Search

This paper minimizes fake news, which has been a hot topic recently, using blockchain and artificial intelligence technology, and verifies it with blockchain. Also, using Artificial Intelligence technology, we want to create an algorithm that predicts how fake news will spread in the future. You can see various attempts at a news media platform based on Blockchain technology. However, the Blockchain news media platform is still not getting the market response we expected. It is questionable whether the reason is simply because it is a new technology, so it takes a long time to gain trust from consumers, whether consumers are not yet expecting an innovative news media platform, or whether the explosive growth of the Blockchain news media platform is difficult for other reasons. Research to answer this or direct research between Blockchain and media platforms is still lacking. In addition, the method of verifying fake news using artificial intelligence was verified, ANN, CBR, and MDA were changed, and the experiment was verified for progress. In addition, the use of 5-fold cross-validation as a comparative method was added as described above to more closely examine the possibility of its usefulness even in general situations. Also, through various fields of artificial intelligence and blockchain, verification work was done with blockchain, and fake news prediction was made using artificial intelligence. Various experiments were conducted and performance tests were performed, while the performance of about 5,000 TTPS was recorded through the third experiment. In the future, we think it is necessary to combine Artificial Intelligence and blockchain technology.

This work describes a self-powered wireless temperature sensor platform that can be used for foot ulceration monitoring for diabetic patients. The proposed self-powered sensor platform consists of a piezoelectric bimorph, a power conditioning circuit, a temperature sensor readout circuit, and a wireless module. The piezoelectric bimorph mounted inside the shoe effectively converts the foot movement into electric energy that can power the entire sensor platform. Furthermore, a sensor platform was designed, considering the energy requirement of 4.826 mJ for transmitting one data packet of 18 bytes. The self-powered sensor platform prototype was evaluated with five test subjects with different weights and foot shapes; the test results show the subjects had to walk an average of 119.6 s to transmit the first data packet and an additional average of 71.2 s to transmit the subsequent data packet. The temperature sensor showed a resolution of 0.1 °C and a sensitivity of 56.7 mV/°C with a power conditioning circuit efficiency of 74.5%.

Photovoltaic power generation is influenced not only by variable environmental factors, such as solar radiation, temperature, and humidity, but also by the condition of equipment, including solar modules and inverters. In order to preserve energy production, it is essential to maintain and operate the equipment in optimal condition, which makes it crucial to determine the condition of the equipment in advance. This paper proposes a method of determining a degradation of efficiency by focusing on photovoltaic equipment, especially inverters, using LSTM (Long Short-Term Memory) for maintenance. The deterioration in the efficiency of the inverter is set based on the power generation predicted through the LSTM model. To this end, a correlation analysis and a linear analysis were performed between the power generation data collected at the power plant to learn the power generation prediction model and the data collected by the environmental sensor. With this analysis, a model was trained using solar radiation data and power data that are highly correlated with power generation. The results of the evaluation of the model’s performance show that it achieves a MAPE of 7.36, an RMSE of 27.91, a MAE of 18.43, and an R2 of 0.97. The verified model is applied to the power generation data of the selected inverters for the years 2020, 2021, and 2022. Through statistical analysis, it was determined that the error rate in 2022, the third year of its operation, increased by 159.55W on average from the error rate of the power generation forecast in 2020, the first year of operation. This indicates a 0.75% decrease in the inverter’s efficiency compared to the inverter’s power generation capacity. Therefore, it is judged that it can be applied effectively to analyses of inverter efficiency in the operation of photovoltaic plants.

Recently, with the remarkable development of deep learning technology, achievements are being updated in various computer vision fields. In particular, the object recognition field is receiving the most attention. Nevertheless, recognition performance for small objects is still challenging. Its performance is of utmost importance in realistic applications such as searching for missing persons through aerial photography. The core structure of the object recognition neural network is the feature pyramid network (FPN). You Only Look Once (YOLO) is the most widely used representative model following this structure. In this study, we proposed an attention-based scale sequence network (ASSN) that improves the scale sequence feature pyramid network (ssFPN), enhancing the performance of the FPN-based detector for small objects. ASSN is a lightweight attention module optimized for FPN-based detectors and has the versatility to be applied to any model with a corresponding structure. The proposed ASSN demonstrated performance improvements compared to the baselines (YOLOv7 and YOLOv8) in average precision (AP) of up to 0.6%. Additionally, the AP for small objects (APS) showed also improvements of up to 1.9%. Furthermore, ASSN exhibits higher performance than ssFPN while achieving lightweightness and optimization, thereby improving computational complexity and processing speed. ASSN is open-source based on YOLO version 7 and 8. This can be found in our public repository: https://github.com/smu-ivpl/ASSN.git

Synthetic oleanane triterpenoids (SOTs) are small molecules with broad anticancer properties. A recently developed SOT, 1‐[2‐cyano‐3,12‐dioxooleana‐1,9(11)‐dien‐28‐oyl]‐4(‐pyridin‐2‐yl)‐1H‐imidazole (CDDO‐2P‐Im or ‘2P‐Im’), exhibits enhanced activity and improved pharmacokinetics over CDDO‐Im, a previous generation SOT. However, the mechanisms leading to these properties are not defined. Here, we show the synergy of 2P‐Im and the proteasome inhibitor ixazomib in human multiple myeloma (MM) cells and 2P‐Im activity in a murine model of plasmacytoma. RNA sequencing and quantitative reverse transcription PCR revealed the upregulation of the unfolded protein response (UPR) in MM cells upon 2P‐lm treatment, implicating the activation of the UPR as a key step in 2P‐Im‐induced apoptosis. Supporting this hypothesis, the deletion of genes encoding either protein kinase R‐like endoplasmic reticulum kinase (PERK) or DNA damage‐inducible transcript 3 protein (DDIT3; also known as CHOP) impaired the MM response to 2P‐Im, as did treatment with ISRIB, integrated stress response inhibitor, which inhibits UPR signaling downstream of PERK. Finally, both drug affinity responsive target stability and thermal shift assays demonstrated direct binding of 2P‐Im to endoplasmic reticulum chaperone BiP (GRP78/BiP), a stress‐inducible key signaling molecule of the UPR. These data reveal GRP78/BiP as a novel target of SOTs, and specifically of 2P‐Im, and suggest the potential broader utility of this class of small molecules as modulators of the UPR.

Green information technology (Green TI/GC) is provided by green computing, one of the emerging computing technologies in field of computer science engineering and technology. The main objectives of green computing are raising energy efficiency and lowering the use of hazardous materials. This Research propose novel technique in energy management with security enhancement based on heuristic green computing technique and optimised routing protocol. The energy management in green computing is carried out for enhancing the energy efficiency of the network using heuristic green computing technique. Then the security analysis has been carried out for secure data transmission of the network using Hybrid Greedy Secure Optimal Routing Protocol (Hy_GSOpRP). The suggested routing method is superior than the existing ones, according to simulation findings, in terms of MAPE of 62%, MLR of 55%, RMSE of 45%, energy consumption of 59%, network lifetime of 95%, delay of 40% and throughput of 98%.

