Single-Phase Velocity Determination Based in Video and Sub-Images Processing:An Optical Flow Method Implemented with Support of a Programmed MatLab Structured Script

Important in many different sectors of the industry, the determination of stream velocity has become more and more important due to measurements precision necessity, in order to determine the right production rates, determine the volumetric production of undesired fluid, establish automated controls based on these measurements avoiding over-flooding or over-production, guaranteeing accurate predictive maintenance, etc. Difficulties being faced have been the determination of the velocity of specific fluids embedded in some others, for example, determining the gas bubbles stream velocity flowing throughout liquid fluid phase. Although different and already applicable methods have been researched and already implemented within the industry, a non-intrusive automated way of providing those stream velocities has its importance, and may have a huge impact in projects budget. Knowing the importance of its determination, this developed script uses a methodology of breaking-down real-time videos media into frame images, analyzing by pixel correlations possible superposition matches for further gas bubbles stream velocity estimation. In raw sense, the script bases itself in functions and procedures already available in MatLab, which can be used for image processing and treatments, allowing the methodology to be implemented. Its accuracy after the running test was of around 97% (ninety-seven percent);the raw source code with comments had almost 3000 (three thousand) characters;and the hardware placed for running the code was an Intel Core Duo 2.13 [Ghz] and 2 [Gb] RAM memory capable workstation. Even showing good results, it could be stated that just the end point correlations were actually getting to the final solution. So that, making use of self-learning functions or neural network, one could surely enhance the capability of the application to be run in real-time without getting exhaust by iterative loops.

On use of the alpha stable self-similar stochastic process to model aggregated VBR video traffic

The alpha stable self-similar stochastic process has been proved an effective model for high variable data traffic. A deep insight into some special issues and considerations on use of the process to model aggregated VBR video traffic is made. Different methods to estimate stability parameter a and self-similar parameter H are compared. Processes to generate the linear fractional stable noise （LFSN） and the alpha stable random variables are provided. Model construction and the quantitative comparisons with fractional Brown motion （FBM） and real traffic are also examined. Open problems and future directions are also given with thoughtful discussions.

Realtime Object Detection Through M-ResNet in Video Surveillance System

Object detection plays a vital role in the video surveillance systems.To enhance security,surveillance cameras are now installed in public areas such as traffic signals,roadways,retail malls,train stations,and banks.However,monitor-ing the video continually at a quicker pace is a challenging job.As a consequence,security cameras are useless and need human monitoring.The primary difficulty with video surveillance is identifying abnormalities such as thefts,accidents,crimes,or other unlawful actions.The anomalous action does not occur at a high-er rate than usual occurrences.To detect the object in a video,first we analyze the images pixel by pixel.In digital image processing,segmentation is the process of segregating the individual image parts into pixels.The performance of segmenta-tion is affected by irregular illumination and/or low illumination.These factors highly affect the real-time object detection process in the video surveillance sys-tem.In this paper,a modified ResNet model(M-Resnet)is proposed to enhance the image which is affected by insufficient light.Experimental results provide the comparison of existing method output and modification architecture of the ResNet model shows the considerable amount improvement in detection objects in the video stream.The proposed model shows better results in the metrics like preci-sion,recall,pixel accuracy,etc.,andfinds a reasonable improvement in the object detection.

An Efficient Method for Underwater Video Summarization and Object Detection Using YoLoV3

Currently,worldwide industries and communities are concerned with building,expanding,and exploring the assets and resources found in the oceans and seas.More precisely,to analyze a stock,archaeology,and surveillance,sev-eral cameras are installed underseas to collect videos.However,on the other hand,these large size videos require a lot of time and memory for their processing to extract relevant information.Hence,to automate this manual procedure of video assessment,an accurate and efficient automated system is a greater necessity.From this perspective,we intend to present a complete framework solution for the task of video summarization and object detection in underwater videos.We employed a perceived motion energy(PME)method tofirst extract the keyframes followed by an object detection model approach namely YoloV3 to perform object detection in underwater videos.The issues of blurriness and low contrast in underwater images are also taken into account in the presented approach by applying the image enhancement method.Furthermore,the suggested framework of underwater video summarization and object detection has been evaluated on a publicly available brackish dataset.It is observed that the proposed framework shows good performance and hence ultimately assists several marine researchers or scientists related to thefield of underwater archaeology,stock assessment,and surveillance.

