[图形图像处理] DAHENG_MER-U_ProgrammingGuide_CN
大恒图像首次登上工博会的舞台，全方位展示了旗下水星系列相机、机器视觉软件解决方案、机器人2D/3D视觉引导解决方案。水星系列相机产品线覆盖USB3.0/USB 2.0/千兆网接口，分辨率从36万像素到500万像素，其中水星系列千兆网工业摄像机荣获了CEC 2016机器视觉类
For the first time, the image of dheng has been on the stage of the industrial fair, ion nd accurately locathich demonstrates the flexible detection and intelligent positioning functions widely used in the 3C (2018-08-30, C#, 511KB, 下载7次)
[图形图像处理] 2
隐含语义索引及其在中文文本处理中的应用研究.信息检索本质上是语义检索，而传统信息检索系统郝走基于独立词索引，因此检索效果并不理想．臆含语义索
引是一种新型的信息检索模型，它通过奇异值分解，将词向量和文档向量投影到一个低苹空间，消减了词和文档之间的
语义模糊度，使得文档之间的语叉关系更为明晰．实验和理论结果证实了臆台语艾索引能够取得更好的检索艘果奉吏
论述了臆舍语叉索引的理论基础t研究了臆舍语叉索引在中文文本处理中的应用，包括中文文本持索、中文丈本分类扣
中文文本聚妾等．
Inforrnatlon retri ise8sentially semanticretri
the contents of documents and queries with a8et of index terms
lndex(LSI)is a new aIgebralc model for information retri ，
However，most classic Inforrnatl⋯ etri 8ystems represent
which can lead to poorretri performance．I。atent semamlc
which map8 document9 and querlesvecto⋯ nto a lower dimpn—
slona【8pacc by 8mgular value decomp。8ldon，so that the lnherent vagueness as80clated with a retnml process based on key一
Ⅷd setsls cons讨mbly reduced and semantic a8sociatbn amo“g the document8 ls h‘ghlLghted con88quently．TbeoretIcanaly—
sesand elperl唧tal resul拈aU show that LSl can impr。ve retrI performance signlficantIy．Thl8 pa卵rIntroduces the ftmda
mental prInclples of LSI and explores i协8pplicatlons in Chine8e text processi“g，mcludi“g Chinese text retn ·Chlnese t}xt
clas虬flcatIon and clusterIng etc． (2016-08-07, Java, 235KB, 下载3次)
[图形图像处理] AMlogic
AMlogic晶晨 S802/S812/M8 固件解包打包编辑工具
替换开机图片，修改分区大小，定制系统参数，删减系统app等个性化定制固件。
AMlogic crystal morning S802/S812/M8 firmware unpack editing tools package
Replace boot image, modify the partition size, customize the system parameters, delete the system app and other personalized custom firmware. (2016-07-09, Windows_Unix, 1848KB, 下载17次)
[图形图像处理] Graphics-programs
<图形程序开发人员指南>迈克尔·亚伯拉什的经典图形编程黑皮书是迈克尔以前的著作汇编语言和图形编程（包括来自于博士道博的日记他的“图形编程”栏）汇编。很多这本书的重点是分析和代码测试，以及性能优化。它还探讨许多末日和Quake的3-D游戏，和3-D图形的问题，例如纹理映射，隐面消除，等后面的技术。
Michael Abrash s classic Graphics Programming Black Book is a compilation of Michael s previous writings on assembly language and graphics programming (including his Graphics Programming column in Dr. Dobb s Journal). Much of the focus of this book is on profiling and code testing, as well as performance optimization. It also explores much of the technology behind the Doom and Quake 3-D games, and 3-D graphics problems such as texture mapping, hidden surface removal, and the like. (2016-04-03, PDF, 27400KB, 下载7次)
[图形图像处理] vbpototexiao
VB图像调速锐化、对比度、亮度等特效处理程序，带图像滤镜效果，是一个实用小巧的VB6.0图像特效处理程序，带有完整源代码。在图片的处理方面，有两个功能模块：一是图像调节，包括亮度、对比度、色彩饱和度的调整，二是图像滤镜，可添加锐化、柔化、扩散、浮雕、轮廓、铅笔画等效果，实属不错。
