光学学报, 2019, 39 (10): 1001002, 网络出版: 2019-10-09   

基于颜色空间融合与上下文显著性的红外偏振图像目标增强 下载: 1377次

Target Enhancement of Infrared Polarization Image Based on Color Space Fusion and Context-Aware Saliency
作者单位
海军航空大学航空作战勤务学院, 山东 烟台 264000
摘要
针对低对比度、岸岛背景、前景遮挡等复杂背景下前视红外图像在检测识别海面弱小目标困难的问题,提出一种基于红外偏振图像的海面弱小目标增强方法。开发了一种符合人视觉系统感知色彩的改进色调-饱和度-强度(HSI)颜色空间融合算法,融合海面图像中的红外偏振度信息与红外强度信息;设计一种基于红外偏振图像的海面区域分割方法,分割海面区域,并将其作为候选目标增强区域;利用上下文显著性算法计算海面HSI颜色空间融合图像的显著性,根据显著图修正海面HSI颜色空间融合图像,得到目标增强图像。采用目标与背景对比度和局部信噪比评价融合图像增强效果。结果表明,与现有方法相比,所提方法在不同场景下能增强目标并抑制背景干扰,评价指标高于现有方法,能够为海面舰船目标检测与识别提供支持。
Abstract
It is difficult to detect and identify dim and small targets on the sea surface in complex backgrounds, such as low contrast, shore island background, and foreground occlusion. This study proposes a method based on infrared polarization images to detect dim and small targets on the sea surface. Herein, we propose an improved hue-saturation-intensity (HIS) color space fusion algorithm based on the perceived color of the human visual system, which combines infrared polarization information and infrared intensity information of sea surface image. We design a sea surface region segmentation method based on the infrared polarization image, and the sea surface region is segmented as a candidate target enhancement region. The context-aware saliency algorithm is used to calculate the saliency of the sea surface HSI color space fusion image. The sea surface HSI color space fusion image is corrected using the saliency map to obtain a dim and small target enhanced image. The contrast of target and background and the local signal-to-noise ratio are used to evaluate the features of the fusion enhanced image. Results obtained show that in comparison with the existing methods, the proposed method can enhance dim and small targets and suppress the background interference. The evaluation index of the proposed method is higher than that of other existing methods, which provides support for the detection and identification of surface ship targets.

宫剑, 吕俊伟, 刘亮. 基于颜色空间融合与上下文显著性的红外偏振图像目标增强[J]. 光学学报, 2019, 39(10): 1001002. Jian Gong, Junwei Lü, Liang Liu. Target Enhancement of Infrared Polarization Image Based on Color Space Fusion and Context-Aware Saliency[J]. Acta Optica Sinica, 2019, 39(10): 1001002.

本文已被 1 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!