Deep Transfer Learning-Enabled Activity Identification and Fall Detection for Disabled People

The human motion data collected using wearables like smartwatches can be used for activity recognition and emergency event detection.This is especially applicable in the case of elderly or disabled people who live self-reliantly in their homes.These sensors produce a huge volume of physical activity data that necessitates real-time recognition,especially during emergencies.Falling is one of the most important problems confronted by older people and people with movement disabilities.Numerous previous techniques were introduced and a few used webcam to monitor the activity of elderly or disabled people.But,the costs incurred upon installation and operation are high,whereas the technology is relevant only for indoor environments.Currently,commercial wearables use a wireless emergency transmitter that produces a number of false alarms and restricts a user’s movements.Against this background,the current study develops an Improved WhaleOptimizationwithDeep Learning-Enabled Fall Detection for Disabled People(IWODL-FDDP)model.The presented IWODL-FDDP model aims to identify the fall events to assist disabled people.The presented IWODLFDDP model applies an image filtering approach to pre-process the image.Besides,the EfficientNet-B0 model is utilized to generate valuable feature vector sets.Next,the Bidirectional Long Short Term Memory(BiLSTM)model is used for the recognition and classification of fall events.Finally,the IWO method is leveraged to fine-tune the hyperparameters related to the BiLSTM method,which shows the novelty of the work.The experimental analysis outcomes established the superior performance of the proposed IWODL-FDDP method with a maximum accuracy of 97.02%.

Fingerprint Liveness Detection from Different Fingerprint Materials Using Convolutional Neural Network and Principal Component Analysis

Fingerprint-spoofing attack often occurs when imposters gain access illegally by using artificial fingerprints,which are made of common fingerprint materials,such as silicon,latex,etc.Thus,to protect our privacy,many fingerprint liveness detection methods are put forward to discriminate fake or true fingerprint.Current work on liveness detection for fingerprint images is focused on the construction of complex handcrafted features,but these methods normally destroy or lose spatial information between pixels.Different from existing methods,convolutional neural network(CNN)can generate high-level semantic representations by learning and concatenating low-level edge and shape features from a large amount of labeled data.Thus,CNN is explored to solve the above problem and discriminate true fingerprints from fake ones in this paper.To reduce the redundant information and extract the most distinct features,ROI and PCA operations are performed for learned features of convolutional layer or pooling layer.After that,the extracted features are fed into SVM classifier.Experimental results based on the LivDet(2013)and the LivDet(2011)datasets,which are captured by using different fingerprint materials,indicate that the classification performance of our proposed method is both efficient and convenient compared with the other previous methods.

Deep Transfer Learning Driven Automated Fall Detection for Quality of Living of Disabled Persons

Mobile communication and the Internet of Things(IoT)technologies have recently been established to collect data from human beings and the environment.The data collected can be leveraged to provide intelligent services through different applications.It is an extreme challenge to monitor disabled people from remote locations.It is because day-to-day events like falls heavily result in accidents.For a person with disabilities,a fall event is an important cause of mortality and post-traumatic complications.Therefore,detecting the fall events of disabled persons in smart homes at early stages is essential to provide the necessary support and increase their survival rate.The current study introduces a Whale Optimization Algorithm Deep Transfer Learning-DrivenAutomated Fall Detection(WOADTL-AFD)technique to improve the Quality of Life for persons with disabilities.The primary aim of the presented WOADTL-AFD technique is to identify and classify the fall events to help disabled individuals.To attain this,the proposed WOADTL-AFDmodel initially uses amodified SqueezeNet feature extractor which proficiently extracts the feature vectors.In addition,the WOADTLAFD technique classifies the fall events using an extreme Gradient Boosting(XGBoost)classifier.In the presented WOADTL-AFD technique,the WOA approach is used to fine-tune the hyperparameters involved in the modified SqueezeNet model.The proposedWOADTL-AFD technique was experimentally validated using the benchmark datasets,and the results confirmed the superior performance of the proposedWOADTL-AFD method compared to other recent approaches.

