首页|基于视觉显著性的主动声呐背景均衡方法

基于视觉显著性的主动声呐背景均衡方法

扫码查看
针对主动声呐检测输出中的混响与辐射噪声泄漏等干扰抑制问题,提出了一种基于视觉显著性检测的背景均衡方法.首先,建立起主动声呐图像的概念,从视觉显著性统计规律上研讨其与自然图像的相似性;然后,在图像处理框架下提出一种基于二维频谱残差显著性检测的主动声呐背景均衡新方法,并引入图像熵作为均衡效果的评价依据.经海试数据验证,该方法能有效抑制混响片状干扰与辐射噪声条带状干扰,增强回波亮点,相较于常规的加权二维均值背景均衡方法图像熵值降低约 2/3,运算速度提高约一个数量级.
Active sonar background equalization method based on visual saliency
A background equalization method based on visual saliency detection was proposed to achieve interferences suppression on the reverberation and radiated noise leakage in the detection out-put of active sonar.Firstly,the conception of the active sonar image was established after investiga-ting its similarity to the natural image from the perspective on the visual saliency statistics.Secondly,a new method for active sonar background equalization was proposed based on two-dimensional spec-tral residual saliency detection in the image processing framework,and the image entropy was intro-duced as the evaluation basis for the equalization effect.Finally,verified by sea trial data,this new method can effectively suppress the sheet-interference from reverberation and the stripinterference from radiation noise,and enhance the echo-likely bright points.The entropy of result image reduces about 2/3 time with this method comparied to the conventional weighted two-dimensional mean equalization method,and its operation speed is one order higher than the latter.

visual saliencyactive sonarbackground equalizationstrip interferencereverberationsonar image

余杰、王平波、蔡志明

展开 >

海军工程大学 电子工程学院,武汉 430033

中国船舶集团公司 第 715 研究所,杭州 310012

视觉显著性 主动声呐 背景均衡 条带状干扰 混响 声呐图像

国家自然科学基金

41976177

2024

海军工程大学学报
海军工程大学

海军工程大学学报

CSTPCD北大核心
影响因子:0.34
ISSN:1009-3486
年,卷(期):2024.36(3)