Abstract Peptidylarginine deiminase (PADI) 2 catalyzes the post-translational conversion of peptidyl-arginine to peptidyl-citrulline in a process called citrullination. However, the precise functions of PADI2 in bone formation and homeostasis remain unknown. In this study, our objective was to elucidate the function and regulatory mechanisms of PADI2 in bone formation employing global and osteoblast-specific Padi2 knockout mice. Our findings demonstrate that Padi2 deficiency leads to the loss of bone mass and results in a cleidocranial dysplasia (CCD) phenotype with delayed calvarial ossification and clavicular hypoplasia, due to impaired osteoblast differentiation. Mechanistically, Padi2 depletion significantly reduces RUNX2 levels, as PADI2-dependent stabilization of RUNX2 protected it from ubiquitin-proteasomal degradation. Furthermore, we discovered that PADI2 binds to RUNX2 and citrullinates it, and identified ten PADI2-induced citrullination sites on RUNX2 through high-resolution LC-MS/MS analysis. Among these ten citrullination sites, the R381 mutation in mouse RUNX2 isoform 1 considerably reduces RUNX2 levels, underscoring the critical role of citrullination at this residue in maintaining RUNX2 protein stability. In conclusion, these results indicate that PADI2 plays a distinct role in bone formation and osteoblast differentiation by safeguarding RUNX2 against proteasomal degradation. In addition, we demonstrate that the loss-of-function of PADI2 is associated with CCD, thereby providing a new target for the treatment of bone diseases.

ABSTRACTFunctional effector T cells in the tumor microenvironment (TME) are critical for successful anti-tumor responses. T cell anti-tumor function is dependent on their ability to differentiate from a naïve state, infiltrate into the tumor site, and exert cytotoxic functions. The factors dictating whether a particular T cell can successfully undergo these processes during tumor challenge are not yet completely understood. Piezo1 is a mechanosensitive cation channel with high expression on both CD4+ and CD8+ T cells. Previous studies have demonstrated that Piezo1 optimizes T cell activation and restrains the CD4+ regulatory T cell (Treg) pool in vitro and under inflammatory conditions in vivo. However, little is known about the role Piezo1 plays on CD4+ and CD8+ T cells in cancer. We hypothesized that disruption of Piezo1 on T cells impairs anti-tumor immunity in vivo by hindering inflammatory T cell responses. We challenged mice with T cell Piezo1 deletion (P1KO) with tumor models dependent on T cells for immune rejection. P1KO mice had the more aggressive tumors, higher tumor growth rates and were unresponsive to immune-mediated therapeutic interventions. We observed a decreased CD4:CD8 ratio in both the secondary lymphoid organs and TME of P1KO mice that correlated inversely with tumor size. Poor CD4+ helper T cell responses underpinned the immunodeficient phenotype of P1KO mice. Wild type CD8+ T cells are sub-optimally activated in vivo with P1KO CD4+ T cells, taking on a CD25loPD-1hi phenotype. Together, our results suggest that Piezo1 optimizes T cell activation in the context of a tumor response.

Ca2+ overload-induced mitochondrial dysfunction is considered a contributing factor alcohol-associated liver disease pathogenesis. Here the authors report that PDK4 promotes Ca2 + -channelling complex formation at the endoplasmic reticulum-mitochondria contact sites, which contributes to the pathogenesis of alcohol-associated liver disease in studies with male mouse and hepatocyte models.

Abstract Background Muscle atrophy, leading to muscular dysfunction and weakness, is an adverse outcome of sustained period of glucocorticoids usage. However, the molecular mechanism underlying this detrimental condition is currently unclear. Pyruvate dehydrogenase kinase 4 (PDK4), a central regulator of cellular energy metabolism, is highly expressed in skeletal muscle and has been implicated in the pathogenesis of several diseases. The current study was designed to investigated and delineate the role of PDK4 in the context of muscle atrophy, which could be identified as a potential therapeutic avenue to protect against dexamethasone‐induced muscle wasting. Methods The dexamethasone‐induced muscle atrophy in C2C12 myotubes was evaluated at the molecular level by expression of key genes and proteins involved in myogenesis, using immunoblotting and qPCR analyses. Muscle dysfunction was studied in vivo in wild‐type and PDK4 knockout mice treated with dexamethasone (25 mg/kg body weight, i.p., 10 days). Body weight, grip strength, muscle weight and muscle histology were assessed. The expression of myogenesis markers were analysed using qPCR, immunoblotting and immunoprecipitation. The study was extended to in vitro human skeletal muscle atrophy analysis. Results Knockdown of PDK4 was found to prevent glucocorticoid‐induced muscle atrophy and dysfunction in C2C12 myotubes, which was indicated by induction of myogenin (0.3271 ± 0.102 vs 2.163 ± 0.192, ****P

Aminoacyl-tRNA synthetases possess unique domains. In this study the structure of the vertebrate IARS1 and EARS1 complex reveals that vertebrate IARS1 protects the DNA repair factor BRCA1 from proteolytic degradation via its UBX-fold domain.

Abstract Cobll1 affects blast crisis (BC) progression and tyrosine kinase inhibitor (TKI) resistance in chronic myeloid leukemia (CML). PACSIN2, a novel Cobll1 binding protein, activates TKI‐induced apoptosis in K562 cells, and this activation is suppressed by Cobll1 through the interaction between PACSIN2 and Cobll1. PACSIN2 also binds and inhibits SH3BP1 which activates the downstream Rac1 pathway and induces TKI resistance. PACSIN2 competitively interacts with Cobll1 or SH3BP1 with a higher affinity for Cobll1. Cobll1 preferentially binds to PACSIN2, releasing SH3BP1 to promote the SH3BP1/Rac1 pathway and suppress TKI‐mediated apoptosis and eventually leading to TKI resistance. Similar interactions among Cobll1, PACSIN2, and SH3BP1 control hematopoiesis during vertebrate embryogenesis. Clinical analysis showed that most patients with CML have Cobll1 and SH3BP1 expression at the BC phase and BC patients with Cobll1 and SH3BP1 expression showed severe progression with a higher blast percentage than those without any Cobll1, PACSIN2, or SH3BP1 expression. Our study details the molecular mechanism of the Cobll1/PACSIN2/SH3BP1 pathway in regulating drug resistance and BC progression in CML.

