Search

The detection of unknown objects is a challenging task in computer vision because, although there are diverse real-world detection object categories, existing object-detection training sets cover a limited number of object categories. Most existing approaches use two-stage networks to improve a model's ability to characterize objects of unknown classes, which leads to slow inference. To address this issue, we proposed a single-stage unknown object detection method based on the contrastive language-image pre-training (CLIP) model and pseudo-labelling, called CLIP-YOLO. First, a visual language embedding alignment method is introduced and a channel-grouped enhanced coordinate attention module is embedded into a YOLO-series detection head and feature-enhancing component, to improve the model's ability to characterize and detect unknown category objects. Second, the pseudo-labelling generation is optimized based on the CLIP model to expand the diversity of the training set and enhance the ability to cover unknown object categories. We validated this method on four challenging datasets: MSCOCO, ILSVRC, Visual Genome, and PASCAL VOC. The results show that our method can achieve higher accuracy and faster speed, so as to obtain better performance of unknown object detection. The source code is available at https://github.com/BJUTsipl/CLIP-YOLO. [ABSTRACT FROM AUTHOR]

Road segmentation is a fundamental task for dynamic map in unmanned aerial vehicle (UAV) path navigation. In unplanned, unknown and even damaged areas, there are usually unpaved roads with blurred edges, deformations and occlusions. These challenges of unpaved road segmentation pose significant challenges to the construction of dynamic maps. Our major contributions have: (1) Inspired by dilated convolution, we propose dilated cross window self-attention (DCWin-Attention), which is composed of a dilated cross window mechanism and a pixel regional module. Our goal is to model the long-range horizontal and vertical road dependencies for unpaved roads with deformation and blurred edges. (2) A shifted cross window mechanism is introduced through coupling with DCWin-Attention to reduce the influence of occluded roads in UAV imagery. In detail, the GVT backbone is constructed by using the DCWin-Attention block for multilevel deep features with global dependency. (3) The unpaved road is segmented with the confidence map generated by fusing the deep features of different levels in a unified perceptual parsing network. We verify our method on the self-established BJUT-URD dataset and public DeepGlobe dataset, which achieves 67.72 and 52.67% of the highest IoU at proper inference efficiencies of 2.7, 2.8 FPS, respectively, demonstrating its effectiveness and superiority in unpaved road segmentation. Our code is available at https://github.com/BJUT-AIVBD/GVT-URS. [ABSTRACT FROM AUTHOR]

Currently, tracking by detection technique is widely used in crowd or vehicle counting. However, it is difficult to meet detection and counting with mixed pedestrian and vehicular traffic of smart park scenes in emergencies, for occluding small objects recognition problem. An improved detection model YOLOv5-FAC is proposed based YOLOv5s. First, a P2 detection layer is added to expand the detection range of the model and improve its detection ability of different sizes. Second, an auxiliary inference network is constructed using programmable gradient information to provide the model with a stronger information fitting capability. Finally, a cascading triplet attention mechanism is added to the head of model to increase the feature fusion capability. Then, a collision line counting method combined with OCSORT track technology is proposed, in which both direction movement of traffic is considered. The experimental results show that YOLOv5s-FAC has a significantly improved detection quality, with a mAP of 69.5%, and the counting accuracy reached 94.4%. [ABSTRACT FROM AUTHOR]

Salinity is a major abiotic stressor that impedes plant growth and negatively affects crop yield. However, Arbuscular mycorrhizal fungi (AMF) can establish a symbiotic relationship with over 80% of terrestrial plant roots. This relationship ultimately results in increased plant growth, improved plant stress resistance, and, consequently, a promising agricultural production and environmental protection solution. The aim of this experiment was to evaluate the impact of arbuscular mycorrhizal fungi on the photosynthetic physiology of strawberries under salinity stress. The greenhouse experiment involved the strawberry cultivar 'Benihoppe', which was inoculated with Glomus mosseae under three salt stress levels (0 mM, 30 mM, and 60 mM). Subsequently, the results showed that salinity stress led to a significant decline in leaf area, fresh biomass, and photosynthetic characteristics of the strawberries. Under salt stress, especially at the concentration of 60 mM. Pn, Gs, Tr, Ci, Fv/Fm, and NPQ showed significant differences. After inoculation of AMF, arbuscular mycorrhiza established a beneficial symbiotic relationship with strawberry roots, which effectively reduced salt damage and promoted the growth of strawberry plants. Leaf area, fresh biomass, and relative chlorophyll content were significantly increased. Pn, Tr, and Gs of mycorrhizal strawberry were significantly higher than those of control group. In addition, the light energy conversion efficiency of strawberry plants inoculated with AMF was improved, thus increasing the potential photosynthetic capacity and photosynthetic rate of strawberry. [ABSTRACT FROM AUTHOR]