Deeply‐Recursive Attention Network for video steganography

Video steganography plays an important role in secret communication that conceals a secret video in a cover video by perturbing the value of pixels in the cover frames.Imperceptibility is the first and foremost requirement of any steganographic approach.Inspired by the fact that human eyes perceive pixel perturbation differently in different video areas,a novel effective and efficient Deeply‐Recursive Attention Network(DRANet)for video steganography to find suitable areas for information hiding via modelling spatio‐temporal attention is proposed.The DRANet mainly contains two important components,a Non‐Local Self‐Attention(NLSA)block and a Non‐Local Co‐Attention(NLCA)block.Specifically,the NLSA block can select the cover frame areas which are suitable for hiding by computing the correlations among inter‐and intra‐cover frames.The NLCA block aims to effectively produce the enhanced representations of the secret frames to enhance the robustness of the model and alleviate the influence of different areas in the secret video.Furthermore,the DRANet reduces the model parameters by performing similar operations on the different frames within an input video recursively.Experimental results show the proposed DRANet achieves better performance with fewer parameters than the state‐of‐the‐art competitors.

智能行车记录仪图像去雾系统的FPGA设计

雾霾天气下,交通道路能见度低,导致所采集到的视频画面退化、图像信息模糊,同时考虑传统系统处理实时性不高等问题,基于ZYNQ平台设计了一种图像去雾系统,并应用于智能行车记录仪系统中.首先,针对传统暗通道去雾算法在天空区域存在失真等问题,提出了一种分割天空区域的策略来修正图像复原参数;然后,针对计算全局大气光值时,需对整幅图像的像素排序消耗大量资源的问题,利用现场可编程门阵列(FPGA)并行运算的优势,提出一种帧迭代方法优化求取大气光值,同时优化了引导滤波的硬件设计;最后,将双路高清多媒体接口(HDMI)资源中,一路作为视频输入,另一路作为视频处理输出,搭建实时交通图像视频处理试验平台.试验结果表明,系统针对雾霾天气下的交通视频具有较好的去雾效果,尤其是可以解决天空区域去雾的失真问题.在对分辨率为1280像素×720像素的交通视频去雾时,可以达到30帧/s的处理速度,满足实时性要求.

快速视频去雾改进算法的FPGA实现

内窥镜去雾算法在医疗领域具有广泛应用,为临床医生提供清晰、实时的图像。去雾技术虽然已经取得较大的进步,但去雾算法的复杂度较高,在内窥镜等复杂情况下硬件实现较为困难。为了在硬件上实现内窥镜实时去雾效果,对暗通道先验算法进行改进,降低硬件资源消耗和时间复杂度。该改进算法选择适合硬件的大气光照强度估计值、透射率补偿值以及采用流水线结构实现有雾图像的处理。采用Xilinx的ZYNQ7020实现该算法硬件电路,实时处理分辨率为640×480的视频图像,速度可达到260 fps,消耗LUT仅为1.28 K,寄存器619个单元。实验结果表明,相比于传统算法,改进算法具有处理速度快、功耗低、可移植性强的特点,满足内窥镜需要实时处理视频的要求。

基于AMP架构的青霉素结晶与发酵检测系统设计

目前,在青霉素工业生产中,人工检测是青霉素结晶状态以及发酵液高度检测的主要方式,这种方式成本高、自动化水平低且易造成误判,对青霉素产量和品质有很大影响。针对此种情况,文中设计了一种基于AMP架构的青霉素结晶状态与发酵液高度检测系统。该系统以ZYNQ-7000作为可重构嵌入式平台。在PL部分,利用Vivado HLS工具实现结晶状态、发酵液高度检测的图像预处理算法IP核设计,并搭建出视觉传感系统的硬件平台。在PS部分,系统采用双核AMP(非对称多处理)运行模式,完成结晶状态检测、发酵液高度检测算法处理与以太网视频传输、图像存储功能的并行处理。实验结果表明:该系统支持2种检测模式的切换,可实现对工业生产中青霉素结晶状态与发酵液高度的精确检测与视频数据的实时传输。

