-conversion in micro/nanofiber.The phase mismatching at one-third pump frequency gives rise to non-degenerate photon triplets,the joint spectral intensity of which has an elliptical locus with a fixed eccentricity of√6/3.Therefore,we propose a frequency-division scheme to separate non-degenerate photon triplets into three channels with high heralding efficiency for the first time.Choosing an appropriate pump wavelength can compensate for the fabrication errors of micro/nanofiber and also generate narrowband,non-degenerate photon triplet sources with a high signal-to-noise ratio.Furthermore,the long-period micro/nanofiber grating introduces a new controllable degree of freedom to tailor phase matching,resulting from the periodic oscillation of dispersion.In this scheme,the wavelength of photon triplets can be flexibly tuned using quasi-phase matching.We study the generation of photon triplets from this novel perspective of spectrum engineering,and we believe that this work will accelerate the practical implementation of photon triplets in quantum information processing.展开更多
Sensitizing molecular triplets by colloidal nanocrystals via triplet energy transfer is important for applications such as upconversion or organic synthesis.Typically two step triplet energy transfer(TET)are included ...Sensitizing molecular triplets by colloidal nanocrystals via triplet energy transfer is important for applications such as upconversion or organic synthesis.Typically two step triplet energy transfer(TET)are included in these applications:firstly the triplet energy stored in nanocrystals are extracted into surface ligands,and then the ligands further transfer triplet energy into molecules in bulk solution.Here we report one-step TET application from CsPbBr_(3) perovskite nanocrystals(NCs)to surface-anchored metalloporphyrin derivative molecules(MP).Compared to conventional two-step TET,the one-step TET mechanism possess lower energy loss and higher TET efficiency which is more generally implementable.In this scheme,photoexcitation of CsPbBr_(3)NCs leads to the sensitization of MP ligands triplets which efficiently emit phosphorescence.The enhanced light absorption of MP ligands and down-shifted photon emission can be useful in devices such as luminescent solar concentrators.展开更多
This work aimed at investigating the crucial factor in building and maintaining the combustion front during in-situ combustion(ISC),oxidized coke and pyrolyzed coke.The surface morphologies,elemental contents,and non-...This work aimed at investigating the crucial factor in building and maintaining the combustion front during in-situ combustion(ISC),oxidized coke and pyrolyzed coke.The surface morphologies,elemental contents,and non-isothermal mass losses of the oxidized and pyrolyzed cokes were thoroughly examined.The results indicated that the oxidized coke could be combusted at a lower temperature compared to the pyrolyzed coke due primarily to their differences in the molecular polarity and microstructure.Kinetic triplets of coke combustion were determined using iso-conversional models and one advanced integral master plots method.The activation energy values of the oxidized and pyrolyzed cokes varied in the range of 130-153 k J/mol and 95-120 kJ/mol,respectively.The most appropriate reaction model of combustion of the oxidized and pyrolyzed cokes followed three-dimensional diffusion model(D_(3)) and random nucleation and subsequent growth model(F_(1)),respectively.These observations assisted in building the numerical model of these two types of cokes to simulate the ISC process.展开更多
We study the stability of vortices pinning and dynamics in a superconducting thin strip containing a square array of antidot triplets by using the nonlinear Ginzburg–Landau(GL)theory.Compared with the regular square ...We study the stability of vortices pinning and dynamics in a superconducting thin strip containing a square array of antidot triplets by using the nonlinear Ginzburg–Landau(GL)theory.Compared with the regular square array of circular holes,the vortices are no longer pinned inside the circular holes,but instead stabilized at the center of the antidot triplets depending on the geometry parameters.Moreover,the influences of the geometry parameters and the polarity of the applied current on the current–voltage(I–V)characteristics are also studied.The critical current for the sample turning into a normal state becomes smaller when the hole diameter D is smaller and the spacing B between the holes is larger.Due to the asymmetric pinning sites,our numerical simulations demonstrate that the positive and negative rectified voltages appear alternately in the resistive state of the sample under an ac current of square pulses.展开更多