The Internet of Things (IoT) technology allows massive devices to connect to the internet for data exchange. It is anticipated that in near future, trillions of IoT devices will be connected to the internet. To deploy these devices, the requirement of the spectrum is increasing day by day. Most of these devices transmit data over unlicensed frequency bands that cause severe interference to each other while exchanging the data as these bands are becoming overcrowded. Therefore, to overcome spectrum scarcity and interference problems among these devices, a novel communication paradigm called cognitive radio-based internet of things (CR-IoT) is evolving at a very fast pace that integrates cognitive radio technology into the IoT devices. The technology has the potential to overcome the spectrum scarcity and interference problem by allowing dynamic spectrum access to conventional IoT networks. Such devices continuously monitor spectrum availability to transmit the data by incorporating an intelligent sensing mechanism into the devices. However, the performance of the sensing unit in terms of the probability of detection and the probability of false alarm, significantly deteriorates due to the noise uncertainties, especially in low signal-to-noise ratio environments. For the efficient utilization of the spectrum, the probability of detection and the probability of false alarm of the spectrum sensor should be high and low, respectively. Both sensing parameters are greatly influenced by the selection of the sensing threshold. Moreover, these IoT devices deal with short packet transmissions, the optimum sensing time is another crucial parameter that governs the performance of these devices.While addressing these two important issues, the convex optimization problem is formed over sensing time and sensing threshold, and the concavity on sensing threshold is proved. Further, an iterative algorithm is proposed for the CR-IoT system that intelligently adapts the sensing threshold to meet desired sensing performance in terms of Pd and Pf especially, in low SNR regions, and also optimize the sensing time to overcome the sensing throughput tradeoff. The simulation results are presented to validate the effectiveness of the proposed algorithm. It is demonstrated that the proposed algorithm increases the CR-IoT system throughput by 95% as compared to the conventional scheme at the received signal to noise ratio equals −20 dB while satisfying the requirement of Pd and Pf as per IEEE 802.22 standard.

The extracellular matrix (ECM) components present within all tissues and organs help to maintain the cytoskeletal architecture and tissue morphology. Although the ECM plays a role in cellular events and signaling pathways, it has not been well studied due its insolubility and complexity. Brain tissue has a higher cell density and weaker mechanical strength than other tissues in the body. When removing cells using a general decellularization method to produce scaffolds and obtain ECM proteins, various problems must be considered because tissues are easily damaged. To retain the brain shape and ECM components, we performed decellularization in combination with polymerization. We immersed mouse brains in oil for polymerization and decellularization via O-CASPER (Oil-based Clinically and Experimentally Applicable Acellular Tissue Scaffold Production for Tissue Engineering and Regenerative Medicine) and then isolated ECM components using sequential matrisome preparation reagents (SMPRs), namely, RIPA, PNGase F, and concanavalin A. Adult mouse brains were preserved with our decellularization method. Western blot and LC-MS/MS analyses revealed that ECM components, including collagen and laminin, were isolated efficiently from decellularized mouse brains using SMPRs. Our method will be useful to obtain matrisomal data and perform functional studies using adult mouse brains and other tissues.

Unmanned Aerial Vehicles have proven to be helpful in domains like defence and agriculture and will play a vital role in implementing smart cities in the upcoming years. Object detection is an essential feature in any such application. This work addresses the challenges of object detection in aerial images like improving the accuracy of small and dense object detection, handling the class-imbalance problem, and using contextual information to boost the performance. We have used a density map-based approach on the drone dataset VisDrone-2019 accompanied with increased receptive field architecture such that it can detect small objects properly. Further, to address the class imbalance problem, we have picked out the images with classes occurring fewer times and augmented them back into the dataset with rotations. Subsequently, we have used RetinaNet with adjusted anchor parameters instead of other conventional detectors to detect aerial imagery objects accurately and efficiently. The performance of the proposed three step pipeline of implementing object detection in aerial images is a significant improvement over the existing methods. Future work may include improvement in the computations of the proposed method, and minimising the effect of perspective distortions and occlusions.

At present, COVID-19 is posing a serious threat to global human health. The features of hand veins in infrared environments have many advantages, including non-contact acquisition, security, privacy, etc., which can remarkably reduce the risks of COVID-19. Therefore, this paper builds an interactive system, which can recognize hand gestures and track hands for palmvein recognition in infrared environments. The gesture contours are extracted and input into an improved convolutional neural network for gesture recognition. The hand is tracked based on key point detection. Because the hand gesture commands are randomly generated and the hand vein features are extracted from the infrared environment, the anti-counterfeiting performance is obviously improved. In addition, hand tracking is conducted after gesture recognition, which prevents the escape of the hand from the camera view range, so it ensures that the hand used for palmvein recognition is identical to the hand used during gesture recognition. The experimental results show that the proposed gesture recognition method performs satisfactorily on our dataset, and the hand tracking method has good robustness.

Various elements, such as evolutions in IoT services resulting from sensoring by vehicle parts and advances in small communication technology devices, have significantly impacted the mass spread of mobility services that are provided to users in need of limited resources. In particular, business models are progressing away from one-off costs towards longer-term costs, as represented by shared services utilizing kick-boards or bicycles and subscription services for vehicle software. Advances in shared mobility services, as described, are calling for solutions that can enhance the reliability of data aggregated by users leveraging mobility services in the next-generation mobility areas. However, the mining process to renew status ensures continued network communication, and block creation demands high performance in the public block chain. Therefore, easing the mining process for state updates in public blockchains is a way to alleviate the high-performance process requirements of public blockchains. The proposed mechanism assigns token-based block creation authority instead of the mining method, which provides block creation authority to nodes that provide many resources. Blocks are created only by a group of participants with tokens, and after creation, tokens are updated and delivered to new nodes to form a new token group. Additionally, tokens are updated in each block after their initial creation, making it difficult to disguise the tokens and preventing resource-centered centralization.

Immune cells and the cytokines they produce are important mediators of the transition from colitis to colon cancer, but the mechanisms mediating this disease progression are poorly understood. Interferon gamma (IFN-γ) is known to contribute to the pathogenesis of colitis through immune modulatory mechanisms, and through direct effects on endothelial and epithelial homeostasis. Here we explore whether IFN-γ influences tumor progression by expanding the effector memory T cells (TEM) population and restricting the expression of tumor suppressors in a preclinical model of spontaneous colitis-associated colorectal cancer (CAC). We show that IFN-γ expression is significantly increased both in the T cells and the colonic mucosal epithelia of mice with a T cell-restricted deletion of the TGF-β intermediate, SMAD4 (Smad4TKO). The increase of IFN-γ expression correlates with the onset of spontaneous CAC in Smad4TKO mice by 6 months of age. This phenotype is greatly ameliorated by the introduction of a germline deletion of IFN-γ in Smad4TKO mice (Smad4TKO/IFN-γKO, DKO). DKO mice had a significantly reduced incidence and progression of CAC, and a decrease in the number of mucosal CD4+ TEM cells, when compared to those of Smad4TKO mice. Similarly, the colon epithelia of DKO mice exhibited a non-oncogenic signature with a decrease in the expression of iNOS and p-STAT1, and a restoration of the tumor suppressor gene, 15-hydroxyprostaglandin dehydrogenase (15-PGDH). In vitro, treatment of human colon cancer cells with IFN-γ decreased the expression of 15-PGDH. Our data suggest that Smad4-deficient T cells promote CAC through mechanisms that include an IFN-γ-dependent suppression of the tumor suppressor 15-PGDH.

