-
水下光学成像是海底探索和目标识别的一个重要方式. 由于海浪、船舶尾流以及海洋生物游动与呼吸等原因, 存在着大量的气泡. 气泡群的光散射作用往往会使水下目标成像效果受限、难以识别, 并且一般的光学技术难以消除气泡对成像的影响. 针对上述问题, 本文先从理论上推导和仿真了入射光线在水下单气泡、气泡群中以及目标表面的光强和偏振信息的变化; 然后在构建了水下气泡实验平台的基础上探究了光源入射角度的改变以及成像波段的变化对气泡环境中目标偏振成像的影响; 研究了不同金属材质目标物的强度和偏振信息的变化趋势; 分析了水下目标在不同气泡群厚度条件下强度和偏振信息的变化趋势; 最后利用偏振特征提取与视觉信息保留的图像融合方法抑制气泡对水下目标成像的影响. 实验结果显示气泡群中目标成像会受到多种因素的影响, 利用偏振图像融合方法会使气泡群受到较好的抑制, 并提高了水下目标的清晰度.Underwater optical imaging is an important way to implement the seabed exploration and target recognition. There occur a lot of bubbles due to the sea wave, ship wake, marine creatures’ swimming and breathing. The underwater target imaging effect is often limited by light scattering effect of bubbles, so it is difficult to identify targets, and the general optical technology is difficult to eliminate the bubbles’ influence on imaging. In this article from the bubble theoretical derivation and the bubble simulation, we investigate the changing trend of target’s polarization information under the condition of different light incident angles in the underwater environment, data gathering, data processing and data analysis, by using the polarimetric image fusion method to suppress the influence of bubbles to build a complete target imaging research system under bubble group environment in line with the above several big aspects. According to the above problem, in this paper, the change of light intensity and polarization information of incoming light in underwater single bubble, bubble group and target’s surface are investigated; the target imaging in the bubble group environment with the change of light incident angle and polarization imaging band on the basis of the construction of experimental platform of underwater bubbles is explored; the change trends of strength and polarization information with different metal targets are studied; the change trends of strength and polarization information of underwater target under thickness of different bubble groups are analyzed; finally the underwater target images under the condition of different imaging resolutions and the using of fusion methods of polarization feature extraction and visual information of image to suppress the bubble influence on underwater target imaging are studied. The experimental results show that the target imaging under bubble group environment is influenced by many factors, and using polarimetric image fusion method can well weaken the bubble group’s influence on imaging, and improve the clarity of underwater target. In view of difficult problems about target identification existing in the high-density bubble group environment, we will use energy loss compensation or machine learning method to realize the target recognition and image restoration in the future.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] -
入射角度/(°) 水中气泡外界面的光强 1 (A点) 2 (B点) 3 (C点) 4 (D点) 5 0.0201 0.9603 0.0193 3.8662 × 10–4 10 0.0201 0.9602 0.0193 3.9015 × 10–4 15 0.0202 0.9599 0.0194 4.0710 × 10–4 20 0.0207 0.9590 0.0198 4.6056 × 10–4 25 0.0220 0.9567 0.0207 6.0192 × 10-4 30 0.0251 0.9509 0.0230 9.5371 × 10-4 35 0.0328 0.9364 0.0287 0.0019 40 0.0542 0.8971 0.0436 0.0045 45 0.1342 0.7570 0.0888 0.0158 46 0.1761 0.6878 0.1062 0.0223 47 0.2466 0.5786 0.1262 0.0334 48 0.3945 0.3782 0.1353 0.0552 48.75(临界值) 0.9372 0.0042 0.0039 0.0037 
下载:
导出CSV
入射角度/(°) 水中气泡外界面的偏振度/% 1 (A点) 2 (B点) 3 (C点) 4 (D点) 5 2.04 0.08 1.96 4.00 10 8.36 0.34 8.02 16.26 15 19.46 0.80 18.69 36.81 20 35.94 1.52 34.61 62.74 25 57.60 2.59 55.84 85.83 30 81.32 4.18 79.86 97.72 35 98.20 6.66 97.95 99.98 40 94.97 10.86 93.78 99.83 45 63.94 19.63 50.67 86.56 46 54.08 22.82 35.66 75.23 47 42.40 27.23 17.15 55.51 48 27.17 34.32 7.89 19.70 48.75(临界值) 1.81 50.25 48.88 47.50 下载:
导出CSV
距离/m 信息熵 平均梯度 边缘强度 0.5 5.9263 1.3499 12.7868 0.6 5.9145 1.4647 14.2792 0.7 5.9311 1.4946 14.6517 0.8 6.0008 1.6703 16.5898 0.9 5.9563 1.7850 17.8742 1.0 5.9595 1.7225 17.2655 下载:
导出CSV
材质类别 图像类别 信息熵 平均梯度 边缘强度 方差 目标1 原强度图 7.5541 2.6007 28.5980 4.9371 × 103 融合结果图 5.3631 14.9552 146.1138 5.6631 × 103 目标2 原强度图 6.0236 1.1077 11.9431 654.3071 融合结果图 5.6483 17.6877 169.1962 5.7606 × 103 目标3 原强度图 6.0648 4.1370 39.8161 508.8038 融合结果图 5.6336 16.3178 156.5486 5.8342 × 103 目标4 原强度图 6.0806 1.1309 12.2131 965.9536 融合结果图 5.7785 17.5954 169.1398 5.7043 × 103 目标5 原强度图 6.5571 1.4227 15.5275 1.0356 × 103 融合结果图 5.7211 15.0098 146.1177 5.8561 × 103 目标6 原强度图 6.2111 1.4299 15.6348 1.2907 × 103 融合结果图 5.5173 17.9909 174.8208 6.0343 × 103 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:4709
- PDF下载量:101
- 被引次数:0