复杂背景下人体检测算法

静态图像的人体识别技术在实际视频监控场景中, 面临背景复杂、图像分辨率低、光照影响和噪声干扰等问题, 这些多变性以及可能发生的遮挡给单一图像中的人体识别技术提出了挑战. 设计和实现一种复杂背景下的人体检测算法, 基于HOG人体检测算法, 使用积分直方图计算HOG特征, 并用级联SVM分类器对样本进行训练. 实验结果表明, 该算法在复杂视频监控场景中进行人体检测比其它人体检测算法具有更高的准确率和更快的检测速度.     

智能化监控系统信息处理研究

智能视觉监视技术在公共保安领域中的应用带来了监控系统的巨大变革。现在可以把人体识别引入到监控系统中,设计能完全替代监控人员的、可用于无人值守情况下的高智能化监控系统。本文详细介绍了人体运动的视觉监视技术的智能化监控系统一般遵从的处理流程,包括运动检测、人体识别和目标跟踪。     

改进YOLOv3的非机动车检测与识别方法

随着交管部门对非机动车监管力度的增强,在道路交通监控视频中检测和识别非机动车将逐渐成为电子交警系统的必备功能。由于非机动车密度大,容易互相遮挡,且在监控视频中所占面积往往较小,容易出现检测定位不准确和漏检等问题。针对非机动车检测定位不准确和漏检问题,基于YOLOv3,提出一种改进的非机动车检测与识别模型,通过设计新的特征融合结构降低非机动车漏检率,使用GIOU损失提高定位准确度。实验结果表明,所提出的改进模型在自建真实复杂场景非机动车数据集上取得了优于YOLOv3的检测结果,将检测的平均检测准确率（mAP）提高了3.6%。     

医院场景下PDA的安全应用

本文旨在探索有效提升PDA(固定使用的手机)应用安全性的方法.通过电子签名、屏幕管理、无线WIFI和VPN(虚拟专用网络)连接管理、基于工作场景的安全检测、网络安全管理、病毒实时监控和查杀等方面讨论,结合一些外部识别工具,打造基于工作全流程的安全监控体系,实现PDA的安全应用.     

基于视觉的人体运动分析综述

目前,人体运动分析已成为计算机视觉研究领域的热点问题,已被广泛地应用于虚拟现实、视频监控、人机接口等不同场合.基于视觉的人体运动分析,主要研究视频场景中人体目标的检测、识别、描述和跟踪,此外还包括人体行为模式的理解等.近些年来,针对这些课题产生了大量的方法和应用,该文对相关的研究做一综述,并给出了本课题所面临的挑战和未...     

基于改进YOLOv3的教室内人物识别

针对教室内人物目标的识别场景,研究了 目标检测算法YOLOv3的改进及测试.通过分析教室内人物目标识别的特点和YOLOv3网络特征提取架构,运用K-means++算法对anchor box进行宽高聚类,增大特征输出层anchor box数量,增加YOLOv3网络架构中104×104尺寸的特征输出层,得到改进后的YOLO...     

基于改进YOLOv3的火灾检测与识别

现阶段火灾频发,需要自动进行火灾的检测与识别,虽然存在温度、烟雾传感器等火灾检测手段,但是检测实时性得不到保证.为了解决这一问题,提出了基于改进YOLOv3的火灾检测与识别的方法.首先构建一个多场景大规模火灾目标检测数据库,对火焰和烟雾区域进行类别和位置的标注,并针对YOLOv3小目标识别性能不足的问题进行了改进.结合深度网络的特征提取能力,将火灾检测与识别形式化为多分类识别和坐标回归问题,得到了不同场景下火焰和烟雾两种特征的检测识别模型.实验表明,本文提出的改进YOLOv3算法对不同拍摄角度、不同光照条件下的火焰和烟雾检测都能得到理想的效果,同时在检测速度上也满足了实时检测的需求.     

基于YOLOV4的车辆检测系统

本文针对多场景下的道路检测算法系统识别类型偏少和识别数量不精准等问题,提出了一种基于YOLOv4算法的轻量级的目标检测方法,模型包含了大部分常见车辆类型。本文方法 MAP为90.0%。采用7:3的分配方法,数据集共2512张图片。拥有较高的车辆识别精度和视频流畅度,同时有效解决了大型道路交通场景下的车辆遮挡位置问题和小距离目标车辆检测时的问题,具有非常好的实际技术应用研究价值。     

基于YOLOv3与MSER的自然场景中文文本检测研究与实现

针对在复杂的自然场景中中文本倾斜、模糊、光照等检测难题,提出一种基于卷积神经网络YOLOv3与最大极值稳定区域MSER的检测方法YOLOv3-M,并针对中文场景分别改善YOLOv3与MSER.通过YOLOv3算法对图像文本区域的矩形坐标进行回归预测,设计一个基于MSER的角度检测方法与之关联,实现倾斜文本行的检测.YOLOv3-M弥补了YOLOv3不能检测倾斜目标的缺点以及MSER检测容易被复杂场景干扰的缺点,并且YOLOv3与MSER都在检测速度上有着优秀的表现.实验结果表明,YOLOv3-M算法准确率达到81.2％,召回率达到69.7％,其检测速度达到45帧/s.     