The electroluminescence (EL) produced by a highly luminescent phosphorescent dye Cu-4(CdropCPh)(4)L-2 (L = 1.8-bis(di-phenylphosphino)-3,6-dioxaoctane, Cu-4) doped polymer as emitting layer is reported. The effects of...The electroluminescence (EL) produced by a highly luminescent phosphorescent dye Cu-4(CdropCPh)(4)L-2 (L = 1.8-bis(di-phenylphosphino)-3,6-dioxaoctane, Cu-4) doped polymer as emitting layer is reported. The effects of the charge injection balance on the polymers, in particular, poly(N-vinylcarbazole) (PVK) have been studied by using photoluminescence and electroluminescence spectroscopy. Changes in the emission spectra demonstrate the influence of the charge injection balance on the formation ratio of triplet and singlet excitons. This provides a new technical approach to realize the color patterning in polymer LEDs.展开更多
Prime integers and their generalizations play important roles in protocols for secure transmission of information via open channels of telecommunication networks. Generation of multidigit large primes in the design st...Prime integers and their generalizations play important roles in protocols for secure transmission of information via open channels of telecommunication networks. Generation of multidigit large primes in the design stage of a cryptographic system is a formidable task. Fermat primality checking is one of the simplest of all tests. Unfortunately, there are composite integers (called Carmichael numbers) that are not detectable by the Fermat test. In this paper we consider modular arithmetic based on complex integers;and provide several tests that verify the primality of real integers. Although the new tests detect most Carmichael numbers, there are a small percentage of them that escape these tests.展开更多
Triplets are seldom and monozygotic triplets are scarcely seen in humanity. There are few reports on triplets at home and abroad and diagnosis of zygosity is rarely made. We did the analysis of clinical material on tw...Triplets are seldom and monozygotic triplets are scarcely seen in humanity. There are few reports on triplets at home and abroad and diagnosis of zygosity is rarely made. We did the analysis of clinical material on two sets of triplets. Diagnosis and analysis of zygosity were made in eleven ways about 57 items including blood type and RBC enzyme type, etc.展开更多
Pt(Ⅱ)-salophen complexes(S-1~S-4) and 9,10-diphenylanthracene(DPA) tethering pillar[5]arene derivatives(A-1 and A-2) were synthesized to act as sensitizers and annihilators for triplet-triplet annihilation upconversi...Pt(Ⅱ)-salophen complexes(S-1~S-4) and 9,10-diphenylanthracene(DPA) tethering pillar[5]arene derivatives(A-1 and A-2) were synthesized to act as sensitizers and annihilators for triplet-triplet annihilation upconversion(TTA-UC), respectively. It turned out that the pyridine cation served as a mask for the excited state of the sensitizer, the triplet states of S-2 and S-3 were significantly quenched by photo-induced electron transfer(PET) with phosphorescence quantum yield quenched from 24.4% for S-4 to 9.3% for S-3,and therefore, both S-2 and S-3 led to negligible UC emissions when traditional annihilator DPA was used as the annihilator. Delightfully, when supramolecular annihilator A-1 and A-2 were employed to include the pyridine cation, PET was significantly inhibited and the triplet states of the sensitizers were activated,TTA-UC emission was therefore boosted. The UC quantum yield of A-2/S-3 system was up to 130 times higher than that of DPA/S-3 system, and the UC emission was switchable by the addition of competitive vips.展开更多
The distributed fiber optic sensing system,known for its high sensitivity and wide-ranging measurement capabilities,has been widely used in monitoring underground gas pipelines.It primarily serves to perceive vibratio...The distributed fiber optic sensing system,known for its high sensitivity and wide-ranging measurement capabilities,has been widely used in monitoring underground gas pipelines.It primarily serves to perceive vibration signals induced by external events and to effectively provide early warnings of potential intrusion activities.Due to the complexity and diversity of external intrusion events,traditional deep learning methods can achieve event recognition with an average accuracy exceeding 90%.However,these methods rely on large-scale datasets,leading to significant time and labor costs during the data collection process.Additionally,traditional methods perform poorly when faced with the scarcity of low-frequency event samples,making it challenging to address these rare occurrences.To address this issue,this paper proposes a small-sample learning model based on triplet learning for intrusion event recognition.The model employs a 6-way 20-shot support set configuration and utilizes the KNN clustering algorithm to assess the model's performance.Experimental results indicate that the model achieves an average accuracy of 91.6%,further validating the superior performance of the triplet learning model in classifying external intrusion events.Compared to traditional methods,this approach not only effectively reduces the dependence on large-scale datasets but also better addresses the classification of low-frequency event samples,demonstrating significant application potential.展开更多
