外文资料翻译

毕业设计(论文)外文资料翻译

系： 专 业： 姓 名： 学 号：

外文出处： An algorithm for image stitching (用外文写)

and blending

附 件： 1.外文资料翻译译文；2.外文原文。 指导教师评语： 签名： 年 月 日 注：请将该封面与附件装订成册。

附件1：外文资料翻译译文

一种图像拼接和融合

摘要

在许多临床研究，包括癌症，这是非常可取的，以获取图像的整个肿瘤节 同时保留一个微小的决议。通常的办法是建立一个综合的形象，适当重叠个人获得的图象在高倍显微镜下。一块镶嵌图，这些图片可以 准确地运用所形成的图像配准，重复搬运和混合技术。我们描述的最优化， 自动，快速和可靠的方法，既形象加入和融合。这些算法可以适用于大多数类型的光学显微镜的成像。从组织学，从体内血管成像和荧光等方面的应用表明，无论是在二维和三维。算法的不同形象重叠的阶段有所不同，但例子综合获得的图象既手动驱动，电脑控制的阶段介绍。重叠迁移算法是基于互相关的方法，这是用来确定和选择最佳的相关点之间的任何新的形象和以前的综合形象。补充图片混合算法的基础上，梯度法，是用来消除强度变化急剧的形象加入，从而逐渐融合到一个图像邻近'复合'的细节，该算法克服强度 差异和几何失调图像之间的缝合将介绍和说明了几个例子。 关键词：图像拼接，共混，镶嵌图像 1 。导言和背景

有许多应用需要高分辨率图像。在明亮的场或落射荧光显微镜[ 1 ] ， 例如，用于生物和医学应用中，通常需要分析一个完整的组织部分已层面几十毫米，在高分辨率。然而，高分辨率单图片不能意识到了低功耗的目标，必要的，以查看大样本，即使使用数码相机时的几万数以百万计的积极像素。最常见的方法是获得一些图片的部分组织在高 放大率和组装成一个复合单一形象保持高分辨率。这一过程的 组装复合图像从一些图片，也被称为'块'或'拼图'需要一个算法为图像拼接（登记）和混合。自动建立大型高分辨率图像的马赛克是一种越来越多的研究领域涉及计算机视觉和图像处理。拼接与融合可以被界定为生产单一edgeless形象整理一套重叠图像[ 2 ] 。自动化这一进程是一个重要的问题，因为它是困难和消耗时间，以实现手动。这样的一个算法的图像拼接和融合是本文介绍了。 图像拼接结合了一些拍摄高分辨率纳入一个综合的形象。综合图像必须包括图像放在正确的位置和目标是使图像边缘之间的无形的。那个质量表示缝合因此通过测量两个相邻的信件