Fluorescent aptasensor for detection of live foodborne pathogens based on multicolor perovskite-quantum-dot-encoded DNA probes and dual-stirring-bar-assisted signal amplification

In this study,a fluorescent(FL)aptasensor was developed for on-site detection of live Salmonella typhimurium(S.T.)and Vibrio parahaemolyticus(V.P.).Complementary DNA(cDNA)of aptamer(Apt)-functionalized multicolor polyhedral oligomeric silsesquioxane-perovskite quantum dots(cDNA-POSSPQDs)were used as encoded probes and combined with dual-stirring-bar-assisted signal amplification for pathogen quantification.In this system,bar 1 was labeled with the S.T.and V.P.Apts,and then bar 2 was functionalized with cDNA-POSS-PQDs.When S.T.and V.P.were introduced,pathogen-Apt complexes would form and be released into the supernatant from bar 1.Under agitation,the two complexes reached bar 2 and subsequently reacted with cDNA-POSS-PQDs,which were immobilized on MXene.Then,the encoded probes would be detached from bar 2 to generate FL signals in the supernatant.Notably,the pathogens can resume their free state and initiate next cycle.They swim between the two bars,and the FL signals can be gradually enhanced to maximum after several cycles.The FL signals from released encoded probes can be used to detect the analytes.In particular,live pathogens can be distinguished from dead ones by using an assay.The detection limits and linear range for S.T.and V.P.were 30 and 10 CFU/mL and 10^(2) -10^(6) CFU/mL,respectively.Therefore,this assay has broad application potential for simultaneous on-site detection of various live pathogenic bacteria in water.

Fingerprint Liveness Detection Based on Multi-Scale LPQ and PCA

Fingerprint authentication system is used to verify users' identification according to the characteristics of their fingerprints.However,this system has some security and privacy problems.For example,some artificial fingerprints can trick the fingerprint authentication system and access information using real users' identification.Therefore,a fingerprint liveness detection algorithm needs to be designed to prevent illegal users from accessing privacy information.In this paper,a new software-based liveness detection approach using multi-scale local phase quantity(LPQ) and principal component analysis(PCA) is proposed.The feature vectors of a fingerprint are constructed through multi-scale LPQ.PCA technology is also introduced to reduce the dimensionality of the feature vectors and gain more effective features.Finally,a training model is gained using support vector machine classifier,and the liveness of a fingerprint is detected on the basis of the training model.Experimental results demonstrate that our proposed method can detect the liveness of users' fingerprints and achieve high recognition accuracy.This study also confirms that multi-resolution analysis is a useful method for texture feature extraction during fingerprint liveness detection.

Integrate Liveness Detection with Iris Verification to Construct Support Biometric System

The probabilities for the technology to be spoofed are widely acknowledged in biometric verification system. Important efforts have been conducted to study such threats and to develop countermeasures to direct attacks to the biometric verification system to ensure the security of these systems against spoof attacks and reduce this risk, by using another module that is added to the biometric verification system called the “liveness detection” which uses different anatomical properties to distinguish between real and fake traits. Thus, the robustness of the system against direct attacks can be improved through increasing the security level offered to the final user. This paper is an attempt to construct support biometric security system to protect the iris biometric verification system from spoof attacks, through integrating the iris verification system with addition module called liveness detection which composed of two sub-modules (static and dynamic). A test has been performed, for iris verification phase performed on two types of database (MMU DB) for 180 samples and (CASIA DB) for 90 samples, and gave accuracy (99.44%) with FAR of (0.0277) and FRR (0.0055) for MMU DB, and accuracy (97.77%) with FAR of (0.0333) and FRR (0.0222) for CASIA DB.

Detecting Iris Liveness with Batch Normalized Convolutional Neural Network

Aim to countermeasure the presentation attack for iris recognition system,an iris liveness detection scheme based on batch normalized convolutional neural network(BNCNN)is proposed to improve the reliability of the iris authentication system.The BNCNN architecture with eighteen layers is constructed to detect the genuine iris and fake iris,including convolutional layer,batch-normalized(BN)layer,Relu layer,pooling layer and full connected layer.The iris image is first preprocessed by iris segmentation and is normalized to 256×256 pixels,and then the iris features are extracted by BNCNN.With these features,the genuine iris and fake iris are determined by the decision-making layer.Batch normalization technique is used in BNCNN to avoid the problem of over fitting and gradient disappearing during training.Extensive experiments are conducted on three classical databases:the CASIA Iris Lamp database,the CASIA Iris Syn database and Ndcontact database.The results show that the proposed method can effectively extract micro texture features of the iris,and achieve higher detection accuracy compared with some typical iris liveness detection methods.

