模糊聚类算法在船舶图像分割中的应用

Application of fuzzy clustering in ship image segmentation

传统船舶图像分割方法存在分割误差大,抗噪声干扰能力差、分割效率低等缺陷,为了解决传统船舶图像分割方法存在的不足,设计了基于模糊聚类算法的船舶图像分割方法。首先对当前船舶图像分割研究进展进行分析,指出不同传统船舶图像分割方法存在的局限性,然后对船舶图像进行去噪处理,提高船舶图像质量,改善抗噪声干扰能力,最后引入模糊聚类算法进行船舶图像分割,并采用多幅标准船舶图像与传统船舶图像分割方法进行对比测试。测试结果表明,本文方法可以对船舶图像进行高精度的准确分割,能够保留船舶图像边缘的重要信息,船舶图像分割速度可以满足实际应用的要求,获得了比传统船舶图像分割方法更优的结果,具有更加广泛的应用范围。

Traditional ship image segmentation methods have some shortcomings,such as segmentation error,poor anti-noise ability and low segmentation efficiency.In order to solve the shortcomings of traditional ship image segmentation methods,a traditional ship image segmentation method based on fuzzy clustering algorithm is designed.Firstly,the research progress of ship image segmentation is studied,and the limitations of different traditional ship image segmentation methods are pointed out.Then,the ship image is denoised to improve the quality of ship image and anti-noise ability.Finally,a fuzzy clustering algorithm is introduced for ship image segmentation,and multi-scale ship image and traditional ship image segmentation methods are used for ship image segmentation to test.This method can segment the ship image accurately and accurately.It can retain the important information of the edge of the ship image.The speed of the ship image segmentation can meet the requirements of practical application.It has obtained better performance than the traditional ship image segmentation method,and has a wider range of applications.