Polarimetric underwater image recovery via deep learning

Abstract Polarimetric imaging is an effective way for clear vision in water, in which deducing the object radiance in clear water from the obtained polarimetric information in turbid water is essential. In this letter, we propose, for the first time to our knowledge, a learning-based method for polarimetric underwater image recovery. It is based on the dense network and can learn well the relation between the object radiance and the polarization information. The experimental results demonstrate that additionally introducing the polarization information is beneficial for improving the image quality. Moreover, the proposed learning-based method can effectively remove the veiling light and outperforms other existing methods, even in dense turbid water.

[1]  Lei Tian,et al.  Deep learning approach to scalable imaging through scattering media , 2019, Imaging and Applied Optics 2019 (COSI, IS, MATH, pcAOP).

[2]  Yun Fu,et al.  Residual Dense Network for Image Restoration , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Tiegen Liu,et al.  Polarimetric underwater image recovery for color image with crosstalk compensation , 2020 .

[4]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[6]  Seong G. Kong,et al.  An Iterative Image Dehazing Method With Polarization , 2019, IEEE Transactions on Multimedia.

[7]  N. Sujatha,et al.  Selective sensitivity of Mueller imaging for tissue scattering over absorption changes in cancer mimicking phantoms , 2018 .

[8]  Jingping Zhu,et al.  Target detection in turbid medium using polarization-based range-gated technology. , 2013, Optics express.

[9]  J. L. Pezzaniti,et al.  Mueller matrix imaging polarimetry , 1995 .

[10]  Mohammed Ghanbari,et al.  Scope of validity of PSNR in image/video quality assessment , 2008 .

[11]  Jian Liang,et al.  Polarimetric dehazing method for dense haze removal based on distribution analysis of angle of polarization. , 2015, Optics express.

[12]  Lin Zhao,et al.  Enhancing Visibility of Polarimetric Underwater Image by Transmittance Correction , 2017, IEEE Photonics Journal.

[13]  Y.Y. Schechner,et al.  Recovery of underwater visibility and structure by polarization analysis , 2005, IEEE Journal of Oceanic Engineering.

[14]  Yoav Y. Schechner,et al.  Active Polarization Descattering , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Kun Li,et al.  Polarimetric image recovery in turbid media employing circularly polarized light. , 2018, Optics express.

[16]  N. C. Flemming,et al.  Archaeology of the continental shelf: Marine resources, submerged landscapes and underwater archaeology , 2008 .

[17]  Guohua Gu,et al.  Micro deep learning profilometry for high-speed 3D surface imaging , 2019, Optics and Lasers in Engineering.

[18]  Tiegen Liu,et al.  Underwater image recovery considering polarization effects of objects. , 2016, Optics express.

[19]  Jian Sun,et al.  Single image haze removal using dark channel prior , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[20]  Tiegen Liu,et al.  Polarimetric image recovery method combining histogram stretching for underwater imaging , 2018, Scientific Reports.

[21]  Xun Wang,et al.  Texture filtering based physically plausible image dehazing , 2016, The Visual Computer.

[22]  Junchao Zhang,et al.  Learning a convolutional demosaicing network for microgrid polarimeter imagery. , 2018, Optics letters.

[23]  Lawrence B. Wolff,et al.  Polarization vision: a new sensory approach to image understanding , 1997, Image Vis. Comput..

[24]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[25]  Bingliang Hu,et al.  Method for enhancing visibility of hazy images based on polarimetric imaging , 2014 .