The gap junction complex functions as a transport channel across the membrane. Among gap junction subunits, gap junction protein α1 (GJA1) is the most commonly expressed subunit. A recent study showed that GJA1 is necessary for the maintenance of motile cilia; however, the molecular mechanism and function of GJA1 in ciliogenesis remain unknown. Here, we examined the functions of GJA1 during ciliogenesis in human retinal pigment epithelium-1 and Xenopus laevis embryonic multiciliated-cells. GJA1 localizes to the motile ciliary axonemes or pericentriolar regions beneath the primary cilium. GJA1 depletion caused malformation of both the primary cilium and motile cilia. Further study revealed that GJA1 depletion affected several ciliary proteins such as BBS4, CP110, and Rab11 in the pericentriolar region and basal body. Interestingly, CP110 removal from the mother centriole was significantly reduced by GJA1 depletion. Importantly, Rab11, a key regulator during ciliogenesis, was immunoprecipitated with GJA1 and GJA1 knockdown caused the mislocalization of Rab11. These findings suggest that GJA1 regulates ciliogenesis by interacting with the Rab11-Rab8 ciliary trafficking pathway.

Abstract The TGF-β signaling pathway governs key cellular processes under physiologic conditions and is deregulated in many pathologies, including cancer. TGF-β is a multifunctional cytokine that acts in a cell- and context-dependent manner as a tumor promoter or tumor suppressor. As a tumor promoter, the TGF-β pathway enhances cell proliferation, migratory invasion, metastatic spread within the tumor microenvironment and suppresses immunosurveillance. Collectively, the pleiotropic nature of TGF-β signaling contributes to drug resistance, tumor escape and undermines clinical response to therapy. Based upon a wealth of preclinical studies, the TGF-β pathway has been pharmacologically targeted using small molecule inhibitors, TGF-β-directed chimeric monoclonal antibodies, ligand traps, antisense oligonucleotides and vaccines that have been now evaluated in clinical trials. Here, we have assessed the safety and efficacy of TGF-β pathway antagonists from multiple drug classes that have been evaluated in completed and ongoing trials. We highlight Vactosertib, a highly potent small molecule TGF-β type 1 receptor kinase inhibitor that is well-tolerated with an acceptable safety profile that has shown efficacy against multiple types of cancer. The TGF-β ligand traps Bintrafusp alfa (a bifunctional conjugate that binds TGF-β and PD-L1), AVID200 (a computationally designed trap of TGF-β receptor ectodomains fused to an Fc domain) and Luspatercept (a recombinant fusion that links the activin receptor IIb to IgG) offer new ways to fight difficult-to-treat cancers. While TGF-β pathway antagonists are rapidly emerging as highly promising, safe and effective anticancer agents, significant challenges remain. Minimizing the unintentional inhibition of tumor-suppressing activity and inflammatory effects with the desired restraint on tumor-promoting activities has impeded the clinical development of TGF-β pathway antagonists. A better understanding of the mechanistic details of the TGF-β pathway should lead to more effective TGF-β antagonists and uncover biomarkers that better stratify patient selection, improve patient responses and further the clinical development of TGF-β antagonists.

Object detection is a fundamental task in computer vision. Over the past several years, convolutional neural network (CNN)-based object detection models have significantly improved detection accuracyin terms of average precision (AP). Furthermore, feature pyramid networks (FPNs) are essential modules for object detection models to consider various object scales. However, the AP for small objects is lower than the AP for medium and large objects. It is difficult to recognize small objects because they do not have sufficient information, and information is lost in deeper CNN layers. This paper proposes a new FPN model named ssFPN (scale sequence (S2) feature-based feature pyramid network) to detect multi-scale objects, especially small objects. We propose a new scale sequence (S2) feature that is extracted by 3D convolution on the level of the FPN. It is defined and extracted from the FPN to strengthen the information on small objects based on scale-space theory. Motivated by this theory, the FPN is regarded as a scale space and extracts a scale sequence (S2) feature by three-dimensional convolution on the level axis of the FPN. The defined feature is basically scale-invariant and is built on a high-resolution pyramid feature map for small objects. Additionally, the deigned S2 feature can be extended to most object detection models based on FPNs. We also designed a feature-level super-resolution approach to show the efficiency of the scale sequence (S2) feature. We verified that the scale sequence (S2) feature could improve the classification accuracy for low-resolution images by training a feature-level super-resolution model. To demonstrate the effect of the scale sequence (S2) feature, experiments on the scale sequence (S2) feature built-in object detection approach including both one-stage and two-stage models were conducted on the MS COCO dataset. For the two-stage object detection models Faster R-CNN and Mask R-CNN with the S2 feature, AP improvements of up to 1.6% and 1.4%, respectively, were achieved. Additionally, the APS of each model was improved by 1.2% and 1.1%, respectively. Furthermore, the one-stage object detection models in the YOLO series were improved. For YOLOv4-P5, YOLOv4-P6, YOLOR-P6, YOLOR-W6, and YOLOR-D6 with the S2 feature, 0.9%, 0.5%, 0.5%, 0.1%, and 0.1% AP improvements were observed. For small object detection, the APS increased by 1.1%, 1.1%, 0.9%, 0.4%, and 0.1%, respectively. Experiments using the feature-level super-resolution approach with the proposed scale sequence (S2) feature were conducted on the CIFAR-100 dataset. By training the feature-level super-resolution model, we verified that ResNet-101 with the S2 feature trained on LR images achieved a 55.2% classification accuracy, which was 1.6% higher than for ResNet-101 trained on HR images.

In this article, we propose a depth data coding method by plane modeling. The plane modeling is a prediction method based on a plane estimation from depth pixels in a block. The plane modeling improves the intra-picture prediction for depth data comparing with intra modes of conventional coding standards. The plane modeling coefficients, which are the information about the estimated plane, are required to be provided during the depth data coding. The plane modeling coefficients are predicted from neighboring depth pixels. A prediction error of the plane modeling coefficients is calculated through the plane modeling for the selected pixels. If the prediction error is below a certain threshold, then the plane modeling coefficients are applied to the plane modeling for the block. From the simulation results, we verify that the proposed method achieves up to 6.76% bit rate saving in the same coding distortion condition compared to VVC test model.

