Search

Similar to previous video coding standards, transform and quantization are employed in the most recent video coding standard High Efficiency Video Coding (HEVC), and there exist a large number of transform coefficients being quantized to zeros. In order to reduce the computations in transform and quantization, a novel prediction algorithm is designed to early detect all-zero 4 × 4 blocks prior to transform and quantization. A theoretical analysis is performed on the 4 × 4 Discrete Cosine Transform (DCT) and the related quantization functions in HEVC; consequently, two sufficient conditions to early detect all-zero 4 × 4 blocks are derived and the proposed prediction algorithm is developed while balancing the detection efficiency and computational overheads. Experimental results demonstrate that the proposed algorithm is able to efficiently predict all-zero 4 × 4 blocks and reduce redundant DCT and quantization computations while maintaining the same coding performance in terms of video quality and bit rates. [ABSTRACT FROM AUTHOR]

In this paper, a new network-transmission-based (NTB) algorithm is proposed for human activity recognition in videos. The proposed NTB algorithm models the entire scene as an error-free network. In this network, each node corresponds to a patch of the scene and each edge represents the activity correlation between the corresponding patches. Based on this network, we further model people in the scene as packages, while human activities can be modeled as the process of package transmission in the network. By analyzing these specific package transmission processes, various activities can be effectively detected. The implementation of our NTB algorithm into abnormal activity detection and group activity recognition are described in detail in this paper. Experimental results demonstrate the effectiveness of our proposed algorithm. [ABSTRACT FROM PUBLISHER]

Abstract: Semi-supervised multi-label classification has been applied to many real-world applications such as image classification, document classification and so on. In semi-supervised learning, unlabeled samples are added to the training set for enhancing the classification performance, however, noises are introduced simultaneously. In order to reduce this negative effect, the nearest neighbor data editing technique is introduced to semi-supervised multi-label classification, and thus an algorithm named Multi-Label Self-Training with Editing (MLSTE) is proposed in this work. The proposed algorithm is able to solve the uncertainty problem in semi-supervised multi-label classification to some extent, by improving the performance of determining the label number and selecting confident samples during the course of semi-supervised learning. Extensive experimental results on several benchmark datasets have been carried out to verify the effectiveness of the proposed MLSTE algorithm. [Copyright &y& Elsevier]

Fast mode decision algorithms have been widely used in the video encoder implementation to reduce encoding complexity yet without much sacrifice in the coding performance. Optimal stopping theory, which addresses early termination for a generic class of decision problems, is adopted in this paper to achieve fast mode decision for the H.264/Scalable Video Coding standard. A constrained model is developed with optimal stopping, and the solutions to this model are employed to initialize the candidate mode list and predict the early termination. Comprehensive simulation results are conducted to demonstrate that the proposed method strikes a good balance between low encoding complexity and high coding efficiency. [ABSTRACT FROM PUBLISHER]

In this paper, we investigate the issues over the smooth quality and the smooth bit rate during rate control (RC) in H.264/AVC. An adaptive quantization-parameter (Qp) clip scheme is proposed to optimize the quality smoothness while keeping the bit-rate fluctuation at an acceptable level. First, the frame complexity variation is studied by defining a complexity ratio between two nearby frames. Second, the range of the generated bits is analyzed to prevent the encoder buffer from overflow and underflow. Third, based on the safe range of the generated bits, an optimal Qp clip range is developed to reduce the quality fluctuation. Experimental results demonstrate that the proposed Qp clip scheme can achieve excellent performance in quality smoothness and buffer regulation. [ABSTRACT FROM AUTHOR]

A novel spatial-layer rate-control algorithm is proposed for scalable video coding (SVC). First, by analyzing the relationship among the best initial quantization parameter (Qp), channel bandwidth, and the initial frames' complexity measure, an adaptive Qp-initialization model is introduced to determine the starting Qp value for not only the base layer but also the spatial enhancement layers of SVC. Then, a two-stage Qp -determination scheme is designed to improve the rate-control performance for the spatial-layer SVC with an efficient frame-complexity prediction method and an adaptive model-parameter update technique employed. Experimental results demonstrate the effectiveness of the proposed Qp -initialization scheme and the two-stage Qp-determination algorithm. By comparison with two other benchmark rate-control algorithms, the proposed rate-control algorithm is able to control constrained bit rates accurately with better rate-distortion performance. [ABSTRACT FROM PUBLISHER]

Abstract: To enable robust video transmission over heterogeneous networks, the hierarchical B-picture prediction structure is employed in the state-of-the-art video coding standard H.264/SVC, aiming to produce scalable bitstreams with various frame rates. However, the exhaustive mode decision process with the hierarchical B-picture structure increases the computational complexity of H.264/SVC encoding dramatically. In this paper, a fast mode decision algorithm is proposed to speed up H.264/SVC encoding with the hierarchical B-picture structure, which is achieved by utilizing macroblock (MB) features, correlation of temporal–spatial neighboring MBs, and the discrepant characteristics of hierarchical layers. Extensive experimental results demonstrate that the proposed algorithm is able to reduce the encoding time of H.264/SVC significantly for video sequences with a wide range of resolutions, and meanwhile the video quality and compression ratio are well preserved. [Copyright &y& Elsevier]

This paper proposes a novel approach to early determination of zero-quantized 8 x 8 discrete cosine transform (DCT) coefficients for fast video encoding. First, with the dynamic range analysis of DCT coefficients at different frequency positions, several sufficient conditions are derived to early determine whether a prediction error block (8 x 8) is an all-zero or a partial-zero block, i.e., the DCT coefficients within the block are all or partially zero-quantized. Being different from traditional methods that utilize the sum of absolute difference (SAD) of the entire prediction error block, the sufficient conditions are derived based on the SAD of each row of the prediction error block. For partial-zero blocks, fast DCTIIDCT algorithms are further developed by pruning conventional 8-point butterflybased DCTIIDCT algorithms. Experimental results exhibit that the proposed early determination algorithm greatly reduces computational complexity in terms of DCT/IDCT, quantization, and inverse quantization, as compared with existing algorithms. [ABSTRACT FROM AUTHOR]

The article informs that an agent-based evolutionary approach is proposed to extract interpretable rule-based knowledge. In the multiagent system, each fuzzy set agent autonomously determines its own fuzzy sets information, such as the number and distribution of the fuzzy sets. The fundamental concept of fuzzy reasoning was first introduced in 1973, and the past few years have witnessed a rapid growth in a number of applications of fuzzy systems. One of the most important motivations for building up a fuzzy model is to let users gain a deep insight into an unknown system through the easily understandable fuzzy rules. Another main attraction undoubtedly lies in the characteristics that fuzzy systems possess, they are capable of handling complex, non-linear, and sometimes mathematically intangible dynamic systems.

