搜索资源列表
NIPS14_final
- 词典是用来描述一张图片的目标区域的信息，和稀疏粒子结合，就能稀疏表达来描述目标。-A dictionary is used to describe a picture information of the target area, is combined with sparse particles, can thin expression to describe the target.
SVGDL
- 词典是用来描述一张图片的目标区域的信息，和稀疏粒子结合，就能稀疏表达来描述目标。-A dictionary is used to describe a picture information of the target area, is combined with sparse particles, can thin expression to describe the target.
test3
- VS2013+OPENCV2.4，OTSU阀值分割确定目标区域，sobel差分算子用来进行边缘检测，可知程序运行时间-VS2013+OPENCV2.4, OTSU segmentation threshold determining target region, sobel operator for performing differential edge detection, it can be seen running time
PeopleTracker-master
- 通过摄像头拍摄来监测目标区域，对目标区域的行人作实时检测-hrough the camera to monitor the target area, pedestrian for real-time detection of the target area
target_positing
- 无人机视觉系统用于飞机着陆时的自主控制，基于图像识别将目标图像的灰度信息提取，校正，最终实现目标定位，指导飞机在目标区域安全降落。-UAV vision system is used for autonomous control of aircraft landing, based on image recognition will be the target image of the gray information extraction, correction, and ultimately
GrabCut
- 用GrabCut算法实现图像的分割，有效的将目标区域从背景区域分离出来，方便有效，可以直接运行。-GrabCut algorithm to achieve the image segmentation, the effective target area the background area separated, convenient and effective, can be run directly.
test5
- 主要测试是否有目标区域的实现，如果有目标区域就上报报警-Main test whether there is target area implementation, if there is target area, report to the police
text1
- 用matlab实现用“+”标记图像中的目标区域中心-Using matlab to achieve with + mark the image of the target area center
fushihepengzhang
- 腐蚀一种消除边界点，使边界向内部收缩的过程，膨胀是将与目标区域接触的背景点合并到该目标物中，使目标边界向外部扩张的处理，利用Visual C++编写实现图像膨胀运算或者腐蚀运算的程序，-Corrosion A process of eliminating the boundary point and contracting the boundary to the inside. The expansion is a process of combining the background poin
tgrkee
- 使用meanshift加kalman预测实现的跟踪系统,目标区域为半自动选定,(Using meanshift kalman prediction to achieve tracking system, the target area for semi-automatic selected,)
mubiaoq0900
- labview 图像处理中有关提取目标区域即图像的ROI，并进行相关的区域判断。这是一些子VI。(LabVIEW image processing, the extraction of the target area, that is, the image of the ROI, and related regional judgments. These are some sub VI.)
KCFTest
- KCF是一种鉴别式追踪方法，这类方法一般都是在追踪过程中训练一个目标检测器，使用目标检测器去检测下一帧预测位置是否是目标，然后再使用新检测结果去更新训练集进而更新目标检测器。而在训练目标检测器时一般选取目标区域为正样本，目标的周围区域为负样本，当然越靠近目标的区域为正样本的可能性越大。(KCF is a novel tracking framework that utilizes properties of circulant matrix to enhance the processing s
boundaries
- 通过检测到目标区域的一个起始点来顺时针寻找图像边界上的所有点(By detecting a starting point of the target area to find all the points on the image boundary clockwise)
前景提取
- 本文展示了一种自动识别视频中移动目标的方法。论文中提取移动目标通过帧序列，这种方法不需要先验知识，比如：时间阈值调整。基于相邻帧的连续对称差分，我们能得到全分辨率显著图；然后利用最大熵方法计算阈值决定候选区域和获得兴趣点的种子；最后用修改的模糊生长方法获得最终的结果。本文中提出的算法是有效的、具有鲁棒性的。实验结果也证明它具有很好的效果。(This paper presents a method for automatic recognition of moving targets in vid
object dection
- 给出一段有关人行为的视频，视频数据来源于KTH数据库，程序可以提取到我们所关心的前景区域。(GMM,C++, object dection)
最大熵阈值法分割图像
- 使选择的阈值分割图像目标区域、背景区域两部分灰度统计的信息量为最大，最终将图像目标区域与图像背景分离(The two parts of the target area and background area are segmented by the selected threshold, and the information amount of the grayscale statistic is the largest. Finally, the target area and the ba
ga (6)
- 基于码本（codebook）的背景建模的背景差分法+级联基于LBK或haar的adaboost和基于hog的svm分类器+快速hough圆变换进行人头识别+基于区域特征的目标跟踪算法。(编程) AdaBoost是一种增强性机器学习算法(Background modeling based on codebook (codebook) background difference method + cascade based on LBK or Haar AdaBoost and hog based
regionGrowthV1.0
- 一个简单的通过区域生长实现对图像中目标边缘的提取(Edge extraction through regional growth)
keyong.zip
- 检测火焰图像颜色rgb分量，并通过背景差分提取出目标区域(Detection of color of flame image)
puju(yiwei)
- 使用谱聚类算法进行图像分割，提取目标区域(pujulei algorithm used to image cut)