Multi-scale Dilated Convolution Research Articles

A Multi-Level Adaptive Lightweight Net for Damaged Road Marking Detection Based on Knowledge Distillation

To tackle the complexity and limited applicability of high-precision segmentation models for damaged road markings, this study proposes a Multi-level Adaptive Lightweight Network (MALNet) based on knowledge distillation. By incorporating multi-scale dilated convolution and adaptive spatial channel attention fusion modules, the MALNet model significantly enhances the precision, integrity, and robustness of its segmentation branch. Furthermore, it employs an intricate knowledge distillation strategy, channeling rich, layered insights from a teacher model to a student model, thus elevating the latter’s segmentation ability. Concurrently, it streamlines the student model by markedly reducing its parameter count and computational demands, culminating in a segmentation network that is both high-performing and pragmatic. Rigorous testing on three distinct data sets for damaged road marking detection—CDM_P (Collective Damaged road Marking—Public), CDM_H (Collective Damaged road Marking—Highways), and CDM_C (Collective Damaged road Marking—Cityroad)—underscores the MALNet model’s superior segmentation abilities across all damage types, outperforming competing models in accuracy and completeness. Notably, the MALNet model excels in parameter efficiency, computational economy, and throughput. After distillation, the student model’s parameters and computational load decrease to only 31.78% and 27.40% of the teacher model’s, respectively, while processing speeds increase to 1.9 times, demonstrating a significant improvement in lightweight design.

Water body extraction from high spatial resolution remote sensing images based on enhanced U-Net and multi-scale information fusion

Employing deep learning techniques for the semantic segmentation of remote sensing images has emerged as a prevalent approach for acquiring information about water bodies. Yet, current models frequently fall short in accurately extracting water bodies from high-resolution remote sensing images, as these images often present intricate details of terrestrial objects and complex backgrounds. Vegetation, shadows, and other objects close to water boundaries have increased similarity to water bodies. Moreover, water bodies in high-resolution images have different boundary complexities, shapes, and sizes. This situation makes it somewhat challenging to accurately distinguish water bodies in high-resolution images. To overcome these difficulties, this paper presents a novel network model named EU-Net, specifically designed to extract water bodies from high-resolution remote sensing images. The proposed EU-Net model, with U-net as the backbone network, incorporates improved residual connections and attention mechanisms, and designs multi-scale dilated convolution and multi-scale feature fusion modules to enhance water body extraction performance in complex scenarios. Specifically, in the proposed model, improved residual connections are introduced to enable the learning of more complex features; the attention mechanism is employed to improve the model's discriminative ability by focusing on important channels and spatial areas. The implemented multi-scale dilated convolution technique enhances the model's receptive field while maintaining the same number of parameters. The designed multi-scale feature fusion module is capable of processing both small-scale details and large-scale structures in images, while simultaneously modeling the spatial context relationships of features at different scales. Experimental results validate the superior performance of EU-Net in accurately identifying water bodies from high-resolution remote sensing images, outperforming current models in terms of water extraction accuracy.

Multiscale dilated convolutional neural network for Atrial Fibrillation detection.

Atrial Fibrillation (AF), a type of heart arrhythmia, becomes more common with aging and is associated with an increased risk of stroke and mortality. In light of the urgent need for effective automated AF monitoring, existing methods often fall short in balancing accuracy and computational efficiency. To address this issue, we introduce a framework based on Multi-Scale Dilated Convolution (AF-MSDC), aimed at achieving precise predictions with low cost and high efficiency. By integrating Multi-Scale Dilated Convolution (MSDC) modules, our model is capable of extracting features from electrocardiogram (ECG) datasets across various scales, thus achieving an optimal balance between precision and computational savings. We have developed three MSDC modules to construct the AF-MSDC framework and assessed its performance on renowned datasets, including the MIT-BIH Atrial Fibrillation Database and Physionet Challenge 2017. Empirical results unequivocally demonstrate that our technique surpasses existing state-of-the-art (SOTA) methods in the AF detection domain. Specifically, our model, with only a quarter of the parameters of a Residual Network (ResNet), achieved an impressive sensitivity of 99.45%, specificity of 99.64% (on the MIT-BIH AFDB dataset), and an [Formula: see text] score of 85.63% (on the Physionet Challenge 2017 AFDB dataset). This high efficiency makes our model particularly suitable for integration into wearable ECG devices powered by edge computing frameworks. Moreover, this innovative approach offers new possibilities for the early diagnosis of AF in clinical applications, potentially improving patient quality of life and reducing healthcare costs.

An improved attention module based on nnU-Net for segmenting primary central nervous system lymphoma (PCNSL) in MRI images1.

