融合Res3D、BiLSTM和注意力机制的羊只行为识别方法

识别动物行为可以为疾病预防和合理喂养提供重要依据,从而有助于更好地关注动物的健康和福利。本文提出了一种融合三维残差卷积神经网络、双向长短期记忆网络和注意力机制的深度学习网络模型(AdRes3D-BiLSTM)。AdRes3D-BiLSTM模型可以直接针对视频流进行识别,在AdRes3D部分引入了深度可分离卷积和注意力机制,不但减少了浮点运算量,提升了网络轻量化程度,还提高了时间和空间两个维度的特征提取能力;提取的特征被输入BiLSTM模块后,从前后2个方向对时序特征向量进行筛选和更新,最后对羊只行为进行准确识别。试验结果表明,AdRes3D-BiLSTM对羊只站立、躺卧、进食、行走和反刍5种行为的综合识别准确率达到了98.72%,帧速率达到52.79 f/s,模型内存占用量为28.03 MB。研究结果为基于视频流的动物行为识别提供了新的方法和思路。

基于Res2Net的人脸表情识别方法

为解决自然条件下人脸表情识别易受角度、光线、遮挡物的影响以及人脸表情数据集各类表情数量不均衡等问题,提出基于Res2Net的人脸表情识别方法。使用Res2Net50作为特征提取的主干网络,在预处理阶段对图像随机翻转、缩放、裁剪进行数据增强,提升模型的泛化性。引入广义平均池化(generalized mean pooling, GeM)方式,关注图像中比较显著的区域,增强模型的鲁棒性;选用Focal Loss损失函数,针对表情类别不平衡和错误分类问题,提高较难识别表情的识别率。该方法在FER2013数据集上准确率达到了70.41%,相较于原Res2Net50网络提高了1.53%。结果表明,在自然条件下对人脸表情识别具有更好的准确性。

基于Ghost-SE-Res2Net的多模型融合语音唤醒词检测方法

语音唤醒词检测(WWD)是语音交互中的关键技术,选择合适大小的检测窗对WWD性能的影响很大。提出一种新的多模型融合方法,通过融合小检测窗和大检测窗的检测结果来提高WWD性能。多模型融合方法包含两个分类模型,分别使用小检测窗和大检测窗,均基于轻量化的挤压与激励残差网络(SE-Res2Net)模块,即GhostSE-Res2Net,SE-Res2Net结构的多尺度机制可显著提升WWD的能力。在Ghost-SE-Res2Net中,首先使用Ghost卷积替换SE-Res2Net中的普通卷积以降低模型参数量,然后使用注意力池化层替换SE-Res2Net中的全局平均池化层进一步提升WWD能力。在实际检测时融合连续3个小检测窗模型的检测结果的最大值和1个大检测窗模型的检测结果,来判断唤醒词是否被触发。在训练时引入困难样本挖掘算法,选择性地学习较难检测的唤醒词信息以提高分类模型的检测性能。在包含2个唤醒词的Mobvoi数据集上评估系统性能,实验结果表明,在每小时0.5次错误唤醒的情况下,该系统在2个唤醒词上的错误拒绝率分别为0.46%和0.43%,实现了与先进基线相似的性能,并且系统参数量比基线少31%。

Infrared and Visible Image Fusion Based on Res2Net-Transformer Automatic Encoding and Decoding

A novel image fusion network framework with an autonomous encoder and decoder is suggested to increase thevisual impression of fused images by improving the quality of infrared and visible light picture fusion. The networkcomprises an encoder module, fusion layer, decoder module, and edge improvementmodule. The encoder moduleutilizes an enhanced Inception module for shallow feature extraction, then combines Res2Net and Transformerto achieve deep-level co-extraction of local and global features from the original picture. An edge enhancementmodule (EEM) is created to extract significant edge features. A modal maximum difference fusion strategy isintroduced to enhance the adaptive representation of information in various regions of the source image, therebyenhancing the contrast of the fused image. The encoder and the EEM module extract features, which are thencombined in the fusion layer to create a fused picture using the decoder. Three datasets were chosen to test thealgorithmproposed in this paper. The results of the experiments demonstrate that the network effectively preservesbackground and detail information in both infrared and visible images, yielding superior outcomes in subjectiveand objective evaluations.