Industrial hemp (Cannabis sativa L.) is a multifaced crop that has the potential to be exploited for many industrial applications, and making use of salt lands is considered to be a sustainable development strategy for the hemp industry. However, no elite salt-tolerant hemp varieties have been developed, and therefore supplementing appropriate exogenous substances to saline soil is one possible solution. Calcium-containing compounds are well-known for their salt tolerance enhancing effects, but the underlying molecular mechanisms remain largely unclear. Here, we first assessed the ameliorative effects of calcium amendments on salt-stressed hemp plants and then investigated these mechanisms on hemp using integrative analysis of proteomics and metabolomics. The stress phenotypes could be lessened by Ca2+ treatment. Certain concentrations of Ca2+ maintained relative electrical conductivity and the contents of malondialdehyde and chlorophyll. Ca2+ treatment also generally led to greater accumulations of soluble proteins, soluble carbohydrates and proline, and enhanced the activities of superoxide dismutase and peroxidase. Through functional classification, pathway enrichment, and network analysis, our data reveal that accumulation of dipeptides is a prominent metabolic signature upon exogenous Ca2+ treatment, and that changes in mitochondrial properties may play an important role in enhancing the salt tolerance. Our results outline the complex metabolic alternations involved in calcium-mediated salt stress resistance, and these data and analyses would be useful for future functional studies.Key message: Integrated omics analyses outline cellular and metabolic processes underlying calcium-mediated salt tolerance in hemp and highlight the important roles of accumulation of dipeptides and changes in mitochondrial properties. [ABSTRACT FROM AUTHOR]

Recent research highlights the significant impact of methionine metabolism on glioma progression. An increasing amount of compelling evidence bridges long non-coding RNAs to abnormal metabolism in gliomas. However, the specific role of long non-coding RNAs in methionine metabolism regulating glioma progression remains unclear. This study reveals that methionine deprivation inhibits the proliferation, migration, and invasion capabilities of gliomas. Interestingly, the expression of TP53TG1, a long non-coding RNA, is also suppressed. TP53TG1 is highly expressed in gliomas and associated with poor patient outcomes. Subsequently, our data proves that inhibition of TP53TG1 suppresses glioma cell proliferation and the epithelial-mesenchymal transition process both in vitro and in vivo. Ultimately, we found that the underlying mechanism involves a competing endogenous RNA regulating network, in which TP53TG1 modulates the target protein STK17B by competitively binding to miR-96-5p, thus regulating glioma progression. These findings suggest that targeting methionine deprivation could be a promising approach for the clinical treatment of glioma. [ABSTRACT FROM AUTHOR]

Highlights: Nonflammable gel polymer electrolyte (SGPE) is developed by in situ polymerizing trifluoroethyl methacrylate (TFMA) monomers with flame-retardant triethyl phosphate (TEP) solvents and LiTFSI–LiDFOB dual lithium salts. Molecular polarity interaction between TEP and PTFMA mitigates interfacial reactions and changes the solvation of Li+. SGPE forms stable inorganic-rich solid electrolyte interface/cathode electrolyte interface layer, exhibiting well compatibility with Li anode and LiCoO2-type high-voltage cathode. The risk of flammability is an unavoidable issue for gel polymer electrolytes (GPEs). Usually, flame-retardant solvents are necessary to be used, but most of them would react with anode/cathode easily and cause serious interfacial instability, which is a big challenge for design and application of nonflammable GPEs. Here, a nonflammable GPE (SGPE) is developed by in situ polymerizing trifluoroethyl methacrylate (TFMA) monomers with flame-retardant triethyl phosphate (TEP) solvents and LiTFSI–LiDFOB dual lithium salts. TEP is strongly anchored to PTFMA matrix via polarity interaction between -P = O and -CH2CF3. It reduces free TEP molecules, which obviously mitigates interfacial reactions, and enhances flame-retardant performance of TEP surprisingly. Anchored TEP molecules are also inhibited in solvation of Li+, leading to anion-dominated solvation sheath, which creates inorganic-rich solid electrolyte interface/cathode electrolyte interface layers. Such coordination structure changes Li+ transport from sluggish vehicular to fast structural transport, raising ionic conductivity to 1.03 mS cm−1 and transfer number to 0.41 at 30 °C. The Li|SGPE|Li cell presents highly reversible Li stripping/plating performance for over 1000 h at 0.1 mA cm−2, and 4.2 V LiCoO2|SGPE|Li battery delivers high average specific capacity > 120 mAh g−1 over 200 cycles. This study paves a new way to make nonflammable GPE that is compatible with Li metal anode. [ABSTRACT FROM AUTHOR]