In recent years,with the rapid development of deep learning technology,relational triplet extraction techniques have also achieved groundbreaking progress.Traditional pipeline models have certain limitations due to er...In recent years,with the rapid development of deep learning technology,relational triplet extraction techniques have also achieved groundbreaking progress.Traditional pipeline models have certain limitations due to error propagation.To overcome the limitations of traditional pipeline models,recent research has focused on jointly modeling the two key subtasks-named entity recognition and relation extraction-within a unified framework.To support future research,this paper provides a comprehensive review of recently published studies in the field of relational triplet extraction.The review examines commonly used public datasets for relational triplet extraction techniques and systematically reviews current mainstream joint extraction methods,including joint decoding methods and parameter sharing methods,with joint decoding methods further divided into table filling,tagging,and sequence-to-sequence approaches.In addition,this paper also conducts small-scale replication experiments on models that have performed well in recent years for each method to verify the reproducibility of the code and to compare the performance of different models under uniform conditions.Each method has its own advantages in terms of model design,task handling,and application scenarios,but also faces challenges such as processing complex sentence structures,cross-sentence relation extraction,and adaptability in low-resource environments.Finally,this paper systematically summarizes each method and discusses the future development prospects of joint extraction of relational triples.展开更多
Accurate defect detection plays a critical role in ensuring product quality and equipment reliability.Small-object detection poses unique challenges due to weak feature representation and significant background interf...Accurate defect detection plays a critical role in ensuring product quality and equipment reliability.Small-object detection poses unique challenges due to weak feature representation and significant background interference.To address these issues,this study incorporates three key innovations into the YOLOv8 framework:the use of GhostNet convolution for lightweight and efficient feature extraction,the addition of a P2 detection layer to enhance small-object detection capabilities,and the integration of the Triplet Attention mechanism to capture comprehensive spatial and channel dependencies.These improvements collectively optimize detection performance for small objects while reducing computational complexity.Experimental results demonstrate that the enhanced model achieves a mean average precision(mAP@0.5)of 97.46%and a mAP@0.5∶0.95 of 61.84%,representing a performance improvement of 1.9%and 3.2%,respectively,compared to the baseline YOLOv8 model.Additionally,the model achieves a frame rate of 158 FPS,maintaining real-time detection capabilities while reducing the parameter count by 50%,further underscoring its efficiency and suitability for smallobject detection in complex scenarios.展开更多
Image-based similar trademark retrieval is a time-consuming and labor-intensive task in the trademark examination process.This paper aims to support trademark examiners by training Deep Convolutional Neural Network(DC...Image-based similar trademark retrieval is a time-consuming and labor-intensive task in the trademark examination process.This paper aims to support trademark examiners by training Deep Convolutional Neural Network(DCNN)models for effective Trademark Image Retrieval(TIR).To achieve this goal,we first develop a novel labeling method that automatically generates hundreds of thousands of labeled similar and dissimilar trademark image pairs using accompanying data fields such as citation lists,Vienna classification(VC)codes,and trademark ownership information.This approach eliminates the need for manual labeling and provides a large-scale dataset suitable for training deep learning models.We then train DCNN models based on Siamese and Triplet architectures,evaluating various feature extractors to determine the most effective configuration.Furthermore,we present an Adapted Contrastive Loss Function(ACLF)for the trademark retrieval task,specifically engineered to mitigate the influence of noisy labels found in automatically created datasets.Experimental results indicate that our proposed