Peak Detection Implementation for Real-Time Signal Analysis Based on FPGA

In this paper a real-time peak detection method based on modified Automatic Multiscale Field Detection (AMPD) algorithm and Field Programmable Gate Arrays (FPGA) technologies of a time series data is studied, and optimum scaling is highlighted after testing several scales. To validate the results obtained from modified algorithm, they are compared with the results of original AMPD method. As data of this study, three-phase voltage values of a power station are used. A detail detective sensitivity analysis of phase-to-phase voltage values is tried at different scales. Moreover, the original algorithm is tested regarding the off-line mode to obtain optimum scaling for real-time peak point detection. It is concluded that the peak detection of minimum and maximum points of data series achieved by modified algorithm is very close to the results of original AMPD algorithm.

Research on Influence of Electromagnetic Interference on X-Ray Digital Detection

X-ray digital imaging technology has found wide application owing to its advantages of real-time, visualization and rapid imaging. In substations the substantial electromagnetic interference has some influence on the live detection by the X-ray digital imaging technology, hindering the promotion of the technology in the detection of electric equipment. Based on a large number of field tests, the author carried out a series of researches on electromagnetic interference protection measures, image de-noising, and image enhancement algorithms.

跨域人脸活体检测的单边对抗网络算法

现有跨域人脸活体检测算法,其特征提取过程容易发生过拟合和缺乏特征聚合所导致的泛化性不足问题。针对该问题,提出了跨域人脸活体检测的单边对抗网络算法,将分组卷积与改进的倒残差结构融合替换普通卷积,降低网络参数同时加强人脸细粒度特征的表达能力,并引入自适应特征归一化模块,强调图像中人脸活体信息区域淡化无关背景区域,有效避免人脸活体信息的过拟合并加强来自不同源域的人脸活体检测能力。基于NetVLAD引入通道注意力机制模块,通道注意力机制模块作为特征聚合网络的分支,学习不同源域中人脸局部特征的语义信息,有效增强对不同源域的人脸活体信息分类的泛化能力。设计两模块融合网络以提高未知场景下跨域人脸活体检测精度。在OULU-NPU、CASIA-FASD、MSU-MFSD和Idiap Replay-Attack数据集上的实验结果表明,该算法在跨数据集测试O&C&M to I、O&C&I to M、I&C&M to O、O&M&I to C均有不错的表现,其中,在O&C&I to M及O&M&I to C性能评估指标分别提升了0.99个百分点和0.5个百分点的精度。

产志贺毒素大肠埃希氏菌活菌检测方法的应用现状及展望

产志贺毒素大肠埃希氏菌作为一种重要的食源性致病菌,会对人体健康造成严重威胁。对其进行快速、准确的鉴定检测是预防和控制相关疾病的有效手段。基于此,文章对目前可用于检测活体产志贺毒素大肠埃希氏菌的方法进行综述,主要包括传统的细胞培养方法、免疫学检测方法和分子生物学方法等,并对这些方法的应用和发展前景展开了深入探讨。随着分子生物学和生物工程技术的发展,不同检测方法联合使用可以克服单一方法的缺陷,从而取得更高效、准确的检测结果。

基于改进YOLO v5的野外实景视频水鸟检测方法

为实现野外视频监控下水鸟的快速准确识别,基于YOLO v5框架提出了一种自动化水鸟实时检测方法YOLO v5_k-mixup.该方法在YOLO v5网络的基础上内置了Mixup数据增强模块,能有效提高YOLO v5的泛化能力,改善水鸟相互遮挡无法识别的问题;同时,针对水鸟体型差异带来的检测框定位困难问题,提出了基于k-means++聚类锚框的方法,提高了检测框定位精度.与未改进的YOLO v5相比,YOLO v5_k-mixup在保持高检测速度的情况下,平均精度由84.8%提升到了87.1%.改进后的模型对复杂环境、密集遮挡等情况下的水鸟均能实现高精度识别与定位,具有较强的鲁棒性.

