以绝对位移帧差作为时域分割的标准；
Temporal segmentation is processed by the absolute displaced frame difference instead of changing detecting mask.
重点研究了三帧差分法和混合高斯模型背景差分法。
The three Frames subtraction and the background subtraction of Gaussian mixture model are researched.
车辆检测常用的有帧差法、背景差法和边缘检测法等。
There are usually three approaches for vehicle detection, that is frame differencing, background differencing and edge detection.
结果显示该算法明显优于帧差法、背景差法、路面标记法。
The result shows this algorithm is better than frame difference method, background difference method and road mark method.
为此，文章提出并实现了一种基于互帧差的视频运动对象分割方法。
This paper proposes and implements a moving object segmentation method, which based on inter-frame difference.
三帧差分法比较简单，对环境的适应能力强，但是检测到的运动目标不精确；
The three Frames subtraction is relatively simple, which has strongly adaptability to the environment. However, the motion objects detected are inaccurate.
其基本思想是：首先根据多帧帧差信息，基于高阶统计检测方法提取运动区域。
The steps are as follows: First, motion area is extracted based on high-order statistics detection method and motion information of multiple frames.
对于待跟踪的运动目标，采用三帧差和区域增长法分割目标并得到主颜色信息。
We used frame difference and region growing method to divide target and get main color information.
其次在相邻帧差分边缘和当前帧边缘的基础上，检测出视频对象的初始边缘模板。
On the basis of difference edge mask and current edge mask, a binary edge model of the video object is initialized.
通过改进的三帧差分法计算运动目标的位置，控制X、Y轴电动位移台跟踪目标。
X axis and Y axis electric moving stages are driven to track object based on its position which is calculated by improved 3 frame difference algorithm.
为了克服了减背景法对光照变化敏感的不足，再结合帧差法对运动目标进行检测。
In order to overcome the lack of sensitivity of light change, combining the adaptive background subtraction with the symmetrical differencing obtains the integrity foreground image.
将边缘检测法与改进帧差法进行有机结合，有效地实现了运动和静止车辆的同时检测。
It combines organically the edge detection method and the improved frame difference method to effectively detect the moving and stagnant vehicles simultaneously.
结合实际拍摄的动态交通图像，提出了基于数学形态学和帧差法的交通图像噪声滤除方法。
A noise-filtering algorithm of real traffic video images based on mathematical morphology and frame difference was proposed.
基于差分的火焰识别方法是根据运动目标的速度进行相邻帧或者隔帧差分以得到运动信息。
The method of flame recognition based on image difference detects the moving information by differencing two sequential frames or every two frames according to the speed of object.
采用简单的三帧差背景剪除策略检测运动目标，合并运动估计和背景补偿以加快系统反应速度。
A simple strategy named 3 frames difference background subtraction is adopted to detect moving object, and movement estimation and compensation are combined to improve processing speed.
描述了采用差分图像法实现对谷物害虫检测判断和三帧差分法实现谷物害虫图像恢复与提取的方法。
A method of using difference images to detect grain pests was described and three - image difference was adopted to re-store and extract the pest images.
在介绍相邻帧差分算法的基本原理的基础上，提出了一个改进的相邻帧差分算法，并给出了实验结果。
Based on adjacent frame difference, we proposed a improved algorithm. An experiment result was given to show this method is efficient.
因此，本文算法首先利用累积帧差，弥补了两帧帧差所得边界的不完整性，得到准确的运动对象的定位；
So in this method, Firstly, using accumulating frame difference, we obtain the exact moving object localization which makes up the incompletely of the edge gotten by the frame difference.
提出了利用三帧差分法和背景差分法对运动目标进行检测的方法，该运动检测的方法实现比较简单、速度快。
The moving target detection method based on the three frame difference and background difference method was proposed, which is simply and fast to realize.
该方法取运动角点为特征量，通过选择灰度相关函数为特征匹配函数，实现了车辆在一定帧差内移动距离的测量。
Motion corner points are chosen to form a feature vector, and the image intensity is used as a feature matching function to measure inter-frame movement of the vehicle.
通常的镜头检测方法都是利用帧间差来计算镜头边界的检测阈值，该文提出了一种基于二次帧差的突变镜头检测方法。
Shot detection method often USES difference between frames to compute boundary detection threshold. In this paper, a quadratic difference based method for abrupt shot detection is presented.
给出连续图像帧差分和二次帧差分改进的图像HSI差分模型，采用自适应分割算法能在任意条件下自动提取运动目标区域。
It has been focused in compute vision research fields. A improved HSI image difference model based on sequences image difference and second image difference is presented.
介绍了在视觉监控领域经常用到的几种运动人体检测算法，如背景减除法、帧差法、对称差分法、基于RGB图像的运动检测法。
The paper introduces several methods for vision-surveillance using background subtraction, frame-difference method, symmetrical differencing and a method based on RGB image.
首先，对待合成视频，通过全局运动估计与补偿，将相邻帧在空间上对齐到当前帧。通过计算帧差，得到当前帧中的运动前景区域信息。
First, the information of the foreground areas in current frame is obtained with global motion estimation and image registration between current frame and temporally adjacent frames.
算法二是针对两帧差分图像之间所提供的差异信息过多或过少这两种极端情况，用差影法粗分割对人脸图像进行预处理时失效而提出来的。
The second algorithm is proposed because the method of difference image is not effective when the different information coming from two frames of difference image is too much or too little.
在对视频监控中运动目标检测识别常用算法进行研究的基础上，本文提出了一种新的基于两帧差分法和背景差分法相结合的运动目标检测方法。
According to the study of main motion detection and moving objects extraction methods, a new method combining frame-difference and background-difference was put forward.
背景差法和帧间差法都是常用的运动目标检测方法。
Background subtraction and frame difference method are commonly used moving objects detection methods.
对于司机面部器官的检测，首先要确定司机的脸部，我们采用帧间差分法和肤色检测相结合的方法。
So the integrated method of differential frames arithmetic and complexion detection can make an instant and exact orientation of the driver's face.
传统的帧间差分法算法简单、便于硬件实现，但在信噪比较低时抗噪性能较差。
Considering the characteristics of infrared image, the frame difference algorithm is simple and easy to be realized by hardware.
传统的帧间差分法算法简单、便于硬件实现，但在信噪比较低时抗噪性能较差。
Considering the characteristics of infrared image, the frame difference algorithm is simple and easy to be realized by hardware.
应用推荐