Abstract: A new scheme based on multi-objective hierarchical genetic algorithm (MOHGA) is proposed to extract interpretable rule-based knowledge from data. The approach is derived from the use of multiple objective genetic algorithm (MOGA), where the genes of the chromosome are arranged into control genes and parameter genes. These genes are in a hierarchical form so that the control genes can manipulate the parameter genes in a more effective manner. The effectiveness of this chromosome formulation enables the fuzzy sets and rules to be optimally reduced. Some important concepts about the interpretability are introduced and the fitness function in the MOGA will consider both the accuracy and interpretability of the fuzzy model. In order to remove the redundancy of the rule base proactively, we further apply an interpretability-driven simplification method to newborn individuals. In our approach, we first apply the fuzzy clustering to generate an initial rule-based model. Then the multi-objective hierarchical genetic algorithm and the recursive least square method are used to obtain the optimized fuzzy models. The accuracy and the interpretability of fuzzy models derived by this approach are studied and presented in this paper. We compare our work with other methods reported in the literature on four examples: a synthetic nonlinear dynamic system, a nonlinear static system, the Lorenz system and the Mackey–Glass system. Simulation results show that the proposed approach is effective and practical in knowledge extraction. [Copyright &y& Elsevier]

In this study, the corrosion electrochemistry and corrosion behavior of two steels were studied under the simulated alumina production conditions. The corrosion rate of 16Mn steel is greater than that of Q235 steel. The effect of S2− concentration on corrosion rate was significantly higher than that of S2O32−. The synergistic corrosion rates of Q235 and 16Mn steels increase at first and then decrease with the sulfur content, and the peak value appears when the concentration of S2− and S2O32− is 4 g/L and 3 g/L respectively. There are two main types of corrosion products: one is surface octahedral grain, which is composed of Fe2O3, Fe3O4 and Al2O3.The other is the interlayer corrosion between the surface layer and the matrix, which is composed of FeS, FeS2 and NaFeO2.The formation mechanism of the corrosion and corrosion mechanism were obtained by analyzing the phenomenon of ion competitive adsorption. Further validation and analysis of ion competition adsorption phenomenon were conducted using first-principles calculations based on density functional theory (DFT). The formation of corrosion products on the steel surface was investigated at an ion level, and the adsorption energies of OH− and S2− at the top site of Fe(110) surface were calculated. It was found that S2− is more likely to be adsorbed on the Fe(110) surface compared to OH−. The corrosion mechanism of steel is discussed preliminarily. [ABSTRACT FROM AUTHOR]

Background: Exosomes are emerging as important mediators of the cross-talk between tumor cells and the microenvironment. The communication between tumor-derived exosomes and macrophages has a critical role in facilitating tumor progression. However, the mechanisms by which exosomes modulate tumor development in lung cancer are not fully understood. Methods: Short hairpin RNA mediated knockdown or exogenous expression of TRIM59 combined with in vitro and in vivo assays were performed to prove the functional significance of TRIM59. Western blotting, real-time PCR, co-immunoprecipitation, immunofluorescence (IF) staining assays, proximity ligation assay (PLA), ubiquitination assays, lactate secretion and lipid droplets content measurement, and rescue experiments were used to evaluate the mechanism. Lewis lung carcinoma (LLC) cells were injected via subcutaneously or tail vein into C57BL/6 wild-type (WT) and transgenic mice to assess the role of TRIM59 in vivo. Results: We demonstrated that tripartite motif-containing 59 (TRIM59) was expressed in lung cancer cells-derived exosomes, and can be transferred to macrophages through the exosomes. Activated macrophages by TRIM59 promote lung cancer progression in vitro and in vivo. Mechanistic investigations revealed that TRIM59 physically interacts with abhydrolase domain containing 5 (ABHD5) and directly induced the ubiquitination of ABHD5 and led to its proteasome-dependent degradation. ABHD5, an lipolytic co-activator, deficiency induced metabolic reprogramming and enabled NLRP3 inflammasome activation in macrophages. Further studies showed that the exacerbation of NLRP3 inflammasome activation by ABHD5 deficiency, provides a positive feedback loop to promote cancer progression by preferentially secrete the proinflammatory cytokine IL-1β. Conclusions: Collectively, these data indicate that tumor-derived exosomal TRIM59 converts macrophages to tumor-promoting functions of macrophages via regulating ABHD5 proteasomal degradation, to activate NLRP3 inflammasome signaling pathway to promote lung cancer progression by IL-1β secretion. Our findings also indicate that tumor-derived exosomal TRIM59 has an important role in intercellular communication for fostering an inflammatory microenvironment and promoting lung metastasis. [ABSTRACT FROM AUTHOR]

The residence time of particles in the flow field is highly sensitive to the fluid's flow characteristics and the particles' motion. We proposed a dynamic entry mode with the constant-pulse (CP) condition in response to the conditions of symmetric impinging streams and asymmetric impinging streams. Single-particle motion behaviors and particle residence time in a novel cross-type structure reactor we designed were studied and discussed in this article. This was done by looking at how these variables changed depending on the impulse inlet velocity conditions, the reactor geometry, and the particle properties. The results demonstrate that when particle relaxation time is short, those particles without oscillatory motions were directly accelerated to leave the active area. When its relaxation time is reasonably set, even a single particle can hugely elevate the oscillatory motions. With the structure of the novel cross type, the particle merely exhibits three oscillatory under the condition of the symmetric pulse inlet flow and asymmetric pulse inlet flow, notwithstanding, CP contains at least three oscillatory motions. The ttot is the duration that an individual particle spends in the working region. The ttot and active area were affected by inlet velocity, parameters of particles and the shape of the reactor parameters of particles and the shape of the reactor. The ttot raised with τ* increased in a certain range. T is the inlet velocity variation period. The particle resides time at T = Tb reaches a maximum in the CP. The maximum residence times Tb was affected by particle relaxation time, under CP with a diameter ratio of 5 and inlet velocity difference of 0.8. The particle retention time in the novel cross-type structure reactor was heightened by 7 times. The CP and the novel cross-type structure reactor can improve the retention time of particles in the flow field. [ABSTRACT FROM AUTHOR]

