图像智能化的目标检测技术(Ⅱ)--数据流程与背景感知

着重探讨了线性预测估计和非线性均值滤波两种常用背景估计技术和自适应空间滤波的形态学背景感知方法.形态学背景感知方法从精细刻画图像的结构特征并基于形状细节进行图像分析入手,将使操作过程得以从近似的线、面传统模式中解脱出来,面向更为真实和丰富的自然结构,改善了处理效果.实验测评表明,形态滤波算法对背景的自适应感知能力最强,输出结果能真实贴切地反映图像背景的起伏变化规律,因而具有抑制杂波噪声和增强目标信号的双重功效.

Image Intelligent Technology of Target Detection(Ⅱ)--Data Flow Chart and Background Estimating

We approach mostly two kinds of common background estimate techniques that include linear predictive estimating and nonlinear mean filtering, and morphological estimating algorithm for adaptive space filtering. Experimental results show that adap-tively perceptive ability of the morphological estimating algorithm is most effective. Its output information can be consistent with the fluctuant change status of image background. Thus, it is helpful to suppress clutter and to enhance signal intensity of target. Finally, the paper emphasizes that mathematical morphology is a powerful tool and has opened new avenues for research in the fields of signal and image processing because of a most attractive feature of morphology being well suited to capturing geometric information.