Water Retrieval Embedded Attention Network With Multiscale Receptive Fields for Hyperspectral Image Refined Classification

Liang, Xuejian; Zhang, Ye; Zhang, Junping

doi:10.1109/tgrs.2021.3091985

Cited by 5 publications

(2 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Atrous convolution [11], multi-level feature structured models like Unet [9], high-resolution models like HRNetV2 [15], and transformer models like SegFormer [16] were developed to deal with this problem. Another problem is the RF imbalance [46], [47], which means multi-scale and multi-level architectures widely used [21], [23] provided unsuitable RF for some objects, negatively impacting the segmentation of objects of varying sizes. For this problem, Liu, et al [48] designed the scale-layer attention module (SLAM) and scale-feature attention module (SFAM) to weigh useful information after ASPP and skip connection, respectively.…”

Section: B Receptive Field In Convnetsmentioning

confidence: 99%

AD-RoadNet: An Auxiliary-Decoding Road Extraction Network Improving Connectivity While Preserving Multiscale Road Details

Luo

Zhou

Tan

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Obtaining Road information from high-resolution remote sensing images (HRSI) is gaining attention in intelligent transportation systems. Existing road extraction methods tend to improve road connectivity with graph convolution or global attention, however, ignore the damage of introduced excessive effective receptive field (ERF) to multi-scale road details. In this study, we propose an Auxiliary-Decoding Road Extraction Network named AD-RoadNet which decouples multi-scale road representation and connectivity improvement based on two modules; the Hybrid Receptive Field Module (HRFM) and the Topological Feature Representation Module (TFRM). The HRFM is introduced in the encoder to emphasize target road features through adaptively matching the receptive field (RF) size for various scale roads, thus beneficial for multi-scale road representation. The TFRM is introduced in an auxiliary decoder to represent topological features with the position information encoded in the shared encoder and then helps the main decoder reason occluded roads, thus improving connectivity. Between the encoder and main decoder. The proposed model has a similar parameter scale as HRNetV2 and outperforms the state-of-the-art ResUnet, D-LinkNet, and HRNetV2 by 3.34%, 2.03%, and 1.53% in the mean Intersection of Union (mIoU) on DeepGlobe Road Dataset. Ablation analysis, inference size matter, and the robustness for unseen occlusion scenarios, low-quality labels, and various quality inference images are further presented to evaluate the proposed AD-RoadNet.

show abstract

Section: B Receptive Field In Convnetsmentioning

confidence: 99%

AD-RoadNet: An Auxiliary-Decoding Road Extraction Network Improving Connectivity While Preserving Multiscale Road Details

Luo

Zhou

Tan

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Dealing with imbalanced datasets can typically be addressed using feature learning methods, where the original data are transformed or reduced to a new feature space that makes it easier to distinguish between different datasets [ 27 , 28 ]. Alternatively, ensemble learning methods can be used to combine multiple individual classifiers into a strong classifier, improving the classification performance [ 29 , 30 ].…”

Section: Introductionmentioning

confidence: 99%

A Method of Water COD Retrieval Based on 1D CNN and 2D Gabor Transform for Absorption–Fluorescence Spectra

et al. 2023

View full text Add to dashboard Cite

Chemical Oxygen Demand (COD) is one of the indicators of organic pollution in water bodies. The rapid and accurate detection of COD is of great significance to environmental protection. To address the problem of COD retrieval errors in the absorption spectrum method for fluorescent organic matter solutions, a rapid synchronous COD retrieval method for the absorption–fluorescence spectrum is proposed. Based on a one-dimensional convolutional neural network and 2D Gabor transform, an absorption–fluorescence spectrum fusion neural network algorithm is developed to improve the accuracy of water COD retrieval. Results show that the RRMSEP of the absorption–fluorescence COD retrieval method is 0.32% in amino acid aqueous solution, which is 84% lower than that of the single absorption spectrum method. The accuracy of COD retrieval is 98%, which is 15.3% higher than that of the single absorption spectrum method. The test results on the actual sampled water spectral dataset demonstrate that the fusion network outperformed the absorption spectrum CNN network in measuring COD accuracy, with the RRMSEP improving from 5.09% to 1.15%.

show abstract

A novel semantic feature enhancement network for extracting lake water from remote sensing images

Hao,

Sun,

Wang

et al. 2024

Int. J. Mach. Learn. & Cyber.

View full text Add to dashboard Cite

Water Retrieval Embedded Attention Network With Multiscale Receptive Fields for Hyperspectral Image Refined Classification

Cited by 5 publications

References 71 publications

AD-RoadNet: An Auxiliary-Decoding Road Extraction Network Improving Connectivity While Preserving Multiscale Road Details

AD-RoadNet: An Auxiliary-Decoding Road Extraction Network Improving Connectivity While Preserving Multiscale Road Details

A Method of Water COD Retrieval Based on 1D CNN and 2D Gabor Transform for Absorption–Fluorescence Spectra

A novel semantic feature enhancement network for extracting lake water from remote sensing images

Contact Info

Product

Resources

About