Visual Question Answering (VQA) aims to output a correct answer based on cross‐modality inputs including question and visual content. In general pipeline, information reasoning plays the key role for a reasonable answer. However, visual information is commonly not fully employed in many popular models nowadays. Facing this challenge, a new strategy is proposed in this work to make the best of visual information during reasoning. In detail, visual information is divided into two subsets: (1) question‐relevant visual set, and (2) question‐irrelevant visual set. Then, both of these two sets are employed by reasoning to generate reasonable outputs. Experiments are conducted on the benchmark VQAv2 dataset, which demonstrate the effectiveness of the proposed strategy. The project page can be found in https://mic.tongji.edu.cn/e6/8d/c9778a190093/page.htm. [ABSTRACT FROM AUTHOR]

Background: Alterations in several tripartite motif-containing (TRIM) family proteins have been implicated in the pathogenesis of lung cancer. TRIM28, a member of the TRIM E3 ligase family, has been associated with tumorigenesis, cell proliferation, and inflammation. However, little is known about TRIM28 expression and its role in the immune microenvironment of non-small cell lung cancer (NSCLC). Methods: We assessed the clinical significance of TRIM28 in tissue microarrays and TCGA cohorts. We investigated the function of TRIM28 in syngeneic mouse tumor models, the KrasLSL−G12D/+; Tp53fl/fl (KP) mouse model, and humanized mice. Immune cell composition was analyzed using flow cytometry and immunohistochemistry. Results: Our findings revealed a positive correlation between TRIM28 expression and the infiltration of suppressive myeloid-derived suppressor cells (MDSCs) in NSCLC. Moreover, silencing TRIM28 enhanced the efficacy of anti-PD-1 immunotherapy by reshaping the inflamed tumor microenvironment. Mechanistically, we demonstrated that TRIM28 could physically interact with receptor-interacting protein kinase 1 (RIPK1) and promote K63-linked ubiquitination of RIPK1, which is crucial for sustaining activation of the NF-κB pathway. Mutagenesis of the E3 ligase domain corroborated the essential role of E3 ligase activity in TRIM28-mediated NF-κB activation. Further experiments revealed that TRIM28 could upregulate the expression of CXCL1 by activating NF-κB signaling. CXCL1 could bind to CXCR2 on MDSCs and promote their migration to the tumor microenvironment. TRIM28 knockdown increased responsiveness to anti-PD-1 therapy in immunocompetent mice, characterized by increased CD8+T tumor-infiltrating lymphocytes and decreased MDSCs. Conclusion: The present study identified TRIM28 as a promoter of chemokine-driven recruitment of MDSCs through RIPK1-mediated NF-κB activation, leading to the suppression of infiltrating activated CD8+T cells and the development of anti-PD-1 resistance. Understanding the regulation of MDSC recruitment and function by TRIM28 provides crucial insights into the association between TRIM28 signaling and the development of an immunosuppressive tumor microenvironment. These insights may inform the development of combination therapies to enhance the effectiveness of immune checkpoint blockade therapy in NSCLC. [ABSTRACT FROM AUTHOR]

As a by-product of the coking industry, coke oven gas (COG) is often burned as fuel gas. However, COG contains a large amount of hydrogen. If hydrogen is recovered, the economic and environmental benefits of the coking industry can be improved. In this work, a novel automotive fuel cell hydrogen production process coupled with membrane separation (PRISM® Membrane Separator, polyimide hollow fiber membrane), pressure swing adsorption (PSA) and methane steam reforming (SMR) is proposed. The process uses membrane separation and PSA to produce high-purity hydrogen for fuel cells. The methane-rich membrane residue gas is used as the feed gas of the reforming reactor. The hydrogen-containing gas at the outlet of the reactor enters the membrane separator to further improve the hydrogen yield. In addition, based on the principle of temperature gradient utilization, the high-temperature gas at the outlet of the reactor is used as the heat source for the system. Through a synergy of separation and reaction, the process supports the low-cost and efficient production of fuel cell hydrogen from COG. The proposed process was established using UniSim Design. The key variables were determined by sensitivity analysis, and the response surface methodology based on Box–Behnken design (BBD) method was further used to optimize the key variables. The optimal operating variables are as follows: the area of the third hydrogen membrane (HM3) is 3000 m2, the reaction temperature is 917 K and the molar ratio of steam to methane (S/C) is 1.12. The cost of the hydrogen production process is 1.56 $/kg, which means it has good prospects for use in industrial applications. • A novel fuel cell hydrogen production process from coke oven gas is proposed. • The process is coupled with membrane separation, PSA and MSR. • The process is strictly simulated and optimized by the response surface methodology. • The cost of the hydrogen production process from coke oven gas is 1.56 $/kg. [ABSTRACT FROM AUTHOR]

Background: Diabetic retinopathy (DR) is one of the most common complications of diabetes and has become a major global cause of blindness. Curcumin, an extract of Curcuma longa (turmeric), is effective in preventing and treating diabetes. Recent studies have shown that curcumin can delay DR development. However, there has been no systematic review of its treatment of DR. This study will conduct a systematic review and meta-analysis of currently published randomized controlled trials (RCT) of curcumin for treating DR patients to evaluate its efficacy and safety. Methods: We will search the relevant studies of curcumin in the treatment of DR in PubMed, Medline, EMBASE, Cochrane Library, China National Knowledge Infrastructure (CNKI), VIP, and Wanfang databases from their respective inception dates to May 2022. A meta-analysis of the data extracted from qualified RCTs will be conducted, including the progression of DR, visual acuity, visual field, macular edema, quality of life, and adverse events. The meta-analysis will be performed using Review Manager 5.4.1 software, and the results will be based on either random-effects or fixed-effects models, depending on the heterogeneity. The Grading of Recommendations, Development, and Evaluation (GRADE) system will be used to evaluate the reliability and quality of evidence. Results: The results of this study will provide sound and high-quality evidence for the efficacy and safety of curcumin in the treatment of DR. Conclusion: This study will be the first meta-analysis to comprehensively assess the efficacy and safety of curcumin in the treatment of DR and will provide helpful evidence for the clinical management of this disease. Systematic review registration: INPLASY202250002. [ABSTRACT FROM AUTHOR]