用于人体检测的YOLOv3改进及压缩算法的研究

针对Yolov3算法应用于人体检测中的准确率低,参数量、计算量和模型体积大难以在资源有限的嵌入式平台上实现等问题,提出了YOLOv3改进及其模型压缩算法.在YOLOv3中通过引入密集连接与多分支结构,增加网络宽度和多尺度感受野,加强特征重用,提高了模型的检测精度;对改进的YOLOv3通过联合优化权重损失函数和BN层缩放...     

基于无线网络的智能监控系统设计与实现

本文设计与实现了运用背景差法的一个基于无线网络的交互的智能监控系统,该系统的基本功能有1)运用背景差的算法检测并跟踪行人,并以文件形式保存下图像,发送该图像到网络服务器上,供用户以Web形式浏览;2)将图像发送到手机终端;3)实现手机与系统交互。实验证明,该系统在复杂的场景中具有良好的性能,且实时性满足要求。     

A model of the relationship between psychological characteristics,mobile phone addiction and use of mobile phones by Taiwanese university female students

While many researches have analyzed the psychological antecedents of mobile phone addiction and mobile phone usage behavior, their relationship with psychological characteristics remains mixed. We investigated the relationship between psychological characteristics, mobile phone addiction and use of mobile phones for 269 Taiwanese female university students who were administered Rosenberg’s self-esteem scale, Lai’s personality inventory, and a mobile phone usage questionnaire and mobile phone addiction scale. The result showing that: (1) social extraversion and anxiety have positive effects on mobile phone addiction, and self-esteem has negative effects on mobile phone addiction. (2) Mobile phone addiction has a positive predictive effect on mobile phone usage behavior. The results of this study identify personal psychological characteristics of Taiwanese female university students which can significantly predict mobile phone addiction; female university students with mobile phone addiction will make more phone calls and send more text messages. These results are discussed and suggestions for future research for school and university students are provided.     

Correlating mobile phone usage and travel behavior – A case study of Harbin,China

Information and communication technologies (ICTs), such as mobile phones and the Internet, are increasingly pervasive in modern society. These technologies provide new resources for spatio-temporal data mining and geographic knowledge discovery. Since the development of ICTs also impacts physical movement of individuals in societies, much of the existing research has focused on examining the correlation between ICT and human mobility. In this paper, we aim to provide a deeper understanding of how usage of mobile phones correlates with individual travel behavior by exploring the correlation between mobile phone call frequencies and three indicators of travel behavior: (1) radius, (2) eccentricity, and (3) entropy. The methodology is applied to a large dataset from Harbin city in China. The statistical analysis indicates a significant correlation between mobile phone usage and all of the three indicators. In addition, we examine and demonstrate how explanatory factors, such as age, gender, social temporal orders and characteristics of the built environment, impact the relationship between mobile phone usage and individual activity behavior.     

Mapping the Patterns of Mobile Phone Usage Among Fishermen in Malaysia

The main attempt of this paper is to identify the patterns of mobile phone usage among fishermen in Pangkor Island in Malaysia. This study is quantitative in nature, where a total of 250 fishermen were selected as the respondents. The findings confirm that although the majority of fishermen do not spend much time and money on making and receiving calls and messaging, the mobile phone is still a crucial tool for them, particularly in its role in enhancing their safety and easing the communication process with their family while conducting their fishing operation. Although the mobile phone has superior functions in marketing, fishermen in Pangkor Island seem to place less importance on it. The findings also show the similarities and differences relevant to the patterns of mobile phone usage in Pangkor Island, and hopefully this can be applied to other fishing areas and further intensify the development in the fisheries industry.     

基于Qtopia的智能手机进程间通信的研究与实现

语音通信、短信收发作为智能手机的基本功能,硬件上由GSM模块提供完备支持,在软件上除了定制应用程序以外,还需要一个后台服务进程对智能手机基本应用事件进行监控及响应.以Qt/Embedded Qtopia作为应用程序开发平台,分析了其内部的进程间通信机制,设计并实现了智能手机基本应用程序的后台服务进程.     