model(Efficient-Net_v21_Siamese)performs best at both True Negative Rate(TNR)threshold levels,TNR 0.9 and TNR 0.95,with==respective True Positive Rates(TPRs)of 77.7%and 70.8%and accuracies of 83.9%and 80.4%.Additionally,when testing on the public trademark dataset METU_v2,our model achieves a normalized average rank(NAR)of 0.0169,outperforming the current state-of-the-art(SOTA)model.Based on these findings,we estimate that considering only approximately 10%of the returned trademarks would be sufficient,significantly reducing the review time.Therefore,the paper highlights the potential of utilizing national trademark data to enhance the accuracy and efficiency of trademark retrieval systems,ultimately supporting trademark examiners in their evaluation tasks.展开更多
Heterostructures of organic semi-conductors and transition metal dichalcogenides(TMDs)are viable candidates for superior optoelec-tronic devices.Photoinduced inter-facial charge transfer is crucial for the performance...Heterostructures of organic semi-conductors and transition metal dichalcogenides(TMDs)are viable candidates for superior optoelec-tronic devices.Photoinduced inter-facial charge transfer is crucial for the performance efficiency of such devices,yet the underlying mecha-nism,especially the roles of optical-ly dark triplets and spatially sepa-rated charge transfer states,is poorly understood.In the present work,we obtain the struc-tures of distinct excited states and investigate how they are involved in the charge transfer process at the Pd-octaethylporphyrin(PdOEP)and WS_(2) interface in terms of their energies and couplings.The results show that electron transfer from the triplet PdOEP formed via intersystem crossing prevails over direct electron transfer from the singlet(two orders of magnitude faster).Further analysis reveals that the relatively higher rate of triplet electron transfer compared to singlet electron transfer is mainly attributed to a smaller reorganization energy,which is dominated by the out-of-plane vibrations of the organic component.The work emphasizes the important roles of the optically dark triplets in the electron transfer of the PdOEP@WS_(2) heterostructure,and provides valuable theoretical insights for further improv-ing the optoelectronic performance of TMD-based devices.展开更多
基金Project supported by the National Natural Science Foundation of China(Grant No.61605249)the Science and Technology Key Project of Henan Province of China(Grant Nos.182102210577 and 232102211086).
文摘Quantum light sources are the core resources for photonics-based quantum information processing.We investigate the spectral engineering of photon triplets generated by third-order spontaneous parametric down-conversion in micro/nanofiber.The phase mismatching at one-third pump frequency gives rise to non-degenerate photon triplets,the joint spectral intensity of which has an elliptical locus with a fixed eccentricity of√6/3.Therefore,we propose a frequency-division scheme to separate non-degenerate photon triplets into three channels with high heralding efficiency for the first time.Choosing an appropriate pump wavelength can compensate for the fabrication errors of micro/nanofiber and also generate narrowband,non-degenerate photon triplet sources with a high signal-to-noise ratio.Furthermore,the long-period micro/nanofiber grating introduces a new controllable degree of freedom to tailor phase matching,resulting from the periodic oscillation of dispersion.In this scheme,the wavelength of photon triplets can be flexibly tuned using quasi-phase matching.We study the generation of photon triplets from this novel perspective of spectrum engineering,and we believe that this work will accelerate the practical implementation of photon triplets in quantum information processing.
基金This work is supported by the National Natural Science Foundation of China(No.21803070).
文摘Sensitizing molecular triplets by colloidal nanocrystals via triplet energy transfer is important for applications such as upconversion or organic synthesis.Typically two step triplet energy transfer(TET)are included in these applications:firstly the triplet energy stored in nanocrystals are extracted into surface ligands,and then the ligands further transfer triplet energy into molecules in bulk solution.Here we report one-step TET application from CsPbBr_(3) perovskite nanocrystals(NCs)to surface-anchored metalloporphyrin derivative molecules(MP).Compared to conventional two-step TET,the one-step TET mechanism possess lower energy loss and higher TET efficiency which is more generally implementable.In this scheme,photoexcitation of CsPbBr_(3)NCs leads to the sensitization of MP ligands triplets which efficiently emit phosphorescence.The enhanced light absorption of MP ligands and down-shifted photon emission can be useful in devices such as luminescent solar concentrators.
