本文選題：長(zhǎng)波紅外視頻 + DM3730�。� 參考：《天津大學(xué)》2014年碩士論文

【摘要】：紅外熱成像技術(shù)具有被動(dòng)成像、夜間成像、隱蔽性好等優(yōu)點(diǎn),在軍事和民用領(lǐng)域得到廣泛的應(yīng)用。其中嵌入式紅外相機(jī)因其體積小、重量輕、便于戶外應(yīng)用等優(yōu)勢(shì),市場(chǎng)需求越來越大。傳統(tǒng)的紅外相機(jī)以通用DSP或者FPGA為主控進(jìn)行設(shè)計(jì)。兩種方案共有的問題就是集成度低,軟件系統(tǒng)簡(jiǎn)單,一般為裸機(jī)或者基本輸入輸出系統(tǒng)(BIOS),不能提供后期軟件開發(fā)的平臺(tái),用戶友好性較差。本課題針對(duì)目前傳統(tǒng)紅外相機(jī)具有的問題,創(chuàng)新性的選擇了德州儀器生產(chǎn)的低功耗移動(dòng)處理芯片DM3730作為主控。DM3730集成了圖像信號(hào)處理器(ISP)作為視頻采集子系統(tǒng),集成了一個(gè)ARM Cortex-A8核心和一個(gè)C64X+DSP核心,還集成了顯示子系統(tǒng)(DSS)進(jìn)行LCD或TV顯示。極高的集成度使該平臺(tái)芯片使用量減少,減小了電路板尺寸,提高了可靠性。與此同時(shí),DM3730作為一款移動(dòng)處理器,具有功耗控制模塊,降低了功耗。圍繞該平臺(tái),本文開展了以下幾個(gè)方面的工作:首先是提取國(guó)外紅外相機(jī)的原始數(shù)據(jù),并將數(shù)據(jù)通過CPLD轉(zhuǎn)換為12位寬同步模式的視頻信號(hào),以適配DM3730 ISP接口的12位寬采集模式;其次,完成了嵌入式系統(tǒng)在相機(jī)上的移植,并且深入研究了基于V4L2的驅(qū)動(dòng)模型,設(shè)計(jì)了紅外相機(jī)驅(qū)動(dòng)和ISP驅(qū)動(dòng),實(shí)現(xiàn)將12位寬的灰度格式視頻采集到內(nèi)存中;然后,研究ARM內(nèi)核和DSP的內(nèi)核交互的Codec engine機(jī)制,尋找便于實(shí)現(xiàn)的紅外圖像處理方法,并調(diào)用DSP內(nèi)核對(duì)采集的圖像進(jìn)行處理;最后,研究DM3730的DSS顯示功能,實(shí)現(xiàn)將視頻以0.5x~8x的縮放顯示。本設(shè)計(jì)在硬件上,芯片使用量少、集成度高,同時(shí)雙核結(jié)構(gòu)使其性能強(qiáng)大;在軟件上,驅(qū)動(dòng)的使用使相機(jī)在靈活性方面更勝一籌,Linux操作系統(tǒng)使其在后期軟件開發(fā)更具優(yōu)勢(shì);在圖像處理上,調(diào)用DSP進(jìn)行處理,Codec engine框架使相機(jī)可以集成多種算法。實(shí)驗(yàn)結(jié)果顯示,本相機(jī)可正確進(jìn)行12位寬灰度格式的視頻采集、調(diào)用DSP算法進(jìn)行處理、用DSS進(jìn)行縮放顯示。同時(shí),非均勻性校正算法進(jìn)行校正后,圖像質(zhì)量有了極大提高,證明了算法的有效性。
[Abstract]:Infrared thermal imaging technology has been widely used in military and civil fields because of its advantages of passive imaging, night imaging and good concealment. The market demand of embedded infrared camera is more and more because of its small size, light weight, convenient outdoor application and so on. The traditional infrared camera is designed with general DSP or FPGA as the main control. The common problem of the two schemes is that the integration level is low, the software system is simple, it is generally bare machine or basic input and output system, it can not provide the platform of software development in the later stage, and the user friendliness is poor. Aiming at the problems of traditional infrared camera, this paper innovatively selects the low power mobile processing chip (DM3730) produced by Texas Instruments as the main control. DM3730 integrates the image signal processor (ISP) as the video acquisition subsystem. A ARM Cortex-A8 core and a C64X DSP core are integrated, and the display subsystem is integrated for LCD or TV display. The high level of integration reduces the usage of the chip, reduces the size of the circuit board and improves the reliability. At the same time, as a mobile processor, DM3730 has a power control module, which reduces power consumption. Around the platform, this paper has carried out the following work: firstly, the original data of foreign infrared camera is extracted, and the data is converted into 12-bit wide synchronous video signal through CPLD to adapt the 12-bit wide acquisition mode of DM3730 ISP interface; Secondly, the transplant of embedded system on camera is completed, and the driving model based on V4L2 is deeply studied. The infrared camera driver and ISP driver are designed to capture 12 bit wide gray-scale video into memory. This paper studies the Codec engine mechanism of the interaction between the ARM kernel and the DSP kernel, looks for the convenient infrared image processing method, and calls the DSP kernel to process the collected image. Finally, the DSS display function of DM3730 is studied to realize the video display by 0.5x~8x scaling. In the hardware, the chip is used less, the integration is high, and the dual-core structure makes its performance strong, in software, the use of driving makes the camera more flexible than the Linux operating system, so that it has more advantages in the later software development. In image processing, DSP is called to process the frame of Codec engine so that the camera can integrate a variety of algorithms. The experimental results show that the camera can correctly capture video with 12 bits wide grayscale format, call the DSP algorithm to process, and use DSS to scale and display. At the same time, the image quality has been greatly improved after the correction of the nonuniformity correction algorithm, which proves the effectiveness of the algorithm.
【學(xué)位授予單位】：天津大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：O434.3;TB852.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王厚;顧國(guó)華;錢惟賢;陳錢;徐富元;;高速紅外視頻處理系統(tǒng)的設(shè)計(jì)研究[J];紅外技術(shù);2013年07期