The grouted bolt, combining rock bolting with grouting techniques, provides an effective solution for controlling the surrounding rock in deep soft rock and fractured roadways. It has been extensively applied in numerous deep mining areas characterized by soft rock roadways, where it has demonstrated remarkable control results. This article systematically explores the evolution of grouted bolting, covering its theoretical foundations, design methods, materials, construction processes, monitoring measures, and methods for assessing its effectiveness. The overview encompassed several key elements, delving into anchoring theory and grouting reinforcement theory. The new principle of high pretensioned high-pressure splitting grouted bolting collaborative active control is introduced. A fresh method for dynamic information design is also highlighted. The discussion touches on both conventional grouting rock bolts and cable bolts, as well as innovative grouted rock bolts and cables characterized by their high pretension, strength, and sealing hole pressure. An examination of the merits and demerits of standard inorganic and organic grouting materials versus the new inorganic–organic composite materials, including their specific application conditions, was conducted. Additionally, the article presents various methods and instruments to assess the support effect of grouting rock bolts, cable bolts, and grouting reinforcement. Furthermore, it provides a foundation for understanding the factors influencing decisions on grouted bolting timing, the sequence of grouting, the pressure applied, the volume of grout used, and the strategic arrangement of grouted rock bolts and cable bolts. The application of the high pretensioned high-pressure splitting grouted bolting collaborative control technology in a typical kilometer-deep soft rock mine in China—the soft coal seam and soft rock roadway in the Kouzidong coal mine, Huainan coal mining area, was introduced. Finally, the existing problems in grouted bolting control technology for deep soft rock roadways are analyzed, and the future development trend of grouted bolting control technology is anticipated. [ABSTRACT FROM AUTHOR]

Hemp (Cannabis sativa L.) is an environmentally friendly crop with a rich history and promising market prospects. To avoid competing with staple crops for land, we propose that utilizing salt-affected soil could be a potential strategy for developing the hemp industry. However, hemp plants are glycophytes, which means that their salt tolerance needs to be improved. Phytohormones are known to play critical roles in regulating plant responses to stress. Thus, in this study, the effects of foliar spraying of abscisic acid (ABA), salicylic acid (SA), gibberellin A3 (GA3) and α-naphthalene acetic acid (NAA) on the salt tolerance of hemp seedlings were evaluated. The seedlings were treated with 200-mmol/L NaCl and different concentrations of ABA, SA, GA3, or NAA through sand culture experiments. The results showed that the growth of salt-treated hemp seedlings (in terms of plant height, stem diameter, and biomass) was significantly improved by spraying 250-mg/L ABA, 250-mg/L SA, 150-mg/L GA3, or 100-mg/L NAA. Spaying the hemp salt-treated seedlings with these phytohormone concentrations decreased the malondialdehyde content, Na+/K+ ratio, and electrolyte leakage but increased the endogenous indole-3-acetic acid, chlorophyll, carotenoid, soluble sugar, and soluble protein contents, K+ concentration, superoxide dismutase activity, and peroxidase activity. This study is the first to show the effectiveness of ABA, SA, GA3, and NAA in alleviating the adverse effects of salt on hemp seedlings. These novel findings suggest that the exogenous application of phytohormones may serve as a potential management tool to improve the salt tolerance of hemp plants. [ABSTRACT FROM AUTHOR]