Background: Recent studies have indicated that some members of the tripartite motif (TRIM) proteins function as important regulators for non-small cell lung cancer (NSCLC), However, the regulatory mechanism underpinning aberrant expression of TRIM in NSCLC remains unclear. Here we report that TRIM15 plays important roles in NSCLC progression through modulating Keap1-Nrf2 signaling pathway. Methods: TRIM15 expression was evaluated by western blot analysis, tissue microarray-based immunohistochemistry analysis. The interactions between TRIM15 and Keap1 were analyzed by co-immunoprecipitation (Co-IP) and immunofluorescence co-localization assay. The correlation between TRIM15 and Keap1 was measured by Co-IP and ubiquitination analysis in vitro. Gain- and lost-of-function experiments were used to detect TRIM15 promotes proliferation and invasion of NSCLC cells both in vitro and vivo. Results: Here, we revealed that TRIM15 was frequently upregulated in NSCLC samples and associated with poor prognosis. Functionally, TRIM15 knockdown resulted in decreased cancer cell proliferation and metastasis, whereas ectopic TRIM15 expression facilitated tumor cancer cell proliferation and metastasis in vitro and in vivo. Moreover, TRIM15 promoted cell proliferation and metastasis depends on its E3 ubiquitin ligase. Mechanistically, TRIM15 directly targeted Keap1 by ubiquitination and degradation, the principal regulator of Nrf2 degradation, leading to Nrf2 escaping from Keap1-mediated degradation, subsequently promoting antioxidant response and tumor progression. Conclusions: Therefore, our study characterizes the pivotal roles of TRIM15 promotes NSCLC progression via Nrf2 stability mediated by promoting Keap1 ubiquitination and degradation and could be a valuable prognostic biomarker and a potential therapeutic target in NSCLC. 44ozmiCbiGVX1qU1k16dGv Video Abstract [ABSTRACT FROM AUTHOR]

• A novel process namely M-EHP was proposed to achieve efficient hydrogen production from hydrogen-doped natural gas. • A BPNN model was constructed for describing the complex relationship between objective functions and decision variables. • Co-optimization of hydrogen recovery rate and separation cost was carried out by NSGA-II. • 90 % overall H 2 recovery rate was achieved with the separation cost at least 31.81 % lower than commercial PSA process. In this work, a novel process of polymer membrane coupled with electrochemical hydrogen pump (EHP) was proposed to achieve efficient hydrogen production from hydrogen-doped natural gas (HDNG), and was simulated by UniSim Design. A hydrogen separation unit of two-stage polymer membrane was used for achieving hydrogen enrichment from low H 2 concentration HDNG. Due to the hydrogen upgrade, EHP was operated at a H 2 concentration much higher than HDNG, resulting in low energy consumption. Benefiting from the synergy between hydrogen upgrade and purification, the coupling process brought substantial effect while improving the hydrogen purity and recovery rate and reducing the hydrogen separation cost. The real natural gas components were considered into HDNG to ensure the effectiveness of process simulation. An optimization strategy considering both product yield and production cost was proposed to achieve the trade-off between hydrogen recovery rate and separation cost. However, the impact of process parameters on the objectives is nonlinear and difficult to describe with mathematical expressions. Therefore, an agent model of the proposed process was established based on back propagation neural network (BPNN), and multi-objective optimization of H 2 recovery rate and total annual cost (TAC) of hydrogen separation unit was achieved using non-dominated sorting genetic algorithm-II (NSGA-II). Investigations with HDNG under various operating conditions (feed pressure of 8–40 bar and feed H 2 concentration of 5–25 mol%) were performed, which proved that 90 % overall H 2 recovery rate and 99.99 mol% H 2 purity of hydrogen product can be achieved at low cost, with the lowest hydrogen separation cost of 1.00 $/kg H 2 achievable at 40 bar feed pressure and 25 mol% feed H 2 concentration. In summary, this coupling process can provide engineering solutions for efficient hydrogen production from HDNG and promote the development of green hydrogen. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a novel rate control scheme with sliding window basic unit is proposed to achieve consistent or smooth visual quality for H.264/AVC based video streaming. A sliding window consists of a group of successive frames and moves forward by one frame each time. To make the sliding window scheme possible for real-time video streaming, the initial encoder delay inherently in a video streaming system is utilized to generate all the bits of a window in advance, so that these bits for transmission are ready before their due time. The use of initial encoder delay does not introduce any additional delay in video streaming but benefits visual quality as compared to traditional one-pass rate control algorithms of H.264/AVC. Then, a Sliding Window Buffer Checking (SWBC) algorithm is proposed for buffer control at sliding window level and it accords with traditional buffer measurement of H.264/AVC. Extensive experimental results exhibit that higher coding performance, consistent visual quality and compliant buffer constraint can be achieved by the proposed algorithm. [Copyright &y& Elsevier]

Facilitated transport membrane with fixed and mobile carriers is essential yet challenging to improve the CO 2 capture. Herein, a facilitated transport membrane with continuous pathway was constructed by the in-situ interlinkage of mobile aniline with the fixed carriers distributing along electrospun polyacrylonitrile (PAN) nanofibers. In this membrane, fixed amine carriers are introduced through the alkalinous hydrolysis of PAN fibers (HPAN), and are distributed along the HPAN fibers to construct a long-range continuous transport pathway for CO 2. The mobile aniline carriers are introduced into the interfibrous voids of the HPAN nanofibers mat (NFM) by in-situ polymerization of Poly (ethylene glycol) diacrylate (PEGDA), which interlink the fixed amine carriers to achieve the facilitated transport pathway continuous completely. And the fixed amine and mobile aniline carriers are demonstrated to co-exist in the An/PEO@HPAN membrane (NFCM) by the FTIR and XPS results. Meanwhile, the XRD and DSC results indicate that the introduction of aniline increases the free volume of PEO matrix, which is beneficial for enhancing CO 2 permeation. Therefore, the 10 wt% An/PEO@HPAN-3h NFCM presents 90.81 % and 13.15 % increases in CO 2 permeability and CO 2 /N 2 selectivity under mixed gas at 0.3 MPa in contrast with the membrane without mobile aniline carriers. Under mixed gas conditions, this NFCM exhibits CO 2 permeability of 202.93 Barrer and CO 2 /N 2 selectivity of 83.89 at 0.1 MPa, reaching the 2019 McKeown upper bound. [Display omitted] • Fixed amine carriers are distributed along the HPAN nanofibers. • Mobile aniline carriers are introduced with in-situ polymerization of PEGDA. • Aniline interlinks the long-range continuous transport pathway in short range. • Continuous facilitated transport pathway improves CO 2 permeability. [ABSTRACT FROM AUTHOR]