图像缝合形成了综合形象和知名度煤层之间的缝合图像[ 3 ] 。图像拼接（注册） 方法已详细解释[ 4 ] 。 [ 5 ]中，互相关证明是最好的方法，自动注册大量的图像。各种登记方法进行了比较研究[ 5 ]这是显示 的互相关法提供了最小的错误。当这些方法进行了比较而言，速度， 交叉相关证明是第二快的，但更准确比速度最快的方法（主轴 法） 。 有一些文件，处理问题的缝合[ 3 ， 6-8 ] 。图像拼接可以用 图像像素直接-相关法，在频域-快速傅里叶变换法;利用水平低功能，如棱角;使用高层次的功能，如部分对象[ 2 ] 。布朗[ 4 ]分类图片登记根据以下标准：类型的功能空间，类型的搜索策略和类型的相似性 措施。 图像拼接方法的优化，寻求最佳的相关点使用以Levenberg - Marquardt 方法给出了[ 2 ， 9 ， 10 ] 。以Levenberg - Marquardt方法使良好的效果，但它是昂贵和计算能停留在局部极小。还有一种方法是运用一个算法搜索最佳的相关点拥有由'粗到细的决议的办法，以减少计算[ 10 ， 11 ] 。 该方法提供了本文所选择的最佳关联点，以下列方式。基于 了解预期重叠当使用机动阶段，这将是直截了当的，以找到最佳的 相关点在理想的情况。然而，重叠面积并非尽善尽美，而且肯定不是一个准确的一 像素，由于偏离阶段的理想位置，而且由于舞台/相机失调。我们的算法提供了一个如何克服这一问题，寻找周围地区的小预期中央重叠像素，以便找到 最佳关联点。定位获得图像手册阶段小得多准确，因此有必要 寻找更广阔的领域，以找到最佳的关联点。 大多数现有方法的图像拼接或者制作一个粗略'缝，不能处理的共同特征 如血管，彗星细胞和组织，或者他们需要一些用户输入[ 12 ] 。新算法的提出 本文介绍了嵌入式代码来处理这种功能。 为了消除边缘，使一个紧凑的形象，有必要适用于更多的图像融合。那个 图像融合过程中限制区的重叠这是确定在缝合过程。这个 也就是说，如果重叠区域图像间大，图像不完全匹配的这些部分， 鬼影或'模糊'是可见的。但是，如果这些地区的小国，接缝将会看到[ 13 ] 。为了避免 这些影响，使模糊效果微乎其微，在互相关函数之间的综合形象和的形象是被缝合需要适当运用。新的方法本文介绍表明， 最好的质量，才能实现图像融合的应用，如果每张图片后，已缝合。这种办法提高 缝合更多图像，因为互相关应用到混合复合形象，给出了 更有力的结果。当获取图像的高度非均匀样品，因为这是在我们的情况在体内研究中，照明条件变化，从而影响互相关期间适用缝合。这些照明变化防止取消文物。为了避免这种影响有可能正常化照明的图像，但它可能会导致有些损失的信息作为一个无法确定的真正原因是什么的变化图像照明。它可

以来自改变照明，而且还从不同的组织的颜色。因此，一些光照补偿是必要的。我们的成就是一个高质量，自动拼接和融合算法，响应的功能，如血船只，彗星细胞和组织样本。光照补偿不纳入提交算法。

本文主办如下。第2节解释了图像采集过程。第3节解释采用的方法开发过程中的图像处理算法，该算法适用于缝合和混合。第4节给出的结果，应用算法选取的图像拼接后只有后两个缝纫和混合，并说明了算法的有效性。结论中提出的 第5款，并指示今后工作的定义。 2 。图像采集 二维图像

所有图片被收购，利用标准的显微镜。所有2D图像获得的样本翻译和收集的手动或自动使用机动阶段。三种类型的二维图像进行收购。这些包括图像，组织学，荧光细胞的一个组成部分彗星阵列（彗星细胞） ，并在体内的血管。那个安装规范，这些图像是摘要列于表1 。 CCD相机用于获取图像要么一个IEEE 1394接口，或使用一个PCI图像采集（类型：由美国国家仪器公司，英国） 。成像面积的命令1x1平方毫米时，用客观x10 。 CCD相机推出两款噪音影响。一个是暗电流和 另一个是一个非均匀像素反应。为了取消了暗电流的影响，获得的图象，没有轻被减去图像的样本。取消的非均匀像素反应，形象的样本除以一个空白图像收购标准照明清洁幻灯片。镜头畸变也存在。全部成像系统，由于这种像差，遭受了或多或少从每桶或枕形失真，或他们的组合。最困难的条件下的图像拼接是那些各种各样的环境照明即强度跨度-空间不同的照明。据推测，旋转和缩放保持不变全国各地的实验和处理图像。只有翻译错误必须纠正在缝合过程。 三维图像

用于购置三维图像下面的安装使用。尼康电子200荧光显微镜的使用修改阶段，以适应啮齿动物。我们在体内的血管图像获得一个窗口商会安排。它包括双面铝框举行两个平行的玻璃窗。它位于中央上述的目标[ 14 ] 。肿瘤血管生成和血管对治疗的反应在这两个形态的血 血管网络的功能和个人进行了调查船只使用窗口厅。多光子显微技术已应用到获取三维图像的肿瘤血管[ 15 ] ，因为这些技术证明是十分有效地获得三维生物的图片。多光子显微镜系统是基于酶标仪湄公河委员会1024MP工作站，由一个固体statepumped （ 10W的年十， Nd ： YVO4晶体，光谱物理） ，自锁模钛：蓝宝石（海啸，光谱物理）激光系统，一个中心扫描头，

