高光谱海岸带区域分割的活动轮廓模型

近年来,高光谱遥感图像的分割作为地物识别和异常目标探测等应用的基础工作而受到重视,而高光谱遥感图像的海量数据和复杂结构使其分割技术成为一项挑战性的工作.在对海岸带高光谱遥感图像的光谱特性进行分析的基础上,提出一种基于光谱特性的海岸带水、陆区域分割的偏微分方程活动轮廓模型:首先以高光谱海岸带图像的海域像元光谱信息为参照点,构建海岸带高光谱图像的能量偏差矩阵;在此基础上建立适应该能量偏差矩阵的水、陆区域分割的活动轮廓模型.模型通过引入基于梯度的边缘引导函数,提升了对水、陆区域边缘的捕捉能力和抗噪声干扰能力.实验结果表明,与传统活动轮廓模型相比,本文模型不仅能够保证水、陆区域分割的精度,而且具有更快的计算速度.

The active contour model for segmentation of coastal hyperspectral remote sensing image

In recent years, the application of segmentation of coastal hyperspectral remote sensing image, which is as ground objects identification or anomaly target detection, are receiving more and more attention, moreover, the massive data and complex construction make the segmentation technology of coastal hyperspectral remote sensing image a challenging work. This article presents a partial differential equation active contour models based on the spectral characteristic of coastal terraqueous region segmentation for reference: First of all, construct an energy deviation matrix of coastal hyperspectral remote sensing image using pixel spectral information of coastal hyperspectral remote sensing image as the reference point; Then, on that basis, construct a terraqueous region segmentation active contour models which adapts to the energy deviation matrix, and thus improve the capture and antinoise capacities of terraqueous marginal region through inducting edge guide function which is on the based of gradient. The experimental results show that, comparing with traditional active contour models, the new model not only enhances the accuracy of terraqueous region segmentation, but also improve the calculation speed.