基于边缘计算的多摄像头视频协同分析方法

为了减少智慧城市场景下多摄像头实时视频数据的处理量，提出了基于机器学习算法的边缘端视频协同分析方法。首先，针对各摄像头检测到的重要目标物体，设计了不同的关键窗口来筛选视频的感兴趣区域，缩减视频数据量并提取其特征。然后，根据提取的数据特征，对不同摄像头视频中的相同目标物体进行标注，并设计了摄像头之间关联程度值的计算策略，用于进一步缩减视频数据量。最后，提出了基于图卷积网络和重识别技术的GC-ReID算法，旨在实现多摄像头视频协同分析。实验结果表明，与现有的视频分析方法相比，所提方法能够有效降低系统时延和提高视频压缩率，并保证较高的准确率。     

大型追踪系统的多摄像头协同

多摄像头间的目标追踪,即利用多摄像头协作完成对目标的监控追踪过程,是大规模智能监控系统中的重要研究方向。针对多摄像头追踪系统对多摄像头协同的实时高效性的需求,本文提出了一种利用多Agent代理多摄像头的分布式计算技术,单Agent层面分析其BDI模型,状态转移模型,多Agent层面分析其在目标追踪过程中完整的通信过程,以及多个摄像头的最佳状态选择。通过开源JADE框架,对基于多Agent的多摄像头追踪系统进行系统搭建,系统实现了多摄像头协同下对目标的持续追踪,实验证实了本文的多Agent系统可有效的实现多摄像头的协同,将通信延迟时间缩小到最短,以便及时更新每个摄像头的状态,从而大大提高了目标追踪的准确率,以及计算资源的利用率,达到对目标进行持续追踪的目的。      

行波管多级降压收集极的计算和实验研究

采用多级降压收集极是提高行波管效率析有效途径之一。本文在ＳＬＡＣ－２６６程序的基础上对一支带有二级降压收集极的国外行波管进行了计算，并且将计算结果同实验结果进行了比较。同时用电子注分析仪对降压收集极区电子注和纵向速度分布进行了实验研究，获得了该二级降压收集极的部分工作曲线，并对实验结果进行了分析讨论。     

基于OV7962的车载全景摄像头设计

为了实现车载摄像头的超广角成像,解决倒车影像系统不能全面照顾周围视角的问题,设计了一款超广角车载摄像头。对该摄像头所采用的图像传感器、超广角成像技术、硬件电路设计及接口EMC防护的应用进行研究。根据当前车厂对摄像头的要求选择了基于美国Omnivision公司的CMOS图像传感器OV7962作为成像芯片。采用一种新的凝视型视场全景成像技术——鱼眼透镜成像,并通过匹配成像芯片的参数计算出镜头的焦距和分辨率。对摄像头硬件电路的设计及接口电磁兼容性(EMC)防护进行测试和改进。实验结果表明:摄像头可实现210°超广角成像,接口EMC防护已通过ISO-7637-2脉冲5b标准测试。满足了对倒车影像系统全面顾及周围视角的要求,另外该车载全景摄像头还具有成本低、性能稳定、分辨率高、夜视效果好等优势。     

基于折反射全景相机的360°车载全景驾驶辅助系统设计

针对目前多摄像头拼接式车载全景系统安装复杂、成本较高的问题,提出了一种基于折反射全景相机的车载障碍物距离估计系统,可估计车辆周围360°障碍物的距离.设计了一种基于同心圆环的标定图案,通过折反射全景相机拍摄标定图案,利用最小二乘法建立地面距离与全景图像上的线段长度的对应关系,实现距离标定.实验结果表明,全景图像经过椭圆...     

迷人的FD-422单端放大器（续1）

要使用好五极管电压放大电路,我们必须要知道五极管在使用上的一些基础知识。在这里,我们先要了解五极管在不同帘栅压工作情况下的特性曲线转换的过程计算方式,这个方法对于发烧友应用五极管作为电压放大电路有莫大的方便,而且非常实用,是发烧友必须掌握的一项基本知识。     

基于CUDA与FAST特征的实时机场全景图像拼接并行算法研究

文章提出了一种基于CUDA与FAST特征的全景图实时生成方法。针对相邻图像的无缝拼接,设计了一种基于动态规划缝合线与三角函数的方法。以贵阳机场全景视频监控系统开发过程中采用的全景拼接技术开发过程中的实验,采用了基于四路摄像头图形的全景拼接构成,由于民航业对图像实时性,准确性要求高,实验通过包括柱面投影,图像特征码抽取和图像耦合,图形梯度变化等进行了同步优化整合,实验成果指出通过本实验构建的算法及平台,有效的在保证图像全景生成实时性的基础上,提升了图象耦合部位的可视性和可用性。     

基于摄像头PTZ控制的人脸识别

提出一种基于PTZ(Pan,Tilt,Zoom)摄像头的人脸识别方法.该方法根据人脸中心位置坐标,计算得到PTZ摄像头的运动参数,控制PTZ摄像头旋转,使人脸中心位置与摄像头的视野中心重合,然后将几何特征和PCA(Principal Component Analysis)提取的特征在特征级进行融合,把融合后的特征作为人脸识别的特征.实验表明,该方法能够减少单一特征的人脸识别的不足,显著提高识别率,且能识别出大范围内的人脸.     