基于改进YOLOV4-tiny的电力设备红外识别研究

红外检测技术因其具有无需直接接触、带电状态进行检测、检测快速等优点,广泛应用于电力设备过热故障诊断领域。但是,红外图像质量不佳和电力设备的复杂分布等问题,对电力设备故障红外检测过程中的电力设备种类检测精度带来不利影响。为了实现电力设备种类识别的快速准确检测,在YOLOV4-tiny目标检测算法的基础上,提出了适用于电力设备过热故障诊断过程设备种类识别的YOLOV4-tiny目标检测模型。通过旋转矩形框机制替换水平矩形框机制、改进激活函数和采用PAN+FPN加强特征提取网络的方式,来改进原检测模型,使其在检测时更精准更快速。通过多次实验验证,优化后的模型相较于YOLOV4、YOLOV4-tiny两种模型来说,检测速度没有明显提升,但其检测精度提升了1.89%,这为电力设备过热故障红外诊断过程中的设备种类识别研究带来了新的思路。

基于K-means高频局放10 kV避雷器快速带电检测方法

针对视觉观察、红外测温、泄漏工频电流和直流分量等方法无法快速发现10 kV架空线路上的氧化锌避雷器(metal oxide arrester, MOA)内部受潮及绝缘缺陷问题,文中提出基于K-means智能识别缺陷类型的方法及原理,分别制作10 kV MOA内部受潮、阀片裂纹等缺陷实验样品,在无局放升压装置加压至额定电压10 kV的条件下,使用高频电流传感器(high frequency current sensor, HFCT)采集局部放电原始数据,提取特征量,建立对应的缺陷类型数据库;通过在带电运行现场不同测试点测得不同10 kV MOA亚稳态下的13组数据,结果表明该方法对10 kV MOA内部绝缘缺陷、受潮缺陷能够准确识别,验证了基于K-means高频局放10 kV MOA快速带电检测方法的实用性,具有较高的经济效应和社会使用价值。

一起110kV油-油套管均压罩固定板断裂缺陷分析

本文中作者介绍了一起室内运行220kV变压器带电检测时发现中压电缆仓内部超声信号超标缺陷的故障分析及处理过程,发现了故障位置和故障原因,并提出了现场解决方案。

基于脉冲耦合神经网络的Live wire新方法的仿真研究

提出了一种基于脉冲耦合神经网络(PCNN)的Live wire分割新方法,Live wire方法是把图像当作一个连通图,在边上定义一个代价函数,然后通过搜索最短路径来找物体的边界,把用户指定的物体边界上的两点之间的最短路径当作物体的边界。最短路径是Live wire方法的重要步骤,本文在介绍脉冲耦合神经网络的基本工作原理和特性的基础上,将改进的脉冲耦合神经网络算法引入到Live wire边缘检测的算法中,用于最短路径算法的研究。并在改进算法中应用路径封锁和在线训练来提高算法的准确性和应用性。

抛物线边界随机幅值的局部放电图谱虚拟仿真培训系统

为了解决变电站气体绝缘组合电器(gas insulated switchgear,GIS)组合电器、高压开关柜局部放电带电检测技能培训装置成本高、危险性高的问题,以虚拟现实(virtual reality,VR)技术构建变电站虚拟工作环境,以3ds Max软件polygon建模方法建立220 kVGIS、10 kV开关柜设备三维模型。基于真实录波数据(data,DAT)文件的数据解析法,获取典型局部放电缺陷的真实局部放电(phase resolved partial discharge,PRPD)图谱及局部放电脉冲信号(phase resolved pluse sequence,PRPS)图谱数据,形成用于培训的局部放电故障数据。同时基于典型局部放电PRPD故障图谱特征,提出一种抛物线边界包络线的随机幅值呈现方法,通过Unity3D引擎实现PRPD与PRPS图谱三维动态呈现。可以在三维仿真第一视角下,开展变电站间隔巡检、局部放电检测仪器操作、故障图谱判断、故障定位等变电站带电检测技能练习。

