Cross-modality pedestrian re-identification has important appli-cations in the field of surveillance.Due to variations in posture,camera per-spective,and camera modality,some salient pedestrian features are difficult ...Cross-modality pedestrian re-identification has important appli-cations in the field of surveillance.Due to variations in posture,camera per-spective,and camera modality,some salient pedestrian features are difficult to provide effective retrieval cues.Therefore,it becomes a challenge to design an effective strategy to extract more discriminative pedestrian detail.Although many effective methods for detailed feature extraction are proposed,there are still some shortcomings in filtering background and modality noise.To further purify the features,a pure detail feature extraction network(PDFENet)is proposed for VI-ReID.PDFENet includes three modules,adaptive detail mask generation module(ADMG),inter-detail interaction module(IDI)and cross-modality cross-entropy(CMCE).ADMG and IDI use human joints and their semantic associations to suppress background noise in features.CMCE guides the model to ignore modality noise by generating modality-shared feature labels.Specifically,ADMG generates masks for pedestrian details based on pose estimation.Masks are used to suppress background information and enhance pedestrian detail information.Besides,IDI mines the semantic relations among details to further refine the features.Finally,CMCE cross-combines classifiers and features to generate modality-shared feature labels to guide model training.Extensive ablation experiments as well as visualization results have demonstrated the effectiveness of PDFENet in eliminating background and modality noise.In addition,comparison experi-ments in two publicly available datasets also show the competitiveness of our approach.展开更多
A new strategy to fabricate oxygen-promoted Cu,N co-doped carbon(OP-CuN@C)composites is reported.The strategy consists of only two simple steps:chemical polymerization and high temperature carbonization.Electrochemica...A new strategy to fabricate oxygen-promoted Cu,N co-doped carbon(OP-CuN@C)composites is reported.The strategy consists of only two simple steps:chemical polymerization and high temperature carbonization.Electrochemical measurements were conducted to investigate the catalytic activity and mechanism of ORR on the resulting samples.All the electrochemical results indicate that OP-CuN@C exhibits the best ORR catalytic activity.The ORR onset potential of OP-CuN@C is slightly lower than that of commercial Pt/C catalyst.The good performance is attributed to the large specific surface area,high content of heteroatoms(pyridinic,graphitic nitrogen,and oxygen atom)and synergistic effect between divalent copper and nitrogen dopant.展开更多
A two-dimensional geometric model is developed for a polymer electrolyte based on the liquid water penetration mechanism in the membrane electrode assemblies under the action of capillary pressure.The effects of the d...A two-dimensional geometric model is developed for a polymer electrolyte based on the liquid water penetration mechanism in the membrane electrode assemblies under the action of capillary pressure.The effects of the diameter,number,and distribution of cracks in the micro-pore layers(MPLs)of the modeled MEA on the performance of the PEMFC are simulated to investigate the influence of mass transfer across the membrane.The results indicate that liquid water in the catalyst layer(CL)of the MEA can be discharged to gas channels through the cracks in MEA under the action of capillary pressure,thereby alleviating the flooding in the CL and enhancing the diffusion of oxygen to the CL.When the proportion of the total area of cracks in the active area of the MEA was 8%-12%,crack diameter was 20-30μm,and cracks were distributed uniformly.MEAs with and without cracks were prepared,fuel cells were assembled,and their performance was measured.The effects of cracks on mass transfer were then verified.This study helps prepare MEAs with controllable cracks.展开更多
In order to overcome the problem that the low conductivity and high content of graphitic N will lead to serious irreversible capacity loss,magnesiothermic denitriding method was employed to fabricate nitrogen deficien...In order to overcome the problem that the low conductivity and high content of graphitic N will lead to serious irreversible capacity loss,magnesiothermic denitriding method was employed to fabricate nitrogen deficient g-C3N4 (ND-g-C3N4).By controlling the reaction conditions,ND-g-C3N4-675 with optimal electrochemical properties was obtained.The ND-g-C3N4-675 has thinner two-dimensional porous structure,with high specific surface area and good conductivity.The ND-g-C3N4-675 showed superior cyclic stability and rate capability (After 500 cycles under 1 000 mA·g-1,2 264.9 mAh·g-1 was obtained).Moreover,it presented high initial coulombic efficiency (42.2%).展开更多
基金supported by the National Natural Science Foundation of China (Grant No.61906168,62202429)Zhejiang Provincial Natural Science Foundation of China (Grant No.LY23F020023)Construction of Hubei Provincial Key Laboratory for Intelligent Visual Monitoring of Hydropower Projects (2022SDSJ01).
