Low Light Image Enhancement Algorithm Based on Detail Prediction and Attention Mechanism

Hui, Yanming; Wang, Jue; Shi, Ying; Li, Bo

doi:10.3390/e24060815

Cited by 11 publications

(7 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The related achievement methods focus selectively on interest regions by reassigning the related weight values of input sequences [ 29 ]. Attention mechanism has many image processing applications, including target detection [ 30 ], image enhancement [ 31 ] and emotion recognition [ 32 ], and it can be divided into local-attention, soft-attention and hard-attention [ 33 ] according to the achievement methods. Hereinto, Soft-attention mechanism can currently be subdivided into channel attention, spatial attention, and their combined module.…”

Section: Relevant Workmentioning

confidence: 99%

FDNet: An end-to-end fusion decomposition network for infrared and visible images

Di,

Ren,

Liu

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

Infrared and visible image fusion can generate a fusion image with clear texture and prominent goals under extreme conditions. This capability is important for all-day climate detection and other tasks. However, most existing fusion methods for extracting features from infrared and visible images are based on convolutional neural networks (CNNs). These methods often fail to make full use of the salient objects and texture features in the raw image, leading to problems such as insufficient texture details and low contrast in the fused images. To this end, we propose an unsupervised end-to-end Fusion Decomposition Network (FDNet) for infrared and visible image fusion. Firstly, we construct a fusion network that extracts gradient and intensity information from raw images, using multi-scale layers, depthwise separable convolution, and improved convolution block attention module (I-CBAM). Secondly, as the FDNet network is based on the gradient and intensity information of the image for feature extraction, gradient and intensity loss are designed accordingly. Intensity loss adopts the improved Frobenius norm to adjust the weighing values between the fused image and the two raw to select more effective information. The gradient loss introduces an adaptive weight block that determines the optimized objective based on the richness of texture information at the pixel scale, ultimately guiding the fused image to generate more abundant texture information. Finally, we design a single and dual channel convolutional layer decomposition network, which keeps the decomposed image as possible with the input raw image, forcing the fused image to contain richer detail information. Compared with various other representative image fusion methods, our proposed method not only has good subjective vision, but also achieves advanced fusion performance in objective evaluation.

show abstract

Section: Relevant Workmentioning

confidence: 99%

FDNet: An end-to-end fusion decomposition network for infrared and visible images

Di,

Ren,

Liu

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…DELLIE [14] When combined with the detail component prediction model, it is possible to extract and fuse image detail features.…”

Section: Prien [12]mentioning

confidence: 99%

“…According to the literature [ 13 ], an end-to-end enhancement network is built with a module stacking method and attention blocks, and the image is then enhanced with fusion. Because most methods focus on enhancing image luminance while ignoring image detail information, the literature [ 14 ] proposes the DELLIE algorithm, which focuses on image detail information extraction and fusion, thus recovering image detail information while maximizing image semantic information retention. However, the method described above does not achieve a balance of light intensity, color retention, and detail information.…”

Section: Introductionmentioning

confidence: 99%

Cyclic Generative Attention-Adversarial Network for Low-Light Image Enhancement

Zhen,

Peng,

2023

Sensors

View full text Add to dashboard Cite

Images captured under complex conditions frequently have low quality, and image performance obtained under low-light conditions is poor and does not satisfy subsequent engineering processing. The goal of low-light image enhancement is to restore low-light images to normal illumination levels. Although many methods have emerged in this field, they are inadequate for dealing with noise, color deviation, and exposure issues. To address these issues, we present CGAAN, a new unsupervised generative adversarial network that combines a new attention module and a new normalization function based on cycle generative adversarial networks and employs a global–local discriminator trained with unpaired low-light and normal-light images and stylized region loss. Our attention generates feature maps via global and average pooling, and the weights of different feature maps are calculated by multiplying learnable parameters and feature maps in the appropriate order. These weights indicate the significance of corresponding features. Specifically, our attention is a feature map attention mechanism that improves the network’s feature-extraction ability by distinguishing the normal light domain from the low-light domain to obtain an attention map to solve the color bias and exposure problems. The style region loss guides the network to more effectively eliminate the effects of noise. The new normalization function we present preserves more semantic information while normalizing the image, which can guide the model to recover more details and improve image quality even further. The experimental results demonstrate that the proposed method can produce good results that are useful for practical applications.

show abstract

“…The attention mechanism has achieved great success in some fields, including target detection [ 22 ], image enhancement [ 23 ], semantic segmentation [ 24 ] and emotion recognition [ 25 ]. Recently, many scholars have applied the attention mechanism to image fusion tasks.…”

Section: Related Workmentioning

confidence: 99%

Infrared and Visible Image Fusion for Highlighting Salient Targets in the Night Scene

Zhan

Wang

Jiang

et al. 2022

Entropy

View full text Add to dashboard Cite

The goal of infrared and visible image fusion in the night scene is to generate a fused image containing salient targets and rich textural details. However, the existing image fusion methods fail to take the unevenness of nighttime luminance into account. To address the above issue, an infrared and visible image fusion method for highlighting salient targets in the night scene is proposed. First of all, a global attention module is designed, which rescales the weights of different channels after capturing global contextual information. Second, the loss function is divided into the foreground loss and the background loss, forcing the fused image to retain rich texture details while highlighting the salient targets. Finally, a luminance estimation function is introduced to obtain the trade-off control parameters of the foreground loss function based on the nighttime luminance. It can effectively highlight salient targets by retaining the foreground information from the source images. Compared with other advanced methods, the experimental results adequately demonstrate the excellent fusion performance and generalization of the proposed method.

show abstract

Low Light Image Enhancement Algorithm Based on Detail Prediction and Attention Mechanism

Cited by 11 publications

References 31 publications

FDNet: An end-to-end fusion decomposition network for infrared and visible images

FDNet: An end-to-end fusion decomposition network for infrared and visible images

Cyclic Generative Attention-Adversarial Network for Low-Light Image Enhancement

Infrared and Visible Image Fusion for Highlighting Salient Targets in the Night Scene

Contact Info

Product

Resources

About