VB image speed sharpening, contrast, brightness and other effects processing procedures, with the image filter effects, is a practical small VB6.0 effects of image processing procedures, with complete source code. In the aspect of image processing, has two functional modules: one is image adjustment, including brightness, contrast, color saturation adjustment, the two is the image filter, sharpen, blur, add diffusion, relief, contour, pencil drawing effect is good. (2014-07-07, Visual Basic, 18KB, 下载9次)
[图形图像处理] vbphototexiao
VB图像调速锐化、对比度、亮度等特效处理程序，带图像滤镜效果，是一个实用小巧的VB6.0图像特效处理程序，带有完整源代码。在图片的处理方面，有两个功能模块：一是图像调节，包括亮度、对比度、色彩饱和度的调整，二是图像滤镜，可添加锐化、柔化、扩散、浮雕、轮廓、铅笔画等效果，实属不错。
VB speed image sharpening, contrast, brightness and other effects processing procedures with image filter effects, is a practical compact VB6.0 image effects processing program, with full source code. In the processing of the picture, there are two functional modules: one image adjustments, including adjustments to brightness, contrast, color saturation, and second, image filters, you can add sharpen, soften, diffusion, relief, contours, pencil drawing, etc. effect, it is good. (2014-04-21, Visual Basic, 18KB, 下载26次)
[图形图像处理] 9
针对传统的空间域纹理图像修复算法计算量大、修复时间长的缺点, 本文提出了一种纹理图像的快速修复算法。该算
法的基本思想是在小波域中利用小波系数的能量来确定待修复块的填充顺序, 并结合纹理合成的方法填充待修复区。实验结
果表明, 该算法不仅可以大大提高纹理图像的修复速度, 而且在峰值信噪比和主观视觉效果上都优于传统的图像修复算
Trad itional spatia-l doma in inpainting algorithm s for tex ture image need comp licated computat ion and consume
a long t ime. A fast inpainting algorithm for tex ture image is proposed by determ in ing the order o f the filling
blocks using the energ ies o f w avelet coeffic ients in w ave let dom ain, and filling the inpa in ting b locks using tex ture
synthesis method. Experimental resu lts show that the a lgorithm not on ly improves inpainting speed g reat ly, but a lso
achieves better peak signa l no ise rat io ( PSNR ) and higher v isual qua lity.
(2014-03-22, matlab, 791KB, 下载21次)
[图形图像处理] 7
本文提出一种基于边缘自适应小波变换的多尺度图像修复算法，对非纹理图像有比较好的修复效
果。边缘自适应小波变换的基本思想是，先检测出图像的主要边缘，这些边缘把图像分割成几个平滑区，然后
对图像进行不跨越边缘的小波分解，即在各平滑区内部进行小波变换，得到图像的多尺度表示，并且同时计算
边缘的多尺度表示。这样的小波分解使高频信息基本都集中在边缘上，而高频系数则非常稀疏，而且都接近
于零。在此基础上进行图像修复，就只需要对低频部分与边缘图像进行修复，然后重构得到修复图像即可。
经过小波分解，低频部分破损区域大大缩小，用比较简单的插值方法就可进行修复，大大降低了计算量。对边
Abstract：A fast inpainting algorithm based on edge—adaptive wavelet transform are in—
troduced in this paper．The main idea of edge-adaptive wavelet transform is to firstly de—
tect the main edges of the image and then carry Out a wavelet transform not cross those ed—
ges．It makes the high—frequency coefficients more sparse and more weak．After this kind
of wavelet transform。inpainting only need to fill in the low—frequency sub—band and con—
nect the broken edges．Because the inpainting field become more small in the Iow—frequen—
cy sub—band，it only need a simple (2014-03-22, matlab, 500KB, 下载34次)
[图形图像处理] grayscale
灰階(gray-scale)圖像處理(60*60 pixel)controller控制各個程式的地址以及開關，input_mem將資料讀進記憶體，grayscale將讀取資料像素的亮度以數值來表示，將24bit的 像素化成四個8bit的值輸出。接著進入sobel，在此將前面的四個值乘上1或-1個別的相加，得出新的四個值，輸入進shiftcase進行threshold的判斷，大於threshold則表現出白色(255)，小於threshold則表現出黑色(0)，最後將結果存入記憶體out_mem。//Verilog
Implement digital image processing through HDL. Design a system to perform simple image processing using mask.