文摘Cross-modality pedestrian re-identification has important appli-cations in the field of surveillance.Due to variations in posture,camera per-spective,and camera modality,some salient pedestrian features are difficult to provide effective retrieval cues.Therefore,it becomes a challenge to design an effective strategy to extract more discriminative pedestrian detail.Although many effective methods for detailed feature extraction are proposed,there are still some shortcomings in filtering background and modality noise.To further purify the features,a pure detail feature extraction network(PDFENet)is proposed for VI-ReID.PDFENet includes three modules,adaptive detail mask generation module(ADMG),inter-detail interaction module(IDI)and cross-modality cross-entropy(CMCE).ADMG and IDI use human joints and their semantic associations to suppress background noise in features.CMCE guides the model to ignore modality noise by generating modality-shared feature labels.Specifically,ADMG generates masks for pedestrian details based on pose estimation.Masks are used to suppress background information and enhance pedestrian detail information.Besides,IDI mines the semantic relations among details to further refine the features.Finally,CMCE cross-combines classifiers and features to generate modality-shared feature labels to guide model training.Extensive ablation experiments as well as visualization results have demonstrated the effectiveness of PDFENet in eliminating background and modality noise.In addition,comparison experi-ments in two publicly available datasets also show the competitiveness of our approach.
基金Funded in Part by the National Key Research and Development Program of China(2017YFB0102801)。
文摘A new strategy to fabricate oxygen-promoted Cu,N co-doped carbon(OP-CuN@C)composites is reported.The strategy consists of only two simple steps:chemical polymerization and high temperature carbonization.Electrochemical measurements were conducted to investigate the catalytic activity and mechanism of ORR on the resulting samples.All the electrochemical results indicate that OP-CuN@C exhibits the best ORR catalytic activity.The ORR onset potential of OP-CuN@C is slightly lower than that of commercial Pt/C catalyst.The good performance is attributed to the large specific surface area,high content of heteroatoms(pyridinic,graphitic nitrogen,and oxygen atom)and synergistic effect between divalent copper and nitrogen dopant.
基金Funded by National Natural Science Foundation of China(No.21676207)National Key Development Project of New Energy Vehicle Test Program of China(Nos.2016YFB0101207 and 2017YFB0102803)。
文摘A two-dimensional geometric model is developed for a polymer electrolyte based on the liquid water penetration mechanism in the membrane electrode assemblies under the action of capillary pressure.The effects of the diameter,number,and distribution of cracks in the micro-pore layers(MPLs)of the modeled MEA on the performance of the PEMFC are simulated to investigate the influence of mass transfer across the membrane.The results indicate that liquid water in the catalyst layer(CL)of the MEA can be discharged to gas channels through the cracks in MEA under the action of capillary pressure,thereby alleviating the flooding in the CL and enhancing the diffusion of oxygen to the CL.When the proportion of the total area of cracks in the active area of the MEA was 8%-12%,crack diameter was 20-30μm,and cracks were distributed uniformly.MEAs with and without cracks were prepared,fuel cells were assembled,and their performance was measured.The effects of cracks on mass transfer were then verified.This study helps prepare MEAs with controllable cracks.
基金the National Natural Science Foundation of China(No.21473128)。
文摘In order to overcome the problem that the low conductivity and high content of graphitic N will lead to serious irreversible capacity loss,magnesiothermic denitriding method was employed to fabricate nitrogen deficient g-C3N4 (ND-g-C3N4).By controlling the reaction conditions,ND-g-C3N4-675 with optimal electrochemical properties was obtained.The ND-g-C3N4-675 has thinner two-dimensional porous structure,with high specific surface area and good conductivity.The ND-g-C3N4-675 showed superior cyclic stability and rate capability (After 500 cycles under 1 000 mA·g-1,2 264.9 mAh·g-1 was obtained).Moreover,it presented high initial coulombic efficiency (42.2%).
