English  |  正體中文  |  简体中文  |  Items with full text/Total items : 889/889 (100%)
Visitors : 14785453      Online Users : 35
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://ccur.lib.ccu.edu.tw/handle/A095B0000Q/338

    Title: 不規則ROI雙鏡頭高動態範圍影像合成;Irregular ROI Dual camera HDR Synthesis
    Authors: 白翊廷;PAI, YI-TING
    Contributors: 電機工程研究所
    Keywords: 特徵點對應;曝光補償;不規則ROI;曝光融合;高動態範圍;ZYNQ-FPGA;ZYNQ-FPGA;brightness compensation;Irregular ROI;Image alignment;Exposure Fusion;High Dynamic Range
    Date: 2016
    Issue Date: 2019-07-17
    Publisher: 電機工程研究所
    Abstract: 傳統的高動態範圍影像合成,是以一系列連續且不同曝光時間的影像來合成,以輻射照度域對應(radiance domain mapping)和曝光融合[1]兩類為主。過去方式大多是假設輸入影像為完美對齊,但實際連續拍攝場景並無法完美對齊。我們提出了針對不規則ROI區域的高動態範圍影像合成技術,以Dual camera同時設置兩張來源影像[2],分別為自動曝光(automatic exposure)與補償曝光(complimentary exposure) 兩張不同曝光與視角的影像。過去的影像對齊方式使用尺度不變特徵轉換(Scale-invariant feature transform),在相同曝光時間影像中找到特徵點並加以對應轉換,但在對應曝光時間差距較大影像會難以找到對應特徵點,導致對齊困難。我們提出區域亮度分部一致性來尋找對應特徵區域,解決以往全域搜尋對齊時間較長的問題,再以細部pixel劃分出ROI邊緣,準確勾勒ROI外形。在針對亮度高反差場景的情況,本篇論文題出了區域曝光補償的方法,增加ROI邊緣與背景的亮度色彩連續性,解決高動態範圍合成時ROI邊緣亮暗色彩不連續之情形,以達成高動態範圍影像合成。最後提出了在ZYNQ-FPGA板上,以linux系統為基礎,透過DMA資料搬移的軟硬體溝通環境模型,以供日後將演算法在FPGA上實現。
    Traditional high dynamic range image synthesis methods are acquired from a sequence of differently exposed images. Most of them are set with static scenes and assume the all input images are perfectly aligned. However, these methods are not effective to cope with practical scenario that the misaligned situation occur.In this paper, we present an effective HDR image synthesis technique for irregular ROI form two different exposed images, one for automatic-exposed and the other is complementary-exposed. For reducing the images capture timing, we make the images captured by dual camera in the same time, so that we can make sure the scenes be static in the least capturer timing. It becomes the problems that derived from the different viewpoint and different exposure timing of two image. We propose a matching method that overcomes those problems, not only match the feature region in two misaligned images but label the ROI edge shape by pixel, which become more faithful to the over exposured region. To over come the scenes within high contrast brightness, we propose the Brightness Compensation Image Synthesis, coping with the color discontinuous and brightness discontinuous around the ROI. In the last chapter we propose the environment on ZYNQ-FPGA,useing DMA with user-space. After understanding the AXI bus in Zedboard, We can conveniently transfer data between the ARM, PL and DDR so that we can realize our algorithm in the future.
    Appears in Collections:[電機工程研究所] 學位論文

    Files in This Item:

    File Description SizeFormat

    All items in CCUR are protected by copyright, with all rights reserved.

    版權聲明 © 國立中正大學圖書館網頁內容著作權屬國立中正大學圖書館


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback