搜索资源列表
Video-Demo
- VC++视频图像运动目标检测,视频演示算法包括: 1. 静态背景下的背景预测法目标检测 2. 静态背景下帧间差分法目标检测 3. Mean Shift目标跟踪方法 4. 重心多目标跟踪方法 视频只限于RGB非压缩Windows AVI格式-VC++ video image moving target detection, video presentations algorithm including: 1. The background prediction target d
beijingfa
- 在vc++6.0下,利用opencv函数库,使用背景差分法检测车辆,并显示背景帧,和前景帧-Use opencv library in vc++6.0, using the background subtraction method detects the vehicle, and the background frame, frame and prospects
BackSub
- 帧间差分法去背景,可以实现图片序列中的任意两帧的相减-Inter-frame difference method to the background, you can achieve any picture sequence subtraction of two
fuzabeijinggenzong
- 介绍了一种常用的Kim目标分割方法,并 针对其不足,对Kim方法进行了改进,将连续两帧的差分图像和背景差分图像直接相乘得到灰度图 像,然后对该灰度图像进行阈值分割来获取目标区域模板,再基于灰度加权图像模板匹配法实现目标 跟踪 -A common Kim target segmentation, and for its shortcomings, Kim method improvements, the two consecutive frames of the different
multitracking
- 基于OpenCV2.4.4+Visual Studio2008下的多目标跟踪代码。基于帧间差分法判断视频的背景和前景。-OpenCV2.4.4+ Visual Studio2008-based multi-target tracking code. Based on inter-frame difference method to determine the background and foreground of the video.
detection-and-tracking
- 视频演示算法包括: 1. 静态背景下的背景预测法目标检测 2. 静态背景下帧间差分法目标检测 3. Mean Shift目标跟踪方法 4. 重心多目标跟踪方法-Algorithm for video presentation include: 1. Static background background prediction target detection 2. Static background frame difference method for
video-processing-
- 视频演示算法包括: 1. 静态背景下的背景预测法目标检测 2. 静态背景下帧间差分法目标检测 3. Mean Shift目标跟踪方法 4. 重心多目标跟踪方法 -Algorithm for video presentation include: 1. Static background background prediction target detection 2. Static background frame difference method for target d
second2
- 静态背景目标跟踪。采用帧间差,去噪处理,区域生长法。还原前景。-Static background target tracking. Using inter-frame difference, denoising, region growing method. Restore outlook.
The-infrared-target-detection-method
- 本程序采用matlab编码,一共有包括帧间差分法、背景差分法、光流法、混合高斯模型法四种方法来实现红外目标检测的功能,代码检测易懂,适合初学者多多借鉴~-This procedure using matlab coding, a total including the inter-frame difference, background subtraction, optical flow method, Gaussian mixture model approach are four ways
Video-Demo
- 视频演示算法包括: 1. 静态背景下的背景预测法目标检测 2. 静态背景下帧间差分法目标检测 3. Mean Shift目标跟踪方法 4. 重心多目标跟踪方法-Algorithm for video presentation include: 1 Static background background prediction target detection 2 Static background frame difference method for target de
Background-subtraction
- 背景差针法提取目标,输出背景和前景图,帧数。-Background subtraction needle extraction of the target, the output background and foreground figure, frames.
frame-differential-method-
- 源代码在matlab实现相邻帧间差分法对于视频中对移动目标的识别。附送视频,在matlab上能打开的视频只有无压缩的AVI视频。视频背景没有变化的情况下,处理效果很好。-The source code in matlab between adjacent frame difference method for the recognition of moving target in the video. Attached to the video, can open the video on th
1
- 利用当前帧图像与背景图像对应象素点的灰度差值来检测车辆。如果当前图像的象素点和背景图像的象素点灰度值差别很大,就认为此象素点有车通过;相反,如果当前图像的象素点和背景图像的象素点灰度值差别较小,在一定的阈值范围内,我们就认为此象素点为背景象素点。-Using the current frame image and the background image corresponding to the difference between the pixel gray vehicle is detec
zhenchajianfa
- 基于帧间差分法的静态背景下目标的识别与追踪,源码包括视频的读取与回放以及二值化后的图像处理。-Based on the identification and tracking under static background frame difference method goals, including the source and playback of video and read the binarized image processing.
Video-Demo
- 静态背景下背景预测法目标检测,静态背景下帧间差分法目标检测,Mean Shift方法目标跟踪方法-Background prediction target detection under static background, frame difference method for target detection under static background, Mean Shift method for target tracking method
gmm
- 混合高斯模型使用K(基本为3到5个) 个高斯模型来表征图像中各个像素点的特征,在新一帧图像获得后更新混合高斯模型,用当前图像中的每个像素点与混合高斯模型匹配,如果成功则判定该点为背景点, 否则为前景点。通观整个高斯模型,他主要是有方差和均值两个参数决定,,对均值和方差的学习,采取不同的学习机制,将直接影响到模型的稳定性、精确性和收敛性。由于我们是对运动目标的背景提取建模,因此需要对高斯模型中方差和均值两个参数实时更新。为提高模型的学习能力,改进方法对均值和方差的更新采用不同的学习率 为提高在繁忙
dangegenzong
- 通过背景差分法实现单个物体的跟踪,可以在物体外围加矩形框,但第一帧必须没有运动物体-By implementing tracking a single object in the background difference method, you can add the object rectangle perimeter, but the first frame must be no moving objects
BackGroundTest
- 帧间差分法:一种比较简单的目标分割方法,在静态背景下,先通过视频序列建立相应的背景图,然后利用当前图像与背景图像的差分来检测运动区域,并进行膨胀、腐蚀等操作,从而提取出运动目标。-Frame difference method: a relatively simple object segmentation, in the static context, first established by video sequences corresponding background image, an
m13
- 针对户外视频监控存在光照变化这一问题, 提出一个用于准确完成目标检测的实时背景建模框架. 考虑到目标检测 的准确性要求, 建立基于帧间像素亮度差统计直方图的像素亮度扰动阈值. 在此基础上, 针对背景建模的实时性要求, 提出一种基于自回归背景模型的参数快速更新方法. 鉴于不同光照变化的适应性要求, 定义对光照变化不敏感的背景纹理模型. 上述模型统称为自回归{ 纹理(Auto regression and texture, ART) 模型, 该模型适应于户外光照变化. 基于该模型构建像素亮度和纹
beijingPzhenjian
- 实现surendra背景差分,并和帧间差分比较效果-Achieve surendra background difference, and the comparative effectiveness of inter-frame difference