Single Image Rain Removal Using Image Decomposition and a Dense Network

Removing rain from a single image is a challenging task due to the absence of temporal information. Considering that a rainy image can be decomposed into the low-frequency(LF) and high-frequency(HF) components, where the coarse scale information is retained in the LF component and the rain streaks and texture correspond to the HF component, we propose a single image rain removal algorithm using image decomposition and a dense network. We design two task-driven sub-networks to estimate the LF and non-rain HF components of a rainy image. The high-frequency estimation sub-network employs a densely connected network structure, while the low-frequency sub-network uses a simple convolutional neural network(CNN).We add total variation(TV) regularization and LF-channel fidelity terms to the loss function to optimize the two subnetworks jointly. The method then obtains de-rained output by combining the estimated LF and non-rain HF components.Extensive experiments on synthetic and real-world rainy images demonstrate that our method removes rain streaks while preserving non-rain details, and achieves superior de-raining performance both perceptually and quantitatively.

Optimization of hydrogen networks with multiple impurities and impurity removal

To explore the effect of removing different impurities to hydrogen networks, an MINLP model is proposed with all matching possibilities and the trade-off between operation cost and capital cost is considered. Furthermore,the impurity remover, hydrogen distribution, compressor and pipe setting are included in the model. Based on this model, the impurity and source(s) that are in higher priority for impurity removal, the optimal targeted concentration, and the hydrogen network with the minimum annual cost can be identified. The efficiency of the proposed model is verified by a case study.

Neural Network Modeling for Ni(II) Removal from Aqueous System Using Shelled Moringa Oleifera Seed Powder as an Agricultural Waste

A single-layer Artificial Neural Network (ANN) model was developed to predict the removal efficiency of Ni(II) ions from aqueous solution using shelled Moringa Oleifera seed (SMOS) powder. Batch experiments resulted into standardization of optimum conditions: biomass dosage (4.0 g), Ni(II) concentration (25 mg/L) volume (200 mL) at pH 6.5. A time of forty minutes was found sufficient to achieve the equilibrium. The ANN model was designed to predict sorption efficiency of SMOS for target metal ion by combining back propagation (BP) with principle component analysis. A sigmoid axon was used as transfer function for input and output layers. The Levenberg–Marquardt Algorithm (LMA) was applied, giving a minimum mean squared error (MSE) for training and cross validation at the ninth place of decimal.

Predicting pollutant removal in constructed wetlands using artificial neural networks(ANNs)

Growth in urban population,urbanisation,and economic development has increased the demand for water,especially in water-scarce regions.Therefore,sustainable approaches to water management are needed to cope with the effects of the urbanisation on the water environment.This study aimed to design novel configurations of tidal-flow vertical subsurface flow constructed wetlands(VFCWs)for treating urban stormwater.A series of laboratory experiments were conducted with semi-synthetic influent stormwater to examine the effects of the design and operation variables on the performance of the VFCWs and to identify optimal design and operational strategies,as well as maintenance requirements.The results show that the VFCWs can significantly reduce pollutants in urban stormwater,and that pollutant removal was related to specific VFCW designs.Models based on the artificial neural network(ANN)method were built using inputs derived from data exploratory techniques,such as analysis of variance(ANOVA)and principal component analysis(PCA).It was found that PCA reduced the dimensionality of input variables obtained from different experimental design conditions.The results show a satisfactory generalisation for predicting nitrogen and phosphorus removal with fewer variable inputs,indicating that monitoring costs and time can be reduced.

Material Removal Rate Prediction of Electrical Discharge Machining Process Using Artificial Neural Network

This article presents an Artificial Neural Network （ANN） architecture to model the Electrical Discharge Machining （EDM） process. It is aimed to develop the ANN model using an input-output pattern of raw data collected from an experimental of EDM process, whereas several research objectives have been outlined such as experimenting machining material for selected gap current, identifying machining parameters for ANN variables and selecting appropriate size of data selection. The experimental data （input variables） of copper-electrode and steel-workpiece is based on a selected gap current where pulse on time, pulse off time and sparking frequency have been chosen at optimum value of Material Removal Rate （MRR）. In this paper, the result has significantly demonstrated that the ANN model is capable of predicting the MRR with low percentage prediction error when compared with the experimental result.

