Cloud detection is one of the important tasks for remote sensing image (RSI) preprocessing. In this paper, we utilize the thumbnail (i.e., preview image) of RSI, which contains the information of original multispectral or panchromatic imagery, to extract cloud mask efficiently. Compared with detection cloud mask from original RSI, it is more challenging to detect cloud mask using thumbnails due to the loss of resolution and spectrum information. To tackle this problem, we propose a cloud detection neural network (CDnet) with an encoder-decoder structure, a feature pyramid module (FPM), and a boundary refinement (BR) block. The FPM extracts the multiscale contextual information without the loss of resolution and coverage; the BR block refines object boundaries; and the encoder-decoder structure gradually recovers segmentation results with the same size as input image. Experimental results on the ZY-3 satellite thumbnails cloud cover validation data set and two other validation data sets (GF-1 WFV Cloud and Cloud Shadow Cover Validation Data and Landsat-8 Cloud Cover Assessment Validation Data) demonstrate that the proposed method achieves accurate detection accuracy and outperforms several state-of-the-art methods.

In underwater imaging scenarios, the scattering media could cause severe image degradation due to the backscatter veiling as well as signal attenuation. In this paper, we consider the polarization effect of the object, and propose a method of retrieving the objects radiance based on estimating the polarized-difference image of the target signal. We show with a real-world experiment that by taking into account the polarized-difference image of the target signal additionally, the quality of the underwater image can be effectively enhanced, which is particularly effective in the cases where both the object radiance and the backscatter contribute to the polarization, such as underwater detection of the artifact objects.

The underwater imaging could be severely degraded by the scattering media because of the backscattered light and signal attenuation, especially in the case of strong scattering for dense turbid medium. In this paper, we propose an improved method for recovering the underwater image combining the histogram stretching and polarimetric recovery in a proper way. In this method, we stretch the histograms of the orthogonal polarization images while maintaining the polarization relation between them, and then, based on the processed orthogonal polarization images, the recovered image with higher quality can be obtained by the traditional polarimetric recovery method. Several groups of experimental results demonstrate that the quality of underwater images can be effectively enhanced by our method, and its performance is better than that of the traditional polarimetric recovery method. In particular, the proposed method is also quite effective in the condition of dense turbid medium.

Imaging in scattering media has been a challenging and important subject in optical science. In scattering media, the image quality is often severely degraded by the scattering and absorption effects owing to the small particles and the resulting nonuniform distribution of the intensity or polarization properties. This study reviews the recent development in polarimetric imaging techniques that address these challenges. Specifically, based on the polarization properties of the backscattering light, polarimetric methods can estimate the intensity level of the backscattering and the transmittance of the media. They can also separate the target signal from the undesired ones to achieve high-quality imaging. In addition, the different designs of the polarimetric imaging systems offer additional metrics, for example, the degree/angle of polarization, to recover images with high fidelity. We first introduce the physical degradation models in scattering media. Secondly, we apply the models in different polarimetric imaging systems, such as polarization difference, Stokes vector, Mueller matrix, and deep learning-based systems. Lastly, we provide a model selection guideline and future research directions in polarimetric imaging.

Circular polarization memory is a well-known phenomenon indicating that the circular polarization light can persist better its polarization property during propagating through turbid media compared with the linear polarization light. Therefore, in principle, using circularly polarized light can probably improve the quality of image recovery in dense turbid media than using the linearly polarized light. In this paper, we propose a new polarimetric image recovery method in dense turbid media with the illumination light of circular polarization, and we realize the image recovery combining the circular polarization information and linearly polarization information. The real-world experiment results demonstrate that the proposed method is more effective than previous methods, including the traditional polarimetric image recovery method by Schechner’s [Appl. Opt. 42, 511 (2003)] based on linear polarization.

The underwater imaging could be severely degraded by the scattering media due the backscatter veiling and signal attenuation. Image recovery based on polarization information has shown its efficiency in improving the quality of image in scattering medium, but few of them considered the effect of non-uniform optical field on the image recovery. In this paper, we consider the condition of non-uniform optical field, and we propose a method of retrieving the objects radiance based on estimating the degree of polarization (DOP) and the intensity of backscatter at different positions of the image. The experimental results demonstrate that the quality of the underwater image can be effectively enhanced by this method. In addition, compared with the previous methods, the method proposed in this paper can maintain a better performance at different positions of the image, thanks to considering the DOP and intensity of backscatter as spatial variables instead of constants.