Abstract The methods combining correlation filters (CFs) with the features of convolutional neural network (CNN) are good at object tracking. However, the high-level features of a typical CNN without residual structure suffer from the shortage of fine-grained information, it is easily affected by similar objects or background noise. Meanwhile, CF-based methods usually update filters at every frame even when occlusion occurs, which degrades the capability of discriminating the target from background. A novel scale-adaptive object-tracking method is proposed in this paper. Firstly, the features are extracted from different layers of ResNet to produce response maps, and then, in order to locate the target more accurately, these response maps are fused based on AdaBoost algorithm. Secondly, to prevent the filters from updating when occlusion occurs, an update strategy with occlusion detection is proposed. Finally, a scale filter is used to estimate the target scale. The experimental results demonstrate that the proposed method performs favorably compared with several mainstream methods especially in the case of occlusion and scale change.

High efficiency video coding (HEVC) has been developed rapidly to support new generation display devices and their ultra high definition (UHD) with high dynamic range (HDR) and wide color gamut (WCG). To support HDR/WCG sequences on the HEVC standard, pre-/post-processing technique has been designed. After an HDR video is compressed, a reconstructed frame exhibits chromatic distortions that resemble color smearing. To remove this color artifact, we herein propose a block-level quantization parameter (QP) offset-control-based efficient compression algorithm for the HDR sequence. First, we extract the candidate coding units (CUs) with the annoying area to the human eye based on the just noticeable distortion (JND) model. Subsequently, the chromatic distorted blocks are verified by the activity function as the chromatic artifact is observed at the nearby strong edge. For the verified artifact blocks, we reassign the QPs for the Cb and Cr chroma components. Our experimental results show that the proposed method yields an average gain in BD-rate of 3.3% for U, and 3.4% for V with a negligible bitrate increase of 0.3% on average.

As the demands of various network-dependent services such as Internet of things (IoT) applications, autonomous driving, and augmented and virtual reality (AR/VR) increase, the fifthgeneration (5G) network is expected to become a key communication technology. The latest video coding standard, versatile video coding (VVC), can contribute to providing high-quality services by achieving superior compression performance. In video coding, inter bi-prediction serves to improve the coding efficiency significantly by producing a precise fused prediction block. Although block-wise methods, such as bi-prediction with CU-level weight (BCW), are applied in VVC, it is still difficult for the linear fusion-based strategy to represent diverse pixel variations inside a block. In addition, a pixel-wise method called bi-directional optical flow (BDOF) has been proposed to refine bi-prediction block. However, the non-linear optical flow equation in BDOF mode is applied under assumptions, so this method is still unable to accurately compensate various kinds of bi-prediction blocks. In this paper, we propose an attention-based bi-prediction network (ABPN) to substitute for the whole existing bi-prediction methods. The proposed ABPN is designed to learn efficient representations of the fused features by utilizing an attention mechanism. Furthermore, the knowledge distillation (KD)- based approach is employed to compress the size of the proposed network while keeping comparable output as the large model. The proposed ABPN is integrated into the VTM-11.0 NNVC-1.0 standard reference software. When compared with VTM anchor, it is verified that the BD-rate reduction of the lightweighted ABPN can be up to 5.89% and 4.91% on Y component under random access (RA) and low delay B (LDB), respectively.

With the continued development of artificial intelligence (AI) technology, research on interaction technology has become more popular. Facial expression recognition (FER) is an important type of visual information that can be used to understand a human's emotional situation. In particular, the importance of AI systems has recently increased due to advancements in research on AI systems applied to AI robots. In this paper, we propose a new scheme for FER system based on hierarchical deep learning. The feature extracted from the appearance feature-based network is fused with the geometric feature in a hierarchical structure. The appearance feature-based network extracts holistic features of the face using the preprocessed LBP image, whereas the geometric feature-based network learns the coordinate change of action units (AUs) landmark, which is a muscle that moves mainly when making facial expressions. The proposed method combines the result of the softmax function of two features by considering the error associated with the second highest emotion (Top-2) prediction result. In addition, we propose a technique to generate facial images with neutral emotion using the autoencoder technique. By this technique, we can extract the dynamic facial features between the neutral and emotional images without sequence data. We compare the proposed algorithm with the other recent algorithms for CK+ and JAFFE dataset, which are typically considered to be verified datasets in the facial expression recognition. The ten-fold cross validation results show 96.46% of accuracy in the CK+ dataset and 91.27% of accuracy in the JAFFE dataset. When comparing with other methods, the result of the proposed hierarchical deep network structure shows up to about 3% of the accuracy improvement and 1.3% of average improvement in CK+ dataset, respectively. In JAFFE datasets, up to about 7% of the accuracy is enhanced, and the average improvement is verified by about 1.5%.

Image-based data hiding methods have been used in the development of various applications in computer vision. At present, there are two main types of data hiding based on secret sharing, namely dual-image data hiding and multi-image data hiding. Dual-image data hiding is a kind of secret sharing-based data hiding in the extreme case. During the image transmission and storage process, the two shadow images are visually highly similar. Multi-image data hiding disassembles the cover image into multiple meaningless secret images through secret sharing. Both of the above two methods can easily attract attackers’ attention and cannot effectively guarantee the security of the secret message. In this paper, through the Shamir threshold scheme for secret sharing, the secret message is disassembled into multiple subsecrets that are embedded in the smooth blocks of multiple different images, by substituting the bitmap of block truncation coding. Thus, the shortcomings of the above two data hiding methods are effectively avoided. The proposed method embeds the secret messages in the compressed images, so it satisfactorily balances the visual quality and the embedding capacity. In our method, the shadow images make sense while they are not visually similar. The compression ratio is four, so the embedding capacity of our method has an obvious advantage under the same storage space.

At the ImageNet Large Scale Visual Re-Conversion Challenge (ILSVRC), a 2012 global image recognition contest, the University of Toronto Supervision team led by Prof [...]

The cyclin-dependent kinase inhibitor p27Kip1 is a tumor suppressor whose intrinsic activity in cancer cells correlates with tumor aggressiveness, invasiveness, and impaired tumor cell differentiation. Here we explore whether p27Kip1 indirectly influences tumor progression by restricting expansion and survival of effector memory T cell (TEM) populations in a preclinical model of spontaneous colitis-associated colorectal cancer (CAC). We show mRNA and protein expression of p27Kip1 to be significantly decreased in the colons of mice with a T cell-restricted deletion of the TGF-β intermediate, SMAD4 (Smad4TKO). Loss of p27Kip1 expression in T cells correlates with the onset of spontaneous CAC in Smad4TKO mice by 8 months of age. This phenotype is greatly accelerated by the introduction of a germline deletion of CDKN1b (the gene encoding p27Kip1) in Smad4TKO mice (Smad4TKO/p27Kip1-/-, DKO). DKO mice display colon carcinoma by 3 months of age and increased mortality compared to Smad4TKO. Importantly, the phenotype in DKO mice is associated with a significant increase in the frequency of effector CD4 T cells expressing abundant IFN-γ and with a concomitant decrease in Foxp3+ regulatory T cells, both in the intestinal mucosa and in the periphery. In addition, induction of inflammatory mediators (IFN-γ, TNF-γ, IL-6, IL-1β, iNOS) and activation of Stat1, Stat3, and IκB is also observed in the colon as early as 1–2 months of age. Our data suggest that genomic alterations known to influence p27Kip1 abundance in gastrointestinal cancers may indirectly promote epithelial malignancy by augmenting the production of inflammatory mediators from a spontaneously expanding pool of TEM cells.