Images are often affected by insufficient illumination and suffer from degradation problems such as low brightness, noise, and color distortion, which results in reduced image quality. Existing low-light image enhancement methods based on Retinex theory decompose images into reflectance and illumination components, which are adjusted separately; however, the intrinsic connection between reflectance and illumination during decomposition is not considered, and multi-scale information during subsequent adjustments is inadequately utilized. In this study, we propose a low-light image enhancement network based on Retinex decomposition and multi-scale adjustment (RDMA), which performs initial decomposition followed by subsequent adjustment. We utilized prior knowledge to design the feature interaction module (FIM) and the feature fusion module (FFM) for image decomposition. Furthermore, a coarse-to-fine multi-scale network with residual channel and spatial attention (RCSA) was designed to remove noise from reflectance, suppress color distortion, preserve image details, and adjust the brightness of illumination. An evaluation of various low-light image datasets and comparisons with state-of-the-art methods showed that the proposed network is superior in terms of enhancement results. [ABSTRACT FROM AUTHOR]

This paper addresses the finite-time control problem of a class of uncertain nonlinear systems subject to input saturation and actuator faults. To approximate the unknown system states, a fuzzy state observer is established where fuzzy logic systems are utilized to estimate the unknown nonlinearities. A nonlinear disturbance observer to estimate the external disturbance of the system. A new finite-time adaptive fuzzy controller is constructed together with the proposed disturbance and state observers, which can guarantee the tracking error into a small neighborhood around zero within finite time. To avoid the tedious and arithmetic problems brought by the traditional backstepping control, the dynamic surface technique is applied, which can effectively reduce the computation burden. It can be proved that not only the boundedness of the closed-loop system states can be guaranteed but also the tracking error is regulated into a small range near the equilibrium in finite time, despite unknown actuator faults and input saturation. Finally, the simulations of two-stage chemical reactor nonlinear system are conducted to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Locoregional recurrence and distant metastasis of breast cancer still pose a significant risk for patients' survival. To address the clinical challenge, functional absorbable sponges (HA-SH/PP-Dox/Lap/COL I (HCNPs)) were constructed by biomimetic extracellular matrix of collagen I / hyaluronic acid complex conjugated with doxorubicin/lapatinib (Dox/Lap)-loaded nanoparticles. The HCNPs sponge exhibited excellent clotting ability and blood absorption rate. Worthily, Dox/Lap-loaded nanoparticles were synchronously endowed with a large number of oligo hyaluronic acid segments after degradation, which thus enhanced the ability of targeting into CD44-overexpressed tumor cells. The implantable HCNPs sponge in resected cavity of postoperative 4T1 models inhibited the spread of scattered tumor cells by absorbing the inevitable bleeding. More importantly, CD44 targeted nanoparticle with suitable Dox/Lap proportion continuously released from sponge to kill tumor cells of surrounding HCNPs and those remaining at surgical margin, thus prevented local recurrence as well as distant metastasis. Therefore, the functional HCNPs sponge might provide a safer and more effective strategy for postoperative treatment of cancer. [ABSTRACT FROM AUTHOR]

Haze reduces the imaging effectiveness of outdoor vision systems, significantly degrading the quality of images; hence, reducing haze has been a focus of many studies. In recent years, decoupled representation learning has been applied in image processing; however, existing decoupled networks lack a specific design for information with different characteristics to achieve satisfactory results in dehazing tasks. This study proposes a heterogeneous decoupling unsupervised dehazing network (HDUD-Net). Heterogeneous modules are used to learn the content and haze information of images individually to separate them effectively. To address the problem of information loss when extracting the content from hazy images with complex noise, this study proposes a bi-branch multi-hierarchical feature fusion module. Additionally, it proposes a style feature contrast learning method to generate positive and negative sample queues and construct contrast loss for enhancing decoupling performance. Numerous experiments confirm that the proposed algorithm achieves higher performance according to objective metrics and a more realistic visual effect when compared with state-of-the-art single-image dehazing algorithms. [ABSTRACT FROM AUTHOR]

Visual tracking is widely used in industrial systems such as vision servo systems and in intelligent robots. However, most tracking algorithms are designed without considering the balance of algorithmic efficiency and accuracy in system applications, making them less preferable for applications. This paper proposes a siamese global location-aware object tracking algorithm (SiamGLA) to address this issue. First, due to the limited performance of efficient lightweight backbone networks, this study designs an internal feature combination (IFC) module that improves feature representation with almost no additional parameters. Second, a global-aware (GA) attention module is proposed to improve the classification ability of foreground and background, which is especially important for trackers. Finally, a location-aware (LA) attention module is designed to improve the regression performance of the tracking framework. Comprehensive experiments show that SiamGLA is effective, and overcomes the drawbacks of poor robustness and weak generalization ability. When the performance reaches state-of-the-art, SiamGLA requires fewer calculations and parameters, making it more likely to be applied in practice. [ABSTRACT FROM AUTHOR]