焦探测器和一个倒置显微镜（尼康TE200 ） [ 15 ] 。多光子显微镜可以准确地找到一个三维荧光量，可成功地应用于分析血管形态。通常是一个小肿瘤（直径几毫米）植入皮肤的窗口中庭。整个肿瘤血管的影像大多数实验。图片10倍的目标而采取的一切，但最小的 肿瘤和图像覆盖约1.3x1.3毫米的组织。成堆的图像所采取的一个典型的堆栈50片。它通常需要13分钟获得的图像为整个堆栈 3 。图像处理-方法

主要有两个阶段在处理这些图像： 1 ）图像拼接

拼接是由滑动的新形象的综合形象和寻找最佳关联点。 2 ）图像融合

配煤是由分离颜色的飞机，在必要情况下，采用混合算法每个彩色带 重组飞机一起获得全彩色图像的输出。混合图像应保持质量输入图像[ 16 ] 。这些过程中有详细的解释，并参阅下文二维图像，除非明确指出，他们提到的三维图像。算法开发了C编程语言LabWindows / CVI的7.0 （美国国家仪器有限公司）开发环境，使用IMAQ图像处理图书馆和Windows XP专业版操作系统。那个算法是完全自动的，他们已经在电脑上测试的处理器速度1.53GHz和448MB的内存。 3.1拼接方法

在该算法的缝合是由图像翻译只。应用程序可以被称为作为拼接，瓦工， montaging或缝合。第一步是生成的相对位置所获得的图像和建立一个空的图像阵列在电脑记忆体，这些图片将放在。下一步是搜索对于这一点的最佳关联是由相邻的图像边缘滑动是双向的，直到最佳比赛的边缘特征发现。这个搜索过程需要选择最佳的搜索空间如图1所示，在其中进行搜索的最佳关联。使用太多像素内使这个方块相关过程耗时太少像素，同时减少比赛的质量。选择若干像素使用密切相关的各个方面的功能预期将显着的形象而这又取决于重点质量，即对目前的最大空间频率的形象。

附件2：外文原文

An algorithm for image stitching and blending ABSTRACT

In many clinical studies, including those of cancer, it is highly desirable to acquire images of whole tumour sectionswhilst retaining a microscopic resolution. A usual approach to this is to create a composite image by appropriatelyoverlapping individual images acquired at high magnification under a microscope. A mosaic of these images can beaccurately formed by applying image registration, overlap removal and blending techniques. We describe an optimised,automated, fast and reliable method for both image joining and blending. These algorithms can be applied to most typesof light microscopy imaging. Examples from histology, from in vivo vascular imaging and from fluorescenceapplications are shown, both in 2D and 3D. The algorithms are robust to the varying image overlap of a manually moved stage, though examples of composite images acquired both with manually-driven and computer-controlled stages are presented. The overlap-removal algorithm is based on the cross-correlation method; this is used to determine and select the best correlation point between any new image and the previous composite image. A complementary image blending algorithm, based on a gradient method, is used to eliminate sharp intensity changes at the image joins, thus gradually blending one image onto the adjacent ‘composite’. The details of the algorithm to overcome both intensity discrepancies and geometric misalignments between the stitched images will be presented and illustrated with several examples.

Keywords: Image Stitching, Blending, Mosaic images 1. INTRODUCTION AND BACKGROUND

There are many applications which require high resolution images. In bright-field or epifluorescence microscopy [1],for example, which are used in biological and medical applications, it is often necessary to analyse