一种基于计算机视觉的低成本室内空间测量系统设计与实现

设计并实现一种基于计算机视觉的室内空间测量系统,通过3个低成本的USB摄像头对室内目标进行图像拍摄,实时测量出目标的空间精确位置。采用张氏平面标定法标定摄像头的内外参数,并针对室内环境特点设计了简单有效的图像处理方法,利用立体视觉定位原理计算出目标点的空间位置信息。实验结果表明：系统测量误差小于15 mm,数据更新率为15 Hz,均可满足室内机器人的定位需要。     

柱面全景图像视图合成中的关键问题研究

图像校正和插值是基于柱面全景图像视图合成中的关键问题。首先分析了极线采样方法在柱面图像校正像素保持率和像素有效率上的优势,通过计算极线的数量、范围及起止位置,有效保留了源图像的信息,减少了以往算法中由于使用单应性映射而导致的弱采样和过采样效应;其次,在新视点插值上,根据柱面全景图像的成像模型,推导了合成视图上像素点的位置和颜色计算公式。最后给出了基于仿真图像和实际图像的视图合成结果。      

Video image mosaic implement based on planar-mirror-based catadioptric system

Planar-mirror-based catadioptric method is one of the most hot topics in recent years. To overcome the disadvantages of the planar-mirror-based catadioptric panoramic camera, described by Nalwa (1996, 2001, 2000), such as the requirement for high-precision optical device designing and the stitching lines in the resulting images, we proposed a planar-mirror-based video image mosaic system with high precision for designing. Firstly, we designed a screw nut on our system, which can be adjusted to locate the viewpoints of the cameras’ mirror images at a single point approximately. It provides a method for those who have difficulties in their designing and manufacturing for high precision. Then, after the image distortion correction and cylinder projection transforms, we can stitch the images to get a wide field of view image by template matching algorithm. Finally, nonlinear weighting fusion is adopted to eliminate the stitching line effectively. The experimental results show that our system has good characteristics such as video rate capture, high resolution, no stitching line and without affection by the depth of field of view.     

全景立体成像技术浅述

大视场场景的立体感知和重现的需求,推动了全景立体成像技术的产生。而折反射全景立体成像具有360°大视场、实时成像、结构和几何计算简单、成本低等特点,近年来发展较快。首先介绍了已有的全景立体成像技术和方法、全景图像映射的三种模型,重点介绍了折反射全景立体成像的基本原理、反射镜类型及其各自特点、图像映射展开原理、典型的折反射全景立体成像装置类型等内容,最后强调了折反射全景立体成像技术中一些关键环节处理方法。     

镜像不变的折反射全向图像与透视图像匹配

该文针对折反射全向摄像机的不同配置可能使全向图像和透视图像存在的镜像关系,而尺度不变特征变换算法并不对镜像保持不变性,提出将透视图像进行一次水平翻转,然后将原图与翻转后的图像分别与全向图像进行匹配,取效果好的匹配作为最终结果,从而实现镜像不变。针对全向图像的环形畸变,提出先将透视图像变换为扇形图像后再进行匹配。实验结果显示,将透视图像变换为扇形图像后再与全向图像进行匹配,匹配点总数增加,而错误匹配点数减少,匹配结果明显好于变换前的匹配结果。     

Scale space analysis and active contours for omnidirectional images.

A new generation of optical devices that generate images covering a larger part of the field of view than conventional cameras, namely catadioptric cameras, is slowly emerging. These omnidirectional images will most probably deeply impact computer vision in the forthcoming years, provided that the necessary algorithmic background stands strong. In this paper, we propose a general framework that helps define various computer vision primitives. We show that geometry, which plays a central role in the formation of omnidirectional images, must be carefully taken into account while performing such simple tasks as smoothing or edge detection. Partial differential equations (PDEs) offer a very versatile tool that is well suited to cope with geometrical constraints. We derive new energy functionals and PDEs for segmenting images obtained from catadioptric cameras and show that they can be implemented robustly using classical finite difference schemes. Various experimental results illustrate the potential of these new methods on both synthetic and natural images.     