The Ten Shades of Silver: Segmentation of Older Adults in the Mobile Phone Market

With the closing of the age-related gap in mobile phone uptake and the aging of the population, older adults have increasingly become a desirable target group for mobile phone providers. Although there is abundant literature describing the use of mobile phones among older adults, segmentation studies on this group of consumers remain scant and inconclusive. Drawing on the benefit and behavioral segmentation, this study presents a segmentation model of older adults in the mobile phone market that incorporates a wide variety of consumer behavior, attitudinal, and acceptance variables relevant for the understanding of mobile phone (non-)usage patterns among users of feature phones and smartphones, as well as among mobile phone nonusers. The model is based on a hybrid approach with a two-step cluster analysis using data collected by a telephone survey from a representative sample (n = 1,581) of adults aged 55 or older in Slovenia in 2015. The analysis yielded two uniquely profiled clusters within each of the five a priori-determined segments, summing to a total of 10 clusters. The results show that older adults still prefer feature phones. Nevertheless, the mobile phone market has become highly heterogeneous, with clusters of older adults having distinguishable feature phone- and smartphone-use patterns as well as socio-demographic and life-course profiles. Since many of these profiles undermine the notion of older adults as “basic” mobile phone users, our findings represent a solid foundation for scholars and marketers to further investigate their specific needs in relation to more advanced smartphone and mobile internet usage.     

The influence of prototype fidelity and aesthetics of design in usability tests: effects on user behaviour, subjective evaluation and emotion

An empirical study examined the impact of prototype fidelity on user behaviour, subjective user evaluation and emotion. The independent factors of prototype fidelity (paper prototype, computer prototype, fully operational appliance) and aesthetics of design (high vs. moderate) were varied in a between-subjects design. The 60 participants of the experiment were asked to complete two typical tasks of mobile phone usage: sending a text message and suppressing a phone number. Both performance data and a number of subjective measures were recorded. The results suggested that task completion time may be overestimated when a computer prototype is being used. Furthermore, users appeared to compensate for deficiencies in aesthetic design by overrating the aesthetic qualities of reduced fidelity prototypes. Finally, user emotions were more positively affected by the operation of the more attractive mobile phone than by the less appealing one.     

针对3G手机病毒的产业链分析

该文简要介绍了3G手机和手机病毒,然后根据3G手机和病毒的特点及其传播方式进行分析,提出防范3G手机病毒的产业一条链服务,并提出一些改进方法和未来展望。     

嵌入注意力机制的轻量级钢筋检测网络

智慧工地中的设备内存和计算能力有限,在现场的设备上通过目标检测对钢筋进行实时检测具有很大的难度,而且其钢筋检测速度慢、模型部署成本高。针对这些问题,在YOLOv3网络的基础上,提出了一个嵌入注意力机制的轻量级钢筋检测网络RebarNet。首先,利用残差块作为网络的基本单元来构建特征提取结构,并用其提取局部和上下文信息;其次,在残差块中添加通道注意力（CA）模块和空间注意力（SA）模块,以调整特征图的注意力权重,并提升网络提取特征的能力;然后,采用特征金字塔融合模块,以增大网络的感受野,并优化中等钢筋图像的提取效果;最后,输出经过8倍下采样后的52×52通道的特征图用于后处理和钢筋检测。实验结果表明,所提网络的参数量仅为Darknet53网络的5%,在钢筋测试集上以106.8 FPS的速度达到了92.7%的mAP。与现有的EfficientDet、SSD、CenterNet、RetinaNet、Faster RCNN、YOLOv3、YOLOv4和YOLOv5m等8个目标检测网络相比,RebarNet具有更短的训练时间（24.5 s）、最低的显存占用（1 956 MB）、最小的模型权重文件（13 MB）。与目前效果最好的YOLOv5m网络相比,RebarNet的mAP略低0.4个百分点,然而其检测速度上升了48 FPS,是YOLOv5m网络的1.8倍。以上结果表明,所提出的网络有助于完成智慧工地中要求实现的高效、准确的钢筋检测任务。     

手机软板自动检测的图像预处理及分割方法

针对自动光学检测在手机软板缺陷检测中对图像处理的要求比较高的特殊性,研究了手机软板图像的预处理及分割的方法;根据手机软板图像存在的噪声多为椒盐噪声的特点,采用中值滤波的平滑方法;而LOG(Laplacian of Gaussian)的锐化方法既可以实现噪声的控制,又能对边缘进行检测,很适合图像的增强;在分割方法的选择上,根据其灰度分布特性,提出了一种改进的自适应阈值选取方法;实验结果证明了算法的有效性和可靠性,为进一步提高手机软板缺陷检测的准确性提供了保障。