Pervaporation Characteristics in Removal of Benzene from Water through Polystyrene-Poly (Dimethylsiloxane) IPN Membranes

This paper focuses on the effects of the PSt content of polystyrene (PSt)-poly (dimethylsiloxane) (PDMS) interpenetrateing network (IPN) polymer membranes, on the pervaporation (PV) characteristics during the removal of benzene from an aqueous solution of dilute benzene. When an aqueous solution of 0.05wt% benzene was permeated through the PSt-PDMS IPN membranes, they showed high benzene/water selectivity. Both the permeability and the benzene/water selectivity of the membranes were enhanced with increasing PSt content in the PSt-PDMS IPN membrane. The physicochemical mechanism of permeation and separation through the PSt-PDMS IPN membranes during PV is also discussed. The best normalized permeation rate, separation factor for benzene selectivity, and PV separation index of the PSt-PDMS IPN membrane were 1.27 × 10-6 kgm (m2hr)-1, 3293, and 41821, respectively. These PV characteristics are discussed from the viewpoint of the chemical and physical structure of the PSt-PDMS IPN membranes.

Influence of network structure on spreading dynamics via tie range

There are various phenomena of malicious information spreading in the real society, which cause many negative impacts on the society. In order to better control the spreading, it is crucial to reveal the influence of network structure on network spreading. Motifs, as fundamental structures within a network, play a significant role in spreading. Therefore, it is of interest to investigate the influence of the structural characteristics of basic network motifs on spreading dynamics.Considering the edges of the basic network motifs in an undirected network correspond to different tie ranges, two edge removal strategies are proposed, short ties priority removal strategy and long ties priority removal strategy. The tie range represents the second shortest path length between two connected nodes. The study focuses on analyzing how the proposed strategies impact network spreading and network structure, as well as examining the influence of network structure on network spreading. Our findings indicate that the long ties priority removal strategy is most effective in controlling network spreading, especially in terms of spread range and spread velocity. In terms of network structure, the clustering coefficient and the diameter of network also have an effect on the network spreading, and the triangular structure as an important motif structure effectively inhibits the spreading.

Fetal MRI Artifacts: Semi-Supervised Generative Adversarial Neural Network for Motion Artifacts Reducing in Fetal Magnetic Resonance Images

This study addresses challenges in fetal magnetic resonance imaging (MRI) related to motion artifacts, maternal respiration, and hardware limitations. To enhance MRI quality, we employ deep learning techniques, specifically utilizing Cycle GAN. Synthetic pairs of images, simulating artifacts in fetal MRI, are generated to train the model. Our primary contribution is the use of Cycle GAN for fetal MRI restoration, augmented by artificially corrupted data. We compare three approaches (supervised Cycle GAN, Pix2Pix, and Mobile Unet) for artifact removal. Experimental results demonstrate that the proposed supervised Cycle GAN effectively removes artifacts while preserving image details, as validated through Structural Similarity Index Measure (SSIM) and normalized Mean Absolute Error (MAE). The method proves comparable to alternatives but avoids the generation of spurious regions, which is crucial for medical accuracy.

Artificial Neural Network Modeling for Sorption of Cadmium from Aqueous System by Shelled Moringa Oleifera Seed Powder as an Agricultural Waste

A two-layer Artificial Neural Network (ANN) model was developed to predict the removal efficiency of Cd(II) ions from aqueous solution using shelled Moringa Oleifera seed (SMOS) powder. Batch experiments re-sulted into standardization of optimum conditions: biomass dosage (4.0 g), Cd(II) concentration (25 mg/L) volume (200 mL) at pH 6.5. A time of forty minutes was found sufficient to achieve the equilibrium. The ANN model was designed to predict sorption efficiency of SMOS for target metal ion by combining back propagation (BP) with principle component analysis. A sigmoid axon was used as transfer function for input and output layer. The Levenberg-Marquardt algorithm (LMA) was applied, giving a minimum mean squared error (MSE) for training and cross validation at the ninth place of decimal.

