一种基于改进 Surendra背景更新算法的运动目标检测

建立背景模型时,采用动态阈值的方法,以减少噪声和光照强度变化等外界环境的影响;在运动目标检测中,针对传统方法检测到的运动目标偏大的问题,提出一种改进的三帧差分检测方法.在形态学处理过程中,采用自适应方法选择结构元素的大小,对二值图像进行去噪处理,以得到平滑图像.实验证明,改进后的方法与传统背景模型建立方法相比,能更准确地检测到运动目标区域,并实时更新背景模型

Moving Objects Detection Based on Improved Background Updating Algorithm of Surendra

When building background model, a method using dynamic threshold is used to reduce the influences of the external environmental factors,such as noises and brightness variations. Furthermore,the moving object detected using traditional methods is bigger than the actual object, so an improved detecting method base on three frame difference is presented to solve this problem. During the morphology operation, the appropriate size of structural element is chosen using adaptive method,and the binary image is denoised to get the smooth image. Compared with traditional methods of building background model, the experimental results show that the improved method can more accurately detects moving objects, and is for real time updating the background model.