基于区域特征的线阵图像分割

Line-Scan CCD Image Segmentation Based on Region Feature

摘要: 图像分割是图像处理中的一个重要问题，也是一个经典难题。图像分割可以降低后续算法的难度和空间复杂度。然而，现有的图像分割方法的阈值难确定、算法复杂、耗时长等问题依然存在。本文针对线阵图像的区域特征，提出一种自适应阈值的图像分割算法。利用线阵图像相对运动和相对静止的物体在图像中差异显著的优良特性，计算线阵图像每行像素的方差。相对运动部分方差差异显著，相对静止部分方差差异较小，以此来实现对线阵图像的分割。

Abstract: Image segmentation is an important and classic problem in image processing. Image segmentation can reduce the difficulty and spatial complexity of subsequent algorithms. However, the existing image segmentation methods still have some problems such as being difficult to determine the threshold, complex algorithm, being time-consuming and others. In this paper, an image segmentation algorithm based on adaptive threshold is proposed. The variance of each row pixel of line-scan image is calculated by using the excellent characteristic that the relative moving object and the relatively stationary object differ significantly in the image. The variance difference of the relative moving part is significant, but the variance difference of the relative stationary part is small, so as to realize the segmentation of the line array image.