Unified Model of Purification Units in Hydrogen Networks

Purification processes are widely used in hydrogen networks of refineries to increase hydrogen reuse. In refineries, hydrogen purification techniques include hydrocarbon, hydrogen sulfide and CO removal units. In addition, light hydrocarbon recovery from the hydrogen source streams can also result in hydrogen purification. In order to simplify the superstructure and mathematical model of hydrogen network integration, the models of different purification processes are unified in this paper, including mass balance and the expressions for hydrogen recovery and impurity removal ratios, which are given for all the purification units in refineries. Based on the proposed unified model, a superstructure of hydrogen networks with purification processes is constructed.

Comparison of mechanisms and efficacies of five formulas for improving blood circulation and removing blood stasis

Objective This study aimed to compare the mechanisms and efficacies of five formulas that improve blood circulation and remove blood stasis.Methods(1)A network pharmacology method was used to determine the targets of five formulas that promote circulation and remove blood stasis.Compounds of the five formulas,namely Danshen Yin(丹参饮,DSY),Huoluo Xiaoling Dan(活络效灵丹,HLXLD),Shixiao San(失笑散,SXS),Taohong Siwu Tang(桃红四物汤,THSWT),and Xuefu Zhuyu Tang(血府逐瘀汤,XFZYT),were retrieved from the Traditonal Chinese Medicine System Pharmacology Database(TCMSP),the Shanghai Institute of Organic Chemistry of CAS,and the TCM Integrated Database.Drug target network was constructed by searching the Swiss Target Prediction database and the STITCH database.The target network of stasis was extracted from the PharmGKB database,the Online Mendelian Inheritance in Man(OMIM)database,the Genetic Association Database(GAD),and the Therapeutic Target Database(TTD).Candidate targets were determined using protein-protein interaction(PPI)network extension and topology selection.Thereafter,Kyoto Encyclopedia of Genes and Genomes(KEGG)enrichment analysis was used to determine the differentiation of the mechanism of the five formulas.(2)Animal experiments were conducted to explore the efficacies of the five formulas in treating blood stasis.Seventy New Zealand rabbits were exposed to high-fat feeding+epinephrine injection to construct a blood stasis syndrome model.The rabbits were evenly divided into control,model,DSY,HLXLD,SXS,THSWT,and XFZYT groups.The latter five groups were orally administered the corresponding formulas[DSY:3.92 g/(kg·d),XFZYT:7.10 g/(kg·d),SXS:1.12 g/(kg·d),HLXLD:5.60 g/(kg·d),THSWT:4.48 g/(kg·d)].Serum lipid and blood rheology were analyzed,and pathology slices were observed.Results(1)A total of 269,358,288,370,and 376 candidate targets of DSY,HLXLD,SXS,THSWT,and XFZYT were obtained among which were 232 shared candidate targets.Fluid shear stress and atherosclerosis were the biological processes common to the five formulas.HLXLD,SXS,DSY,and THSWT regulated lipolysis in adipocytes,and XFZYT,HLXLD,SXS,and THSWT regulated the AGE-RAGE signaling pathway and complement and coagulation cascades.HLXLD,SXS,and XFZYT regulated the HIF-1 signaling pathway,DSY regulated the cGMP-PKG signaling pathway,HLXLD reduced platelet activation,SXS regulated the calcium signaling pathway,and XFZYT regulated the PPAR signaling pathway.(2)In the animal experiments,the values of total cholesterol(TC),triglyceride(TG),low density lipoprotein cholesterol(LDL),high density lipoprotein cholesterol(HDL),TC/HDL,and TG/HDL in each group decreased,among which the ones seen in XFZYT,HLXLD,and SXS groups were statistically significant(P<0.05).XFZYT presented the best effect,followed by HLXLD and SXS.XFZYT and HLXLD decreased apolipoprotein B100(apoB100)and increased apolipoprotein A1/apoB100(P<0.05).XFZYT decreased all the values of hematocrit(HCT),plasma viscosity,whole blood viscosity(WBV);HLXLD and SXS affected HCT;and DSY and THSWT regulated WBV(P<0.05).All the five formulas decreased the values of optical density and area of plaque,among which XFZYT and HLXLD showed statistical significance(P<0.05).Conclusion Adjusting fluid shear stress and alleviating the injury of endothelial cells might be the common mechanisms by which the five formulas promote blood circulation and remove blood stasis.Different formulas also have unique targets,which may provide guidance for clinical drug selection.By regulating different indices,the five formulas can regulate blood lipid and hemorheology,improve the state of blood stasis,and decrease the degree of aortic plaque in the blood stasis model rabbits.XFZYT and HLXLD had higher efficacies than DSY,THSWT,and SXS.

Bangla Handwritten Character Recognition Using Extended Convolutional Neural Network

The necessity of recognizing handwritten characters is increasing day by day because of its various applications. The objective of this paper is to provide a sophisticated, effective and efficient way to recognize and classify Bangla handwritten characters. Here an extended convolutional neural network (CNN) model has been proposed to recognize Bangla handwritten characters. Our CNN model is tested on “BanglalLekha-Isolated” dataset where there are 10 classes for digits, 11 classes for vowels and 39 classes for consonants. Our model shows accuracy of recognition as: 99.50% for Bangla digits, 93.18% for vowels, 90.00% for consonants and 92.25% for combined classes.

Revealing the material basis of MMP9-mediated activating blood and removing blood stasis drugs on Danshen-Ligusticum chuanxiong anti- vascular effect

Objective:Reveal the material basis and mechanism of angiogenesis effect of Danshen-chuanxiong herb-partners.Methods:The effective components of Danshen and chuanxiong were searched in the TCM System Pharmacology Database(TCMSP),and the drug targets were searched in the Swiss Target Prediction database;in the Disease-Gene Database(DisGeNET))Search for angiogenesis-promoting and inhibiting angiogenesis targets;use the protein interaction database(STRING)database and Cytoscape software to analyze the network core targets;use the DAVID database to enrich the drug-disease target intersection for GO and KEGG;RT-qPCR experiments verified the effect of luteolin on MMP9 gene expression;in vitro tube formation experiments analyzed the effect of luteolin on endothelial cell neovascularization.Results:After analysis,it is concluded that Danshen has 10 active ingredients and 50 angiogenesis targets;Chuanxiong has 2 active ingredients and 4 angiogenesis targets.The intersection of drug-taking targets yielded 9 targets,including AR,PARP1,MMP9,MMP2,MMP12,AKR1B10,ABCC1,CDK6,and STAT3.After analyzing the active ingredient-target network graph of Danshen/Chuanxiong,it was concluded that luteolin is the key chemical ingredient.After protein interaction(PPI)analysis,the results showed that MMP9 is the core target.The KEGG pathway is mainly enriched in 5 signaling pathways,the most important pathway is cancer-related pathways.There are 22 main biological processes of GO,which mainly involve collagen catabolism,protein extracellular matrix,metal endopeptidase activity,etc.The preliminary analysis of tubule formation experiment shows that luteolin inhibits angiogenesis and down-regulates the expression of MMP9 gene.Conclusion:Luteolin,the key chemical component of the Danshen-Ligusticum chuanxiong drug pair,inhibits angiogenesis by regulating the MMP9 target,and develops the reverse regulation of angiogenesis for the traditional Chinese medicine for promoting blood circulation and removing blood stasis on the basis of the known treatment of ischemic diseases.Provide a new perspective and theoretical basis for the use of Danshen and Chuanxiong medicine in clinical cancer patients.Reveal the material basis and mechanism of angiogenesis effect of Danshen-chuanxiong herb-partners.

基于多尺度生成对抗网络的锥束CT图像耦合伪影校正方法

针对锥束CT(CBCT)图像存在的耦合伪影难以完全校正问题,提出1种基于多尺度生成对抗网络的锥束CT图像耦合伪影校正方法。首先,根据CT图像的伪影特征构建了一套包括仿真图像和实际图像的CBCT耦合伪影数据集,用于提高模型泛化能力。再将特征金字塔结构(FPN)和基于卷积块的注意机制(CBAM)融入网络的生成器结构中,帮助网络捕获更全面特征信息,并配合多尺度判别器(MSD)搭建生成对抗网络框架,使得生成的去伪影图像更加清晰和真实。实验分析显示,经本文方法校正图像的PSNR和SSIM在仿真数据集中提高了21.595 dB、0.541,在实际数据集中提高了14.072 dB、0.274。实验结果表明,本文方法可有效校正耦合伪影。

