-
无先验水下主动偏振成像方法能够实现目标信息光偏振度和后向散射光偏振度的自动获取, 但该方法在反演过程中仅追寻高对比度这一单一指标, 有时会导致自动获取的两项偏振反演参数过于相近, 使图像复原效果不理想, 且常伴有大量噪声. 针对上述仅追求单一指标导致复原图像质量不理想的问题, 本文提出一种基于双层多指标优化的水下偏振成像方法. 首先, 第1层以互信息和对比度为目标函数, 基于多目标遗传优化算法自动获取偏振参数最优解集; 其次, 选择信息熵为第2层目标函数, 遍历最优解集, 获取偏振参数最终解, 并将其代入成像模型, 获取复原图像; 最终, 根据所获偏振参数之差, 选取适当数字图像处理手段进一步提升复原图像质量. 实验结果表明, 无论背景区域存在与否, 无论目标物偏振度高低, 本方法均能有效增强图像细节, 平衡各项图像质量评价指标, 得到综合质量较高的复原图像.
Underwater imaging is of great significance in exploring seabed resource , monitoring marine environment, implementing underwater rescue and military reconnaissance, etc. by providing clear vison. Among various underwater imaging techniques, the polarization imaging is considered to be an effective way to improve the quality of underwater imaging. It can realize underwater image restoration by using the difference in polarization characteristic between the target light and backscattered light. A classical underwater active polarization imaging method was presented by Treibitz [Treibitz T, Schechner Y Y 2009 IEEE Trans. Pattern Anal. Mach. Intell. 31385 ], in which the degrees of linear polarization (DoLPs) of target light and backscattered light are used to recover clear image. A variety of improved methods have been derived from this, but most of them require background areas and human-computer interaction. Then, a new underwater active polarization imaging method without prior knowledge was presented by Zhao [Zhao Y, He W, Ren H, Li Y, Fu Y2022 Opt. Lasers Eng. 148106777 ], in which the DoLPs of target light and backscattered light can be automatically obtained without background region. However, sometimes the above two parameters are very close and thus introduce a lot of noise into the restored images, for this method takes only the contrast into account.In this work, an underwater active polarization imaging method based on two-layer multi-index optimization is proposed. First, the mutual information and contrast are taken as the upper objective functions, and the Pareto optimal solution set is obtained by the multi-objective genetic optimization algorithm. Second, the information entropy is taken as the lower objective function to obtain the optimal parameters from this optimal solution set. Based on the optimal parameters, the restored images are obtained. According to the difference between the DoLPs of target light and backscattered light, these restored images are further improved by the digital image processing method. The experimental results indicate that our method can not only enhance image details effectively but also balance various evaluation indexes of the imaging quality to obtain high-quality restored images. The proposed algorithm is suitable for underwater targets with low and high DoLPs, with or without background regions. [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] -
Index (a) (b) (c) (d) MI(B,S) 2.529 2.969 2.062 3.093 C(S) 0.203 0.279 0.067 0.306 Entropy(S) 4.834 4.695 4.455 4.953 
下载:
导出CSV
Parameter Toy Ruler Cube-coin Treibitz Zhao Our Treibitz Zhao Our Treibitz Zhao Our Pobj 0.190 0.249 0.248 1 0.503 0.501 1 0.668 0.649 Pscat 0.291 0.271 0.272 0.479 0.495 0.445 0.659 0.651 0.586 |Pdif| 0.101 0.022 0.024 0.521 0.008 0.056 0.341 0.017 0.063 下载:
导出CSV
Image SNR Entropy Contrast NIQE Toy-Intensity 15.607 5.373 0.033 9.208 Toy-Treibitz 4.120 5.949 0.096 11.008 Toy-Zhao 3.269 5.183 0.319 11.889 Toy-Our 3.653 7.279 0.287 8.989 Ruler-Intensity 24.047 3.826 0.011 10.126 Ruler-Treibitz 3.368 3.059 0.008 8.337 Ruler-Zhao 2.954 2.079 0.448 15.465 Ruler-Our 6.318 6.971 0.138 14.719 Cube-coin-Intensity 17.751 4.951 0.018 7.982 Cube-coin-Treibitz 1.281 3.294 0.014 7.046 Cube-coin-Zhao 2.535 2.532 0.407 12.732 Cube-coin-Our 7.028 7.635 0.178 12.154 下载:
导出CSV
Image SNR Entropy Contrast NIQE Toy-Intensity 17.101 5.319 0.029 9.006 Toy-CLAHE 13.283 6.214 0.059 10.357 Toy-Zhao 6.314 6.083 0.127 15.829 Toy-Our 5.269 6.914 0.145 16.152 Ruler-Intensity 15.794 6.284 0.062 9.391 Ruler-CLAHE 15.389 6.442 0.067 10.395 Ruler-Zhao 3.424 1.154 0.492 15.277 Ruler-Our 9.889 7.421 0.128 19.501 Cube-coin-Intensity 16.477 5.313 0.023 7.727 Cube-coin-CLAHE 14.139 6.048 0.041 9.149 Cube-coin-Zhao 2.210 1.828 0.438 12.893 Cube-coin-Our 3.477 6.753 0.187 12.269 下载:
导出CSV
Parameter Toy Ruler Cube-coin Zhao Our Zhao Our Zhao Our Pobj 0.223 0.199 0.328 0.334 0.656 0.687 Pscat 0.271 0.254 0.319 0.265 0.643 0.613 |Pdif| 0.048 0.055 0.009 0.069 0.013 0.074 下载:
导出CSV
-
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
下载:
计量
- 文章访问数:3737
- PDF下载量:104
- 被引次数:0