针对视频帧图像背景复杂、字体大小变化较大等特点，提出一种改进的视频帧中文本区域定位算法。
Contraposing the characters of images in video, such as complicated backdrop and big diversification of font, this paper proposes an improved algorithm for text region localization in video.
提出一种在检测到人脸区域的前提下，对人眼进行准确定位的算法。
An eye location algorithm was developed on the premise of that the face region has been detected.
实验结果表明，该算法不仅具有良好的定位篡改区域功能，而且能够有效地克服“拼贴攻击”。
The experimental results have demonstrated that this algorithm can locate the tampered area as well as detecting "collage-attack" to the watermarked image.
针对车牌区域内字符串具有丰富的纹理特征，本文提出了一种基于行扫描的车牌定位算法。
An algorithm of license plate location based on line scanning is proposed according to the abundant texture features in the domain of vehicle license plate.
由于减弱了背景区域的干扰，该算法具有较高定位精度和较低时间复杂度。
The algorithm has high location accuracy and low time complexity due to weakening the interference of background areas.
针对车牌区域难以定位的问题，提出了一种基于固定颜色搭配的车牌定位算法。
Aiming at the difficulty in locating vehicle plate, an algorithm of vehicle plate's location based on fixed color collocation is put forward in this paper.
利用该算法，本文给出了罐号区域定位部分试验结果。
Some experimental results of character region location are given.
本文针对该篡改技术，提出一种有效快速的检测与定位篡改区域算法。
In this paper, we describe an efficient and robust algorithm for detecting and localizing this type of malicious tampering.
然而，用现有的人脸特征点定位算法进行人脸形状估计时，嘴巴区域特征点的定位误差相对较大。
However, when estimating facial shapes using current facial landmarks detecting methods, the locating error of feature points around the mouth region is relatively large.
在定位分割出上半人脸运动单元子区域图像之后，提出了采用KPCA算法提取它们的特征。
After upper facial action unit location and segmentation, we present the facial action unit feature extraction algorithm based on KPCA.
首先利用一个基于颜色的两步定位算法，利用增强了的R分量R1信息同时在照片四个角区域进行搜索以确定出四个可能包含日期的候选矩形区域。
The R component of the photo image is enhanced to the R1 component so that four potential regions can be found in the four original corner areas.
跟踪算法分为两步进行，首先应用均值漂移算法实现手部区域的定位，然后基于自适应活动轮廓模型提取手部轮廓。
The proposed method consists of two steps: hand location using mean shift and hand extraction based on adaptive active contour model.
本文采用该算法训练人脸面部特征检测器，提取区域的几何中心进行面部特征点定位，获取以双眼和嘴巴为特征顶点的人脸特征三角形。
The thesis uses the method to train face feature classifier, to get the face feature points and build face feature triangle that uses eyes and mouth as vertexes.
结果本方法在边缘区域的定位效果优于二维最大似然定位算法。
Results Experiments showed that the positioning result of this method was better than those of two-dimensional ML.
该算法将边缘检测应用于区域生长规则中，充分发挥了边缘方法和区域生长方法两者的优点，实现了三维曲面边界的精确定位。
In this way, this proposed algorithm takes merits of the combination of edge detection and region growth to realize the accurate boundary location of 3d curved surface.
实验表明，该算法可以提取视频帧中90%以上的文本区域，并且文本区域定位准确率达92.0%。
The experiments show that the algorithm can extract more than 90% of text regions contained in video frames and the precision can reach up to 92.0%.
实验结果表明。该算法不仅能够精确定位图像中旋转篡改的区域，而且能估算出区域旋转角度。
Experimental results demonstrate that the proposed approach can not only localize the duplicated regions accurately, but also evaluate the rotation Angle effectively.
实验结果表明。该算法不仅能够精确定位图像中旋转篡改的区域，而且能估算出区域旋转角度。
Experimental results demonstrate that the proposed approach can not only localize the duplicated regions accurately, but also evaluate the rotation Angle effectively.
应用推荐