a complete tissue section which has dimensions of several tens of millimetres, at high resolution. However, the high resolution single image cannot be realised with a low power objective, necessary to view a large sample, even if using cameras with tens of millions of active pixels. The most common approach is to acquire several images of parts of the tissue at high magnification and assemble them into a composite single image which preserves the high resolution. This process of assembling the composite image from a number of images, also known as ‘tiling’ or ‘mosaicing’ requires an algorithm for image stitching (registration) and blending. The automatic creation of large high resolution image mosaics is a growing research area involving computer vision and image processing. Mosaicing with blending can be defined as producing a single edgeless image by putting together a set of overlapped images [2]. Automating this process is an important issue as it is difficult and time consuming to achieve it manually. One such algorithm for image stitching and blending is presented in this paper. Image stitching combines a number of images taken at high resolution into a composite image. The composite image must consist of images placed at the right position and the aim is to make the edges between images invisible. The quality of stitching is therefore expressed by measuring both the correspondence between adjacent stitched images that form the composite image and the visibility of the seam between the stitched images [3]. Image stitching (registration) methods have been explained in detail in [4]. In [5], cross-correlation is shown to be the preferred method for automatic registration of large number of images. Various registration methods were compared in this paper [5] and it was showed that the cross-correlation method provided the smallest error. When these methods were compared in terms of speed, the cross-correlation was shown to be the second fastest but much more accurate than the fastest method (principal axes method). There are a number of papers that deal with the stitching problem [3, 6-8]. Image stitching can be performed using image pixels directly - correlation method;

in frequency domain - fast Fourier transform method; using low level Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XII, Jose-Angel Conchello, Carol J. Cogswell, Tony Wilson, Editors, March 2005 191 features such as edges and corners; using high level features such as parts of objects [2]. Brown [4] classifies image registration according to following criteria: type of feature space, type of search strategies and type of similarity measure.Approaches for image stitching that optimise the search for the best correlation point by using Levenberg-Marquardt method are given in [2, 9, 10]. Levenberg-Marquardt method gives good results, but it is computationally expensive and can get stuck at local minima. An alternative way is to apply an algorithm which searches for the best correlation point by employing a ‘coarse to fine’ resolution approach in order to reduce the number of calculations [10, 11]. The approach offered in this paper makes the selection of the best correlation point in the following way. Based on knowledge about the expected overlap when using the motorised stage, it would be straightforward to find the best correlation point in the ideal case. However, the overlap area is not perfect, and certainly not to an accuracy of one pixel, due to deviations in stage position from the ideal and due to stage/camera misalignment. Our algorithm offers away to overcome this problem by searching the small area around the expected central overlap pixel in order to find the best correlation point. Positioning of acquired images with a manual stage is much less accurate, so there is a need to search a wider area in order to find the best cross-correlation point. Most of the existing methods of image stitching either produce a ‘rough’ stitch that cannot deal with common features such as blood vessels, comet cells and histology, or they require some user input [12]. The new algorithm presented in this paper has embedded code to deal with such features. In order to remove the edges and make one compact image it is necessary to apply additional image blending. The process of image blending is restricted to zones of overlap which are

determined during the stitching process. This means that if the overlap regions between images are large, and images are not perfectly matched on these parts, ghosting or ‘blurring’ is visible. However, if these regions are small, the seams will be visible [13]. In order to avoid these effects and make the blurring effect negligible, the cross-correlation function between the composite image and the image which is to be stitched needs to be applied appropriately. The new method presented in this paper shows that the best quality image can be achieved if blending is applied after each image has been stitched. This approach improves the stitching of additional images because the cross-correlation is applied to a blended composite image which gives a more robust result. When acquiring images of highly non-uniform samples, as it is the case in our in vivo studies, the lighting conditions change and thus influence the cross-correlation applied during stitching. These lighting changes prevent the removal of artefacts. In order to avoid this effect it may be possible to normalise the illumination of the images, but it could cause some loss of information as one cannot be sure what the real cause for the variation in the image illumination is. It can come from the changes in the lighting but also from the different tissue colour. Hence, some illumination compensation is necessary. Our achievement is a high-quality, automatic stitching and blending algorithm that responds to features such as blood vessels, comet cells and histology samples. The illumination compensation is not incorporated in the presented algorithm. This paper is organised as follows. Section 2 explains the image acquisition process. Section 3 explains the methodology followed during the development of the image processing algorithm that applies both the stitching and blending. Section 4 gives the results of the applied algorithm on the selected images after the stitching only and after both stitching and blending and illustrates the effectiveness of the proposed algorithm. Conclusions are presented in Section 5 and directions for the future work are defined.