Refer to the presentation slides “Simple Image Processor”. (2014-03-06, Others, 75KB, 下载4次)
[图形图像处理] Floyd_Steinberg_Dithering
有序抖动也是一种常用的半色调技术方法，可以说是最简单的领域处理过程的方
法。它是用一个固定大小的阈值矩阵在原始灰度图像上面一边“移动”，一边比较。如
果原始图像中像素的灰度值大于相应阈值矩阵中的阈值，那么抖动后该像素的值为1 （白
色），否则为0（黑色）。
Ordered dithering is also a common halftoning method, is the simplest way to process the fields. It is a fixed size in the threshold matrix at the side of the raw grayscale image " move" , while comparison. If the gray scale value of the pixel in the original image is larger than the threshold value matrix corresponding to a threshold value, then the dithering of the pixel value of 1 (white), otherwise as 0 (black). (2013-12-09, matlab, 1KB, 下载13次)
[图形图像处理] cun
通過實現前人的研究成果，首先進行簡單的視屏圖像分割並對分割結果進行運動目標跟蹤；然後通過對跟蹤目標和監控區域的位置狀態信息進行分析。實現提取視屏幀，圖像預處理，運動信息提取分析，運動目標檢測這些流程完成目標檢測。最後比較視屏提取人數的魯棒性，準確性，及時間上的差異。
Simple Screen image segmentation and segmentation results by the results of previous studies, the first moving object tracking then analyzed by tracking the location of the target and control area status information. Screen frame extraction, image preprocessing, sports information extraction, analysis, moving target detection process is complete target detection. Finally, we compare the Screen extracted number of robustness, accuracy, and time differences. (2013-04-13, matlab, 1KB, 下载6次)
[图形图像处理] Active-Contour-Models-
传统Snake 模型存在的缺点是, 其初始轮廓必须靠近图像中感兴趣目标的真实边缘,否则会得到错误结
果,且由于Snake 模型的非凸性,结果不能进入感兴趣目标的深凹部分,很容易陷入局部极小点. 由此该文提出一
种基于力场分析的主动轮廓模型,详细分析了基于欧氏距离变换的距离势能力场分布,归纳出感兴趣目标上真轮
廓点与假轮廓点的判别标准. 建立了由曲线能量到最终结果的有效方法,避免了Snake 陷入局部极小点. 实验结果
表明,该模型具有较大的捕获区域,能够进入感兴趣目标的深凹部分,准确提取感兴趣目标的轮廓. 与GVFSnake
模型相比, 该模型具有很小的计算量.
The t raditional snake initial contour should be close to the t rue boundary of interested ob2
ject in an image , or else it would converge to the wrong result . Next , active contours have difficulties
progressing into boundary concavities. Moreover , the t raditional snake and it s almost kinds of im2
proved methods are easy to get into local minimum because snake models are non2convex. An active
contour model based on force field analysis , namely FFASnake model , is presented in this paper.
Based on analyzing force dist ribution rules of distance potential force field , a standard is int roduced
here to distinguish the false one f rom contour point s. (2012-12-28, PDF, 318KB, 下载21次)
[图形图像处理] 09871
均值滤波、中值滤波、维纳滤波和基于小波变换的软阈值法和硬阈值法是常用的几种图像去噪方
法，通过对添加了不同噪声的图像进行了仿真实验。实验结果进一步验证了均值滤波在去除噪声的同时造成了
图像边缘一定程度的模糊，中值滤波能相对较好的保留了图像的边缘，维纳滤波对高斯白噪声去噪效果较好但
计算量大，相比较小波阈值法去噪是目前较好的的去噪方法，通过选取合适的阈值可以得到较为理想的去噪效
果。
Average filter，median filter，Wiener filter，and sofl and hard threshold method based on wavelet
transform are most commonly of several denoising methods ． A simulated experiment was carried by adding different
noise images． The experimental results demonstrate that the average filter can remove noise but it cause the
image edge blurred in a certain degree， the median filter can give a relatively fine retained image edge，and the