Hunga Tonga-Hunga Ha'apai climactic eruption on January 15, 2022, released enormous energy that affected the ionosphere over the Pacific Rim. We analyzed ionospheric disturbance following volcanic eruptions using near-field (<1000km), regional (1000–5000 km), and far-field (5000–12000 km) global positioning system (GPS) observations. The results indicate that the near-field ionospheric perturbation that occurred 8–15 min after the cataclysmic eruption was mainly derived from the shock wave (~1000 m/s) generated by the blast, while the low-frequency branch with long-distance propagation characteristics over the regional and the far-field was mainly associated with atmospheric Lamb waves (~330 m/s). Moreover, the amplitude of disturbance and background total electron content (TEC) are related proportionally. The intensity of the volcanic eruption and the background ionospheric conditions determine the magnitude of ionospheric responses. TEC perturbations were invisible on the reference days. Furthermore, the source location and onset time were calculated using the ray tracing technique, which confirms that the Tonga event triggered the ionospheric anomaly beyond the crater. Finally, the change in the frequency of the perturbations coincided with the arrival of the initial tsunami, implying the generation of a meteotsunami. [ABSTRACT FROM AUTHOR]

Fast and accurate prohibited object detection in X-ray images is great challenging. Based on YOLOv6 object detection framework, in this paper, Channel-Target Attention Feature Pyramid Network (CTA-FPN) is proposed for prohibited object detection in X-ray images. It includes two key components: TAAM (Target Aware Attention Module) and CAM (Channel Attention Module). TAAM is to generate the target attention map to enhance the features of prohibited object regions and suppress those of the background regions, so as to solve the problems of object occlusion and cluttered background in X-ray images. CAM is to highlight the feature channels important to the detection tasks, and suppress the irrelevant ones. The target-wise and channel-wise feature enhancement can effectively strengthen the feature representation capability of the network. The proposed CTA-FPN is incorporated into S, M and L models of YOLOv6 respectively, obtaining three X-ray prohibited object detection models. The experimental results on two publicly available benchmark datasets of SIXray and CLCXray show that, CTA-FPN can effectively improve the detection performance of YOLOv6. Especially, YOLOv6-CTA-FPN-L can achieve the state-of-the-arts detection accuracy. [ABSTRACT FROM AUTHOR]

In HTTP adaptive video streaming service, video clarity and fluency are the two most important Influencing Factors (IFs) affecting user's Quality of Experience (QoE). In this paper, a Video Clarity-Fluency Network (VCFNet) is proposed to establish a QoE evaluation model, which focuses on characterizing the clarity-fluency characteristics of video. Firstly, the Harmonic-ResNeXt101 network is constructed by introducing the Harmonic Network into ResNeXt101 to capture the clarity information of video frames. The output of the Fully Connected (FC) layer of the Harmonic-ResNeXt101 network is fed into the Gated Recurrent Unit (GRU), which is used to perform short-term temporal modeling to capture the fluency information of video chunk. Then, the final output of GRU is extracted as the clarity-fluency features, which are concatenated with the statistical features of other IFs (including video quality level, re-buffering duration, re-buffering frequency, etc.) to form the feature parameter vector of IFs. Finally, a neural network composed of One-Dimensional Convolutional Neural Network (1D CNN) layer and two FC layers is designed to establish the mapping relationship model between the feature parameter vector of IFs and Mean Opinion Score (MOS) to predict user's QoE. Experimental results on SQoE-III and SQoE-IV datasets demonstrate that the proposed VCFNet can effectively capture the clarity-fluency information of video, and the resulted QoE model can achieve the state-of-the-art performance compared with the existing QoE evaluation models. [ABSTRACT FROM AUTHOR]

