现代显示
現代顯示
현대현시
ADVANCED DISPLAY
2012年
8期
36-42
,共7页
深度图像%提升算法%小波变换%现场可编程门阵列(FPGA)
深度圖像%提升算法%小波變換%現場可編程門陣列(FPGA)
심도도상%제승산법%소파변환%현장가편정문진렬(FPGA)
depth image%lifting arithmetic%wavelet transform%FPGA
由2D图像转换到3D图像,关键是从原始已知的2D图像(images of limited depth of field(DOF))中生成深度图像。文章提出的是如何在FPGA上实时的对图像进行小波分析,以便后续利用小波分析的结果得到深度图像。与传统的卷积方法相比,该硬件框架提升了运算速度,简化了结构。
由2D圖像轉換到3D圖像,關鍵是從原始已知的2D圖像(images of limited depth of field(DOF))中生成深度圖像。文章提齣的是如何在FPGA上實時的對圖像進行小波分析,以便後續利用小波分析的結果得到深度圖像。與傳統的捲積方法相比,該硬件框架提升瞭運算速度,簡化瞭結構。
유2D도상전환도3D도상,관건시종원시이지적2D도상(images of limited depth of field(DOF))중생성심도도상。문장제출적시여하재FPGA상실시적대도상진행소파분석,이편후속이용소파분석적결과득도심도도상。여전통적권적방법상비,해경건광가제승료운산속도,간화료결구。
One of the most important factors which convert the 2D image to 3D image is to generate depth images from the original 2D images (images of limited depth of field(DOF)). This paper is how to realize wavelet analysis of the real-time image in FPGA for the depth image. Compared to the traditional convolution method, the hardware frame mentioned in the paper improves arithmetic speed and facilitate the structure.