Wiener filter require much of calculation． Comparatively the threshold denoising in wavelet domain is a better denoising
method at present． It can be obtained an ideal effect in denoising by choosing appropriate threshold values． (2012-12-19, C/C++, 642KB, 下载62次)
[图形图像处理] OpenCV_CalcHist
直方图是计算机视觉中一个很重要的工具，OpenCV里面提供了不少有关直方图处理的函数。其中最基本的是计算直方图的函数calcHist()。有关直方图在前面的博文中也有所介绍：基础学习笔记之opencv(4)：直方图使用学习。不过目前由于本人课题研究上需要计算多张图片的一维直方图特性，且每张图片对应有自己不同的掩膜矩阵，开始以为OpenCV中提供的calcHist()函数能够实现这个功能，因为其中一个重载函数中有一个参数为InputArrayOfArrays，咋一看会觉得这不就是可以处理多张图片的吗？后面仔细研究后发现其实它实现的功能和我的需求是不同的，因此需要自己单独写函数来完成这一功能。
开发环境：OpenCV2.4.3+Qtcreator2.5.1
The histogram is a very important tool in computer vision, OpenCV which provides a lot of information on the function of the histogram processing. One of the most basic is calculated histogram function calcHist (). The histogram has also been introduced in front of the Bowen: basic study notes of opencv (4): histogram learning. However, I research needs to be calculated multiple pictures of a one-dimensional histogram characteristics, and each image corresponds to a different mask matrix, and began to think that OpenCV provided calcHist () function to enable this functionality, because one of the a parameter InputArrayOfArrays, at first glance Is not that can handle multiple pictures you will feel overloaded functions? Behind careful study to find out that it implements the functionality and my needs are different, and therefore requires its own separate write a function to accomplish this function. Development Environment: OpenCV2.4.3+Qtcreator2.5.1 (2012-12-01, C/C++, 737KB, 下载58次)
[图形图像处理] Edge-image-registration
图像配准是多传感器图像融合研究中的一项关键技术。多传感器图像特别是波段相距较远的相
关性较小的图像，要实现图像配准存在很大的困难。对于能够用仿射变换模型来描述的图像，图像之间比较
明显的特征是各个物体之间的边缘．该文研究利用小波变换的方法提取图像的边缘，并对边缘图像作交互方
差分析，搜索出最佳交互方差的配准参数。通过对SPOT和TM图像的处理，达到了精度较高的配准效
果．
Image registration is a key technology in the study of the multi-sensor image fusion. Band distant correlation smaller image, to achieve image registration, there are great difficulties in multi-sensor image. For affine transformation model to describe the image, the more obvious features in the image between the edge between the respective objects. This paper studies the use of wavelet transform to extract image edge, and the edge of the image for interactive analysis of variance, and search out the best interactive variance of the registration parameters. SPOT and TM image processing, to achieve a higher accuracy with prospective effect. (2012-11-15, Visual C++, 653KB, 下载15次)
[图形图像处理] 042325
背单词程序：
程序使用词汇表文件，其中每个词条由单词和解释两部分组成，编程完成：用户输入一个单词，程序将给出单词的词性和中文意思。
建立一个词汇表测试用词汇表名为新东方红宝书.txt，用c语言文件操作打开该文件，并将文件读入命名为dictionary[DICTSIZE]的数组内。读入数据存储于dictionary[DICTSIZE]所分配的地址空间内，以8字节表示一个信息。