Multi-Criteria Wildfire Risk Hazard Assessment in GIS Environment: Projection for the Future and Impact on RES Projects Installation Planning

It is alarming for the fact that Wildfires number, severity and consequently impact have significantly increased during the last years, an aftermath of the Climate Change. One of the most affected areas worldwide is Mediterranean, due to the unique combination of its type of vegetation and demanding climatic conditions. This research is focused on the Region of Epirus in Greece, an area with significant natural vegetation and a range of geomorphological aspects. In order to estimate the Wildfire Risk Hazard, several factors have been used: geomorphological (slope, aspect, elevation, TWI, Hydrographic network), social (Settlements and landfils, roads, overhead lines and substations), environmental (land cover) and climatic (Fire Weather Index). Through a multi-criteria decision analysis (MCDA) and an analytic hierarchy process (AHP) in a GIS environment, the Wildfire Risk Hazard has been estimated not only for current conditions but also for future projections for the near future (2031-2060) and the far future (2071-2100). The selected case study includes the potential impact of the Wildfires to the installed (or targeted to be installed) RES projects in the studied region.

基于改进的ResNet在甘蔗病害识别中的研究

针对传统Res Net在甘蔗病害识别中的局限性,如模型泛化能力不足、训练收敛速度慢和容易过拟合等问题,提出一种改进型Res Net模型的解决方案。对原始Res Net模型实施了多方面的优化,重点包括对残差模块的改进以减少梯度消失,扩展模型深度以提高表达能力,以及引入注意力机制以更有效地捕捉局部信息和关键特征。试验结果表明:改进型Res Net模型在甘蔗病害识别任务中实现了较高的准确率、召回率和F1值;训练过程中改进型Res Net模型展现出较快的收敛速度,同时通过采用早停策略能够有效地避免过拟合问题,使得模型在测试集上的性能更加稳定,从而进一步提高了甘蔗病害识别的准确性和可靠性。

SAMHD1基因突变导致Aicardi-Goutières综合征5型一例并文献复习

目的探讨SAMHD1基因突变导致Aicardi-Goutières综合征(Aicardi-Goutières syndrome,AGS)5型患儿的临床特点。方法选取北京协和医院儿科收治的AGS 5型患儿1例,回顾性分析其临床资料,并复习相关文献。结果本例患儿女性,4岁;1岁时以皮疹起病、形态多样,主要表现为冻疮样皮疹、环形红斑,2岁时出现肌张力增高、步态异常;体格检查提示身材矮小。辅助检查提示血沉62 mm/h,干扰素评分为19.6分,双侧基底节及额叶钙化,双侧额叶及脑室前角脑白质病变。Griffiths神经发育评估提示整体发育落后。基因检测提示SAMHD1基因纯合突变,诊断AGS 5型明确。以SAMHD1 AND Aicardi-Goutières syndrome为主题词在pubmed中进行文献检索,共检出17篇文献,包括AGS 5型患者42例,最常见临床特点包括颅内钙化(95.45%)、发育落后(82.92%)、矮小(70.73%)、皮疹(63.41%)、颅内血管狭窄(62.96%)和四肢肌张力增高(60.71%)等。结论冻疮样皮疹、颅内钙化、脑白质病变、生长发育落后、颅内血管狭窄对于该疾病可能具有提示意义,该疾病可持续进展多年,诊断后需密切监测身体各系统,及时干预。

21例Aicardi-Goutières综合征的临床表现和遗传学分析

目的:总结我国经基因测序诊断Aicardi-Goutières综合征(Aicardi-Goutières syndrome,AGS)患者的临床表现和遗传学特征。方法:收集符合纳入标准的AGS患者的临床资料和基因测序结果并进行回顾性分析。结果:本研究纳入21例我国AGS患者,临床表现为智力障碍(90.0%)、运动障碍(89.5%)、肌张力障碍(73.7%)和小头畸形(70.6%)等;头颅影像学多表现为基底节双侧对称钙化、进行性脑萎缩(95.2%);皮肤表现以冻疮样皮疹为主(84.2%)。21例患者中,TREX1基因突变8例,RNASEH2C基因突变5例,IFIH1基因突变4例,ADAR基因突变2例,RNASEH2A和RNASEH2B基因突变各1例,未涉及SAMHD1基因突变。84.2%为家系遗传,父母携带者均无症状,15.8%为新发突变。结论:建议产前加强对胎儿生长受限、小头畸形和侧脑室增宽等软指标异常胎儿的重视,加强产检管理,及时多学科讨论和产前诊断,减少AGS患儿的出生。