低温环境条件下电力设备红外测温缺陷识别方法

输变电系统中采用红外线设备进行运维诊断时要求环境温度大于0℃,但对于东北地区冬季低温环境下带电设备缺少红外诊断相关依据,因此需对典型电气设备/连接件的金属接触点温度与环境温度的关系开展研究。本文在实验室环境下搭建了低温环境红外测试系统,获得了不同环境温度时设备金属接触点温度的变化规律,揭示了3种接触电阻在不同负荷电流、环境温度时对测试温度的影响规律。测试结果表明:当接触电阻小于20μΩ时,环境温度、负载电流对接触点温度梯度的变化几乎无影响;而当接触电阻大于20μΩ时,负载电流的影响显著增加,且0℃可作为温度梯度变化的分界点。基于此在考虑接触电阻下提出了低温红外测温的修正方法,可修正低温环境对红外测温结果的影响,文中的分析及试验结果为低温环境下红外检测时结果的修正提供了实践依据,同时准确判断电气设备接触异常等问题,为高寒地区电气设备运维策略的制定提供重要理论支撑。

1000 kV GIS断路器局放带电检测异常分析与处理措施

文中介绍了一起1000 kV GIS断路器局部放电带电检测异常的详细经过、缺陷处理、原因分析以及后续整改措施。检修人员通过对1000 kV GIS设备开展局部放电带电检测、SF6分解物检测等工作,分析超声波和特高频局部放电异常信号特征,锁定异常信号的位置,判断出GIS内部可能存在轻微故障。然后经过现场检查确认了故障的存在,对故障断路器进行整体更换处理,并将故障断路器保存原样返厂进行分析。最终通过现场开罐检查、返厂解体、异物样品检测等手段找出局部放电异常的原因,本次1000 kV GIS断路器局部放电信号异常的原因为屏蔽罩开裂、紧固螺丝松动导致。屏蔽罩开裂主要原因是屏蔽罩紧固螺钉松动,屏蔽罩受力不均匀,导致屏蔽罩安装孔附近金属疲劳开裂,而螺丝松动的原因为锁固剂涂覆不规范、螺栓紧固力矩不符合要求。后续对同类设备进行整改,优化静侧屏蔽罩结构,增加屏蔽罩的整体强度。在屏蔽罩螺栓连接处增加半圆垫圈,并将屏蔽罩紧固螺丝垫圈更换为外径更大的特殊垫圈,提高屏蔽罩紧固的可靠性。通过对该缺陷的处理和分析,在GIS安装、验收、运维检修等方面制定相应对策防止该类缺陷再次发生。

2019-2022年上海市长宁区退休和生活困难妇女妇科常见病普查结果分析

目的:描述长宁区退休和生活困难妇女妇科常见病检出情况,为妇女病防治提供科学依据。方法:对2019-2022年在上海市长宁区退休和生活困难妇女妇科常见病普查结果进行分析。总人数41416人,年龄40~69岁。普查项目包括:妇科双/三合诊、阴道分泌物、宫颈细胞学、阴道超声、乳房超声和乳腺外科检查。2021年将原有的宫颈刮片调整为宫颈液基薄层细胞学检查(TCT),并新增乳房彩超检查。结果:2019-2022年筛查患病人数15569例,平均检出率为37.59%。妇女常见病前五位:子宫肌瘤、卵巢肿块、宫颈息肉、乳房肿块、急性子宫颈炎。恶性肿瘤共检出86例,检出率为207.65/10万,其中乳腺癌的检出率为193.16/10万;宫颈癌检出率为9.66/10万。自2021年上海市出台妇女常见病普查方法调整政策,乳腺癌及宫颈病变检出率明显提高(93.98/10万比322.10/10万,0.09%比0.42%,均P<0.01)。结论:乳腺癌、宫颈癌仍威胁着女性的健康,应加强落实普查实效。