Video description is challenging due to the high complexity of translating visual content into language. In most popular attention‐based pipelines for this task, visual features and previously generated words are usually concatenated as a vector to predict the current word. However, the errors caused by the inaccuracy of the predicted words may be accumulated, and the gap between visual features and language features may bring noises into the description model. Facing these problems, a variant of recurrent neural network is designed in this work, and a novel framework is developed to enhance the visual clues for video description. Moreover, a multi‐objective reinforcement learning strategy is implemented to build a more comprehensive reward with multiple metrics to improve the consistency and semantics of the generated description sentence. The experiments on the benchmark MSR‐VTT2016 and MSVD datasets demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

[Display omitted] • The micro/nano structure of FPI nanofiber membranes was regulated by electrospinning combined electrospraying techniques. • The first application of FPI nanofiber membranes in vacuum membrane distillation. • The prepared FPI nanofibrous membrane had excellent hydrophobicity, durability and desalination performance. In this study, we firstly proposed a fluorinated polyimide (FPI) /polytetrafluoroethylene (PTFE) nanofiber membrane for vacuum membrane distillation (VMD) by electrospinning FPI and double eletcrospraying PTFE particles. The resultant FPI /PTFE nanofiber membrane possessed desirable superhydrophobicity (water contact angle (WCA) of 155.3°), anti-wettability and durability due to the rough micro/nano structure of the membrane surface. The surface morphology and chemical composition of obtained FPI nanofiber membranes were characterized by SEM, FTIR, XRD and XPS, and the effects of PTFE content on the VMD performance were investigated. Results showed that the liquid entry pressure (LEP) of FPI/PTFE nanofiber membranes increased significantly after PTFE particles were introduced (LEP of 158 kPa). In the VMD process, the FPI/PTFE nanofiber membrane showed stable MD flux (33.49 L·m2·h−1) and excellent salt rejection (99.96%). Furthermore, the FPI/PTFE nanofiber membrane maintained its favorable performance, resulting in stable MD flux (30.21 L·m2·h−1) and salt rejection (99.96%) after the 120 h continuous long-term operation. It shows a promising future for VMD of FPI/PTFE nanofiber membranes. [ABSTRACT FROM AUTHOR]

This paper proposes a novel descriptor called Maximal Granularity Structure Descriptor (MGSD) for feature representation and an effective metric learning method called Generalized Multi-view Discriminant Analysis based on representation consistency (GMDA-RC) for person re-identification (Re-ID). The proposed descriptor of MGSD captures rich local structural information from overlapping macro-pixels in an image, analyzes the horizontal occurrence of multi-granularity and maximizes the occurrence to extract a robust representation for viewpoint changes. As a result, the proposed descriptor of MGSD can obtain rich person appearance whilst being robust against different condition changes. Besides, considering multi-view information, we present a new GMDA-RC for different views, inspired by the observation that different views share similar data structures. The proposed metric learning method of GMDA-RC seeks multiple discriminant common spaces for multiple views by jointly learning multiple view-specific linear transforms. Finally, we evaluate the proposed method of (MGSD+GMDA-RC) on three publicly available person Re-ID datasets: VIPeR, CUHK-01 and Wide Area Re-ID dataset (WARD). For the VIPeR and CUHK-01, the experimental results show that our method significantly outperforms the state-of-the-art methods, achieving the rank-1 matching rates of 67.09%, 70.61%, and the improvements of 17.41%, 5.34%, respectively. For the WARD, we consider different pairwise camera views (camera 1–2, camera 1–3, camera 2–3) and our method can achieve the rank-1 matching rates of 64.33%, 59.42%, 70.32%, increasing of 5.68%, 11.04%, 9.06% compared with the state-of-the-art methods, respectively. [ABSTRACT FROM AUTHOR]

• A partial element stage cut electrochemical hydrogen pump (EHP) model was proposed. • Three real factors were embedded into the model to describe EHP performance accurately. • The acuracy of EHP model was verified for four aspects under four different feedstock systems. • The variation law of current density caused by GDL mass transfer and PEM resistance was simulated and analyzed quantitatively. • The back-diffusion behavior of hydrogen under cathode high pressure was simulated and analyzed quantitatively. In this work, a partial element stage cut electrochemical hydrogen pump (EHP) model for the separation of multiple H 2 -contaning gases and hydrogen compression was proposed to realize fast prediction of the EHP's performance (nearly 10 s for one I EHP – E point). Three real factors, namely, anode impurity diffusion, hydrogen back-diffusion and anode catalyst deactivation, were embedded into the model. The model accuracy was verified for four aspects (current density distribution, polarization curve, hydrogen recovery and purity) under four feedstock systems (H 2 /N 2 , H 2 /CH 4 , H 2 /He and H 2 /CO 2) and a wide cathode pressure range of 0.1–1.0 MPa. The model had good accuracy ( R 2 ≥ 0.90). Benefiting from the embedding of the real factor of the hydrogen back-diffusion, hydrogen back-diffusion ratio was simulated accurately under a wide cathode pressure range of 0.13–20 MPa. By considering the dual- effect on current density caused by PEM proton conduction and GDL mass transformation resistance, optimal region of hydrogen recovery rate, energy efficiency and hydrogen purity can be simulated. About 64 % and 53 % for hydrogen recovery rate and energy efficiency can be achieved under low hydrogen feedstock of 20 mol%. In summary, it is promising to realize the "hydrogen economy" in the future under the utilization of EHP. [ABSTRACT FROM AUTHOR]