Accurate volumetric segmentation of primary central nervous system lymphoma (PCNSL) is essential for assessing and monitoring the tumor before radiotherapy and the treatment planning. The tedious manual segmentation leads to interindividual and intraindividual differences, while existing automatic segmentation methods cause under-segmentation of PCNSL due to the complex and multifaceted nature of the tumor. To address the challenges of small size, diffused distribution, poor inter-layer continuity on the same axis, and tendency for over-segmentation in brain MRI PCNSL segmentation, we propose an improved attention module based on nnUNet for automated segmentation. We collected 114 T1 MRI images of patients in the Huashan Hospital, Shanghai. Then randomly split the total of 114 cases into 5 distinct training and test sets for a 5-fold cross-validation. To efficiently and accurately delineate the PCNSL, we proposed an improved attention module based on nnU-Net with 3D convolutions, batch normalization, and residual attention (res-attention) to learn the tumor region information. Additionally, multi-scale dilated convolution kernels with different dilation rates were integrated to broaden the receptive field. We further used attentional feature fusion with 3D convolutions (AFF3D) to fuse the feature maps generated by multi-scale dilated convolution kernels to reduce under-segmentation. Compared to existing methods, our attention module improves the ability to distinguish diffuse and edge enhanced types of tumors; and the broadened receptive field captures tumor features of various scales and shapes more effectively, achieving a 0.9349 Dice Similarity Coefficient (DSC). Quantitative results demonstrate the effectiveness of the proposed method in segmenting the PCNSL. To our knowledge, this is the first study to introduce attention modules into deep learning for segmenting PCNSL based on brain magnetic resonance imaging (MRI), promoting the localization of PCNSL before radiotherapy.

Blind recognition of channel codes based on a multiscale dilated convolution neural network

Channel coding is an essential part of communication. In a non-cooperative communication system, the receiver has to recover the signal without any priori knowledge. Therefore, it is important to recognize channel code types and parameters. At present, only a few studies focus on channel code type recognition. In this paper, a multiscale dilated convolution neural network (MSDCNN) is proposed to achieve fine-grained recognition of seven commonly used channel code types. MSDCNN has three multiscale modules. Based on the characteristics of each channel code type, the first two identical modules are designed to extract the overall features of codewords with unit convolution and three parallel regular convolution layers. To lower complexity, the third module uses dilated convolution instead of regular convolution to extract features between codewords. Simulation results show that MSDCNN has higher average recognition accuracy than other neural networks. More importantly, MSDCNN lowers complexity, making it more suitable for deployment so that it can be applied in practical non-cooperative receivers.

Instance-aware image dehazing

The existing deep learning-based image dehazing algorithms commonly employ an encoder–decoder structure to learn a direct mapping from hazy images to haze-free images. However, these state-of-the-art methods often fail to consider the varying contents of hazy images across different scenes, resulting in unsatisfactory dehazing outcomes. To address this issue, this paper attempts to integrate a novel instance-aware subnet into the classic encoder–decoder structure in order to achieve a clear separation between figure and background, conducting the selective incorporation of instance features into the dehazing network. Specifically, we introduce a novel architecture called the hybrid residual attention network, which is capable of separately extracting full-image features and instance-level features. This architecture incorporates attention mechanisms and a multi-scale dilated convolution structure, enabling adaptive perception of haze density in different scenes. Additionally, we introduce a global feature fusion subnet that employs a pixel attention structure to fuse features from the entire image and multiple individual instances, thus being aware of instance features. Compared to existing methods, our approach offers a major advantage in accurately estimating the haze density of individual instances, reducing color distortion, and mitigating noise amplification in the output images. Experimental results demonstrate that our method outperforms existing methods across different evaluation metrics and testing benchmarks. Therefore, we believe that our method will serve as a valuable addition to the current collection of artificial intelligence models and will benefit engineering applications in video surveillance and high-level computer vision tasks.

Attention-based ConvNeXt with a parallel multiscale dilated convolution residual module for fault diagnosis of rotating machinery

Convolutional Neural Networks have promoted development of the fault diagnosis in the machine prognostics and health management. However, the existing methods have limited applicability under strong noisy conditions. We propose an attention-based ConvNeXt with a parallel multiscale dilated convolution residual module for the rotor fault diagnosis. The parallel multiscale dilated convolution residual module is firstly introduced to filter noise and extract multiscale discriminative features intelligently. The multi-head attention module is then embedded in ConvNeXt for global discriminative features adaptively. Focal Loss is finally implemented for identifying hard-to-classify samples to further improve the diagnostic accuracy. The application results demonstrate the superiority and robustness of our method in two case studies.

Hyperspectral image classification method based on narrowing semantic gap convolutional neural network

ABSTRACT In recent years, numerous hyperspectral image classification methods based on convolutional neural network (CNN) have been proposed in recent years. However, the majority of CNN-based classification models fail to take into account the semantic gap between shallow and deep features, and instead, just sum up or splice these two types of features, which leads to subpar feature fusion. To address the above problems, the paper proposes a hyperspectral image classification method based on Narrowing Semantic Gap Convolutional Neural Network (NSGCNN). We developed the feature fusion sub-network, which consists of Feature Semantic Enhancement Module (FSEM), Spatial Detail Supplement Module (SDCM), and Cross-layer Feature Fusion Module (CFFM). FSEM extracts rich global semantic information based on multi-scale dilated convolution. Reverse attention mechanism is introduced in the SDCM to complement deep-layer features with fine spatial features like edges and textures. The fusing of hierarchical features is then achieved using CFFM. Additionally, we can regulate the model’s parameter count by regulating the number of output feature maps. Experimental results on three benchmark hyperspectral datasets demonstrate that NSGCNN outperforms other state-of-the-art methods in terms of classification performance and significantly outperforms them in terms of model complexity.