基于Res2-UNet模型的皮带煤量检测

为了能够检测煤矿井下的煤量,预测和提高煤的利用率,同时节省电能,减少人力的监管和资源成本。利用煤矿安装的视频监控系统,采用非接触的方式通过Camshift算法对快速运动皮带上的煤量捕捉和跟踪,然后建立Res2-UNet模型来获得显著性信息,融合灰度、纹理、边缘等特征到单一的网络中,实现了皮带煤量的检测。模型利用U-Net网络的思想以编码器-解码器为架构,编码器以Res2Net网络为骨干网络提取煤流不同层次特征的信息,解码器通过反卷积上采样操作恢复图像尺寸。经过构建皮带数据集训练和测试模型,实验结果表明,提出的方法能够快速准确地检测出皮带上的煤料,精确率达到95.5%,每张图像从输入网络到输出的运行时间为4.8s。表明该方法具有一定的实用性和有效性。

Aicardi-Goutières综合征2例及文献复习

报道2例Aicardi-Goutières综合征的临床资料,并进行文献复习。

基于Res2Net-IDCN-SCF算法的多模态医学图像融合

利用多尺度特征策略进行特征提取的有效性不足是多模态医学图像融合领域存在的问题。为了增加融合结果的多尺结构信息,提出了一种基于残差多尺度网络(residual multi-scale network,Res2Net)、交错稠密网络和空间通道融合算法的多模态医学图像融合算法。Res2Net的编码器在提取多尺度特征时能保留更多语义信息;交错稠密网络减少了解码器和编码器之间的语义差异,丰富了融合图像的结构和细节信息;掩码鉴别器约束了脑瘤病灶区域,进一步提高了融合图像的质量;特征图通过空间通道融合算法融合减少了多模态图像之间的信息冗余。该算法在信息熵(entropy of information,EN)、互信息(mutual information,MI)、结构相似性(structure similarity index measure,SSIM)、多尺度结构相似性(multi scale structural similarity index measure,MI_SSIM)指标上拥有较高水平的性能表现,EN提高了6%,MI提高了3%。结果显示,所提出的算法在视觉感知和指标评估上达到了较高的融合质量。

基于Res2net和金字塔池化的图像去雾算法

在计算机视觉的高级任务中,对图像的清晰度有很高的要求,目前基于深度学习的图像去雾算法仍存在一些问题,如细节丢失、色彩失真、去雾不完全等。为解决这些问题,设计了一种基于Res2net和金字塔池化的端到端图像去雾算法。该网络中,通过使用Res2net模块提取上下文特征,并利用金字塔池化模块融合不同尺度的特征信息。为了得到更好的网络模型,采用RESIDE数据集对提出的模型分别进行训练和测试。结果表明:该模型在主客观评价中都取得了不错的效果,极大地改善了去雾后图片色彩失真和去雾不够彻底的问题。

Assessment of China’s forest fi re occurrence with deep learning, geographic information and multisource data

Considerable economic losses and ecological damage can be caused by forest fi res,and compared to suppression,prevention is a much smarter strategy.Accordingly,this study focuses on developing a novel framework to assess forest fi re risks and policy decisions on forest fi re management in China.This framework integrated deep learning algorithms,geographic information,and multisource data.Compared to conventional approaches,our framework featured timesaving,easy implementation,and importantly,the use of deep learning that vividly integrates various factors from the environment and human activities.Information on 96,594 forest fi re points from 2001 to 2019 was collected on Moderate Resolution Imaging Spectroradiometer(MODIS)fi re hotspots from 2001 to 2019 from NASA’s Fire Information Resource Management System.The information was classifi ed into factors such as topography,climate,vegetation,and society.The prediction of forest fi re risk was generated using a fully connected network model,and spatial autocorrelation used to analyze the spatial aggregation correlation of active fi re hotspots in the whole area of China.The results show that high accuracy prediction of fi re risks was achieved(accuracy 87.4%,positive predictive value 87.1%,sensitivity 88.9%,area under curve(AUC)94.1%).Based on this,it was found that Chinese forest fi re risk shows signifi cant autocorrelation and agglomeration both in seasons and regions.For example,forest fi re risk usually raises dramatically in spring and winter,and decreases in autumn and summer.Compared to the national average,Yunnan Province,Guangdong Province,and the Greater Hinggan Mountains region of Heilongjiang Province have higher fi re risks.In contrast,a large region in central China has been recognized as having a long-term,low risk of forest fi res.All forest risks in each region were recorded into the database and could contribute to the forest fi re prevention.The successful assessment of forest fi re risks in this study provides a comprehensive knowledge of fi re risks in China over the last 20 years.Deep learning showed its advantage in integrating multiple factors in predicting forest fi re risks.This technical framework is expected to be a feasible evaluation tool for the occurrence of forest fi res in China.

