SSAU-Net: A Spectral–Spatial Attention-Based U-Net for Hyperspectral Image Fusion

Liu, Shuaiqi; Liu, Siyuan; Zhang, Shichong; Li, Bing; Hu, Weiming; Zhang, Yudong

doi:10.1109/tgrs.2022.3217168

Cited by 20 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the traditional and machine-learning schemes, the deep-learning procedures efficiently provide a better result on moderate and large datasets. Furthermore, most of these methods can be practically implementable in a chosen hardware system, improving its performance ( 25 – 27 ).…”

Section: Methodsmentioning

confidence: 99%

A framework to distinguish healthy/cancer renal CT images using the fused deep features

Rajinikanth

Vincent

Srinivasan

et al. 2023

Front. Public Health

View full text Add to dashboard Cite

IntroductionCancer happening rates in humankind are gradually rising due to a variety of reasons, and sensible detection and management are essential to decrease the disease rates. The kidney is one of the vital organs in human physiology, and cancer in the kidney is a medical emergency and needs accurate diagnosis and well-organized management.MethodsThe proposed work aims to develop a framework to classify renal computed tomography (CT) images into healthy/cancer classes using pre-trained deep-learning schemes. To improve the detection accuracy, this work suggests a threshold filter-based pre-processing scheme, which helps in removing the artefact in the CT slices to achieve better detection. The various stages of this scheme involve: (i) Image collection, resizing, and artefact removal, (ii) Deep features extraction, (iii) Feature reduction and fusion, and (iv) Binary classification using five-fold cross-validation.Results and discussionThis experimental investigation is executed separately for: (i) CT slices with the artefact and (ii) CT slices without the artefact. As a result of the experimental outcome of this study, the K-Nearest Neighbor (KNN) classifier is able to achieve 100% detection accuracy by using the pre-processed CT slices. Therefore, this scheme can be considered for the purpose of examining clinical grade renal CT images, as it is clinically significant.

show abstract

Section: Methodsmentioning

confidence: 99%

A framework to distinguish healthy/cancer renal CT images using the fused deep features

Rajinikanth

Vincent

Srinivasan

et al. 2023

Front. Public Health

View full text Add to dashboard Cite

show abstract

“…Combining refraction distortion and color enhancement of underwater images is more suitable for engineering applications. Some researchers use artificial intelligence methods for image enhancement processing [11,12], but the adaptability of image enhancement artificial intelligence for underwater environments needs further improvement. Akkaynak and Treibitz [13] modified the traditional atmospheric imaging model for the color restoration of underwater images and their sea-thru method successfully eliminated the influence of underwater blue and green water on the image, resulting in the color of underwater image being consistent with that in the air.…”

Section: Underwater Image Color Enhancement Methodsmentioning

confidence: 99%

Conversion of underwater concrete images to air in detection of hydraulic structures

Lin¹,

Zhang²,

Li³

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Detection of apparent concrete defects in hydraulic buildings is an important measure to ensure their safe and stable operation. When the optical camera is fixed in a waterproof housing for underwater concrete surface defect detection, the dark scenes and refraction of light on different media surfaces result in poor image quality and accompanying refractive distortion. In order to make the collected underwater image data better applied to underwater concrete inspection, we propose an equivalent air image conversion method that combines underwater image color enhancement and refractive distortion correction. This paper, proposes 2 models for underwater image-to-air conversion, one of which, using the light projection method, establishes a refractive distortion model for camera received light penetrating a multilayer medium underwater for correcting the refractive distortion of underwater images and the conversion of image field of view (FOV), and the other, by analyzing the problem of low image imaging quality and loss of edge information due to underwater uneven illumination environment, proposes a dark channel priority adaptive enhancement algorithm for improving underwater image quality. The experimental results in real scenes show that the imaging quality of underwater images is improved by converting underwater images into the air. The pixel error of images converted into the air is ≤1 pixel, and the FOV error of images is ≤8.5%. High precision underwater image conversion provides strong support for subsequent underwater measurement.

show abstract

“…The U-Net architecture, known for its wide applicability, has found extensive adoption across diverse domains. Furthermore, its versatility has led to the creation of various modified versions tailored to specific image-related tasks [27,28]. Moreover, Y-Net is a segmentation and classification network based on the U-Net segmentation network that excels in breast cancer segmentation and diagnosis tasks.…”

Section: Methodology 21 Y-net Architecturementioning

confidence: 99%

CloudY-Net: A Deep Convolutional Neural Network Architecture for Joint Segmentation and Classification of Ground-Based Cloud Images

Hu,

Hou,

Zhu

et al. 2023

Atmosphere

View full text Add to dashboard Cite

Ground-based cloud images contain a wealth of cloud information and are an important part of meteorological research. However, in practice, ground cloud images must be segmented and classified to obtain the cloud volume, cloud type and cloud coverage. Existing methods ignore the relationship between cloud segmentation and classification, and usually only one of these is studied. Accordingly, our paper proposes a novel method for the joint classification and segmentation of cloud images, called CloudY-Net. Compared to the basic Y-Net framework, which extracts feature maps from the central layer, we extract feature maps from four different layers to obtain more useful information to improve the classification accuracy. These feature maps are combined to produce a feature vector to train the classifier. Additionally, the multi-head self-attention mechanism is implemented during the fusion process to enhance the information interaction among features further. A new module called Cloud Mixture-of-Experts (C-MoE) is proposed to enable the weights of each feature layer to be automatically learned by the model, thus improving the quality of the fused feature representation. Correspondingly, experiments are conducted on the open multi-modal ground-based cloud dataset (MGCD). The results demonstrate that the proposed model significantly improves the classification accuracy compared to classical networks and state-of-the-art algorithms, with classification accuracy of 88.58%. In addition, we annotate 4000 images in the MGCD for cloud segmentation and produce a cloud segmentation dataset called MGCD-Seg. Then, we obtain a 96.55 mIoU on MGCD-Seg, validating the efficacy of our method in ground-based cloud imagery segmentation and classification.

show abstract

SSAU-Net: A Spectral–Spatial Attention-Based U-Net for Hyperspectral Image Fusion

Cited by 20 publications

References 42 publications

A framework to distinguish healthy/cancer renal CT images using the fused deep features

A framework to distinguish healthy/cancer renal CT images using the fused deep features

Conversion of underwater concrete images to air in detection of hydraulic structures

CloudY-Net: A Deep Convolutional Neural Network Architecture for Joint Segmentation and Classification of Ground-Based Cloud Images

Contact Info

Product

Resources

About