Improved single image dehazing methods for resource-constrained platforms

Yang, Gengqian; Evans, A.N.

doi:10.1007/s11554-021-01143-6

Cited by 19 publications

(14 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, in Figure 1, the output of inception-like block 1 is the input of the attention gate, and inception-like block 4, after the transposed convolution, is the gate signal. The attention coefficient is calculated using Equation (8). The activation function adopts the ReLU function to ensure that the attention coefficient is greater than 0.…”

Section: 𝑥 ̂𝑖𝑐mentioning

confidence: 99%

“…Ju et al [7] added an absorption function to the ASM and solved the image dehazing problem with the gray-world assumption. Based on DCP, Yang et al [8] improved the estimation method of atmospheric light to avoid mistaking the sky region as containing haze and smoothed the transmission map with two morphological operators, namely, dilation and erosion.…”

Section: Introductionmentioning

confidence: 99%

“…Li et al [10] integrated the transmission map and atmospheric light into a function as the learning target of the proposed AOD-Net. Furthermore, Yang et al [8] redesigned the network architecture of AOD-Net [10], inheriting the advantages of few trainable parameters and real-time capabilities. Based on the idea of integrating the transmission map and atmospheric light as the learning target, Zhang et al [11] designed a multi-scale network architecture using 1 × 1 convolution and pooling.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Attention-Gate-Based Model with Inception-like Block for Single-Image Dehazing

Tsai

Chen

2022

Applied Sciences

View full text Add to dashboard Cite

In recent decades, haze has become an environmental issue due to its effects on human health. It also reduces visibility and degrades the performance of computer vision algorithms in autonomous driving applications, which may jeopardize car driving safety. Therefore, it is extremely important to instantly remove the haze effect on an image. The purpose of this study is to leverage useful modules to achieve a lightweight and real-time image-dehazing model. Based on the U-Net architecture, this study integrates four modules, including an image pre-processing block, inception-like blocks, spatial pyramid pooling blocks, and attention gates. The original attention gate was revised to fit the field of image dehazing and consider different color spaces to retain the advantages of each color space. Furthermore, using an ablation study and a quantitative evaluation, the advantages of using these modules were illustrated. Through existing indoor and outdoor test datasets, the proposed method shows outstanding dehazing quality and an efficient execution time compared to other state-of-the-art methods. This study demonstrates that the proposed model can improve dehazing quality, keep the model lightweight, and obtain pleasing dehazing results. A comparison to existing methods using the RESIDE SOTS dataset revealed that the proposed model improves the SSIM and PSNR metrics by at least 5–10%.

show abstract

Section: 𝑥 ̂𝑖𝑐mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Attention-Gate-Based Model with Inception-like Block for Single-Image Dehazing

Tsai

Chen

2022

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…"Where β is the scattering coefficient of the atmosphere, and d(x) is the distance between the object and the camera" [9]. And the following figure 2 shows (ASM).…”

Section: ░ 2 Model Of Atmospheric Scattering (Asm)mentioning

confidence: 99%

A Review of Methods of Removing Haze from An Image

Karoon

Nemer

2022

IJEER

View full text Add to dashboard Cite

A literature review aids in comprehending and gaining further information about a certain area of a subject. The presence of haze, fog, smoke, rain, and other harsh weather conditions affects outdoor photos. Images taken in unnatural weather have weak contrast and poor colors. This may make detecting objects in the produced hazy pictures difficult. In computer vision, scenes and images taken in a foggy atmosphere suffer from blurring. This work covers a study of many remove haze algorithms for eliminating haze collected in real-world weather scenarios in order to recover haze-free images rapidly and with improved quality. The contrast, viewing range, and color accuracy have been enhanced. All of these techniques it is used in countless fields. Some of the applications that use this technology outdoor surveillance, object recognition, underwater photography, and so on.

show abstract

“…The RESIDE dataset has been used across several works in the recent past [2,28,64], and this work follows the same standard train-test split.…”

Section: Rich Feature Distillation: Feature Affinity Modulementioning

confidence: 99%

Rich Feature Distillation with Feature Affinity Module for Efficient Image Dehazing

Mitheran¹,

Anushri²,

Nisha³

et al. 2022

Preprint

View full text Add to dashboard Cite

Single-image haze removal is a long-standing hurdle for computer vision applications. Several works have been focused on transferring advances from image classification, detection, and segmentation to the niche of image dehazing, primarily focusing on contrastive learning and knowledge distillation. However, these approaches prove computationally expensive, raising concern regarding their applicability to on-the-edge use-cases. This work introduces a simple, lightweight, and efficient framework for single-image haze removal, exploiting rich "darkknowledge" information from a lightweight pre-trained super-resolution model via the notion of heterogeneous knowledge distillation. We designed a feature affinity module to maximize the flow of rich feature semantics from the super-resolution teacher to the student dehazing network. In order to evaluate the efficacy of our proposed framework, its performance as a plug-and-play setup to a baseline model is examined. Our experiments are carried out on the RESIDE-Standard dataset to demonstrate the robustness of our framework to the synthetic and real-world domains. The extensive qualitative and quantitative results provided establish the effectiveness of the framework, achieving gains of upto 15% (PSNR) while reducing the model size by ∼20 times.

show abstract

Improved single image dehazing methods for resource-constrained platforms

Cited by 19 publications

References 48 publications

Attention-Gate-Based Model with Inception-like Block for Single-Image Dehazing

Attention-Gate-Based Model with Inception-like Block for Single-Image Dehazing

A Review of Methods of Removing Haze from An Image

Rich Feature Distillation with Feature Affinity Module for Efficient Image Dehazing

Contact Info

Product

Resources

About