基金supported by Chinese Postdoctoral Science Foundation (2021M692696)the National Science and Technology Project (2016ZX05058-003-017)Sichuan Science and Technology Program (2021YFH0081)。
文摘This work aimed at investigating the crucial factor in building and maintaining the combustion front during in-situ combustion(ISC),oxidized coke and pyrolyzed coke.The surface morphologies,elemental contents,and non-isothermal mass losses of the oxidized and pyrolyzed cokes were thoroughly examined.The results indicated that the oxidized coke could be combusted at a lower temperature compared to the pyrolyzed coke due primarily to their differences in the molecular polarity and microstructure.Kinetic triplets of coke combustion were determined using iso-conversional models and one advanced integral master plots method.The activation energy values of the oxidized and pyrolyzed cokes varied in the range of 130-153 k J/mol and 95-120 kJ/mol,respectively.The most appropriate reaction model of combustion of the oxidized and pyrolyzed cokes followed three-dimensional diffusion model(D_(3)) and random nucleation and subsequent growth model(F_(1)),respectively.These observations assisted in building the numerical model of these two types of cokes to simulate the ISC process.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.11702034,11702218,and 11421062)Fundamental Research Funds for the Central Universities,China(Grant Nos.310812171011 and G2016KY0305)the National Key Project of Magneto-Constrained Fusion Energy Development Program,China(Grant No.2013GB110002)
文摘We study the stability of vortices pinning and dynamics in a superconducting thin strip containing a square array of antidot triplets by using the nonlinear Ginzburg–Landau(GL)theory.Compared with the regular square array of circular holes,the vortices are no longer pinned inside the circular holes,but instead stabilized at the center of the antidot triplets depending on the geometry parameters.Moreover,the influences of the geometry parameters and the polarity of the applied current on the current–voltage(I–V)characteristics are also studied.The critical current for the sample turning into a normal state becomes smaller when the hole diameter D is smaller and the spacing B between the holes is larger.Due to the asymmetric pinning sites,our numerical simulations demonstrate that the positive and negative rectified voltages appear alternately in the resistive state of the sample under an ac current of square pulses.
基金This work was supported by the National Nature Foundation of China (No. 597905006).
文摘The electroluminescence (EL) produced by a highly luminescent phosphorescent dye Cu-4(CdropCPh)(4)L-2 (L = 1.8-bis(di-phenylphosphino)-3,6-dioxaoctane, Cu-4) doped polymer as emitting layer is reported. The effects of the charge injection balance on the polymers, in particular, poly(N-vinylcarbazole) (PVK) have been studied by using photoluminescence and electroluminescence spectroscopy. Changes in the emission spectra demonstrate the influence of the charge injection balance on the formation ratio of triplet and singlet excitons. This provides a new technical approach to realize the color patterning in polymer LEDs.
文摘Prime integers and their generalizations play important roles in protocols for secure transmission of information via open channels of telecommunication networks. Generation of multidigit large primes in the design stage of a cryptographic system is a formidable task. Fermat primality checking is one of the simplest of all tests. Unfortunately, there are composite integers (called Carmichael numbers) that are not detectable by the Fermat test. In this paper we consider modular arithmetic based on complex integers;and provide several tests that verify the primality of real integers. Although the new tests detect most Carmichael numbers, there are a small percentage of them that escape these tests.
文摘Triplets are seldom and monozygotic triplets are scarcely seen in humanity. There are few reports on triplets at home and abroad and diagnosis of zygosity is rarely made. We did the analysis of clinical material on two sets of triplets. Diagnosis and analysis of zygosity were made in eleven ways about 57 items including blood type and RBC enzyme type, etc.
