Semantic Segmentation on Remotely-Sensed Images Using Enhanced Global Convolutional Network with Channel Attention and Domain Specific Transfer Learning

Panboonyuen, Teerapong; Jitkajornwanich, Kulsawasd; Lawawirojwong, Siam; Srestasathiern, Panu; Vateekul, Peerapon

doi:10.20944/preprints201812.0090.v2

Cited by 5 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the past few years, deep learning models have achieved considerable successes in both natural and biomedical image processing. [35][36][37][38][39] For cerebrovascular segmentation, researchers followed this trend and proposed various excellent frameworks. For example, Phellan et al 21 presented a feasible analysis by applying deep CNN for automatic cerebrovascular segmentation and achieved promising results in terms of accuracy.…”

Section: Deep Learning-based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A nested parallel multiscale convolution for cerebrovascular segmentation

et al. 2021

View full text Add to dashboard Cite

Purpose: Cerebrovascular segmentation in magnetic resonance imaging (MRI) plays an important role in the diagnosis and treatment of cerebrovascular diseases. Many segmentation frameworks based on convolutional neural networks (CNNs) or U‐Net‐like structures have been proposed for cerebrovascular segmentation. Unfortunately, the segmentation results are still unsatisfactory, particularly in the small/thin cerebrovascular due to the following reasons: (1) the lack of attention to multiscale features in encoder caused by the convolutions with single kernel size; (2) insufficient extraction of shallow and deep‐seated features caused by the depth limitation of transmission path between encoder and decoder; (3) insufficient utilization of the extracted features in decoder caused by less attention to multiscale features. Methods: Inspired by U‐Net++, we propose a novel 3D U‐Net‐like framework termed Usception for small cerebrovascular. It includes three blocks: Reduction block, Gap block, and Deep block, aiming to: (1) improve feature extraction ability by grouping different convolution sizes; (2) increase the number of multiscale features in different layers by grouping paths of different depths between encoder and decoder; (3) maximize the ability of decoder in recovering multiscale features from Reduction and Gap block by using convolutions with different kernel sizes. Results: The proposed framework is evaluated on three public and in‐house clinical magnetic resonance angiography (MRA) data sets. The experimental results show that our framework reaches an average dice score of 69.29%, 87.40%, 77.77% on three data sets, which outperform existing state‐of‐the‐art methods. We also validate the effectiveness of each block through ablation experiments. Conclusions: By means of the combination of Inception‐ResNet and dimension‐expanded U‐Net++, the proposed framework has demonstrated its capability to maximize multiscale feature extraction, thus achieving competitive segmentation results for small cerebrovascular.

show abstract

Section: Deep Learning-based Methodsmentioning

confidence: 99%

“…In the past few years, deep learning models have achieved considerable successes in both natural and biomedical image processing 35‐39 . For cerebrovascular segmentation, researchers followed this trend and proposed various excellent frameworks.…”

Section: Related Workmentioning

confidence: 99%

A nested parallel multiscale convolution for cerebrovascular segmentation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Some works use the attention mechanism for the segmentation of RSIs. In [36], a channel attention block is designed to enhance the decoding branch of the CNN. In [37], the attention mechanism is used to match the caption nouns with the objects in RSIs.…”

Section: Attention Mechanismmentioning

confidence: 99%

LANet: Local Attention Embedding to Improve the Semantic Segmentation of Remote Sensing Images

Ding

Tang

Bruzzone

2021

IEEE Trans. Geosci. Remote Sensing

245

View full text Add to dashboard Cite

The trade-off between feature representation power and spatial localization accuracy is crucial for the dense classification/semantic segmentation of remote sensing images (RSIs). High-level features extracted from the late layers of a neural network are rich in semantic information, yet have blurred spatial details; low-level features extracted from the early layers of a network contain more pixel-level information but are isolated and noisy. It is therefore difficult to bridge the gap between high-and low-level features due to their difference in terms of physical information content and spatial distribution. In this article, we contribute to solve this problem by enhancing the feature representation in two ways. On the one hand, a patch attention module (PAM) is proposed to enhance the embedding of context information based on a patchwise calculation of local attention. On the other hand, an attention embedding module (AEM) is proposed to enrich the semantic information of low-level features by embedding local focus from high-level features. Both proposed modules are lightweight and can be applied to process the extracted features of convolutional neural networks (CNNs). Experiments show that, by integrating the proposed modules into a baseline fully convolutional network (FCN), the resulting local attention network (LANet) greatly improves the performance over the baseline and outperforms other attention-based methods on two RSI data sets.

show abstract

“…Currently, these three types of attention mechanisms have been widely used in the field of RSI segmentation and have achieved good results. For example, Panboonyuen et al 33 . proposed a novel CNN using a global convolutional network, channel attention mechanism and domain-specific transfer learning for RSI segmentation.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised domain adaptation using modified cycle generative adversarial network for aerial image classification

Ren

Jia

Yue

et al. 2022

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

.The unsupervised domain adaptation (UDA) of aerial image classification is of great significance in the traffic management and public safety applications. Due to the differences in the distribution of source and target domain datasets from different sensors and cities, and the acquisition difficulty and high annotation cost of target domain datasets, existing domain adaptive methods have difficulty obtaining high accuracy and stability. In this work, a modified cycle generative adversarial network (CycleGAN) (MGAN) based on the multiscale residual block (MSRB) and multipooling coordinate attention (MPCA) modules is proposed to generate fake images, which is the most critical stage of the UDA method. In the MGAN, the MSRB module can extract the multiscale information of the image, and the MPCA module can fuse the position and channel information in the image. These two modules better extract the features of edges, regions and semantic classes, which makes the generated fake images better retain the information of the source domain and possess the style of the target domain. Otherwise, the Deeplabv3+ model is used to obtain a pretrained and final segmentation model in the first and third stages. In the experiment, the proposed method is validated on the aerial images of the Potsdam and Vaihingen datasets in the International Society for Photogrammetry and Remote Sensing (ISPRS) two-dimensional semantic labeling benchmark and improves overall accuracy from 52% to 60% on the segmentation of the target domain, compared with the existing optimal method; it also achieves an accuracy of 77% for buildings. The results demonstrate that the proposed method can achieve a higher accuracy and stability.

show abstract

Semantic Segmentation on Remotely-Sensed Images Using Enhanced Global Convolutional Network with Channel Attention and Domain Specific Transfer Learning

Cited by 5 publications

References 26 publications

A nested parallel multiscale convolution for cerebrovascular segmentation

A nested parallel multiscale convolution for cerebrovascular segmentation

LANet: Local Attention Embedding to Improve the Semantic Segmentation of Remote Sensing Images

Unsupervised domain adaptation using modified cycle generative adversarial network for aerial image classification

Contact Info

Product

Resources

About