2. IMAGE ACQUISITION 2D images

All images were acquired using a standard microscope. All 2D images were acquired by sample translation and collected either manually or automatically using the motorised stage. Three types of 2D images were acquired. These include the images of histology, fluorescent cells as part of a comet array (comet cells) and in vivo blood vessels. The setup specification for these images is summarised in Table 1. CCD cameras were used to acquire the images with either a IEEE 1394 interface or using a PCI frame grabber (type: by National Instruments, UK). The imaging area is of the order of 1x1 mm2 when using objective x10. CCD cameras introduce two noise effects. One is a dark current and another is a non-uniform pixel response. In order to cancel out the dark current effect, images acquired with no light were subtracted from images of the sample. For cancellation of the non-uniform pixel response, the image of the sample is divided by a blank image acquired with standard illumination of a clean slide. Lens aberrations are also present. All imaging systems, due to such aberrations, suffer to a greater or lesser extent from barrel or pincushion distortion, or their combination. The most difficult conditions for image stitching are those with wide range of ambient lighting i.e. Proceedings of SPIE -- Volume 5701

Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XII,Jose-Angel Conchello, Carol J. Cogswell, Tony Wilson, Editors, March 2005 192 with a large intensity span - spatially varying illumination. It is assumed that rotation and scaling stay the same throughout both the experiments and processing the images. Only translation errors need to be corrected during the stitching process. Table 1 Summary of the setup specification used for the image acquisition 3D images

For the acquisition of 3D images the following setup was used. A Nikon TE

200 fluorescence microscope was used with a modified stage to accommodate rodents. Our in vivo blood vessel images were acquired through a window chamber arrangement. It consists of double sided aluminium frame holding two parallel glass windows. It is located centrally above the objectives [14]. Tumour angiogenesis and vascular response to treatment in both the morphology of blood vessel networks and the function of individual vessels have been investigated using the window chamber. Multi-photon microscopy techniques have been applied to obtain 3D images of tumour vasculature [15], as these techniques are shown to be highly effective in obtaining three-dimensional biological images. The multi-photon microscope system is based on Bio-Rad MRC 1024MP workstation and consists of a solid-statepumped (10W Millennia X, Nd:YVO4, Spectra-Physics), self-mode-locked Ti:Sapphire (Tsunami, Spectra-Physics) laser system, a focal scan-head, confocal detectors and an inverted microscope (Nikon TE200) [15]. Multi-photon microscopy can accurately locate fluorescence within a 3D volume and can be successfully applied to the analysis of vascular morphology. Usually a small tumour (few millimetres in diameter) was implanted into the skin in the window chamber. The whole tumour vasculature was imaged for most experiments. Images were taken with 10X objective for all but the smallest tumours and image covered approximately 1.3x1.3 mm tissue. Stacks of images are taken with a typical stack of 50 slices. It takes typically 13 minutes to acquire images for an entire stack. 3. IMAGE PROCESSING - METHODOLOGY

There are two main stages in processing these images: 1) Image stitching

Stitching is performed by sliding the new image over the composite image and finding the best cross-correlation point. 2) Image blending

Blending was done by separating colour planes, where necessary, applying blending algorithm for each colour band and

recomposing planes together to get full colour image at the output. The blended images should maintain the quality of the input images [16]. These processes are explained in detail below and refer to 2D images unless specified that they refer to 3D images. Algorithms were developed in C programming language under LabWindows/CVI 7.0 (National Instruments) development environment, using IMAQ Image Processing Library and Windows XP Professional operating system. The algorithms are completely automated and they have been tested on a PC with processor speed 1.53GHz and 448MB of RAM.

3.1 Stitching method

In the presented algorithm the stitching is performed by image translation only. The applied procedure can be referred to as mosaicing, tiling, montaging or stitching. The first step is the generation of relative positions of acquired images and the creation of an empty image array in computer memory where these images will be placed. The next step is a search for the point of best correlation which is performed by sliding adjacent image edges in both directions until the best match of edge features is found. This search process requires the choice of an optimum search space shown in Figure 1, in which a search is performed for the best correlation. The use of too many pixels inside this box makes the correlation process time consuming whilst too few pixels reduce the quality of match. The choice of number of pixels used is strongly related to the dimensions of features expected to be visible in the image which in turn depends on focus quality, i.e. on the maximum spatial frequencies present in the image.

本文来源：https://www.bwwdw.com/article/5h6g.html