The current systematic review and meta-analysis examined the effect of racemic ketamine or esketamine on suicidal ideation in individuals with uni- or bipolar depression. We searched the MEDLINE, Embase, Central, PsycINFO, and Web of Science databases to identify randomized controlled trials that examined the effect of racemic ketamine or esketamine monotherapy on suicidal ideation (SI) in individuals with uni- or bipolar depression. The two monotherapies were compared; the primary outcome was the rate of remission of SI, and the secondary outcome was the SI score. The risk ratio was used as an effect size measure for binary variables, while the standardized mean difference was used as an effect size measure for continuous variables. Our meta-analysis included 13 randomized controlled trials involving 1,1109 individuals with uni- or bipolar depression. Patients receiving racemic ketamine monotherapy had a significantly higher acute SI remission rate than those receiving placebo or midazolam (RR = 2.06, 95% CI 1.47 to 2.91, P < 0.0001). Racemic ketamine also led to significantly lower SI scores than placebo or midazolam (SMD = -0.36, 95% CI -0.71 to -0.01, P = 0.04). The evidence for the treatment of SI with esketamine was inconsistent. The pooled effect sizes for long-term anti-SI effects did not reveal significant differences between therapies. Our study indicated the efficacy of racemic ketamine monotherapy for rapidly and transiently reducing SI in individuals with uni- or bipolar depression, but the efficacy of racemic ketamine monotherapy against long-term suicidal ideation remains unclear. There is not -sufficient evidence to support the anti-suicidal effects of esketamine monotherapy.Protocol registration: Prospero registration number: CRD42023434380. [ABSTRACT FROM AUTHOR]

It is highly important and challenging to develop donor-acceptor-donor structured small-molecule second near-infrared window (NIR-II) dyes with excellent properties such as water-solubility and chem/photostability. Here, we discovery an electron acceptor, 6,7-di(thiophen-2-yl)-[1,2,5]thiadiazolo[3,4-g]quinoxaline (TQT) with highest stability in alkaline conditions, compared with conventional NIR-II building block benzobisthiadiazole (BBT) and 6,7-diphenyl-[1,2,5] thiadiazolo[3,4-g]quinoxaline (PTQ). The sulfonated hydrophilic dye, FT-TQT, is further synthesized with 2.13-fold increased quantum yield than its counterpart FT-BBT with BBT as acceptor. FT-TQT complexed with FBS is also prepared and displays a 16-fold increase in fluorescence intensity compared to FT-TQT alone. It demonstrates real-time cerebral and tumor vessel imaging capability with µm-scale resolution. Dynamic monitoring of tumor vascular disruption after drug treatment is achieved by NIR-II fluorescent imaging. Overall, TQT is an efficient electron acceptor for designing innovative NIR-II dyes. The acceptor engineering strategy provides a promising approach to design next generation of NIR-II fluorophores which open new biomedical applications. Small molecule NIR-II fluorophores are of interest for a range of applications but can suffer from chemical and photostability issues. Here, the authors report on the development of an acceptor molecule with improved stability in alkaline conditions expanding the range of possible applications. [ABSTRACT FROM AUTHOR]

For mobile streaming media service providers, it is necessary to accurately predict the quality of experience (QoE) to formulate appropriate resource allocation and service quality optimization strategies. In this paper, a QoE evaluation model is proposed by considering various influencing factors (IFs), including perceptual video quality, video content characteristics, stalling, quality switching and video genre attribute. Firstly, a no-reference video multimethod assessment fusion (VMAF) model is constructed to measure the perceptual quality of the video by the deep bilinear convolutional neural network. Then, the deep spatio-temporal features of video are extracted using a TSM-ResNet50 network, which incorporates temporal shift module (TSM) with ResNet50, obtaining feature representation of video content characteristics while balancing computational efficiency and expressive ability. Secondly, video genre attribute, which reflects the user's preference for different types of videos, is considered as a IF while constructing the QoE model. The statistical parameters of other IFs, including the video genre attribute, stalling and quality switching, are combined with VMAF and deep spatio-temporal features of video to form an overall description parameters vector of IFs for formulating the QoE evaluation model. Finally, the mapping relationship model between the parameters vector of IFs and the mean opinion score is established through designing a deep neural network. The proposed QoE evaluation model is validated on two public video datasets: WaterlooSQoE-III and LIVE-NFLX-II. The experimental results show that the proposed model can achieve the state-of-the-art QoE prediction performance. [ABSTRACT FROM AUTHOR]