With the increasingly extensive application of membrane technology, facing the wastewater system with complex components, it is urgent to develop a porous membrane with excellent pollution-resistant and easily cleaned after pollution. Herein, a membrane with dual pH-responsiveness effects was successfully fabricated via a facile and eco-friendly method. The isoelectric point of zeta potential of the modified membrane (M-D-IA) appeared at pH = 4.31. It exhibited positive-charged and the water intake rate reached 427% under strongly acidic conditions, while it showed negative-charged with the water intake rate reaching 492% under strongly alkaline conditions, showing an excellent dual pH-responsiveness. Meanwhile, the membrane M-D-IA possessed good super-hydrophilic and underwater super-oleophobic surface characteristics. Furthermore, the membrane M-D-IA displayed an excellent oil-fouling resistance in the treatment of hexadecane/water mixed emulsion, with the water flux decay rate (FDR) only 17.5% after 7 cycles of operation and the water flux recovery rate (FRR) remained above 99.7% after alkali cleaning. Notably, the membrane M-D-IA also had excellent dual pH-responsiveness when treating typical polluted wastewater HA/CaCO 3 feed solution. The FRR was as high as 98.3% after washing with alkali and acid solution in turn, indicating that the membrane not only possessed a significant resistance to organic/inorganic pollution but also had an outstanding recovery of membrane flux after cleaning. Moreover, the water flux of the membrane M-D-IA was maintained at 850 L m−2 h−1 during 30 days of storage, displaying an excellent resistance to biological pollution. [Display omitted] • PVDF-CTFE membrane with dual pH-responsiveness was successfully fabricated. • Super-hydrophilic and underwater super-oleophobic endowed the membranes with excellent anti-fouling performance. • The dual-charge characteristic played a synergistic role in the process of cleaning. • The M-D-IA possessed excellent anti-bacterial property, retarding membrane bio-fouling during use. [ABSTRACT FROM AUTHOR]

In recent years, membrane separation technology has been widely used in various fields. And for the increasingly severe oily wastewater system, there is an increasing demand for the anti-fouling properties of the membranes. Based on the concept of fluorine-free, non-toxic, environmental-friendly and low-cost, this study constructed a discontinuous silicon-island structure with low surface energy on the surface of the hydrophilic membrane, which endows the PVDF-CTFE membrane with fouling resistance and cleaning pH-responsiveness. The discontinuous silicon-island structures were constructed by introducing SiO 2 nanoparticles on the super-hydrophilic layer of the PVDF-CTFE membrane modified by itaconic acid, and then hydrophobically treated with polydimethylsiloxane (PDMS). The experimental results exhibited that the modified membrane reached the hydrophobic and underwater super-oleophobic state when treated by PDMS for 1.0 h. The water contact angle (WCA) and the underwater oil contact angle (OCA) were 124° and 151°, respectively. Furthermore, the water flux remained at 186 L m−2 h−1 with no serious pore-plugging. Moreover, the water flux decay rate (FDR) remained below 19.6%, when modified membrane treating soybean oil/water emulsion for 6 cycles, presenting an excellent anti-fouling performance. Additionally, the water flux recovery rate (FRR) was over 99.3% after alkali cleaning, indicating the modified membrane possessed excellent cleaning pH-responsiveness. The above analysis exhibited that the synergistic effect of hydrophilic micro-domains and oleophobic micro-domains not only slows down the attachment of hydrophobic pollutants but also promotes the separation of hydrophobic pollutants during chemical cleaning, which greatly improves its anti-fouling performance and expands its application prospect. [Display omitted] • The discontinuous silicon-island structure with low surface energy was constructed based on the super-hydrophilic layer. • The anti-oil-fouling performance of the modified membrane was significantly improved. • Hydrophilic base membrane retained pH-responsiveness for cleaning. • Silicon-island structure was constructed with SiO 2 -PDMS to avoid the use of toxic compounds containing "fluorine". [ABSTRACT FROM AUTHOR]

The toxic effect of zearalenone (ZEA) is not fully understood and there is an urgent need for the development of effective agents to protect against the toxic effects of ZEA. In this study, we detected whether curcumin (CUR) can reduce Leydig cells apoptosis induced by ZEA. The Effects of ZEA and CUR on cell viability was evaluated using a Cell Counting kit-8 assay (CCK-8). Apoptosis was detected by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay and flow cytometry. The protective effect of CUR on oxidative stress induced by ZEA was determined by ROS, MDA, T-SOD, GSH and GSH-Px levels. In addition, we also determined effects of key signaling pathways and proteins involved in the apoptotic, PI3K-AKT, Nrf2 and endoplasmic reticulum stress signaling pathways by Western blotting. The expressions of proteins (PTEN, AKT, p-AKT, Bax, Bcl-2, GRP78, CHOP, JNK, P-JNK, Caspase-12, Caspase-9, Caspase-3, Nrf2, Keap1 and HO-1) were measured. The experimental results showed that CUR can alleviate oxidative stress and apoptosis caused by ZEA through the PI3K-AKT, Nrf2 and endoplasmic reticulum stress signaling pathways. Our results provide a theoretical basis for molecular studies of ZEA toxicology and clinical application of CUR. • Curcumin can protect Leydig cell from zearalenone-induced apoptosis and oxidative stress. • This specific mechanism of the protective effect of curcumin on germ cells is via PTEN/Nrf2/Bip signaling pathway. • Curcumin can ameliorate the inhibitory effect of zearalenone on testosterone secretion in Leydig cells [ABSTRACT FROM AUTHOR]

Ultrasound image super-resolution (SR) aims to transform low-resolution images into high-resolution ones, thereby restoring intricate details crucial for improved diagnostic accuracy. However, prevailing methods relying solely on image modality guidance and pixel-wise loss functions struggle to capture the distinct characteristics of medical images, such as unique texture patterns and specific colors harboring critical diagnostic information. To overcome these challenges, this paper introduces the Multi-Modal Regularized Coarse-to-fine Transformer (M2Trans) for Ultrasound Image SR. By integrating the text modality, we establish joint image-text guidance during training, leveraging the medical CLIP model to incorporate richer priors from text descriptions into the SR optimization process, enhancing detail, structure, and semantic recovery. Furthermore, we propose a novel coarse-to-fine transformer comprising multiple branches infused with self-attention and frequency transforms to efficiently capture signal dependencies across different scales. Extensive experimental results demonstrate significant improvements over state-of-the-art methods on benchmark datasets, including CCA-US, US-CASE, and our newly created dataset MMUS1K, with a minimum improvement of 0.17dB, 0.30dB, and 0.28dB in terms of PSNR. Our code and dataset will be available at: https://github.com/eezkni/M2Trans.