Identifying anthropogenic and natural causes of wildfires by maximum entropy method-based ignition susceptibility distribution models

Turkey has a high potential for wildfires along its Mediterranean coast because of its dense forest cover and mild climate.An average of 250 wildfires occurs every year with more than 10,000 hectares destroyed due to natural and human-related causes.The study area is sensitive to fires caused by lightning,stubble burning,discarded cigarette butts,electric arcing from power lines,deliberate fire setting,and traffic accidents.However,52%of causes could not be identified due to intense wildfires occurring at the same time and insufficient equipment and personnel.Since wildfires destroy forest cover,ecosystems,biodiversity,and habitats,they should be spatially evaluated by separating them according to their causes,considering environmental,climatic,topographic and forest structure variables that trigger wildfires.In this study,wildfires caused by lightning,the burning of agriculture stubble,discarded cigarette butts and power lines were investigated in the provinces of Aydin,Mugla and Antalya,where 22%of Turkey’s wildfires occurred.The MaxEnt method was used to determine the spatial distribution of wildfires to identify risk zones for each cause.Wildfires were used as the species distribution and the probability of their occurrence estimated.Additionally,since the causes of many wildfires are unknown,determining the causes is important for fire prediction and prevention.The highest wildfire occurrence risks were 9.7%for stubble burning,30.2%for lightning,4.5%for power lines and 16.9%by discarded cigarette butts.In total,1,266 of the 1,714 unknown wildfire causes were identified by the analysis of the cause-based risk zones and these were updated by including cause-as signed unknown wildfire locations for verification.As a result,the Area under the ROC Curve(AUC)values were increased for susceptibility maps.

Smoldering charcoal detection in forest soil by multiple CO sensors

Cleaning up residual fires is an important part of forest fire management to avoid the loss of forest resources caused by the recurrence of a residual fire.Existing residual fire detection equipment is mainly infrared temperature detection and smoke identification.Due to the isolation of ground,temperature and smoke characteristics of medium and large smoldering charcoal in some forest soils are not obvious,making it difficult to identify by detection equipment.CO gas is an important detection index for indoor smoldering fire detection,and an important identification feature of hidden smoldering ground fires.However,there is no research on locating smoldering fires through CO detection.We studied the diffusion law of CO gas directly above covered smoldering charcoal as a criterion to design a detection device equipped with multiple CO sensors.According to the motion decomposition search algorithm,the detection device realizes the function of automatically searching for smoldering charcoal.Experimental data shows that the average CO concentration over the covered smoldering charcoal decreases exponentially with increasing height.The size of the search step is related to the reliability of the search algorithm.The detection success corresponding to the small step length is high but the search time is lengthy which can lead to search failure.The introduction of step and rotation factors in search algorithm improves the search efficiency.This study reveals that the average ground CO concentration directly above smoldering charcoal in forests changes with height.Based on this law,a CO gas sensor detection device for hidden smoldering fires has been designed,which enriches the technique of residual fire detection.

FIR-YOLACT:Fusion of ICIoU and Res2Net for YOLACT on Real-Time Vehicle Instance Segmentation

Autonomous driving technology has made a lot of outstanding achievements with deep learning,and the vehicle detection and classification algorithm has become one of the critical technologies of autonomous driving systems.The vehicle instance segmentation can perform instance-level semantic parsing of vehicle information,which is more accurate and reliable than object detection.However,the existing instance segmentation algorithms still have the problems of poor mask prediction accuracy and low detection speed.Therefore,this paper proposes an advanced real-time instance segmentation model named FIR-YOLACT,which fuses the ICIoU(Improved Complete Intersection over Union)and Res2Net for the YOLACT algorithm.Specifically,the ICIoU function can effectively solve the degradation problem of the original CIoU loss function,and improve the training convergence speed and detection accuracy.The Res2Net module fused with the ECA(Efficient Channel Attention)Net is added to the model’s backbone network,which improves the multi-scale detection capability and mask prediction accuracy.Furthermore,the Cluster NMS(Non-Maximum Suppression)algorithm is introduced in the model’s bounding box regression to enhance the performance of detecting similarly occluded objects.The experimental results demonstrate the superiority of FIR-YOLACT to the based methods and the effectiveness of all components.The processing speed reaches 28 FPS,which meets the demands of real-time vehicle instance segmentation.

