搜索资源列表
The-detection-of-traffic-lights
- 交通信号灯的检测是复杂场景下交通灯识别的重点。采用了色彩分割与关联滤波方案进行交通灯的检测。首先，建立交通信号灯的高斯模型，利用高斯向量与多色彩空间结合的方法进行图像分割。然后，采用基于区域增长与相似性判定的关联滤波，对色彩分割后的图像进行处理。最后，使用基于canny算子的边缘提取算法获取方向指示灯轮廓特征，并使用基于改进hu不变矩和马氏距离对方向指示信号灯进行分类。-The detection of traffic lights is the focus of traffic lights
DlgImageMatchMoment
- 基于不变矩特征的图像匹配算法的c++代码的实现-Invariant moment-based image matching the algorithm c++ code
colorMoment
- 在图像识别的时候用到的颜色矩特征的提取算法 是用C++实现的-Color moment feature extraction algorithm is used when the image recognition using C++
IVisualPatterm
- 图象不变矩理论提出的论文,1962年Hu在IEEE上发变,被2700多骗IEEE论文引用,奠定了图象不变矩在模式识别中的理论基础文献资资料详-Image paper presented at the same moment theory, 1962 Hu in IEEE transformer, reference IEEE papers over 2700 cheat, laid the theoretical basis of the image of the same moments in
Hu_moment-
- Hu不变矩。本程序用于计算图像的Hu矩特征。该程序不能在5.3版本下运行，必须在6.5以上版本中运行。-Hu moment invariants. This procedure is used to calculate the image of Hu moment features. The program can not run under version 5.3 must be running version 6.5 above.
bujianju
- 用matlab对图像的7个不变矩特征进行提取，并进行归一化。亲测可用，适合初学者。-Matlab for image seven moment invariant features were extracted and normalized. Pro-test is available for beginners.
6434esdfsdf
- 针对标准均值滤波存在的问题,提出自适应均值滤波算法。算法首先计算窗口的四阶累积量和二阶中心矩并确定噪声 点阈值 然后根据窗口内噪声点个数自适应调整滤波窗口,自适应计算权值 最后对噪声点逐点滤波。该方法既能有效去除图 像噪声点,又能较好保持图像细节。论文最后给出实验和分析,结果表明该方法是有效的-Standard mean filter exists, adaptive mean filtering algorithms. The algorithm calculates the wind
HmomenttU
- HU的七阶不变矩，具有旋转，平移不变性，对图像描述稳定，，对于图像分类，匹配具有重要的作用，已通过测试。 -HU seven-order moment invariants, rotation, translation invariance, stable image descr iption for image classification match has an important role, has been tested.
Color-Feature-of-images
- 实现了图像的颜色直方图，颜色矩，颜色相关向量CCV，以及颜色相关图的提取，有详尽的文档，基于window7+vc2010+opencv2.3.1-The color histogram of the image, color moment, color vector CCV, as well as color correlogram extraction, a detailed document based window7+vc2010+opencv2.3.1
Object-Recognition-Based-on
- 一篇图像不变性识别的好文章，基于不变矩的图像几何变换不变性识别研究，我的导师写的，大家不妨看一下啊-An image invariant recognition based on moment invariants image geometric transformation invariant recognition research, written by my mentor, you may wish to look at ah
statxture
- 产生图像的纹理信息，包括平均亮度，熵，三阶矩，一致性-Generating texture of the image information, including an average brightness, entropy, the third moment, consistency. . .
zernike
- 求zernike矩，反应图像高阶信息，比hu矩稍微好点，用opencv写的-Seeking zernike moments, reaction image high-end, slightly better than the hu moment point, written with opencv
hujv_recognization_classification
- 矩特征主要表征了图像区域的几何特征,又称为几何矩， 由于其具有旋转、平移、尺度等特性的不变特征，所以又称其为不变矩。 利用不变矩进行识别与分类-Moment feature is mainly characterized by the geometric features of the image area, also known as geometric moments, due to its rotation, translation, scaling, and other characte
feature-extracion
- 实现对图像的分形维数，7个Hu矩，目标长宽，转动惯量，周长，图像均值，标准差，最大值的特征提取-The fractal dimension of the image, seven Hu moments, target length and width, the moment of inertia, the perimeter of the image mean, standard deviation, and the maximum value of feature extraction
Sub Pixel Detection
- 使用空间矩算法实现图像亚像素边缘检测，程序精度能达到0.1个像素级别-Spatial moment algorithm to image sub-pixel edge detection
statxture
- 评价图像质量的函数 包括图像的熵、三阶矩、标准差等6个参数 代码有详细注释-Evaluation of image quality functions including image entropy, the third moment, the standard deviation of 6 parameters code detailed notes
Zernike
- Zernike矩是一种具有尺度、移位和旋转不变性的正交不变矩，本设计的目的就是利用Zernike不变矩设计一种图像检索系统，该系统能够充分验证Zerinike矩的不变性及其在图像检索中的优良性能。具体内容包括： (1) 图像特征提取、统计特征提取； (2) Zernike不变矩及其应用方法； (3) 基于Zernike不变矩的图像检索系统。 -Zernike moments is a scale, shift and rotation invariant orthogonal i
DLGPro_OpenCV
- 读入图像并显示,实现直方图均衡化,读入一幅图像, 进行直方图统计, 求图像的均值和二阶矩,读入两幅图像, 分别进行直方图统计, 以其中一个为规定化直方图实现另一图像的直方图规定化-Read into the image and displayed, histogram equalization, reads an image histogram, find the mean value of the image and second moment, and read into the two i
DLGPro_Matlab
- 《数字图像处理》课程Matlab项目框架 已测试，可适用于Matlab8.0及以后版本 Project 1. 实践主要内容： 1) 实现直方图均衡化. (3分) 读入一幅图像, 进行直方图统计, 求图像的均值和二阶矩. 2) 实现直方图规定化. 读入两幅图像, 分别进行直方图统计, 以其中一个为规定化直方图实现另一图像的直 方图规定化.-" Digital Image Processing" course the Matla
hu-bubianju
- 基于不变矩的图像识别。本程序的作用是通过训练样本，对三维图像的一帧或几帧进行特征匹配以确定图形的类别（模式识别）。-Image recognition based on moment invariants. The role of this procedure by training samples, one or several frames of the three-dimensional image feature matching to determine the category of