Methods based on deep learning have achieved great success in the field of video action recognition. When these methods are applied to real-world scenarios that require fine-grained analysis of actions, such as being tested on a tea ceremony, limitations may arise. To promote the development of fine-grained action recognition, a fine-grained video action dataset is constructed by collecting videos of tea ceremony actions. This dataset includes 2745 video clips. By using a hierarchical fine-grained action classification approach, these clips are divided into 9 basic action classes and 31 fine-grained action subclasses. To better establish a fine-grained temporal model for tea ceremony actions, a method named TSM-ConvNeXt is proposed that integrates a TSM into the high-performance convolutional neural network ConvNeXt. Compared to a baseline method using ResNet50, the experimental performance of TSM-ConvNeXt is improved by 7.31%. Furthermore, compared with the state-of-the-art methods for action recognition on the FineTea and Diving48 datasets, the proposed approach achieves the best experimental results. The FineTea dataset is publicly available.

As a technically challenging topic, visual storytelling aims at generating an imaginary and coherent story with narrative multi-sentences from a group of relevant images. Existing methods often generate direct and rigid descriptions of apparent image-based contents, because they are not capable of exploring implicit information beyond images. Hence, these schemes could not capture consistent dependencies from holistic representation, impairing the generation of reasonable and fluent stories. To address these problems, a novel knowledge-enriched attention network with group-wise semantic model is proposed. Three main novel components are designed and supported by substantial experiments to reveal practical advantages. First, a knowledge-enriched attention network is designed to extract implicit concepts from external knowledge system, and these concepts are followed by a cascade cross-modal attention mechanism to characterize imaginative and concrete representations. Second, a group-wise semantic module with second-order pooling is developed to explore the globally consistent guidance. Third, a unified one-stage story generation model with encoder-decoder structure is proposed to simultaneously train and infer the knowledge-enriched attention network, group-wise semantic module and multi-modal story generation decoder in an end-to-end fashion. Substantial experiments on the visual storytelling datasets with both objective and subjective evaluation metrics demonstrate the superior performance of the proposed scheme as compared with other state-of-the-art methods. The source code of this work can be found in https://mic.tongji.edu.cn.

Dynamic point cloud is a volumetric visual data representing realistic 3D scenes for virtual reality and augmented reality applications. However, its large data volume has been the bottleneck of data processing, transmission, and storage, which requires effective compression. In this paper, we propose a Perceptually Weighted Rate-Distortion Optimization (PWRDO) scheme for Video-based Point Cloud Compression (V-PCC), which aims to minimize the perceptual distortion of reconstructed point cloud at the given bit rate. Firstly, we propose a general framework of perceptually optimized V-PCC to exploit visual redundancies in point clouds. Secondly, a multi-scale Projection based Point Cloud quality Metric (PPCM) is proposed to measure the perceptual quality of 3D point cloud. The PPCM model comprises 3D-to-2D patch projection, multi-scale structural distortion measurement, and fusion model. Approximations and simplifications of the proposed PPCM are also presented for both V-PCC integration and low complexity. Thirdly, based on the simplified PPCM model, we propose a PWRDO scheme with Lagrange multiplier adaptation, which is incorporated into the V-PCC to enhance the coding efficiency. Experimental results show that the proposed PPCM models can be used as standalone quality metrics, and they are able to achieve higher consistency with the human subjective scores than the state-of-the-art objective visual quality metrics. Also, compared with the latest V-PCC reference model, the proposed PWRDO-based V-PCC scheme achieves an average bit rate reduction of 13.52%, 8.16%, 10.56% and 9.54%, respectively, in terms of four objective visual quality metrics for point clouds. It is significantly superior to the state-of-the-art coding algorithms. The computational complexity of the proposed PWRDO increases by 1.71% and 0.05% on average to the V-PCC encoder and decoder, respectively, which is negligible. The source codes of the PPCM and PWRDO schemes are available at https://github.com/VVCodec/PPCM-PWRDO.

It is challenging to generate temporal action proposals from untrimmed videos. In general, boundary-based temporal action proposal generators are based on detecting temporal action boundaries, where a classifier is usually applied to evaluate the probability of each temporal action location. However, most existing approaches treat boundaries and contents separately, which neglect that the context of actions and the temporal locations complement each other, resulting in incomplete modeling of boundaries and contents. In addition, temporal boundaries are often located by exploiting either local clues or global information, without mining local temporal information and temporal-to-temporal relations sufficiently at different levels. Facing these challenges, a novel approach named multi-level content-aware boundary detection (MCBD) is proposed to generate temporal action proposals from videos, which jointly models the boundaries and contents of actions and captures multi-level (i.e., frame level and proposal level) temporal and context information. Specifically, the proposed MCBD preliminarily mines rich frame-level features to generate one-dimensional probability sequences, and further exploits temporal-to-temporal proposal-level relations to produce two-dimensional probability maps. The final temporal action proposals are obtained by a fusion of the multi-level boundary and content probabilities, achieving precise boundaries and reliable confidence of proposals. The extensive experiments on the three benchmark datasets of THUMOS14, ActivityNet v1.3 and HACS demonstrate the effectiveness of the proposed MCBD compared to state-of-the-art methods. The source code of this work can be found in https://mic.tongji.edu.cn.

Convolutional Neural Networks (CNNs) dominate image processing but suffer from local inductive bias, which is addressed by the transformer framework with its inherent ability to capture global context through self-attention mechanisms. However, how to inherit and integrate their advantages to improve compressed sensing is still an open issue. This paper proposes CSformer, a hybrid framework to explore the representation capacity of local and global features. The proposed approach is well-designed for end-to-end compressive image sensing, composed of adaptive sampling and recovery. In the sampling module, images are measured block-by-block by the learned sampling matrix. In the reconstruction stage, the measurements are projected into an initialization stem, a CNN stem, and a transformer stem. The initialization stem mimics the traditional reconstruction of compressive sensing but generates the initial reconstruction in a learnable and efficient manner. The CNN stem and transformer stem are concurrent, simultaneously calculating fine-grained and long-range features and efficiently aggregating them. Furthermore, we explore a progressive strategy and window-based transformer block to reduce the parameters and computational complexity. The experimental results demonstrate the effectiveness of the dedicated transformer-based architecture for compressive sensing, which achieves superior performance compared to state-of-the-art methods on different datasets. Our codes is available at: https://github.com/Lineves7/CSformer.