Program uses the vocabulary file, where each entry and explain the words of two parts, programming to complete: user enters a word, the program will give part of speech and the Chinese words mean. Test to establish a glossary of new vocabulary called Eastern Red Book. Txt, file operations with the c language to open the file and read the file named dictionary [DICTSIZE] within the array. Read the data stored in the dictionary [DICTSIZE] within the allocated address space, a message of 8 bytes. (2011-05-16, Visual C++, 405KB, 下载6次)
[图形图像处理] worldwind
：World Wind是NASA发布的一个开放源代码的地理科普软件，由NASA Research开发，NASA Learning Technologies来发展，它是一个可视化地球仪，将NASA、USGS以及其它WMS服务商提供的图像通过一个三维的地球模型展现，还包含了火星和月球的展现。软件用C#编写，调用微软SQL Server影像库Terrain Server来进行全球地形三维显示。它通过将遥感影像与SRTM高程（航天飞机雷达拓扑测绘）叠加生成三维地形。
: World Wind is NASA released an open source geographic science software developed by the NASA Research, NASA Learning Technologies to develop, it is a visual globe, will be NASA, USGS and other WMS services provided by a three-dimensional image Earth model shows, also includes display of Mars and the moon. Software written using C#, call the Microsoft SQL Server Image Database Terrain Server to display the global three-dimensional terrain. It is through the elevation of remote sensing images and SRTM (Shuttle Radar Topology mapping) stack generate three-dimensional terrain. (2010-06-22, Visual C++, 1800KB, 下载81次)
[图形图像处理] 300
了一种基于融合和广义高斯模型的遥感影像变化检测算法。该算法融合差值影像和比值影像的结
果构造差异影像，然后运用灰度形态学对差异影像进行顾及空间邻域关系的处理，再对处理后的结果运用广义高
斯分布模型估计变化与非变化像元的概率密度参数，最后采用改进的O2 算法计算最佳分割阈值，提取变化区域。
实验结果表明，所提出的变化检测算法稳健、高效，具有较大的实用价值。
0LDI TCTUG TGUIUNVI C MLCNEU WUVUMVD5N CTTG5CML 5X GUB5VU IUNIDNE DBCEUI YCIUW 5N XFID5N CNW
EUNUGC6 3CFII B5WU6（ 33/）7 0LDI BUVL5W M5NIVGFMVI WDXXUGUNMU DBCEU YK XFIDNE VLU GUIF6VI 5X IFYVGCMVD5N
5TUGCVD5N CNW WD8DID5N 5TUGCVD5N，VLUN，TG5MUIIUI WDXXUGUNMU DBCEU YK EGCK B5GTL565EK，CECDN，UIVDBCVUI VLU
TG5YCYD6DVK WUNIDVK 5X MLCNEU TDQU6I CNW N5:MLCNEU TDQU6I YK 33/，XDNC66K，MC6MF6CVUI VLU 5TVDBC6 VLGUIL56W YK
VLU DBTG58UW O2 CTTG5CML CNW UQVGCMVI VLU MLCNEUW GUED5NI7 .QTUGDBUNVI IL5Z VLCV VLU BUVL5W 5X XFID5N CNW
33/ DI I56DW CNW UXXDMCMD5FI，ZLDML LCI TGUXUGCY6U CTT6DMCVD5N 8C6FU7 (2009-10-19, Others, 297KB, 下载120次)
[图形图像处理] str_01
國立台灣大學數學形態學及碎形維度於X光醫學影像分析數學形態學及碎形維度為影像分割及描述的新方法。研究分為以下三個部份來評估數學形態學及碎形維度於X光醫學影像分析時之可行性及限制性。第一部份:利用豬脊骨試體之Micro CT數位影像，模擬X光照射時可能產生的變異。討論照射角度、影像對比及幾何誤差的變異對分析結果所產生的影響。第二部份:以材料試驗機求得試體的材料性質(楊氏系數及極限強度)，探討影像特徵之碎形維度與楊氏系數及極限強度之相關性。第三部份以臨床根管治療為例，以數學形態學擷取影像特徵，並配合碎形維度評估根管治療前後影像特徵的差異，以期做為早期根管治療成效的評估。另外，碎形維度常見的評估方法-方格計數法，於非整數分割時，所產生的平台效應，本研究亦提出修正公式來解決此一問題。
mathematical morphological approach to edge detection (2009-03-04, Visual C++, 829KB, 下载41次)
[图形图像处理] DIP
需要matlab7.0或更高版本以及matlab图像处理工具箱。
1) 文件菜单
• 程序可处理具有不同宽高的灰度图像，显示区域多余的部分以白色填补；
注意：在做哈达马和沃氏变换时图像的尺寸先按2的幂次截断，在按最大的边长补成方形；
• 重载可以重新载入最近一次通过“文件——〉打开”打开的图像，这时将对重新载入的图像进行处理；
• 对于彩色图像，可以显示但如果想要使程序能正确处理，必须将其先转换成灰度图像，使用“文件——〉RGB转换成灰度图”；
• 对于每一次的处理结果都可以被保存为bmp或jpg格式；
• 关闭选项可清除两个显示区域的被处理和处理结果的图像；
• 退出则退出程序；
2) 两个操作的说明
“灰度处理”子菜单中“对数非线性变换”与“指数非线性变换”所使用的公式在参数输入对话框弹出的同时也会显示在另一个对话框中，以方便理解参数的意义。
3) 图像编码
霍夫曼编码和行程编码产生的结果将分别保存在“HuffmanCodeTable.txt”和“RLC.txt”文件中，由于MATLAB无法对位进行操作，因此编码结果实际上还是十进制数值。其中“HuffmanCodeTable.txt”文件中第一列表示灰度值，第二列表示码值，第三列表示编码的二进制位数。
err (2008-03-03, matlab, 348KB, 下载63次)