2 代少升;張強(qiáng);;基于伽馬壓縮的紅外圖像非均勻性校正算法研究[J];半導(dǎo)體光電;2013年02期

3 呂士欽;朱曉軍;張剛;;基于OMAP3530的SIP視頻監(jiān)控客戶端設(shè)計(jì)與實(shí)現(xiàn)[J];計(jì)算機(jī)工程與應(yīng)用;2012年20期

4 呂雷;張學(xué)峰;;基于FPGA的紅外圖像實(shí)時(shí)非均勻性校正[J];激光與紅外;2011年06期

5 張輝;李新華;劉波;錢翔;;基于V4L2的視頻設(shè)備驅(qū)動(dòng)開發(fā)與移植[J];電腦知識(shí)與技術(shù);2010年15期

6 韓慧英;潘婭;;基于Davinci的嵌入式Web視頻監(jiān)控系統(tǒng)[J];兵工自動(dòng)化;2010年04期

7 賴睿;楊銀堂;王炳健;周慧鑫;劉上乾;;一種基于S曲線模型的紅外焦平面陣列非均勻性自適應(yīng)校正算法[J];光學(xué)學(xué)報(bào);2009年04期

8 劉爽;馮林;趙凱生;鐘智勇;;一種紅外圖像非均勻性校正新方法的DSP實(shí)現(xiàn)[J];儀器儀表學(xué)報(bào);2008年05期

9 潘小冬;陳澤祥;高升久;黃自力;;FPGA+DSP的紅外圖像數(shù)據(jù)采集與顯示[J];紅外與激光工程;2007年06期

10 胡明鵬;李宏壯;馬冬梅;;凝視型紅外光電系統(tǒng)噪聲等效溫差測(cè)量[J];光電工程;2007年08期

相關(guān)博士學(xué)位論文 前2條

1 楊鶴猛;遠(yuǎn)紅外實(shí)時(shí)成像樣機(jī)系統(tǒng)關(guān)鍵技術(shù)研究[D];天津大學(xué);2012年

2 寇小明;紅外成像觀測(cè)系統(tǒng)性能評(píng)價(jià)方法研究[D];西安電子科技大學(xué);2011年

相關(guān)碩士學(xué)位論文 前7條

1 杜一穎;640×512元非制冷紅外焦平面讀出電路的研究與設(shè)計(jì)[D];電子科技大學(xué);2013年

2 張寧;基于三維噪聲的非制冷紅外探測(cè)器測(cè)試系統(tǒng)的研究[D];電子科技大學(xué);2012年

3 劉柳;基于大容量NAND閃存文件系統(tǒng)關(guān)鍵技術(shù)研究[D];浙江工業(yè)大學(xué);2011年

4 袁懿;紅外圖像校正算法研究[D];北方工業(yè)大學(xué);2010年

5 許燕文;基于DSP的便攜式電子鼻系統(tǒng)的研究[D];西安理工大學(xué);2008年

6 趙巖;紅外焦平面陣列非均勻校正算法研究[D];國(guó)防科學(xué)技術(shù)大學(xué);2007年

7 爐慶洪;基于紅外熱成像技術(shù)的健康青年體表溫度分布特征[D];福建師范大學(xué);2006年

，

本文編號(hào)：1795748