强伪装下的航空图像边缘分割方法仿真

Edge Segmentation Method Simulation of Aviation Image in Strong Guise

研究强伪装下的航空图像边缘准确分割问题。随着伪装技术的广泛应用,航空图像中的背景色彩已经高度接近目标,造成目标和伪装物在像素色差、灰度等特征上高度一致。传统的分割算法,在分割航空图像特征时,会随着灰度的接近陷入无穷分割的境地,造成阀值设定过程无穷化,很难保证完整分割出特征。提出了一种曲线演化模型的航空图像边缘分割方法。利用P&M算法,对采集的航空图像进行平滑滤波处理,从而去除航空图像中的噪声,提高图像的分辨率。建立曲线演化模型,将航空图像的监测目标中的能量进行由粗到细的演变,从而扩大图像边缘捕获区域,使其对捕获过程更加敏感,对强伪装下的航空图像边缘进行分割处理。实验结果表明,算法进行强伪装下的航空图像边缘分割,能够在航空图像中准确的提取边缘区域,为航空图像应用领域提供了决策依据。

The accurate segmentation of aviation image edge in the strong guise was researched.In this paper,we presented a curve evolution model for aviation image edge segmentation.By using P ＆ M method,the collected air images were processed through a smoothing filter,thereby removed the noise in the image and improved the image resolution.In addition,we established a curve evolution model,made the energy evolution of monitoring target in a viation images from coarse to fine,thereby expanded the edges of the image capture area,made it more sensitive for the capture process,performed image segmentation of the edges of the aviation images in the strong guise.The experimental results show that the algorithm used in aviation image edge segmentation in the strong guise can extract the edge region accurately from the aviation image.It also provides a basis for decision making in the aviation image application areas.