权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Restoration of the Image Signal Lost by Analog to Digital Conversion

模数转换丢失的图像信号的恢复

基本信息

批准号：
18500140
负责人：
TAGUCHI Akira
金额：
$ 2.18万
依托单位：
Musashi Institute of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2006
资助国家：
日本
起止时间：
2006 至 2007
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-18500140/
关键词：
sampling quantization enlargement of digital image bit-length expansion frame rate up-conversion

项目摘要

Resolution and bit-length of digital images are decided by Analog to Digital Conversion. In the case of video signals, frame rate is also decided by video recoding. Image enragement, frame rate up-conversion and bit-length expansion techniques are necessary for many digital image/video equipments, such as high-quality display such as the LCD display and the plasma display. In this research, three techniques(i.e., image enragement, frame rate up-conversion and bit-length expansion) are studied.Image enragement : The Laplacian pyramid representation(LP) method, which has been proposed by ourselves, is implemented on FPGA(Xilinx Virtex-II XC2V6000). The LP method is realized by 5654 slices, 7715 FFs and 42 block RAMs. We achieve 206.9[MHz] for maximum operating frequency and only 20 lines for system latency.Frame rate up-conversion : In this research, a novel frame rate up-conversion(FRUC) method using the motion compensation(MC) based on block matching algorithm(BMA) is proposed. The pro … More perty of the BMA is changed by the size of block. It is desirable for small moving object to set block size small. On the other hand, we would like to set the block size large in the global motion region. Therefore, the block size is decided depend on the property of the local region of frame image. A novel FRUC technique based on the motion compensated interpolation with multi-size blocks is proposed. At first, the frame image is divided into 8x8 blocks and motion vectors are estimated for each 8x8 block. Next, the similar motion blocks are merged into one block, up to 32x32 block. Thus the proposed method reduces block artifacts and realizes the clear interpolation of moving objects regardless of the size of those.Bit-length expansion : In this research, we propose a bit-length expansion method from M gray levels to N(=3(M-1) +1) gray levels. This operation means that two gray-levels are introduced between existing levels. The bit-length expansion method results in the estimation problem of an error image between an M gray level's image and an N gray level's image. The error image is constructed by only three values. If the difference between the existing level and the inserted new level is defined as Δ, the N level image is obtained by adding+Δ, 0 or-Δ to the M level image. We show that the proposed method can remove pseudo-contour of digital images through a lot of application results. Less

数字图像的分辨率和位长是由模数转换决定的。在视频信号的情况下，帧速率也由视频编码决定。图像增强、帧频上变频和比特长度扩展技术是许多数字图像/视频设备所必需的，例如高质量显示器，如LCD显示器和等离子体显示器。在这项研究中，三种技术（即，图像增强：在FPGA（Xilinx Virtex-II XC 2 V6000）上实现了我们提出的拉普拉斯金字塔表示（LP）方法。LP方法由5654个片、7715个FF和42个块RAM实现。帧率上转换：提出了一种基于块匹配算法（BMA）的运动补偿（MC）帧率上转换（FRUC）方法。职业 ...更多信息 BMA的性能随块的大小而改变。对于小移动对象，期望将块大小设置得小。另一方面，我们希望在全局运动区域中将块大小设置得大。因此，分块大小取决于帧图像的局部区域的属性。提出了一种基于多块运动补偿插值的FRUC算法。首先，将帧图像划分为8 × 8块，并为每个8 × 8块估计运动矢量。接下来，相似的运动块被合并成一个块，直到32 x32块。因此，该方法减少了块效应，并实现了清晰的运动物体的插值，而不管那些的大小。位长度扩展：在这项研究中，我们提出了一种从M个灰度级到N（=3（M-1）+1）个灰度级的位长度扩展方法。该操作意味着在现有的灰度级之间引入两个灰度级。位长扩展方法导致了M灰度级图像和N灰度级图像之间的误差图像的估计问题。误差图像仅由三个值构成。如果现有级别与插入的新级别之间的差被定义为Δ，则通过将+Δ、0或-Δ加到M级别图像来获得N级别图像。大量的应用结果表明，该方法可以有效地去除数字图像中的伪轮廓。少