Scholarlink

<p>Continuous monitoring in maritime areas using the latest remote sensing technologies is vital for the security, environmental protection, and supervision of these areas. SAR space technology has been used since 1985 to detect ships. However, the high cost, poor accessaibility, and complicated processing of SAR images limited their application. This research sought to extract the vessels and ships on the surface of international waters by processing the Sentinel-1 radar images. This research aims to detect marine targets using high-resolution radar images based on the constant false alarm rate (CFAR) method. The proposed algorithm is a combination of the CFAR algorithm and the Poisson distribution process (PDP) (i.e., PDP-CFAR). To this end, naval vessels were extracted through CFAR and the results were improved by the developed Poisson algorithm. Four different values of the k constant in the Poisson distribution were examined, where k = 4 gave the best result with an ENL index value of 6.52, a&#xa0;coefficient of variation of 0.3917, and a maximum error of 0.1306. The results have shown that the proposed algorithm is very effective in objects extraction from the Sentinel-1 radar images.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A New Method Based on CFAR and Poisson Distribution Process (PDP_CFAR) for Naval Vessels Detection Using Sentinel-1 Radar Images

  • 期刊论文

  • 发表日期：

    2025年2月25日

查看全文
  • Jafar Jafarzadeh,
  • Sara Attarchi

摘要

Continuous monitoring in maritime areas using the latest remote sensing technologies is vital for the security, environmental protection, and supervision of these areas. SAR space technology has been used since 1985 to detect ships. However, the high cost, poor accessaibility, and complicated processing of SAR images limited their application. This research sought to extract the vessels and ships on the surface of international waters by processing the Sentinel-1 radar images. This research aims to detect marine targets using high-resolution radar images based on the constant false alarm rate (CFAR) method. The proposed algorithm is a combination of the CFAR algorithm and the Poisson distribution process (PDP) (i.e., PDP-CFAR). To this end, naval vessels were extracted through CFAR and the results were improved by the developed Poisson algorithm. Four different values of the k constant in the Poisson distribution were examined, where k = 4 gave the best result with an ENL index value of 6.52, a coefficient of variation of 0.3917, and a maximum error of 0.1306. The results have shown that the proposed algorithm is very effective in objects extraction from the Sentinel-1 radar images.