婴幼儿头皮医用粘胶相关性皮肤损伤防护方案的构建与应用

目的降低视频脑电图监测婴幼儿头皮医用粘胶相关性皮肤损伤。方法采用历史对照设计,将2018年1月至2019年12月入住癫痫监测单元行头皮视频脑电图监测的患儿856例纳入对照组,实施常规皮肤护理;将2021年1月至2022年12月入住癫痫监测单元行头皮视频脑电图监测的患儿851例纳入观察组,构建和实施医用粘胶相关性皮肤损伤防护方案。比较两组头皮医用粘胶相关性皮肤损伤发生率。结果观察组头皮医用粘胶相关性皮肤损伤发生率显著低于对照组(P<0.05)。结论婴幼儿头皮医用粘胶相关性皮肤损伤防护方案的应用,可有效降低婴幼儿头皮医用粘胶相关性皮肤损伤发生率。

视频侦查技术关键及其发展展望

视频侦查技术的核心是通过视频图像的提取、查看、分析和研判来获取侦查线索、固定涉案证据。多年来经过在实战应用中不断地发展创新,视频侦查技术形成了自己独特的关键技术体系:视频信息分析解读技术是侦查应用和证据固定的基础和核心;视频证据固定保全技术的规范是审判中心主义的客观要求,可以获取视频侦查记录报告、视频检验鉴定报告或视频数据关联报告;低质量视频图像的增强恢复技术专业性强,应用范围窄,技术成熟度高,然而它不断面临新的挑战。目前,视频数据的智能应用在大数据背景下变得越来越重要,仍需进一步突破视频自动识别技术的应用范畴,建立完善多层次的视频数据综合应用体系,打造适应不同业务需要的视频数据实战应用模型。

基于注意力机制的短视频助力导论类课程教学改革研究——以人工智能导论课程为例

以人工智能导论为代表的导论类课程具有课时少、知识点多、算法难、学生基础差异大等特点,给教学带来诸多难点,如难以平衡深度和广度,难以持续吸引注意力等。利用短视频的多形式表达、多通道传输、高密度信息等优点可以解决导论类课程教学难题。该文在剖析课堂注意力特点和短视频吸引注意力机制的基础上,运行注意分配的认知理论,提出一套基于注意力机制的短视频助力导论类课程的教学方法,可为导论类教学工作者提供参考。

基于深度学习的监控视频图像增强处理方法

传统的图像增强处理方法常存在图像清晰度较低、失真度较高的问题。因此,提出基于深度学习的增强处理方法。在获取监控视频图像信息后,对图像进行序列对齐处理。通过图像量化处理的方式显示单个图像中的关键像素点和邻近像素点之间的分布。采用深度学习技术,用8个方向的梯度算子提取反射图像的方向梯度,得到细节增强的反射图像,再通过Retinex反变换完成增强处理。设计对比实验,证明该方法可以改善监视视频的图像增强效果,获得更清楚的景物图片。

新工科背景下数字图像与视频处理课程混合式教学的探索与实践

新工科发展背景下数字图像与视频处理课程传统单一的教学模式和教学方法已无法有效地适应新时代高素质人才培养要求,新工科要求课程的教学内容、教学模式和考核方式等要素不断更新。为改进陈旧的教学模式、激发学生的学习主动性、提升学生的创新思维及编程实践能力,提出通过开展线上线下混合式教学、线下课堂实施六步教学法、课程资源建设、融合课程思政、构建验证性基础实验+综合实验+创新性项目实践的阶梯型实践教学体系、改进教学评价方式等多种改进措施,深化课程教学改革,提高了课程教学质量,取得了良好的教学效果。

基于USB Video Camera的自动对焦系统

介绍一种应用于USB video camera中的自动对焦系统。由USB video camera获取的视频图像经计算机进行FFT运算或微分运算,得到其频谱幅值数据或微分幅值数据,计算机根据所得数据判断USB video camera中的镜头是否处于离焦位置并控制电机将镜头移到对焦位置。文章还进一步讨论了提高自动对焦准确度的措施。实验结果表明该自动对焦系统能很好地实现USB video camera的自动对焦,该系统将使具有USB接口的video camera使用更简单方便。

