【摘要】 提出了一种具有全局运动的视频运动对象时空联合检测算法。针对传统时间分割使用主观固定阈值的缺点,采用了对差分图像进行噪声参数自适应学习的算法获取自动阈值,并利用形态学运算获取修正的时间分割模板;考虑传统分水岭空间分割的不足,提出了基于人眼视觉特征的改进分水岭算法,包括基于形态重建滤波的图像降噪、形态梯度变换以及基于韦伯感知原理的视同灰度非线性变换,有效地解决了过分割问题;对时、空间分割结果进行信息融合处理,从而得到完整的运动对象。仿真实验结果表明,本文算法可以快速准确地分割视频运动对象。更多还原

【Abstract】 A new method to detect moving objects from video sequences with global motion is proposed based on spatio-temporal joint detection.In order to overcome the drawback of subjective fixed threshold of traditional temporal segmentation,the characteristic parameters of background noise in difference image are adaptively evaluated and the automatic threshold is obtained.Then the temporal segmentation mask is amended by morphological operations; Considering the lack of traditional spatial segmentation algorithm of watershed,an improved watershed algorithm in accord with the human vision characteristics is proposed,which includes the noise reduction by morphological reconstruction,gradient transformation and gray level non-linear transformation based on Web perception principle.The improved watershed can restrain over-segmentation effectively;the temporal and spatial information fusion is fulfilled by ratio operation,and the video moving object is obtained.The emulation experiments show that the proposed algorithm can detect the moving object quickly and effectively.更多还原