基金supported by the National Natural Science Foundation of China (Nos. 22171194, 21971169, 92056116 and 21871194)the Fundamental Research Funds for the Central Universities (No. 20826041D4117)the Science & Technology Department of Sichuan Province (Nos. 2022YFH0095 and 2021ZYD0052)。
文摘Pt(Ⅱ)-salophen complexes(S-1~S-4) and 9,10-diphenylanthracene(DPA) tethering pillar[5]arene derivatives(A-1 and A-2) were synthesized to act as sensitizers and annihilators for triplet-triplet annihilation upconversion(TTA-UC), respectively. It turned out that the pyridine cation served as a mask for the excited state of the sensitizer, the triplet states of S-2 and S-3 were significantly quenched by photo-induced electron transfer(PET) with phosphorescence quantum yield quenched from 24.4% for S-4 to 9.3% for S-3,and therefore, both S-2 and S-3 led to negligible UC emissions when traditional annihilator DPA was used as the annihilator. Delightfully, when supramolecular annihilator A-1 and A-2 were employed to include the pyridine cation, PET was significantly inhibited and the triplet states of the sensitizers were activated,TTA-UC emission was therefore boosted. The UC quantum yield of A-2/S-3 system was up to 130 times higher than that of DPA/S-3 system, and the UC emission was switchable by the addition of competitive vips.
基金Supported by the Scientific Research and Technology Development Project of Petrochina Southwest Oil and Gas Field Company(20230307-02)。
文摘The distributed fiber optic sensing system,known for its high sensitivity and wide-ranging measurement capabilities,has been widely used in monitoring underground gas pipelines.It primarily serves to perceive vibration signals induced by external events and to effectively provide early warnings of potential intrusion activities.Due to the complexity and diversity of external intrusion events,traditional deep learning methods can achieve event recognition with an average accuracy exceeding 90%.However,these methods rely on large-scale datasets,leading to significant time and labor costs during the data collection process.Additionally,traditional methods perform poorly when faced with the scarcity of low-frequency event samples,making it challenging to address these rare occurrences.To address this issue,this paper proposes a small-sample learning model based on triplet learning for intrusion event recognition.The model employs a 6-way 20-shot support set configuration and utilizes the KNN clustering algorithm to assess the model's performance.Experimental results indicate that the model achieves an average accuracy of 91.6%,further validating the superior performance of the triplet learning model in classifying external intrusion events.Compared to traditional methods,this approach not only effectively reduces the dependence on large-scale datasets but also better addresses the classification of low-frequency event samples,demonstrating significant application potential.
基金funding from Key Areas Science and Technology Research Plan of Xinjiang Production And Construction Corps Financial Science and Technology Plan Project under Grant Agreement No.2023AB048 for the project:Research and Application Demonstration of Data-driven Elderly Care System.
文摘In recent years,with the rapid development of deep learning technology,relational triplet extraction techniques have also achieved groundbreaking progress.Traditional pipeline models have certain limitations due to error propagation.To overcome the limitations of traditional pipeline models,recent research has focused on jointly modeling the two key subtasks-named entity recognition and relation extraction-within a unified framework.To support future research,this paper provides a comprehensive review of recently published studies in the field of relational triplet extraction.The review examines commonly used public datasets for relational triplet extraction techniques and systematically reviews current mainstream joint extraction methods,including joint decoding methods and parameter sharing methods,with joint decoding methods further divided into table filling,tagging,and sequence-to-sequence approaches.In addition,this paper also conducts small-scale replication experiments on models that have performed well in recent years for each method to verify the reproducibility of the code and to compare the performance of different models under uniform conditions.Each method has its own advantages in terms of model design,task handling,and application scenarios,but also faces challenges such as processing complex sentence structures,cross-sentence relation extraction,and adaptability in low-resource environments.Finally,this paper systematically summarizes each method and discusses the future development prospects of joint extraction of relational triples.