Biometric signals can be acquired with different sensors and recognized in secure identity management systems. However, it is vulnerable to various attacks that compromise the security management in many applications, such as industrial IoT. In a real-world scenario, the target template stored in the database of a biometric system can possibly be leaked, and then used to reconstruct a fake image to fool the biometric system. As such, many reconstruction attacks have been proposed, yet unsatisfactory naturalness, poor visual quality or incompleteness remains as major limitations. Thus, two reinforced palmprint reconstruction attacks are proposed. Any palmprint image, which can be easily obtained, is used as the initial image, and the region of interest is iteratively modified with deep reinforcement strategies to reduce the matching distance. In the first attack, Modification Constraint within Neighborhood (MCwN) limits the modification extent and suppresses the reckless modification. In the second attack, Batch Member Selection (BMS) selects the significant pixels (SPs) to compose the batch, which are simultaneously modified to a slighter extent to reduce the matching number and the visual-quality degradation. The two reinforced attacks can satisfy all the requirements, which cannot be simultaneously satisfied by the existing attacks. The thorough experiments demonstrate that the two attacks have a highly successful attack rate for palmprint systems based on the most state-of-the-art coding-based methods.

Metabolic disorders: TRIM-ming the fat A protein that targets peroxisome proliferator-activated receptor γ (PPARγ) (a key regulator of fat cell formation) for degradation suppresses the formation of fat. Excess fat can lead to obesity and cause type 2 diabetes and cardiovascular disease, yet little is known about the mechanisms through which fat cells arise from progenitor cells. Jang Hyun Choi and colleagues at the Ulsan National Institute of Science and Technology, Korea, have found that TRIM25, a protein implicated in cancer and antiviral immune responses, reduces the activity of PPARγ by adding a small regulatory protein that acts as a signal for degradation. Reducing the levels of TRIM25 in progenitor cells increases their ability to differentiate into fat cells. These findings suggest that TRIM25 could be a useful target for developing new therapies against obesity and metabolic disorders.

The chromatin remodeler RSF1 is enriched at mitotic centromeres but its function there is poorly understood. Here, the authors show that RSF1 regulates H2A phosphorylation and acetylation at mitotic centromeres and contributes to chromosome stability.

A web-based search system recommends and gives results such as customized image or video contents using information such as user interests, search time, and place. Time information extracted from images can be used as a important metadata in the web search system. We present an efficient algorithm to classify time period into day, dawn, and night when the input is a single image with a sky region. We employ the Mask R-CNN to extract a sky region. Based on the extracted sky region, reference color histograms are generated, which can be considered as the ground-truth. To compare the histograms effectively, we design the windowed-color histograms (for RGB bands) to compare each time period from the sky region of the reference data with one of the input images. Also, we use a weighting approach to reflect a more separable feature on the windowed-color histogram. With the proposed windowed-color histogram, we verify about 91% of the recognition accuracy in the test data. Compared with the existing deep neural network models, we verify that the proposed algorithm achieves better performance in the test dataset.

Cognitive workload is a crucial factor in tasks involving dynamic decision-making and other real-time and high-risk situations. Neuroimaging techniques have long been used for estimating cognitive workload. Given the portability, cost-effectiveness and high time-resolution of EEG as compared to fMRI and other neuroimaging modalities, an efficient method of estimating an individual’s workload using EEG is of paramount importance. Multiple cognitive, psychiatric and behavioral phenotypes have already been known to be linked with “functional connectivity”, i.e., correlations between different brain regions. In this work, we explored the possibility of using different model-free functional connectivity metrics along with deep learning in order to efficiently classify the cognitive workload of the participants. To this end, 64-channel EEG data of 19 participants were collected while they were doing the traditional n-back task. These data (after pre-processing) were used to extract the functional connectivity features, namely Phase Transfer Entropy (PTE), Mutual Information (MI) and Phase Locking Value (PLV). These three were chosen to do a comprehensive comparison of directed and non-directed model-free functional connectivity metrics (allows faster computations). Using these features, three deep learning classifiers, namely CNN, LSTM and Conv-LSTM were used for classifying the cognitive workload as low (1-back), medium (2-back) or high (3-back). With the high inter-subject variability in EEG and cognitive workload and recent research highlighting that EEG-based functional connectivity metrics are subject-specific, subject-specific classifiers were used. Results show the state-of-the-art multi-class classification accuracy with the combination of MI with CNN at 80.87%, followed by the combination of PLV with CNN (at 75.88%) and MI with LSTM (at 71.87%). The highest subject specific performance was achieved by the combinations of PLV with Conv-LSTM, and PLV with CNN with an accuracy of 97.92%, followed by the combination of MI with CNN (at 95.83%) and MI with Conv-LSTM (at 93.75%). The results highlight the efficacy of the combination of EEG-based model-free functional connectivity metrics and deep learning in order to classify cognitive workload. The work can further be extended to explore the possibility of classifying cognitive workload in real-time, dynamic and complex real-world scenarios.

In recent years, the importance of catching humans’ emotions grows larger as the artificial intelligence (AI) field is being developed. Facial expression recognition (FER) is a part of understanding the emotion of humans through facial expressions. We proposed a robust multi-depth network that can efficiently classify the facial expression through feeding various and reinforced features. We designed the inputs for the multi-depth network as minimum overlapped frames so as to provide more spatio-temporal information to the designed multi-depth network. To utilize a structure of a multi-depth network, a multirate-based 3D convolutional neural network (CNN) based on a multirate signal processing scheme was suggested. In addition, we made the input images to be normalized adaptively based on the intensity of the given image and reinforced the output features from all depth networks by the self-attention module. Then, we concatenated the reinforced features and classified the expression by a joint fusion classifier. Through the proposed algorithm, for the CK+ database, the result of the proposed scheme showed a comparable accuracy of 96.23%. For the MMI and the GEMEP-FERA databases, it outperformed other state-of-the-art models with accuracies of 96.69% and 99.79%. For the AFEW database, which is known as one in a very wild environment, the proposed algorithm achieved an accuracy of 31.02%.