活血化瘀法治疗盆腔炎性疾病后遗症的临床用药及网络药理学研究

盆腔炎性疾病(Pelvic inflammation disease,PID)是妇科临床常见病和多发病,如果治疗不当或病程迁延较久,则会发展成为盆腔炎性疾病后遗症(Sequelae of PID,SPID),盆腔炎性疾病后遗症可致不孕、输卵管妊娠和盆腔充血综合征等,严重影响妇女健康和生活质量。本文对黑龙江中医药大学附属第一医院门诊电子诊疗系统收集的以“活血化瘀法”治疗盆腔炎性疾病后遗症(Sequelae of pelvic inflammatory disease,SPID)的临床中药方剂进行收集整理和数据挖掘,得到核心药物及组方,为临床用药提供指导及参考。

基于定性网络模型评价生态调控情景对海洋牧场生态系统的影响——以獐子岛海洋牧场为例

在海洋牧场建设过程中,生态调控方式实施后对海洋牧场生态系统的影响通常难以预测,这对海洋牧场的生态安全和高质量发展提出了严峻挑战。为此,建立了一种基于定性网络模型(Qualitative network model,QNM)的海洋牧场生态系统模拟评价方法,并以獐子岛海洋牧场近岸增殖海域作为研究区域,构建以增殖目标种为核心的定性网络模型,模拟评估海洋牧场3种不同类型的生态调控情景(增殖目标种、移除捕食者、海藻场修复)及其复合条件下,牧场群落范围内的响应,分析海洋牧场生态调控策略与生物功能群变化之间潜在关系。结果显示:目标种增殖(仿刺参和虾夷扇贝)产生的上行效应导致其捕食者呈现积极响应,产生的下行效应导致其它底栖动物、浮游植物和有机碎屑等功能群呈现消极响应,移除捕食者海星产生的下行效应导致虾夷扇贝呈现积极响应,表明在增殖区清除敌害生物的重要性,海藻场修复对整个群落有明显的积极影响,体现了海藻场在养护近岸生态系统的重要生态意义。研究表明:QNM可有效识别生态系统潜在的营养级联效应,评估生物功能群的响应,基于QNM的海洋牧场生态调控模拟评价方法,突破了定量食物网模型在数据有限系统中使用的局限性,可为海洋牧场建设的生态调控策略制定提供科学参考。

图像去雾中深度学习数据增强方法

图像去雾是图像处理领域中非常重要的问题。深度学习可以有效提高图像清晰度,但训练过程中由于缺少相对应的真实雾匹配数据对,多采用合成雾作为数据集。现有合成雾多依赖于深度信息、大气散射系数等参数,针对由此作为数据集训练容易造成颜色失真和去雾不彻底的问题,提出基于循环生成对抗网络(CycleGAN)合成雾方法。通过该网络进行不匹配数据对训练学习有雾图像的特征,然后赋予清晰图片真实雾特征并与其自身构成匹配数据对,最后再用此类数据集进行去雾训练。结果表明,这些数据集可以有效解决颜色失真和去雾不彻底等问题。

电源外析原理与拟功率平衡原理及双端口网络等效原理

戴维宁定理和诺顿定理合称戴诺定理,是电路原理中最重要的定理之一,也是求解电路最重要的方法之一,但戴诺定理仅适用于单端口网络。分析总结了实电路和复电路的基本原理,对端口网络进行了完整分类、对双端口网络参数表达进行了分析研究,提出了含激端口网络的反激去激原理和电源外析原理,梳理了平稳电路的拟功率平衡原理和互易性原理,并提出了平稳电路无激双端口网络与含激双端口网络的紧致等效原理,有助于完善实电路和复电路的知识体系。