文摘Accurate defect detection plays a critical role in ensuring product quality and equipment reliability.Small-object detection poses unique challenges due to weak feature representation and significant background interference.To address these issues,this study incorporates three key innovations into the YOLOv8 framework:the use of GhostNet convolution for lightweight and efficient feature extraction,the addition of a P2 detection layer to enhance small-object detection capabilities,and the integration of the Triplet Attention mechanism to capture comprehensive spatial and channel dependencies.These improvements collectively optimize detection performance for small objects while reducing computational complexity.Experimental results demonstrate that the enhanced model achieves a mean average precision(mAP@0.5)of 97.46%and a mAP@0.5∶0.95 of 61.84%,representing a performance improvement of 1.9%and 3.2%,respectively,compared to the baseline YOLOv8 model.Additionally,the model achieves a frame rate of 158 FPS,maintaining real-time detection capabilities while reducing the parameter count by 50%,further underscoring its efficiency and suitability for smallobject detection in complex scenarios.
基金funded by the Institute of InformationTechnology,VietnamAcademy of Science and Technology(project number CSCL02.02/22-23)“Research and Development of Methods for Searching Similar Trademark Images Using Machine Learning to Support Trademark Examination in Vietnam”.
文摘Image-based similar trademark retrieval is a time-consuming and labor-intensive task in the trademark examination process.This paper aims to support trademark examiners by training Deep Convolutional Neural Network(DCNN)models for effective Trademark Image Retrieval(TIR).To achieve this goal,we first develop a novel labeling method that automatically generates hundreds of thousands of labeled similar and dissimilar trademark image pairs using accompanying data fields such as citation lists,Vienna classification(VC)codes,and trademark ownership information.This approach eliminates the need for manual labeling and provides a large-scale dataset suitable for training deep learning models.We then train DCNN models based on Siamese and Triplet architectures,evaluating various feature extractors to determine the most effective configuration.Furthermore,we present an Adapted Contrastive Loss Function(ACLF)for the trademark retrieval task,specifically engineered to mitigate the influence of noisy labels found in automatically created datasets.Experimental results indicate that our proposed model(Efficient-Net_v21_Siamese)performs best at both True Negative Rate(TNR)threshold levels,TNR 0.9 and TNR 0.95,with==respective True Positive Rates(TPRs)of 77.7%and 70.8%and accuracies of 83.9%and 80.4%.Additionally,when testing on the public trademark dataset METU_v2,our model achieves a normalized average rank(NAR)of 0.0169,outperforming the current state-of-the-art(SOTA)model.Based on these findings,we estimate that considering only approximately 10%of the returned trademarks would be sufficient,significantly reducing the review time.Therefore,the paper highlights the potential of utilizing national trademark data to enhance the accuracy and efficiency of trademark retrieval systems,ultimately supporting trademark examiners in their evaluation tasks.
基金supported by the Fundamental Re-search Funds for the Central Universities(Ganglong Cui)and National Key Research and Development Pro-gram of China(No.2021YFA1500703 to Ganglong Cui)National Natural Science Foundation of China(No.22103067 to Xiao-Ying Xie)and Natural Science Foundation of Shandong Province(No.ZR2021QB105 to Xiao-Ying Xie).
文摘Heterostructures of organic semi-conductors and transition metal dichalcogenides(TMDs)are viable candidates for superior optoelec-tronic devices.Photoinduced inter-facial charge transfer is crucial for the performance efficiency of such devices,yet the underlying mecha-nism,especially the roles of optical-ly dark triplets and spatially sepa-rated charge transfer states,is poorly understood.In the present work,we obtain the struc-tures of distinct excited states and investigate how they are involved in the charge transfer process at the Pd-octaethylporphyrin(PdOEP)and WS_(2) interface in terms of their energies and couplings.The results show that electron transfer from the triplet PdOEP formed via intersystem crossing prevails over direct electron transfer from the singlet(two orders of magnitude faster).Further analysis reveals that the relatively higher rate of triplet electron transfer compared to singlet electron transfer is mainly attributed to a smaller reorganization energy,which is dominated by the out-of-plane vibrations of the organic component.The work emphasizes the important roles of the optically dark triplets in the electron transfer of the PdOEP@WS_(2) heterostructure,and provides valuable theoretical insights for further improv-ing the optoelectronic performance of TMD-based devices.