With the sharp booming of online live streaming platforms, some anchors seek profits and accumulate popularity by mixing inappropriate content into live programs. After being blacklisted, these anchors even forged their identities to change the platform to continue live, causing great harm to the network environment. Therefore, we propose an anchor voiceprint recognition in live streaming via RawNet-SA and gated recurrent unit (GRU) for anchor identification of live platform. First, the speech of the anchor is extracted from the live streaming by using voice activation detection (VAD) and speech separation. Then, the feature sequence of anchor voiceprint is generated from the speech waveform with the self-attention network RawNet-SA. Finally, the feature sequence of anchor voiceprint is aggregated by GRU to transform into a deep voiceprint feature vector for anchor recognition. Experiments are conducted on the VoxCeleb, CN-Celeb, and MUSAN dataset, and the competitive results demonstrate that our method can effectively recognize the anchor voiceprint in video streaming. [ABSTRACT FROM AUTHOR]

Purpose: Critically ill COVID-19 patients have significantly increased risk of death. Although several circulating biomarkers are thought to be related to COVID-19 severity, few studies have focused on the characteristics of critically ill patients with different outcomes. The objective of this study was to perform a longitudinal investigation of the potential mechanisms affecting the prognosis of critically ill COVID-19 patients. Methods: In addition to clinical data, 113 whole blood samples and 85 serum samples were collected from 33 severe and critical COVID-19 patients without selected comorbidities. Multi-omics analysis was then performed using longitudinal samples. Results: Obvious transcriptional transitions were more frequent in critical survivors than in critical non-survivors, indicating that phase transition may be related to survival. Based on analysis of differentially expressed genes during transition, the erythrocyte differentiation pathway was significantly enriched. Furthermore, clinical data indicated that red blood cell counts showed greater fluctuation in survivors than in non-survivors. Moreover, declining red blood cell counts and hemoglobin levels were validated as prognostic markers of poor outcome in an independent cohort of 114 critical COVID-19 patients. Protein–metabolite–lipid network analysis indicated that tryptophan metabolism and melatonin may contribute to molecular transitions in critical COVID-19 patients with different outcomes. Conclusions: This study systematically and comprehensively depicted the longitudinal hallmarks of critical COVID-19 patients and indicated that multi-omics transition may impact the prognosis. Take home message: Frequent transcriptional phase transitions may contribute to outcome in critically ill COVID-19 patients. Furthermore, fluctuation in red blood cell and hemoglobin levels may relate to poor prognosis. The biological function of melatonin was suppressed in COVID-19 non-survivors, which may provide a potential theoretical basis for clinical administration. [ABSTRACT FROM AUTHOR]

Although image enhancement methods have been widely applied in various outdoor vision systems, the existing methods still face two critical problems. On the one hand, the existing methods only consider a single degradation. However, in practical applications, image quality is usually degraded by multiple factors. The methods designed for the single degradation factor cannot achieve good performance when facing multi-degraded images. On the other hand, the imaging model-based enhancement methods which use prior knowledge or handcrafted features to perform image enhancement may bring some fitting errors. Therefore, considering multiple degradations in images, an image enhancement method is proposed in this paper. Firstly, a new image degradation model based on the multiple scattering model is proposed, which is used to characterize multiple degradations caused by haze, mixed with blur and noise. Then, an image enhancement convolutional neural network (CNN) based on ResNet is proposed to learn the implicit mapping model between low-quality and high-quality images in the pixel domain directly. The CNN network has been trained with an end-to-end learning manner. Experimental results on the synthetic dataset and real-world hazy images verify the superiority of the proposed method, while compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Heap biooxidation method was used to evaluate the availability of Paodaoling gold ore in Anhui province, China. 15,000 tons of gold ores (≤ 10 mm in diameter) were bioxidized under mesophilic conditions. Under the synergistic effect of microbial community, arsenic and sulfur were oxidized by 42% and 38% after 80 days. Relatively, leaching of gold was improved from 36 to 78% after heap biooxidation. The sequencing results showed there were 28 operational taxonomic units identified the microbial community in the heap. The main genera were Acidithiobacillus, Ferroplasma, Acidiferrobacter and Nitrospira. According to stoichiometry, the content of microorganisms with various functions tended to be balanced. The biomass production rate was 10 g/s, the CO2 fixation rate was 18 g/s, and the oxygen consumption rate was 60 g/s. This study provides a good basis for the further design and application of heap biooxidation technology. [ABSTRACT FROM AUTHOR]

