节点文献

基于异面直线夹角激光检测系统的智能相机研究

Research of the Smart Camera Based on Laser System for Measuring the Angle Formed by Non-uniplanar Lines

【作者】 陈波

【导师】 裘祖荣;

【作者基本信息】 天津大学 , 仪器科学与技术, 2012, 硕士

【摘要】 大尺寸空间异面直线夹角的测量在现代工业及国防现代化建设中具有重要的现实意义。要在数米甚至数十米的范围内对两条空间异面直线的夹角实现直接测量是比较困难的,通过线结构激光可以为两者建立可靠的公共基准,从而将两个独立的测量坐标系准确地联系起来。当前的实现方法是通过CCD摄像头采集图像,用上位机机进行处理来实现的。本文在上述的测量方案基础上进行了处理系统微型化的研究,提出了应用于该领域的智能相机设计方案,主要工作及具体内容如下:1、通过分析比较多种智能相机设计方案,确立了以DSP+FPGA为核心的硬件结构,选取了型号为EP4CE30F23C8N的FPGA和型号为TMS320DM642的DSP作为系统核心处理芯片,并搭建了外围电路。2、遵循高速PCB的绘制原则,进行了PCB绘制,并对系统内关键信号的信号进行了信号完整性分析。3、结合系统的实际使用环境,确定了以SOBEL边缘检测算子为核心的图像预处理过程,分析了BT.656的数据格式,对该类视频信号的有效信号提取及其格式整合进行了系统的研究,并利用Verilog HDL语言实现了预处理功能。4、对IIC的配置过程进行了研究,并分别利用FPGA和DSP实现了对视频解码芯片和视频编码芯片的IIC配置。5、提出了一种夹角计算的核心算法,并通过DSP编程实现。6、学习了TCP/IP的网络传输协议,完成了DSP同上位机的以太网通讯。在系统调试部分,搭建了简易的实验平台,阐述了实验过程,给出了实验结果,并对实验结果进行了分析,给出了评价。实验表明,该系统体积小,处理速度高,能够对大小为720×576的图像进行25f/s的实时采集处理,并可以利用显示器显示边缘提取后的图像,通过以太网传输实现处理数据的显示,各方面均能够达到设计指标。

【Abstract】 In the modern industry and the modernization of national defense, the anglemeasurement of non-uniplanar lines in large-scale space is of important practicalsignificance. It’s quite difficult to measure the space angle of two non—uniplanarlines with a distance of several even tens of meters directly. To do so, line structuredlaser was used to build a common reference for both of the lines, making the twoindependent measuring-coordinate systems connected to each other accurately. Atpresent, a CCD camera was use to complete the image acquisition, and an uppercomputer was used to finish the data processing.Based on the method above, the micromation of the processing system wasstudied in this thesis. A new approach with smart camera was proposed, applied in thisfield. What have been done are as follows:1. After several schemes with smart camera were analyzed and compared to eachother, the structural scheme using DSP+FPGA was chosen. EP4CE30F23C8N, aFPGA chip, and TMS320DM642, a DSP chip, was selected to be the kernelprocessor chips of the system, and the peripheral circuit was established.2. The design of the specialized PCB was finished, abiding by the principles ofhigh-speed PCB’s design. And the integrity of the key signals in the system wasanalyzed.3. Taking the real working environment of the system into account, the algorithmadopting SOBEL was used to deal with edge detection in the pretreatment of theimages. After the data format of BT.656was analyzed, the extraction of the valid partof the video signal in this format and its format integration was studied. Verilog HDLwas used to achieve the pretreatment.4. The procedure of IIC’s configuration was studied, and then the IICconfiguration of the video decoder chips was implemented in FPGA, and the videoencoder chip’s in DSP.5. A new algorithm applying to the calculation of the angle between two lines wasput forward, and was implemented in DSP.6. The TCP/IP protocols were used to realize the communication between DSPand the upper computer. In the sector of system debugging, a simple experimental platform was set up,and how the experiments were carried out was given. The results of those experimentswere presented, and were also analyzed and evaluated. The experiments showed thatthe system with high processing speed was small in size, and could acquire images inthe size of720*576and processed them in real time at the rate of25f/s. After edgedetection, the images could be displayed via the monitor, and the processing datacould be displayed in real time through the Ethernet. All these proved that the systemhad met the design indexes.

【关键词】 智能相机FPGADSP边缘检测夹角测量
【Key words】 smart cameraFPGADSPedge detectionangle measurement
  • 【网络出版投稿人】 天津大学
  • 【网络出版年期】2014年 08期
  • 【分类号】TN249;TP274;TP391.41
  • 【被引频次】4
  • 【下载频次】87
  • 攻读期成果
节点文献中: 

本文链接的文献网络图示:

本文的引文网络