活血化瘀类中成药辅助治疗神经根型颈椎病有效性的网状Meta分析

目的:系统评价不同活血化瘀类中成药辅助治疗神经根型颈椎病(cervical spondylotic radiculopathy,CSR)的临床疗效。方法:应用计算机检索中国知网、维普网、万方数据库、PubMed、Web of Science和Cochrane Library中关于活血化瘀类中成药辅助治疗CSR的随机对照试验,检索时限均为建库至2023年9月20日。对照组单纯采用常规非手术方法治疗,试验组在对照组的基础上联合活血化瘀类中成药口服治疗。依据文献检索及筛选方案筛选出符合要求的文献后,由2名研究人员分别独立进行数据提取和质量评价。采用Stata18.0软件绘制网状关系图,对治疗有效率、颈部疼痛视觉模拟量表(visual analogue scale,VAS)评分、颈椎功能障碍指数(neck disability index,NDI)进行网状Meta分析,并根据累计概率排名曲线下面积(the surface under the cumulative ranking curve,SUCRA)对治疗方法的疗效进行排序。采用比较-校正漏斗图进行发表偏倚检验。结果:共检索到6768篇文献,通过逐层筛选最终纳入28篇文献,涉及颈痛颗粒、颈舒颗粒、脊痛消胶囊、痹祺胶囊、舒筋通络颗粒、颈复康颗粒共6种活血化瘀类中成药。网状Meta分析结果显示,在治疗有效率方面,不同活血化瘀类中成药辅助治疗CSR的疗效排序为颈痛颗粒联合常规治疗(SUCRA=78.4%)>痹祺胶囊联合常规治疗(SUCRA=71.6%)>颈舒颗粒联合常规治疗(SUCRA=62.1%)>舒筋通络颗粒联合常规治疗(SUCRA=53.7%)>脊痛消胶囊联合常规治疗(SUCRA=43.3%)>颈复康颗粒联合常规治疗(SUCRA=40.5%)>常规治疗(SUCRA=0.4%);在颈部疼痛VAS评分方面,不同活血化瘀类中成药辅助治疗CSR的疗效排序为舒筋通络颗粒联合常规治疗(SUCRA=96.3%)>痹祺胶囊联合常规治疗(SUCRA=86.2%)>颈复康颗粒联合常规治疗(SUCRA=51.6%)>颈痛颗粒联合常规治疗(SUCRA=50.8%)>颈舒颗粒联合常规治疗(SUCRA=37.4%)>脊痛消胶囊联合常规治疗(SUCRA=27.4%)>常规治疗(SUCRA=0.4%);在NDI方面,不同活血化瘀类中成药辅助治疗CSR的疗效排序为痹祺胶囊联合常规治疗(SUCRA=95%)>颈复康颗粒联合常规治疗(SUCRA=71.3%)>颈痛颗粒联合常规治疗(SUCRA=69.2%)>脊痛消胶囊联合常规治疗(SUCRA=38.7%)>颈舒颗粒联合常规治疗(SUCRA=20.8%)>常规治疗(SUCRA=5.0%)。结论:现有的证据表明,颈痛颗粒辅助治疗CSR的综合疗效最佳,舒筋通络颗粒在缓解颈部疼痛方面效果最佳,痹祺颗粒在改善颈椎功能方面效果最佳。但由于对照组采用的非手术疗法不尽相同及文献数量、质量等的限制,本研究的结论尚需进一步验证。

耐酸反硝化复合菌群的构建及强化生物脱氮

为增强酸性含氮废水生物脱氮效果和提高水处理微生物对酸性水质条件的适应性,采用“自下而上”的脱氮菌群构建策略,利用统计分析和部分因子实验筛选核心菌株,并围绕核心菌株理性设计构建了一组反硝化复合菌群S119-S136-L22,其在p H值为4.0,NO_(3)^(-)-N初始质量浓度200mg/L的条件下,36h硝氮去除率为97.32%.通过全基因组测序并结合各菌株与氮代谢、反硝化作用相关基因分析,推测菌群中菌株S136和L22主要负责将NO_(3)^(-)-N还原成NO_(2)^(-)-N,S119负责将NO_(2)^(-)-N还原为N_(2).序批式反应器(SBR)研究显示,实验组较空白组NO_(3)^(-)-N、TN的去除率分别提高28.99%、28.67%.高通量测序与系统微生物群落互作网络分析表明,相较于空白组,除了Enterobacter、Alcaligenes和Klebsiella,实验组还出现了2个差异功能菌属Cloacibacterium、Serratia.实验组呈现出更为复杂多样的正相关路径,不仅复合菌群中各菌属间呈协同促进作用,Serratia与Acinetobacter、Achromobacter,Cloacibacteriu与Macellibacteroides和Pleomorphomonas也呈共线性关系.出水TN、NO_(3)^(-)-N与微生物菌群互作网络分析,Cloacibacterium主要驱动TN、NO_(3)^(-)-N的去除,Pleomorphomonas、Microbacterium、Propionicicella在TN、NO_(3)^(-)-N的去除过程中也发挥了重要作用.