Color is one of the most stable attributes of vehicles and often used as a valuable cue in some important applications. Various complex environmental factors, such as illumination, weather, noise and etc., result in the visual characteristics of the vehicle color being obvious diversity. Vehicle color recognition in complex environments has been a challenging task. The state-of-the-arts methods roughly take the whole image for color recognition, but many parts of the images such as car windows; wheels and background contain no color information, which will have negative impact on the recognition accuracy. In this paper, a novel vehicle color recognition method using local vehicle-color saliency detection and dual-orientational dimensionality reduction of convolutional neural network (CNN) deep features has been proposed. The novelty of the proposed method includes two parts: (1) a local vehicle-color saliency detection method has been proposed to determine the vehicle color region of the vehicle image and exclude the influence of non-color regions on the recognition accuracy; (2) dual-orientational dimensionality reduction strategy has been designed to greatly reduce the dimensionality of deep features that are learnt from CNN, which will greatly mitigate the storage and computational burden of the subsequent processing, while improving the recognition accuracy. Furthermore, linear support vector machine is adopted as the classifier to train the dimensionality reduced features to obtain the recognition model. The experimental results on public dataset demonstrate that the proposed method can achieve superior recognition performance over the state-of-the-arts methods. [ABSTRACT FROM AUTHOR]

Vehicle classification plays an important role in intelligent transport system. However, because the conventional vehicle classification methods are not robust to variations such as illumination, weather, noise, and the classification accuracy cannot meet the requirements of practical applications. Therefore, a new vehicle classification method using Convolutional Neural Networks is proposed in this paper, which consists of two steps: pre-training and fine-tuning. In pre-training, GoogLeNet is pre-trained on ILSVRC-2012 dataset to obtain the initial model with the corresponding connection weights. In fine-tuning, the initial model is further fine-tuned on VehicleDataset which is constructed with 13,700 images in this paper to obtain the final classification model. All images in the VehicleDataset are extracted from real highway surveillance videos, including variations of illumination, noise, resolution, angle of video cameras and weather. The vehicles are divided into six categories, i.e., bus, car, motorcycle, minibus, truck and van. The performance evaluation is carried out on the VehicleDataset. The experimental results show that the proposed method can avoid the complicated process of manually extracting features and the average classification accuracy is up to 98.26%, which is 3.42% higher than the conventional methods using 'Feature + Classifier'. [ABSTRACT FROM AUTHOR]

This has been corrected in both the PDF and HTML versions of the Article. Correction to: I Nature Communications i https://doi.org/10.1038/s41467-022-31521-y, published online 02 July 2022 The original version of this Article contained an error in the figure legend of Figure 2L where "TPA-TQT" was mistakenly written as "TPA-BBT". [Extracted from the article]

In the original publication the section heading "Classification of microorganisms" appearing above the sub-section "Air and liquid circulation in the heap" in page four is incorrect. The correct section heading should be read as "Results and discussion". [ABSTRACT FROM AUTHOR]

Brasenia schreberi J. F. Gmel. (Cabombaceae), a perennial freshwater macrophyte characterized by a thick mucilage on all underwater organs and especially young buds, has been widely cultivated as an aquatic vegetable in China for many years but is now listed as an endangered species due to anthropogenic impacts and habitat loss. Recent studies have demonstrated that different B. schreberi populations in China have low levels of genetic diversity but significantly different mucilage contents (MucC). Considering the importance of mucilage on both economic and ecological aspects, we examined mucilage-environment relationships in three B. schreberi cultivation sites. The results indicated that water permanganate index (CODMn), total N (TNw), electrical conductivity (ECw), dissolved oxygen (DOw), sediment organic carbon (SOC) and total N (TNs) were significant factors, which explained 82.2% of the variation in mucilage accumulation. The MucC and mucilage thickness (MucT) as well as single bud weight (SBW) of B. schreberi showed negative relationships with CODMn, TNw and ECw but positive relationships with SOC and TNs. Besides, high temperature may have a negative impact on mucilage accumulation of the species. Our study demonstrated that the mucilage accumulation of B. schreberi required good water quality and nutrient-enriched sediments, suggesting that habitat conservation, especially the quality of water, is important for maintaining B. schreberi populations. [ABSTRACT FROM AUTHOR]