CodeBus
www.codebus.net
Search
Sign in
Sign up
Hot Search :
Source
embeded
web
remote control
p2p
game
More...
Location :
Home
Search - c select
Main Category
SourceCode
Documents
Books
WEB Code
Develop Tools
Other resource
Sub Category
Picture Viewer
2D Graphic
3D Graphic
Graph Drawing
Wavelet
GIS program
Fractal program
Special Effects
OpenGL program
Graph Recognize
CAD
OpenCV
VTK
Search - c select - List
[
Graph Recognize
]
车牌定位
DL : 0
车牌定位 使用时打开此例题目录下pic中的图片,然后依次单击按钮“转”、“1”、“2”、“3”、“4”和“5”,就可以实现精确的车牌定位。 具体步骤 1.24位真彩色->256色灰度图。 2.预处理:中值滤波。 3.二值化:用一个初始阈值T对图像A进行二值化得到二值化图像B。 初始阈值T的确定方法是:选择阈值T=Gmax-(Gmax-Gmin)/3,Gmax和Gmin分别是最高、最低灰度值。 该阈值对不同牌照有一定的适应性,能够保证背景基本被置为0,以突出牌照区域。 4.削弱背景干扰。对图像B做简单的相邻像素灰度值相减,得到新的图像G,即Gi,j= Pi,j-Pi,j-1 i=0,1,…,439 j=0,1,…,639Gi,0=Pi,0,左边缘直接赋值,不会影响整体效果。 5.用自定义模板进行中值滤波 区域灰度基本被赋值为0。考虑到文字是由许多短竖线组成,而背景噪声有一大部分是孤立噪声,用模板(1,1,1,1,1)T对G进行中值滤波,能够得到除掉了大部分干扰的图像C。 6.牌照搜索:利用水平投影法检测车牌水平位置,利用垂直投影法检测车牌垂直位置。 7.区域裁剪,截取车牌图像。-positioning plates used to break pic examples directory of images, and then click the button "turn", "1", "2", "3" and "4" and "5", can achieve precise positioning of the plates. Concrete steps 1.24 true color-gt; 256-color grayscale. 2. Pretreatment : median filter. 3. Binary : an initial threshold value T A pair of images for two to be two binary images B value. T initial threshold value to determine the method is : Select threshold T = Tc - (Tc-Gmin) / 3, respectively Gmin saturated and is the highest and the lowest gray value. The threshold values for different licenses are certain adaptability, to ensure that basic background was home to 0, to highlight regional licenses. 4. Weakened background interference. The images do sim
Date
: 2008-10-13
Size
: 717.34kb
User
:
何问宇
[
Graph Recognize
]
车牌定位
DL : 1
车牌定位 使用时打开此例题目录下pic中的图片,然后依次单击按钮“转”、“1”、“2”、“3”、“4”和“5”,就可以实现精确的车牌定位。 具体步骤 1.24位真彩色->256色灰度图。 2.预处理:中值滤波。 3.二值化:用一个初始阈值T对图像A进行二值化得到二值化图像B。 初始阈值T的确定方法是:选择阈值T=Gmax-(Gmax-Gmin)/3,Gmax和Gmin分别是最高、最低灰度值。 该阈值对不同牌照有一定的适应性,能够保证背景基本被置为0,以突出牌照区域。 4.削弱背景干扰。对图像B做简单的相邻像素灰度值相减,得到新的图像G,即Gi,j= Pi,j-Pi,j-1 i=0,1,…,439 j=0,1,…,639Gi,0=Pi,0,左边缘直接赋值,不会影响整体效果。 5.用自定义模板进行中值滤波 区域灰度基本被赋值为0。考虑到文字是由许多短竖线组成,而背景噪声有一大部分是孤立噪声,用模板(1,1,1,1,1)T对G进行中值滤波,能够得到除掉了大部分干扰的图像C。 6.牌照搜索:利用水平投影法检测车牌水平位置,利用垂直投影法检测车牌垂直位置。 7.区域裁剪,截取车牌图像。-positioning plates used to break pic examples directory of images, and then click the button "turn", "1", "2", "3" and "4" and "5", can achieve precise positioning of the plates. Concrete steps 1.24 true color-gt; 256-color grayscale. 2. Pretreatment : median filter. 3. Binary : an initial threshold value T A pair of images for two to be two binary images B value. T initial threshold value to determine the method is : Select threshold T = Tc- (Tc-Gmin)/3, respectively Gmin saturated and is the highest and the lowest gray value. The threshold values for different licenses are certain adaptability, to ensure that basic background was home to 0, to highlight regional licenses. 4. Weakened background interference. The images do sim
Date
: 2025-12-23
Size
: 717kb
User
:
何问宇
[
Graph Recognize
]
imageproj
DL : 0
原创图片检索程序。选择一张图片以及一个图片文件夹,提取图片特征,使用R树进行查询。请在vc8以上环境打开此工程。-Original image retrieval process. Select a picture and a picture folder, extract image features, using the R tree query. Please open this vc8 over environmental engineering.
Date
: 2025-12-23
Size
: 182kb
User
:
王妙一
[
Graph Recognize
]
SVM-and--Face-Recognition
DL : 0
支持向量机及其在人脸识别中的应用研究 上海交通大学博士论文,在知网上面付费下载得到的。本文从应用的角度出发,较为全面地对一些相关问题进行探讨,并使用Visual C++实现了一个基于支持向量机的人脸识别软件—idTeller。 论文的主要工作和创新点包括: ·提出了两种基于VC边界的支持向量机参数选择算法—固定C算法和VC-CV算法。VC边界是两类支持向量机参数选择的一个理想准则,但它的一些固有缺点使其应用变得困难。本文通过将VC边界转化为VC指标,最终把问题归结为对最小包围体的求解,从理论上和计算上为VC边界的使用铺平了道路。在此基础之上,本文提出了两种基于VC边界的参数选择算法—固定C算法和VC-CV算法。在数个基准数据集上的实验表明,相比交叉验证算法,VC-CV算法不仅能获得性能更好的分类器,而且具有较低的计算复杂度。 ·使用序贯最小优化算法解决了最小包围体求解问题。最小包围体求解是计算VC指标的一个关键步骤,本文使用序贯最小优化算法对其求解,并对算法初始化、参数选择及更新等若干实现问题进行了深入地研究。在多个基准数据集上的实验表明,序贯最小优化算法能够快速而准确地解决最小包围体求解问题。- Support vector machine and its application to face recognition Shanghai Jiaotong University doctoral thesis, in HowNet above pay to get the download. From the application point of view, to more fully explore some related issues, and using Visual C-idTeller a support vector machine-based face recognition software. The main work and innovation of the paper include: two kinds of parameters of support vector machine based on the VC boundary selection algorithm- fixed-C algorithm and the VC-CV algorithm. VC boundaries are two types of support vector machine parameters to select the ideal criteria, but some of its inherent shortcomings make it difficult. This article by VC boundary for the VC index, and ultimately the problem is reduced to the solution of the minimum bounding volume, and paved the way for the use of the VC boundary from the theory and calculations. On this basis, we propose two parameter selection algorithm based on the VC boundary- fixed-C algorithm and the VC-CV algorit
Date
: 2025-12-23
Size
: 9.9mb
User
:
Jessicaying
CodeBus
is one of the largest source code repositories on the Internet!
Contact us :
1999-2046
CodeBus
All Rights Reserved.