Unveiling localized electronic properties of ReS2 thin layers at nanoscale using Kelvin force probe microscopy combined with tip-enhanced Raman spectroscopy

Electronic properties of two-dimensional(2D) materials can be strongly modulated by localized strain. The typical spatial resolution of conventional Kelvin probe force microscopy(KPFM) is usually limited in a few hundreds of nanometers, and it is difficult to characterize localized electronic properties of 2D materials at nanoscales. Herein, tip-enhanced Raman spectroscopy(TERS) is proposed to combine with KPFM to break this restriction. TERS scan is conducted on ReS2bubbles deposited on a rough Au thin film to obtain strain distribution by using the Raman peak shift. The localized contact potential difference(CPD) is inversely calculated with a higher spatial resolution by using strain measured by TERS and CPD-strain working curve obtained using conventional KPFM and atomic force microscopy. This method enhances the spatial resolution of CPD measurements and can be potentially used to characterize localized electronic properties of 2D materials.

Aicardi-Goutières syndrome type 7 in a Chinese child:A case report

BACKGROUND IFIH1 is a protein-coding gene.Disorders associated with IFIH1 include Aicardi-Goutières syndrome(AGS)type 7 and Singleton-Merten syndrome type 1.Related pathways include RIG-I/MDA5-mediated induction of the interferon(IFN)-α/βpathway and the innate immune system.AGS type 7 is an autosomal dominant inflammatory disorder characterized by severe neurological impairment.In infancy,most patients present with psychomotor retardation,axial hypotonia,spasticity,and brain imaging changes Laboratory assessments showed increased IFN-αactivity with upregulation of IFN signaling and IFN-stimulated gene expression.Some patients develop normally in the early stage,and then have episodic neurological deficits.CASE SUMMARY The 5-year-old girl presented with postpartum height and weight growth retardation,language retardation,brain atrophy,convulsions,and growth hormone deficiency.DNA samples were obtained from peripheral blood from the child and her parents for whole-exome sequencing and test of genome-wide copy number variation.Heterozygous mutations in the IFIH1 gene were found.Physical examination at admission found that language development was delayed,the reaction to name calling was average,there was no communication with people,but there was eye contact,no social smile,and no autonomous language.However,the child had rich gesture language and body language,could understand instructions,had bad temper.When she wants to achieve something,she starts crying or shouting.Cardiopulmonary examination showed no obvious abnormality,and abdominal examination was normal.Bilateral muscle strength and muscle tone were symmetrical and slightly decreased.Physiological reflexes exist,but pathological reflexes were not elicited.CONCLUSION We reported the clinical characteristics of a Chinese child with a clinical diagnosis of AGS type 7,which expanded the mutational spectrum of the IFIH1 gene.

现在分词时间状语功能的RES时体理论阐释

RES时体理论在语言学界产生了深远的影响,为解释时体类型和关系研究提供了一个新的范式。该文以RES时体理论为框架,探究英语现在分词作时间状语成分的时体关系。研究表明,“时、体”是现在分词充当时间状语的根本动因;同时也是现在分词作时间状语成分与when、while和after时间状语复句之间转换生成动因;语境对现在分词当时间状语具有调控作用。研究结果为习得现在分词作时间状语结构提供理据。

基于RES2DMOD高密度电阻率法在采空区勘查中的应用

采用RES2DMOD高密度电阻率正演软件和RES2DINV高密度电阻率反演软件,选取有限差分法正演和最小二乘法反演的处理方法,对建立的三个不同采空区模型进行正、反演计算,研究地质体(采空区)的电阻率响应,并以在焦家金矿望儿山尾矿库区的具体应用为例,验证了高密度电阻率法在采空区勘查中的有效性和可靠性,同时对反演结果做出解释和分析,结果表明高密度电阻率法在采空区勘查中具有一定的效果,最小二乘反演效果较好,可以进一步推广和使用。