Epipolar Geometry of Line Cameras Moving with Constant Velocity and Attitude

Image resampling according to epipolar geometry is an important prerequisite for a variety of photogrammetric tasks. Established procedures for resampling frame images according to epipolar geometry are not suitable for scenes captured by line cameras. In this paper, the mathematical model describing epipolar lines in scenes captured by line cameras moving with constant velocity and attitude is established and analyzed. The choice of this trajectory is motivated by the fact that many line cameras can be assumed to follow such a flight path during the short duration of a scene capture (especially when considering space‐borne imaging platforms). Experimental results from synthetic along‐track and across‐track stereo‐scenes are presented. For these scenes, the deviations of the resulting epipolar lines from straightness, as the camera's angular field of view decreases, are quantified and presented.     

不同结构地基光电探测系统的杂散光抑制

分析了不同结构形式的地基折反式光电探测系统的杂散光特性,讨论了相应的杂散光抑制方法.针对有无中间像面的光学系统,分别给出了两种不同的杂散光抑制方案,并对其中的长波红外系统进行了自身辐射分析.结果表明:对于无中间像面的折反系统,外遮光罩可以更好地抑制大离轴角度的杂散光;对于有中间像面的折反系统,采取光阑组合和设置主镜内遮光罩的方式就可以很好地抑制外部杂散光,在满足自身辐射要求的前提下,有效减少因设置外遮光罩而增加的长度和质量,有利于小型化和轻量化.     

双曲凹面折反射全景成像系统

反射镜是折反射全景成像系统的重要部件之一。针对不同场合,出现了双曲面镜、椭球面镜、抛物面镜等反射镜构成的折反射全景成像系统,但是它们成像畸变大、边缘分辨率低,为了使之适合路面勘测,提出了基于前投影模型,采用双曲凹面镜作为反射镜的方法搭建系统,分析了该系统对水平场景的成像特点,对新系统进行了仿真,并与普通双曲面镜、水平等比例镜的成像系统做了比较。结果表明,通过优化相机透视中心到反射镜几何焦点的距离,可以使得新系统在水平场景的成像上具有较小的畸变。     

Geometry-based distributed scene representation with omnidirectional vision sensors

This paper addresses the problem of efficient representation of scenes captured by distributed omnidirectional vision sensors. We propose a novel geometric model to describe the correlation between different views of a 3-D scene. We first approximate the camera images by sparse expansions over a dictionary of geometric atoms. Since the most important visual features are likely to be equivalently dominant in images from multiple cameras, we model the correlation between corresponding features in different views by local geometric transforms. For the particular case of omnidirectional images, we define the multiview transforms between corresponding features based on shape and epipolar geometry constraints. We apply this geometric framework in the design of a distributed coding scheme with side information, which builds an efficient representation of the scene without communication between cameras. The Wyner-Ziv encoder partitions the dictionary into cosets of dissimilar atoms with respect to shape and position in the image. The joint decoder then determines pairwise correspondences between atoms in the reference image and atoms in the cosets of the Wyner-Ziv image in order to identify the most likely atoms to decode under epipolar geometry constraints. Experiments demonstrate that the proposed method leads to reliable estimation of the geometric transforms between views. In particular, the distributed coding scheme offers similar rate-distortion performance as joint encoding at low bit rate and outperforms methods based on independent decoding of the different images.     

Building 3-D Human Face Models from Two Photographs

This paper presents a working system for building 3-D human face models from two photographs. Rather than using expensive 3-D scanners, we show that frontal face models can be faithfully reconstructed from two photographs taken by consumer digital cameras in a totally non-invasive setup. We first rectify the image pair so that corresponding epipolar lines become coincident, by computing a dual point transformation. We then address the correspondence problem by converting it into a maximal surface extraction problem, which is then solved efficiently. The method effectively removes local extrema. Finally, a Euclidean reconstruction is achieved with the help of a novel factorization method for perspective cameras. Most of the computational steps are conducted in projective space. Euclidean information is introduced only at the last stage. This sets apart our system from the traditional ones which begin with metric information by using carefully calibrated cameras. We have collected a bank of face pairs to test our system, and are satisfied with its performance. Results from this image database are demonstrated.     

薄型折反系统布局及金属主镜安装影响分析

对于含口径80 mm 的高次非球面金属主镜、主镜顶点到像面距离仅为17.5 mm 的薄型折反式光学系统,设计了带3 安装耳的主镜,并利用组件合并及周向同环交错前向安装的方式,在保证所有光机件顺利布局的基础上,有效增加了主镜镜体厚度提高了结构刚度。为分析金属主镜在安装应力下的面形,利用ANSYS 预紧力单元模拟螺钉连接的预紧力,在模态校核的基础上分析了主镜柔性支撑环板在不同厚度以及安装面不平时的波面变形。提出的轴向受限的主镜组件结构方案、安装应力变形的集成分析方法及解决方案,为提高主镜的装配面形精度以保证光学系统的成像质量提供了光机设计参考。