Since high quality realistic media are widely used in various computer vision applications, image compression is one of the essential technologies to enable real-time applications. Image compression generally causes undesired compression artifacts, such as blocking artifacts and ringing effects. In this study, we propose a densely cascading image restoration network (DCRN), which consists of an input layer, a densely cascading feature extractor, a channel attention block, and an output layer. The densely cascading feature extractor has three densely cascading (DC) blocks, and each DC block contains two convolutional layers, five dense layers, and a bottleneck layer. To optimize the proposed network architectures, we investigated the trade-off between quality enhancement and network complexity. Experimental results revealed that the proposed DCRN can achieve a better peak signal-to-noise ratio and structural similarity index measure for compressed joint photographic experts group (JPEG) images compared to the previous methods.

In this study, we investigated the anti-obesity properties of the novel peptide Ala-Gly-Leu-Gln-Phe-Pro-Val-Gly-Arg (AGL9), isolated from the enzymatic hydrolysate of Allomyrinadichotoma larvae. To investigate the preventive effects of AGL9 against hepatic steatosis and its possible mechanisms of action, we established an nonalcoholic fatty liver disease (NAFLD) model by feeding C57BL/6 mice a high-fat diet. NAFLD mice were administered 100 mg/kg AGL9 and 60 mg/kg orlistat via gavage (10 mL/kg) for 5 weeks, followed by the collection of blood and liver tissues. We found that AGL9 normalized the levels of serum alanine aminotransferase, aspartate aminotransferase, triglyceride, total cholesterol, high-density lipoprotein, very low-density lipoprotein (LDL)/LDL, adiponectin, and leptin in these mice. Additionally, AGL9 activated the protein-level expression of 5′ AMP-activated protein kinase and acetyl-CoA carboxylase phosphorylation and the transcript-level expression of sterol regulatory element-binding protein-1c, fatty acid synthase, superoxide dismutase, glutathione peroxidase, glucocorticoid receptor, nuclear respiratory factor 2, tumor necrosis factor-α, interleukin-1β, interleukin-6, and monocyte chemoattractant protein-1 in hepatocytes. These results showed that AGL9 exhibited hepatoprotective effects by attenuating lipid deposition, oxidative stress, and inflammation via inhibition of AMPK/Nrf2 signaling, thereby reducing the production of hepatic proinflammatory mediators and indicating AGL9 as a potential therapeutic strategy for NAFLD.

A wide range of multimedia applications must be supported by the modern fifth generation (5G) wireless communication systems for realizing the diverse applications in smart cities. The diverse applications such as real-time monitoring of roads, smart homes, smart industries, etc., for a sustainable smart city emphasizes a robust and efficient image transmission. In this paper, the influence of maximal ratio combining (MRC) on the reception of images with different orthogonal frequency division multiplexing (OFDM) versions is studied. The different OFDM versions considered here are the fast Fourier transform (FFT) based OFDM and discrete cosine transform (DCT) based OFDM. A comparison between diverse modulation levels for the images transmitted through different OFDM methodologies, along with variation in a number of receiving antennas for MRC, is proposed for additive white gaussian noise (AWGN) and Rayleigh fading channels. The diverse modulation levels used are binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), 8-PSK, and 16-PSK. The parameters that are used to compare different versions of OFDM for MRC antenna configurations are signal-to-noise ratio (SNR) vs. bit error rate (BER) and peak signal-to-noise ratio (PSNR) at the receiver as an estimation parameter for the received image quality.

Summary: Proliferating cell nuclear antigen (PCNA) is a DNA clamp essential for DNA replication. During DNA synthesis, PCNA is continuously loaded onto and unloaded from DNA. PCNA recruits various proteins to nascent DNA to facilitate chromosome duplication. Therefore, timely PCNA unloading is crucial for high-fidelity DNA replication. The ATAD5-RFC-like complex (ATAD5-RLC) unloads PCNA from replicated DNA. It is unclear how ATAD5-RLC activity is regulated to prevent premature PCNA unloading. Here, we find that BRD4, an acetyl-histone-binding chromatin reader, inhibits the PCNA-unloading activity of ATAD5-RLC. The BRD4 ET domain interacts with a region upstream of the ATAD5 PCNA-unloading domain. BRD4-ATAD5 binds to acetyl-histones in nascent chromatin. BRD4 release from chromatin correlates with PCNA unloading. Disruption of the interaction between BRD4 and acetyl-histones or between BRD4 and ATAD5 reduces the PCNA amount on chromatin. In contrast, the overexpression of BRD4 increases the amount of chromatin-bound PCNA. Thus, acetyl-histone-bound BRD4 fine-tunes PCNA unloading from nascent DNA. : Kang et al. demonstrate how PCNA unloading is regulated on nascent chromatin. Timely PCNA unloading from replicated DNA is crucial for faithful DNA replication. BRD4 binds to ATAD5, a subunit of PCNA-unloading complex. Acetyl-histone-bound BRD4 inhibits the activity of ATAD5 complex on nascent chromatin to prevent premature PCNA unloading. Keywords: ATAD5, BRD4, PCNA, PCNA unloading, DNA replication, histone acetylation, nascent chromatin, BET protein, RFC-like complex

Metabolic reprogramming during macrophage polarization supports the effector functions of these cells in health and disease. Here, we demonstrate that pyruvate dehydrogenase kinase (PDK), which inhibits the pyruvate dehydrogenase-mediated conversion of cytosolic pyruvate to mitochondrial acetyl-CoA, functions as a metabolic checkpoint in M1 macrophages. Polarization was not prevented by PDK2 or PDK4 deletion but was fully prevented by the combined deletion of PDK2 and PDK4; this lack of polarization was correlated with improved mitochondrial respiration and rewiring of metabolic breaks that are characterized by increased glycolytic intermediates and reduced metabolites in the TCA cycle. Genetic deletion or pharmacological inhibition of PDK2/4 prevents polarization of macrophages to the M1 phenotype in response to inflammatory stimuli (lipopolysaccharide plus IFN-γ). Transplantation of PDK2/4-deficient bone marrow into irradiated wild-type mice to produce mice with PDK2/4-deficient myeloid cells prevented M1 polarization, reduced obesity-associated insulin resistance, and ameliorated adipose tissue inflammation. A novel, pharmacological PDK inhibitor, KPLH1130, improved high-fat diet-induced insulin resistance; this was correlated with a reduction in the levels of pro-inflammatory markers and improved mitochondrial function. These studies identify PDK2/4 as a metabolic checkpoint for M1 phenotype polarization of macrophages, which could potentially be exploited as a novel therapeutic target for obesity-associated metabolic disorders and other inflammatory conditions.

Super resolution (SR) enables to generate a high-resolution (HR) image from one or more low-resolution (LR) images. Since a variety of CNN models have been recently studied in the areas of computer vision, these approaches have been combined with SR in order to provide higher image restoration. In this paper, we propose a lightweight CNN-based SR method, named multi-scale channel dense network (MCDN). In order to design the proposed network, we extracted the training images from the DIVerse 2K (DIV2K) dataset and investigated the trade-off between the SR accuracy and the network complexity. The experimental results show that the proposed method can significantly reduce the network complexity, such as the number of network parameters and total memory capacity, while maintaining slightly better or similar perceptual quality compared to the previous methods.

