论文部分内容阅读
简要介绍了有关独立分量分析的基本理论和算法 ;探讨了独立分量分析在序列图象处理方面的应用 ;提出了基于独立分量分析的运动目标检测新方法 .同时用独立分量分析方法对含有运动目标的序列图象进行了独立分量分离的试验 ,试验中 ,首先获取序列图象的独立分量和模型混合矩阵 ,然后将含有背景干扰的独立分量置零 ,并用混合矩阵进行逆运算 ,从而获得了非常清晰的运动目标轨迹 .试验结果表明 ,这种独立分量分析方法具有良好的盲源分离性能 ,而且在运动目标检测等方面 ,基于独立分量分析的检测方法较传统的检测方法更有效
The basic theories and algorithms of independent component analysis are briefly introduced. The application of independent component analysis in sequence image processing is discussed. A new method of moving object detection based on independent component analysis is proposed. In the meantime, the method of independent component analysis In this experiment, the independent components of the sequence image and the model mixing matrix are obtained. Then, the independent components containing the background noise are set to zero and the inverse operation is performed by using the mixed matrix, thereby obtaining the very Clear trajectory of moving target.The experimental results show that this independent component analysis method has good performance of blind source separation and the detection method based on independent component analysis is more effective than the traditional detection methods in the detection of moving targets