摘要
针对传统运动背景补偿算法难以适应大角度旋转的应用场合,提出了一种改进的圆投影向量的背景补偿方法。该算法首先利用图像熵自动选择合适的图像参考块,以增强算法的实用性;其次对传统的圆投影向量进行了改进,增加了相位向量以获取背景的旋转参量,并采用核函数定义各像素点的距离权值,从而可以获得更加准确的结果。最后给出了算法步骤,并利用实测数据进行验证。通过仿真实验证明了该算法在各种场景、各种运动情况下的有效性以及对噪声、光照等因素具有鲁棒性。
As the traditional background compensation methods can not deal with the background's rotating,an improved method of background compensation based on circular projection is proposed.Firstly,the compatible reference block is selected based on the local image entropy so as to enhance the practicability of the algorithm.Secondly,the traditional circular projection is improved,and the physic vector is used to get the rotate parameter;in addition,the kernel function is adopted to express the distance of each pixel so as to result in a more nicety result.Finally,the algorithm's steps are given and the real images are used to test its performance.The experimental results show that the algorithm is not only suitable for various scenes and movements but also robust to gray change and noise.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2010年第6期1339-1342,共4页
Systems Engineering and Electronics
关键词
目标检测
运动背景补偿
圆投影
图像熵
核函数
target detection
moving background compensation
circular projection
image entropy
kernel function