Sign language is a visual language for communication used by hearing-impaired people with the help of hand and finger movements. Indian Sign Language (ISL) is a well-developed and standard way of communication for hearing-impaired people living in India. However, other people who use spoken language always face difficulty while communicating with a hearing-impaired person due to lack of sign language knowledge. In this study, we have developed a 3D avatar-based sign language learning system that converts the input speech/text into corresponding sign movements for ISL. The system consists of three modules. Initially, the input speech is converted into an English sentence. Then, that English sentence is converted into the corresponding ISL sentence using the Natural Language Processing (NLP) technique. Finally, the motion of the 3D avatar is defined based on the ISL sentence. The translation module achieves a 10.50 SER (Sign Error Rate) score.

There are many studies that seek to enhance a low resolution image to a high resolution image in the area of super-resolution. As deep learning technologies have recently shown impressive results on the image interpolation and restoration field, recent studies are focusing on convolutional neural network (CNN)-based super-resolution schemes to surpass the conventional pixel-wise interpolation methods. In this paper, we propose two lightweight neural networks with a hybrid residual and dense connection structure to improve the super-resolution performance. In order to design the proposed networks, we extracted training images from the DIVerse 2K (DIV2K) image dataset and investigated the trade-off between the quality enhancement performance and network complexity under the proposed methods. The experimental results show that the proposed methods can significantly reduce both the inference speed and the memory required to store parameters and intermediate feature maps, while maintaining similar image quality compared to the previous methods.

Scene or place classification is one of the important problems in image and video search and recommendation systems. Humans can understand the scene they are located, but it is difficult for machines to do it. Considering a scene image which has several objects, humans recognize the scene based on these objects, especially background objects. According to this observation, we propose an efficient scene classification algorithm for three different classes by detecting objects in the scene. We use pre-trained semantic segmentation model to extract objects from an image. After that, we construct a weight matrix to determine a scene class better. Finally, we classify an image into one of three scene classes (i.e., indoor, nature, city) by using the designed weighting matrix. The performance of our scheme outperforms several classification methods using convolutional neural networks (CNNs), such as VGG, Inception, ResNet, ResNeXt, Wide-ResNet, DenseNet, and MnasNet. The proposed model achieves 90.8% of verification accuracy and improves over 2.8% of the accuracy when comparing to the existing CNN-based methods.

Rainfall prediction is an important task due to the dependence of many people on it, especially in the agriculture sector. Prediction is difficult and even more complex due to the dynamic nature of rainfalls. In this study, we carry out monthly rainfall prediction over Simtokha a region in the capital of Bhutan, Thimphu. The rainfall data were obtained from the National Center of Hydrology and Meteorology Department (NCHM) of Bhutan. We study the predictive capability with Linear Regression, Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), and Bidirectional Long Short Term Memory (BLSTM) based on the parameters recorded by the automatic weather station in the region. Furthermore, this paper proposes a BLSTM-GRU based model which outperforms the existing machine and deep learning models. From the six different existing models under study, LSTM recorded the best Mean Square Error (MSE) score of 0.0128. The proposed BLSTM-GRU model outperformed LSTM by 41.1% with a MSE score of 0.0075. Experimental results are encouraging and suggest that the proposed model can achieve lower MSE in rainfall prediction systems.

Understanding a person’s feelings is a very important process for the affective computing. People express their emotions in various ways. Among them, facial expression is the most effective way to present human emotional status. We propose efficient deep joint spatiotemporal features for facial expression recognition based on the deep appearance and geometric neural networks. We apply three-dimensional (3D) convolution to extract spatial and temporal features at the same time. For the geometric network, 23 dominant facial landmarks are selected to express the movement of facial muscle through the analysis of energy distribution of whole facial landmarks.We combine these features by the designed joint fusion classifier to complement each other. From the experimental results, we verify the recognition accuracy of 99.21%, 87.88%, and 91.83% for CK+, MMI, and FERA datasets, respectively. Through the comparative analysis, we show that the proposed scheme is able to improve the recognition accuracy by 4% at least.

Video data for the Internet traffic is increasing, and video data transmission is important for consideration of real-time process in the Internet of Things (IoT). Thus, in the IoT environment, video applications will be valuable approach in networks of smart sensor devices. High Efficiency Video Coding (HEVC) has been developed by the Joint Collaborative Team on Video Coding (JCT-VC) as a new generation video coding standard. Recently, HEVC includes range extensions (RExt), scalable coding extensions, and multiview extensions. HEVC RExt provides high resolution video with a high bit-depth and an abundance of color formats. In this paper, a fast intraprediction unit decision method is proposed to reduce the computational complexity of the HEVC RExt encoder. To design intramode decision algorithm, Local Binary Pattern (LBP) of the current prediction unit is used as texture feature. Experimental results show that the encoding complexity can be reduced by up to 12.35% on average in the AI-Main profile configuration with only a small bit-rate increment and a PSNR decrement, compared with HEVC test model (HM) 12.0-RExt4.0 reference software.

Various applications based on IoT real-time multimedia are under the spotlight. To implement real-time multimedia service, motion estimation in video compression service has a high computational complexity. In this paper, an efficient motion search method based on content awareness is proposed consisting of three steps. The first step is motion classification using the center position cost distribution. The second step is calculation of a predictor based motion classification. The third step is setting the arm size of the search pattern based on adaptive use of the distance between the predictor and the center position. Experimental results show that the proposed algorithm achieves speed-up factors of up to 48.57% and 16.03%, on average, with good bitrate performance, compared with fast integer-pel and fractional-pel motion estimation for H.264/AVC (UMHexagonS), and an enhanced predictive zonal search for single and multiple frame motion estimation (EPZS) methods using JM 18.5, respectively. In addition, the proposed algorithm achieves a speed-up factor of up to 42.61%, on average, with negligible bitrate degradation, compared with the TZ search motion estimation algorithm for the multiview video coding (TZS) method on HM 10.0.

Jini security framework that is being maintained in Apache River project requires roughly 3 sequential tasks in every connection to cope with security threats in communication between clients and servers in the IoT sensor application. These tasks are sequentially proxy preparation, certification of identity and its credential for mutual authentication, and cryptographic operation for session encryption. Since the proxy preparation task is time spent on preparing a secure and trustable stub for both client/server sides, it is not one of the substantial communication-delay factors. We propose a method of improving proxy preparation reducing service connection delay by completing the preparation of mutual authentication data and session key while on proxy preparation task. Through experiments on a test-bed, we have confirmed that service connection delay time can be reduced by about 2.6 times in proportion to the message sizes through our proposed method. Our work presents an approach to reduce the additional overhead that would be paid for by applying the Jini security framework to the development of sensor application based on Jini IoT.