共查询到20条相似文献,搜索用时 314 毫秒
1.
陈佳 《南阳师范学院学报》2008,7(6)
给出一种基于眼睛特征图的眼睛定位方渚。首先通过肤色检测得到人脸区域,然后在限定的人脸区域内建立眼睛色度图,并利用眼睛在人脸区域内的几何分布对眼睛区域进行局部的光线补偿,突出眼睛的色度信息,得到眼睛的二值图像,最终利用灰度投影算法定位眼睛。 相似文献
2.
给出了一种用于固定场景的快速人脸检测方法。在背景差的情况下,利用SHEN滤波进行二值图像并提取出运动目标区域,基于肤色聚类模型对运动目标子图像进行肤色分割,结合椭圆模板检测出人脸区域,进行人脸定位。实验数据表明,该方法对固定场景中的人脸检测具有较好的鲁棒性,对姿态、表情、年龄都有较强的适应性,并具有一定的实时性。 相似文献
3.
提出一种基于肤色模型的快速人脸检测方法,能同时平衡算法速度和精度要求.首先对图像进行肤色区域提取,然后对肤色区域进行基于偏微分方程的形态学处理,对区域形状进行大小粗过滤;接着对粗过滤后的肤色区域对应的原图灰度图运用人眼的几何特征进行人眼粗定位;最后提出一种相关系数判断的方法精确定位人眼,从而确定出人脸区域. 相似文献
4.
介绍了一种简单、快速的实时视频人脸对象跟踪算法.在使用肤色检测算法进行人脸区域定位的基础上,采用差分运动估计,确定最大的搜索区域,在此区域使用两个正交的边缘梯度跟踪模型和内部肤色像素统计直方图跟踪模型跟踪人脸.本算法较好地解决了跟踪的连续性、实时性和精确度间的矛盾.实验结果表明本算法能够对人脸进行实时跟踪,具有一定实际应用价值. 相似文献
5.
6.
介绍了一种基于肤色模型与改进Adaboost算法相结合的人脸检测方法。该方法首先利用肤色在YcbCr空间中的聚类特性,对图像进行预检测,得到候选人脸区域,进而采用改进的Adaboost方法对弱分类器进行级联,得到最终的人脸分类器。在特征选择上,使用基于像素的多层特征(Pixel Based Hierarchical Feature,PBHF),以解决传统Adaboost方法检测时间过长的问题。实验结果表明,该人脸检测方法比单纯采用Haar like 特征的人脸检测方法更加有效。 相似文献
7.
提出了基于肤色建模和眼睛亮度检测的方法对彩色图像中的人脸进行检测。在检测前,先对图像进行光线补偿,再通过肤色模型获得可能的脸部区域,最后根据眼睛在人脸固有位置亮度检测人眼,最终确定人脸区域。通过实验测试说明,该方法对人脸的检测达到了较好的效果。 相似文献
8.
9.
10.
研究了一种基于正面人脸照片的真实感三维人脸自动重建方法,并运用计算机视觉图OpenCV和图形开发库OpenGL,在VC++6.0环境下开发了三维人脸自动建模系统。该系统对输入的人脸照片首先进行人脸检测,在检测到的区域进行人脸关键特征提取,并根据检测到的特征点的几何信息对CANDIDE-3模型进行整体和局部调整,得到个性化的三维几何人脸,最后从人脸图像上获取面部纹理信息并得到真实感的三维人脸。 相似文献
11.
A closed-loop algorithm to detect human face using color information and reinforcement learning is presented in this paper.
By using a skin-color selector, the regions with color “like” that of human skin are selected as candidates for human face.
In the next stage, the candidates are matched with a face model and given an evaluation of the match degree by the matching
module. And if the evaluation of the match result is too low, a reinforcement learning stage will start to search the best
parameters of the skin-color selector. It has been tested using many photos of various ethnic groups under various lighting
conditions, such as different light source, high light and shadow. And the experiment result proved that this algorithm is
robust to the varying lighting conditions and personal conditions.
Project supported by the National Natural Science Foundation of China (No. 60105003) and Zhejiang Provincial Natural Science
Foundation of China (No. 600025). 相似文献
12.
对对称度公差值的计算公式进行了推导;对对称度公差应用最大实体原则时的计算进行了分析,对不能直接测量出对称度公差值的零件面对线的对称度,按最小条件用相似三角形进行计算的方法进行了分析。 相似文献
13.
A framework of region-based dynamic image fusion 总被引:1,自引:0,他引:1
WANG Zhong-hua QIN Zheng LIU Yu 《浙江大学学报(A卷英文版)》2007,8(1):56-62
INTRODUCTION Image fusion is a sub-area of more general topics of Information Fusion and Data Fusion and deals with static images and dynamic images (image sequences or video data) (Maitre and Bloch, 1997). By com-bining the information from a range of image data obtained by different sensors or the same sensor with different imaging schemes, the composite image al-ways presents the complementary information of different input images at the same time and provides us the integrated descr… 相似文献
14.
