| 摘要: |
| 开发了一种基于偏振成像和图像融合的新型水下成像技术。该技术是利用小波的多尺度和多分辨率特性，将 一组偏振方向相互垂直的两幅源图像分别分解后进行加权融合处理，最后利用小波逆变换重构融合后的偏振图像。 用图像的均值、标准差、信息熵和方差等参量对融合效果进行了评价。实验和仿真结果表明该方法十分有效，可极 大地提高图像质量。 |
| 关键词: 偏振成像 小波变换 图像融合 水下物体 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金项目(4071) |
|
| Underwater imaging technology based on image fusion |
|
LI Dai-lin, ZHU Hua-feng,PEI Hong-yan
|
|
(College of Physics Science and Technology in China University of Petroleum, Dongying 257061,Shandong Province, China)
|
| Abstract: |
| A novel underwater imaging technology based on polarization imaging and image fusion Vas developed. According to the characteristics of rautil-scale and rautil-resolution of wavelet transform, the technology can be applied to decompose a pair of orthogonally polarized images into basic blocks respectively, and then to fuse the wavelet coefficients to finally obtain the fused image by inverse wavelet transform. Tlie fused images were evaluated by calculating the mean, standard variation, information entropy and variance to quantitatively examine the method- The results from experiments and simulations show that this method is effective and may improve the image quality substantially. |
| Key words: polarization imaging wavelet transform image fusion underwater object |
