Article
Geochemistry & Geophysics
Jingchun Zhou, Boshen Li, Dehuan Zhang, Jieyu Yuan, Weishi Zhang, Zhanchuan Cai, Jinyu Shi
Summary: We propose an efficient and fully guided information flow network (UGIF-Net) for enhancing underwater images, which overcomes limitations in color recovery accuracy and resilience against irrelevant feature interference.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Hua Zou, Xiao Lin, Huanhuan Wu, Yeh-Cheng Chen
Summary: In this paper, a predictive intelligence approach of a multi-task framework is proposed to enhance low-light images. The framework consists of coarse recovery sub-networks and cross refinement sub-networks. Experimental results demonstrate that the proposed method achieves significant improvements in various metrics compared to state-of-the-art alternatives.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Jiarun Fu, Lingyu Yan, Yulin Peng, KunPeng Zheng, Rong Gao, HeFei Ling
Summary: With the development of deep learning, image recognition and enhancement have become widely used. However, dark lighting environments in reality result in low brightness, severe noise, and loss of details, hindering further analysis and use. This paper proposes a method of low-light image enhancement using the luminance attention mechanism and generative adversarial networks, which improves the enhancement efficiency and has practical significance for solving this problem.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Hao-Tian Wu, Xin Cao, Ruoyan Jia, Yiu-Ming Cheung
Summary: A new 2D histogram based contrast enhancement with reversible data hiding (CE-RDH) scheme is proposed in this paper, taking brightness preservation into account. The scheme can preserve image color and brightness while achieving better image quality than existing schemes.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Zhiying Jiang, Zhuoxiao Li, Shuzhou Yang, Xin Fan, Risheng Liu
Summary: This paper proposes a target-oriented perceptual adversarial fusion network, TOPAL, to improve the quality of underwater images. By enhancing visual contrast and performing color correction, and utilizing multi-scale attention module and adaptive fusion, TOPAL shows superior performance in enhancing the quality of underwater images.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Xuping Huang, Qian Wang, Junxi Chen, Lingna Chen, Zhiyi Chen
Summary: Ultrasound image segmentation is crucial for early disease diagnosis, but ultrasound images have lower resolution and clarity compared to CT and MRI, and are sensitive to external interference. In this paper, we propose a deep convolutional neural network that incorporates a pseudo-color enhancement algorithm and hybrid attention modules to improve feature extraction and modeling capabilities. Experimental results demonstrate that our method outperforms other segmentation methods for ultrasound image lesions.
IMAGE AND VISION COMPUTING
(2023)
Article
Engineering, Electrical & Electronic
Hengshuai Cui, Jinjiang Li, Zhen Hua, Linwei Fan
Summary: In this article, a progressive dual-branch network (PDBNet) is proposed for low-light image enhancement. It improves the loss of details, color imbalance, and artifacts in low-light images through the design of components such as assisted recovery module, large kernel attention block, attention fusion block, and fusion reconstruction module. The experimental results demonstrate that our method outperforms other state-of-the-art low-light image enhancement methods in terms of visual quality and metric evaluation scores.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Information Systems
Yinhua Su, Mian Wu, Ying Yan
Summary: This study proposes two algorithms to improve the visual effect of images captured in low illumination environments and preserve image information integrity. The image enhancement algorithm combines the WASOBI denoising method and adaptive gamma correction for noise reduction and brightness adjustment. The image brightness equalization algorithm divides the image based on illumination area and uses guided filtering and adaptive gamma correction to equalize brightness. Experimental results demonstrate the effectiveness and stability of the proposed image enhancement algorithm. The subjective evaluation score of the brightness equalization algorithm is 4.7, indicating good visual effects. Objective evaluation results show the capability of the equalization algorithm, with the shortest operation time of 0.9452 seconds compared to other algorithms. Hence, the proposed algorithms have certain application value.
Article
Computer Science, Artificial Intelligence
Lintao Peng, Chunli Zhu, Liheng Bian
Summary: Underwater impurities cause poor imaging quality due to light absorption and scattering. Existing data-driven underwater image enhancement techniques lack a large-scale dataset and high-fidelity reference images. In this work, a large-scale underwater image dataset is built, and a Transformer network is introduced for the UIE task. The proposed method shows state-of-the-art performance in terms of enhancement and is validated through extensive experiments.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Automation & Control Systems
Arshiana Shamir, Nokap Park, Bumshik Lee
Summary: A novel deep-learning network is proposed for brightness enhancement of old images, which combines curve map estimation and attention-guided illumination map to adjust the dynamic range and illumination of the images. Experimental results show that the proposed method outperforms existing methods in brightness enhancement on old photo and video datasets.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Chemistry, Multidisciplinary
Zhe Yang, Fangjin Liu, Jinjiang Li
Summary: This paper proposes a low-light image enhancement method called EFCANet, which recovers normal light images from a single exposure-corrected image. It estimates the exposure-corrected image using an Exposure Image Generator (EIG), converts the color space to YCbCr, and fuses the images using a Cross-Attention Fusion Module (CAFM) in the YCbCr color space.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Electrical & Electronic
Yaozu Kang, Qiuping Jiang, Chongyi Li, Wenqi Ren, Hantao Liu, Pengjun Wang
Summary: This paper proposes a perception-aware decomposition and fusion framework for underwater image enhancement (UIE). Two complementary pre-processed inputs are fused in a perception-aware and conceptually independent image space through a structural patch decomposition and fusion (SPDF) approach. The main advantage of SPDF is that it can fuse different components separately without any interactions and information loss. Comprehensive comparisons demonstrate that SPDF outperforms state-of-the-art UIE algorithms.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Zhaorun Zhou, Zhenghao Shi, Wenqi Ren
Summary: This study proposes a linear contrast enhancement network (LCENet) to improve the quality of images captured under low-illumination conditions. LCENet consists of three subnets for restoring gradient maps, enhancing brightness, and adjusting adaptive brightness and contrast. It also introduces a linear contrast enhancement adaptive instance normalization (LCEAIN) module to address the problem of ignoring contrast enhancement during brightness enhancement. Evaluations on synthetic and real low-illumination images demonstrate the superior performance of the proposed method compared to existing methods. The method can handle complex low-illuminance conditions and has good generalization for scenes with backlighting, light sources in night scenes, and low illuminance in underwater scenes.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Electrical & Electronic
Yufeng Li, Zhentao Fan, Jiyang Lu, Xiang Chen
Summary: A Task-Adaptive Operation Network (TAO-Net) is proposed to assign intensity weights for different degradation factors in image restoration and enhancement tasks using a supervised attention mechanism. The image prior is utilized for accurate detail recovery, and a feature aggregation block is adopted to avoid feature interference.
IEEE SIGNAL PROCESSING LETTERS
(2022)
Article
Computer Science, Information Systems
Xin Zhang, Xia Wang
Summary: This study presents a multi-scale attention Retinex network for low-light image enhancement, which improves image quality by introducing detailed inverse illumination map and illuminance-attention map, and formulates a novel loss function to synthetically measure illumination, detail, and colorfulness effects.