视频关键帧提取是视频信号处理中的一个重要内容。由于一个镜头中视觉内容的变化具有连续性,本文采用了距离累加的算法;同时,为了提高检测相邻帧间的相似度,本文将颜色特征与运动变化信息相结合,提出一种关键帧提取方法。实验证明,与过去的关键帧提取方法相比,本文算法提取的关键帧能较完整地表现序列图像的运动过程,更有利于对视频内容的理解。 相似文献
15.
For the realtime classification of moving vehicles in the multi-lane traffic video sequences, a length-based method is proposed. To extract the moving regions of interest, the difference image between the updated background and current frame is obtained by using background subtraction, and then an edge-based shadow removal algorithm is implemented. Moreover, a thresholding segmentation method for the region detection of moving vehicle based on location search is developed. At the estimation stage, a registration line is set up in the detection area, then the vehicle length is estimated with the horizontal projection technique as soon as the vehicle leaves the registration line. Lastly, the vehicle is classified according to its length and the classification threshold. The proposed method is different from traditional methods that require complex camera calibrations. It calculates the pixel-based vehicle length by using uncalibrated traffic video sequences at lower computational cost. Furthermore, only one registration line is set up, which has high flexibility. Experimental results of three traffic video sequences show that the classification accuracies for the large and small vehicles are 97.1% and 96.7% respectively, which demonstrates the effectiveness of the proposed method. 相似文献
16.
A color based system using multiple templates was developed and implemented for detecting human faces in color images. The
algorithm consists of three image processing steps. The first step is human skin color statistics. Then it separates skin
regions from non-skin regions. After that, it locates the frontal human face(s) within the skin regions. In the first step,
250 skin samples from persons of different ethnicities are used to determine the color distribution of human skin in chromatic
color space in order to get a chroma chart showing likelihoods of skin colors. This chroma chart is used to generate, from
the original color image, a gray scale image whose gray value at a pixel shows its likelihood of representing the skin. The
algorithm uses an adaptive thresholding process to achieve the optimal threshold value for dividing the gray scale image into
separate skin regions from non skin regions. Finally, multiple face templates matching is used to determine if a given skin
region represents a frontal human face or not. Test of the system with more than 400 color images showed that the resulting
detection rate was 83%, which is better than most color-based face detection systems. The average speed for face detection
is 0.8 second/image (400×300 pixels) on a Pentium 3 (800 MHz) PC.
Project (No. 60203013) supported by National Natural Science Foundation of China 相似文献
17.
窦仲四 《淮南职业技术学院学报》2010,12(4)
以皖北五沟矿井1011采掘工作面的测井钻孔资料及岩土试验报告为依据,按照该采掘面的实际情况,按照一定建模原则,设计两个计算模型:走向模型Ⅰ和倾向模型Ⅱ,分析开采后上覆岩层移动演化特征,分别从覆岩横向、竖向移动规律进行探讨,从而预测冒落带的分布、导水裂隙带的高度。 相似文献
18.
近年来Adaboost算法被成功地用于人脸检测中,本文给出了一种基于加权最小平方误差boosting算法的人脸检测。首先本方法在每一次循环中用加权最小平方误差准则训练弱假设,与原始Adaboost算法不同的是弱假设的生成不仅用于预测分类,而且用于估计每次预测的自信率,然后由这组合自信率的弱假设集成构造出强分类器。实践表明基于加权最小平方误差boosting算法的分类器有较高的检测率和较低的正样本误检率。 相似文献
19.
生成运算及其在证明多元对称不等式中的应用 总被引:11,自引:6,他引:11
刘保乾 《广东教育学院学报》2005,25(3):10-14
证明多元对称多项式不等式的生成法和这种方法下的生成运算;定义差分代换、对称多项式不等式取等的比值和扩展基本不等式等重要概念;发现若干含参多元对称不等式,并给出了应用方面的例子. 相似文献
20.
针对跳水运动视频中的人体运动目标分割问题,提出一种简单有效的基于颜色特征运动目标检测及分割算法。首帧进行图像预处理,利用色相和亮度两个分量加强颜色差异,然后使用OTSU算法对图像进行阈值分割。结合RGB颜色空间启发式肤色聚类结果确定运动目标所在的连通区域,最终完成首帧运动目标的自动检测。有效地克服了复杂背景变化的影响,能够快速实现跳水运动目标的分割,且对运动对象的快速整体运动有较强的鲁棒性。 相似文献