基于3D卷积视频分析的猪步态评分方法

【目的】猪肢蹄病是种猪淘汰的重要原因之一,给养殖场带来巨大的经济损失。猪蹄疾病判断通常依赖人工肉眼观察猪只步态进行排查,存在效率低、人力成本高等问题。本文旨在实现自动化猪步态评分,高效判断猪只肢蹄健康状况。【方法】本文提出一种“端到端”的猪步态评分方法,在单头种猪经过测定通道时采集视频,并制作四分制步态数据集。采用深度学习技术分析视频,设计了一种基于3D卷积网络的时间注意力模块(Time attention module,TAM),有效提取视频帧图像之间的特征信息。将TAM与残差结构结合,构建猪步态评分模型TA3D,对步态视频进行特征提取与步态分类评分。为进一步提升模型性能并实现自动化处理,本文设计了步态关注模块(Gait focus module,GFM),能够自动从实时视频流中提取有效信息并合成高质量步态视频,在提高模型性能的同时降低计算成本。【结果】试验结果表明,GFM可以实时运行,步态视频大小可以减少90%以上,显著降低存储成本,TA3D模型步态评分准确率达到96.43%。与其他经典的视频分析模型的对比测试结果表明,TA3D的准确率和推理速度均达到最佳水平。【结论】本文提出的方案可应用于猪只步态自动评分,为猪肢蹄病的自动检测提供参考。

基于VideoCharger视频点播系统的设计与实现

探讨了VideoCharger的逻辑组成、交互与配置方式 ,及详细的资源需求分析 ,给出了据此建立VOD系统的系列解决方案。

模拟混合视频处理电路单粒子效应测试系统设计

为开展模拟混合视频处理电路的单粒子效应研究,通过改变伪CCD信号源的输出信号,建立一款模拟混合视频处理电路的单粒子效应测试系统,实现该器件的单粒子闭锁、单粒子翻转与单粒子功能中断等单粒子效应的评估。通过在中科院近物所回旋加速器和中国原子能院串列加速器的试验测试,获取有效数据。试验结果表明该试验方法及测试系统可以有效评估器件的单粒子性能,为器件的加固设计提供有力数据支撑。

基于双向特征金字塔的密集视频描述生成方法

针对当前方法普遍存在较为严重的细节结构信息丢失与事件间重叠的问题,提出一种基于双向特征金字塔的密集视频描述生成方法(dense video captioning with bilateral feature pyramid net,BFPVC)。BFPVC通过带有自底向上、自顶向下、横向链接3条分支的双向特征金字塔强化视频多尺度特征图,兼顾对时序信息、空间信息、语义信息的特征表示,解码器从强化后的视频特征中捕获更加全面的事件候选集,从而为对应的视频事件生成更加丰富、详尽的文本描述。在ActivityNet Captions数据集和YouCook2数据集上的实验结果表明,BFPVC与同类模型相比生成的文本描述更详细、丰富,验证了双向特征金字塔在密集视频描述领域的有效性。

多种分辨率HDMI接口的TMDS信号转为LVDS格式实现方法研究

为了将图像仿真系统输出的高清多媒体接口(High Definition Multimedia Interface,HDMI)最小化传输差分信号(Transition Minimized Differential Signaling,TMDS)转换为摄像头模组接口低电压差分信号(Low-Voltage Differential Signaling,LVDS),文章联合使用ITE68051芯片和现场可编程逻辑门阵列芯片(Field Programmable Gate Array,FPGA)实现了相应的转换。首先,将高清、超高清HDMI接口的TMDS信号通过ITE68051视频处理芯片进行TMDS解码,以双端口方式输出并行RGB数字信号;然后,应用FPGA易编程性的特点,将多种分辨率的RGB信号转换为相应的LVDS信号。联合应用ITE68051和FPGA芯片架构的转换系统,便于视频设备之间的兼容和系统后续升级,为自动驾驶仿真系统的视频信号输入提供了可靠的硬件接口通道。

视觉传感器提取面部运动特征的抑郁症检测算法研究

尽管抑郁症自动诊断系统已经取得了重大进展,但大部分工作集中在结合多种模态的特征来提高分类精度,这会产生大量的时空开销和特征同步问题。提出了一种基于面部表情和面部运动特征的单模态抑郁症检测框架。提出了一种基于面部标志比的鲁棒特征提取方法,并从理论上证明了该特征具有上下、左右平移、深度平移、旋转和翻转不变性。基于该方法提取的特征保持了面部标志点在空间上的拓扑结构关系,并保持了面部标志点前后帧的时间相关性。然后,提出了一种新颖的思路来解决大单元抑郁症视频的分类任务,将大单元视频的抑郁症分类任务分解为多个短序列单元的评分任务,然后通过定义的评分聚合函数得到最终的抑郁症分类结果。在DAIC-WOZ数据集上,所提出的检测框架提高了分类性能,F1得分为0.85,优于当前其他基于单模态的抑郁症检测模型。