采用自适应高斯混合方法为背景建模的难点是对背景模型的维持与更新。
Taking the method of adaptive Gaussian mixture method can make model for background meanwhile it is a difficult point to maintain and update background model.
该算法使用混合高斯模型表示粒子，在每个时刻的修正步骤之后，采用EM算法对粒子进行重新拟合。
It USES Gassian mixture model to represent particles and adopts EM algorithm to refit particles after correction step at each time.
为了解决传统高斯混合模型（GMM）对初值敏感，在实际训练中极易得到局部最优参数的问题，提出了一种采用微粒群算法优化GMM参数的新方法。
The traditional training methods of Gaussian Mixture Model(GMM) are sensitive to the initial model parameters, which often leads to a local optimal parameter in practice.
对于静态背景层，采用基于颜色特征的个数较少的混合高斯模型对背景建模；
A Gaussian Mixture Model based on color feature is adopted in static layer.
对于GMM模型，采用高斯混合数为64时有较好的识别率。
For the model of GMM, there is a good result for choosing 64 mixtures GMM.
由于室内存在多种物体，背景不断变化，且光照条件可能不断变化，提出采用人脸肤色的标准混合高斯模型与人眼特征相结合的人脸检测法，无需对原始图像进行尺度变换。
An effective and fast method of face detection for a service robot is proposed, which combines a mixture of Gaussian distribution model of skin tone color with eyes features.
由于室内存在多种物体，背景不断变化，且光照条件可能不断变化，提出采用人脸肤色的标准混合高斯模型与人眼特征相结合的人脸检测法，无需对原始图像进行尺度变换。
An effective and fast method of face detection for a service robot is proposed, which combines a mixture of Gaussian distribution model of skin tone color with eyes features.
应用推荐