An Underwater Image Enhancement Algorithm Based on Generative Adversarial Network and Natural Image Quality Evaluation Index

نویسندگان

چکیده

When underwater vehicles work, images are often absorbed by light and scattered diffused floating objects, which leads to the degradation of images. The generative adversarial network (GAN) is widely used in image enhancement tasks because it can complete image-style conversions with high efficiency quality. Although GAN converts low-quality into high-quality (truth images), dataset truth also affects However, an lacks enhancement, a poor effect generated image. Thus, this paper proposes add natural quality evaluation (NIQE) index provide higher contrast make them more line perception human eye, at same time, grant better than set existing dataset. In paper, several groups experiments compared, through subjective objective indicators, verified that enhanced algorithm

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Generative Adversarial Network based on Resnet for Conditional Image Restoration

The GANs promote an adversarive game to approximate complex and jointed example probability. The networks driven by noise generate fake examples to approximate realistic data distributions. Later the conditional GAN merges prior-conditions as input in order to transfer attribute vectors to the corresponding data. However, the CGAN is not designed to deal with the high dimension conditions since...

متن کامل

Generative Adversarial Network based Synthesis for Supervised Medical Image Segmentation*

Modern deep learning methods achieve state-ofthe-art results in many computer vision tasks. While these methods perform well when trained on large datasets, deep learning methods suffer from overfitting and lack of generalization given smaller datasets. Especially in medical image analysis, acquisition of both imaging data and corresponding ground-truth annotations (e.g. pixel-wise segmentation...

متن کامل

Comparing Generative Adversarial Network Techniques for Image Creation and Modification

Generative adversarial networks (GANs) have demonstrated to be successful at generating realistic real-world images. In this paper we compare various GAN techniques, both supervised and unsupervised. The effects on training stability of different objective functions are compared. We add an encoder to the network, making it possible to encode images to the latent space of the GAN. The generator,...

متن کامل

An Efficient Underwater Image Enhancement Using Color Constancy Deskewing Algorithm

The objective of image enhancement is to recover the perception of information in underwater images for human viewers, or to provide recovered input for other automated image processing techniques. In underwater conditions, the image clarities are corrupted by light inclusion and diffusion. This causes one colour to dominate the image. In order to enhance the observation of underwater images, w...

متن کامل

High-Quality Face Image SR Using Conditional Generative Adversarial Networks

We propose a novel single face image superresolution method, which named Face Conditional Generative Adversarial Network(FCGAN), based on boundary equilibrium generative adversarial networks. Without taking any facial prior information, our method can generate a high-resolution face image from a low-resolution one. Compared with existing studies, both our training and testing phases are end-toe...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Journal of Marine Science and Engineering

سال: 2021

ISSN: ['2077-1312']

DOI: https://doi.org/10.3390/jmse9070691