Wholly aromatic polyimide (PI) films with good solution processability, light colors, good optical transparency, high storage modulus, and improved heat resistance were prepared and characterized. For this purpose, a multi-component copolymerization methodology was performed from a fluoro-containing dianhydride, 4,4'-(hexafluoroisopropylidene)diphthalic anhydride (6FDA), a rigid dianhydride, 3,3',4,4'-biphenyltetracarboxylic dianhydride (BPDA), and a fluoro-containing diamine, 2,2'-bis(trifluoromethyl)-4,4'-bis [4-(4-amino-3-methyl)benzamide]biphenyl (MABTFMB). One homopolymer, FPI-1 (6FDA-MABTFMB), and five copolymers, FPI-2~FPI-6, containing the BPDA units from 10 mol% to 50 mol% in the dianhydride moieties, were prepared, respectively. The derived PI resins showed good solubility in the polar aprotic solvents, such as N-methyl-2-pyrrolidone (NMP) and N,N-dimethylacetamide (DMAc). The flexible PI films obtained by the solution casting procedure showed good optical properties with the transmittances higher than 74.0% at the wavelength of 450 nm. The PI films exhibited excellent thermal properties, including 5% weight loss temperatures (T 5% ) over 510 °C, together with glass transition temperatures (T g ) over 350.0 °C according to the peak temperatures of the loss modulus in dynamical mechanical analysis (DMA) measurements. The FPI-6 film also showed the lowest linear coefficient of thermal expansion (CTE) value of 23.4 × 10 -6 /K from 50 to 250 °C according to the thermomechanical analysis (TMA) measurements, which was obviously lower than that of FPI-1 (CTE = 30.6 × 10 -6 /K).

Zearalenone (ZEA) is a mycotoxin with estrogenic activity whose main effect is to impair the reproductive systems of animals. It leads to reproductive disorders in livestock and thus causes serious losses to agriculture and animal husbandry. This study aims to examine whether ZEA induces toxicity in Leydig cells through the PI3K/AKT signaling pathway and also to investigate the role played by the upstream phosphatase and tensin homolog (PTEN) gene. An adenovirus vector model was constructed to interfere with the PTEN gene to investigate whether ZEA promotes the apoptosis of TM3 cells through the PI3K/AKT pathway. Apoptosis was detected cytometrically and the protein expression levels of PTEN, AKT, p-AKT, Bax, and Bcl-2 were evaluated via western blot analysis. The results show that ZEA induces apoptosis of TM3 cells. PTEN expression is significantly increased (P < 0.01), Bax expression is increased (P < 0.05), AKT and p-AKT expression of anti-apoptotic protein is significantly decreased (P < 0.01), and Bcl-2 protein expression is decreased (P < 0.05) in the ZEA group compared with the control group. In the shRNA+ZEA group, the expression levels of PTEN and Bax proteins are significantly decreased (P < 0.01), AKT protein is significantly increased (P < 0.01), and p-AKT protein is increased (P < 0.05) compared with the ZEA group. This study thus demonstrates that ZEA promotes apoptosis of TM3 cells by targeting PTEN and thus inhibiting the PI3K/AKT signal pathway., (© 2021. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

The amount of videos over the Internet and electronic surveillant cameras is growing dramatically, meanwhile paired sentence descriptions are significant clues to select attentional contents from videos. The task of natural language moment retrieval (NLMR) has drawn great interests from both academia and industry, which aims to associate specific video moments with the text descriptions figuring complex scenarios and multiple activities. In general, NLMR requires temporal context to be properly comprehended, and the existing studies suffer from two problems: (1) limited moment selection and (2) insufficient comprehension of structural context. To address these issues, a multi-agent boundary-aware network (MABAN) is proposed in this work. To guarantee flexible and goal-oriented moment selection, MABAN utilizes multi-agent reinforcement learning to decompose NLMR into localizing the two temporal boundary points for each moment. Specially, MABAN employs a two-phase cross-modal interaction to exploit the rich contextual semantic information. Moreover, temporal distance regression is considered to deduce the temporal boundaries, with which the agents can enhance the comprehension of structural context. Extensive experiments are carried out on two challenging benchmark datasets of ActivityNet Captions and Charades-STA, which demonstrate the effectiveness of the proposed approach as compared to state-of-the-art methods. The project page can be found in https://mic.tongji.edu.cn/e5/23/c9778a189731/page.htm.

Objective: To investigate the effects of airway epithelial cells on macrophages chemotaxis and inflammatory cytokine expression under hypoxic conditions.
 Methods: Human bronchial epithelial cells (HBE) treated with different concentrations (0, 100, 200, 400, 800 μmol/L) of CoCl2 or transfected with HIF-1α siRNA were co-cultured with THP-1-derived M1 macrophages or M2 macrophages. The chemotactic effects on macrophages were analyzed by Transwell assay. The levels of TNF-α, IFN-γ, IL-4, IL-13 and IL-10 in the supernatants of macrophages were detected by ELISA, and HIF-1α or Cav-1 mRNA expression in HBE or macrophages was detected by RT-qPCR.
 Results: HBE cells promoted macrophages chemotaxis in a time- and concentration-dependent manner. Compared to un-transfected group, the chemotactic ability of HBE transfected with HIF-1α siRNA was significantly weakened (P<0.01). Under the same culture conditions, the chemotaxis of M2 macrophages was greater than that in THP1-derived M1 macrophages. The concentrations of TNF-α, IFN-γ, IL-4, IL-13 and IL-10 in the supernatants of macrophages were increased in a time-and concentration-dependent manner. The concentrations of TNF-α and IFN-γ were increased further after co-culturing for 8 and 12 h; while IL-4, IL-13 and IL-10 concentrations were increased further during 24 h of co-culture. The levels of cytokines in the supernatants of macrophages co-cultured with HBE and transfected with HIF-1α siRNA were significantly lower than those in un-transfected cells (P<0.05 or P<0.01). The reduction of TNF-α or IFN-γ was more obvious. The expression of HIF-1α or Cav-1 mRNA in HBE or macrophages was increased in a concentration-dependent manner after 8 or 12 h co-culture, which was significantly reduced when HBE was transfected with HIF-1α siRNA.
 Conclusion: Airway epithelial cells can enhance macrophages chemotaxis and pro-inflammatory cytokines expressions under hypoxic condition. HIF-1α and Cav-1 may be the important mediators in these processes.