MMDB: Multimodal dual-branch model for multi-functional bioactive peptide prediction

Bioactive peptides can hinder oxidative processes and microbial spoilage in foodstuffs and play important roles in treating diverse diseases and disorders. While most of the methods focus on single-functional bioactive peptides and have obtained promising prediction performance, it is still a significant challenge to accurately detect complex and diverse functions simultaneously with the quick increase of multi-functional bioactive peptides. In contrast to previous research on multi-functional bioactive peptide prediction based solely on sequence, we propose a novel multimodal dual-branch (MMDB) lightweight deep learning model that designs two different branches to effectively capture the complementary information of peptide sequence and structural properties. Specifically, a multi-scale dilated convolution with Bi-LSTM branch is presented to effectively model the different scales sequence properties of peptides while a multi-layer convolution branch is proposed to capture structural information. To the best of our knowledge, this is the first effective extraction of peptide sequence features using multi-scale dilated convolution without parameter increase. Multimodal features from both branches are integrated via a fully connected layer for multi-label classification. Compared to state-of-the-art methods, our MMDB model exhibits competitive results across metrics, with a 9.1% Coverage increase and 5.3% and 3.5% improvements in Precision and Accuracy, respectively.

Fault Diagnosis of the Rolling Bearing by a Multi-Task Deep Learning Method Based on a Classifier Generative Adversarial Network.

Accurate fault diagnosis is essential for the safe operation of rotating machinery. Recently, traditional deep learning-based fault diagnosis have achieved promising results. However, most of these methods focus only on supervised learning and tend to use small convolution kernels non-effectively to extract features that are not controllable and have poor interpretability. To this end, this study proposes an innovative semi-supervised learning method for bearing fault diagnosis. Firstly, multi-scale dilated convolution squeeze-and-excitation residual blocks are designed to exact local and global features. Secondly, a classifier generative adversarial network is employed to achieve multi-task learning. Both unsupervised and supervised learning are performed simultaneously to improve the generalization ability. Finally, supervised learning is applied to fine-tune the final model, which can extract multi-scale features and be further improved by implicit data augmentation. Experiments on two datasets were carried out, and the results verified the superiority of the proposed method.

Multiscale dilated denoising convolution with channel attention mechanism for micro-seismic signal denoising

Denoising micro-seismic signals is paramount for ensuring reliable data for localizing mining-related seismic events and analyzing the state of rock masses during mining operations. However, micro-seismic signals are commonly contaminated by various types of complex noise, which can hinder micro-seismic accurate P-wave pickup and analysis. In this study, we propose the Multiscale Dilated Convolutional Attention denoising method, referred to as MSDCAN, to eliminate complex noise interference. The MSDCAN denoising model consists of an encoder, an improved attention mechanism, and a decoder. To effectively capture the neighborhood features and multiscale features of the micro-seismic signal, we construct an initial dilated convolution block and a multiscale dilated convolution block in the encoder, and the encoder focuses on extracting the relevant feature information, thus eliminating the noise interference and improving the signal-to-noise ratio (SNR). In addition, the attention mechanism is improved and introduced between the encoder and decoder to emphasize the key features of the micro-seismic signal, thus removing the complex noise and further improving the denoising performance. The MSDCAN denoising model is trained and evaluated using micro-seismic data from Stanford University. Experimental results demonstrate an impressive increase in SNR by 11.237 dB and a reduction in root mean square error (RMSE) by 0.802. Compared to the denoising results of the DeepDenoiser, CNN-denoiser and Neighbor2Neighbor methods, the MSDCAN denoising model outperforms them by enhancing the SNR by 2.589 dB, 1.584 dB and 2dB, respectively, and reducing the RMSE by 0.219, 0.050 and 0.188, respectively. The MSDCAN denoising model presented in this study effectively improves the SNR of micro-seismic signals, offering fresh insights into micro-seismic signal denoising methodologies.

Deep multi-scale dilated convolution network for coronary artery segmentation

Automatic segmentation of coronary arteries is of great significance for the rapid and accurate detection of cardiovascular diseases. Currently, deep learning has been successfully applied in the field of coronary artery segmentation. However, the branch structure of coronary arteries is thin, and the contrast between the blood vessels and the background is relatively low, making branches difficult to identify and the false positive rate is high. In response to these challenges, we proposed a multi-scale dilated convolution and deep information extraction network based on unet, which we called 3D-MDCNET. Firstly, adaptive scale expansion convolution modules are designed based on different layers. The advantage is to expand the receptive field and extract a larger range of information, thereby improving the continuity of small branches, while avoiding excessive computational costs. Secondly, the information from different layers of the decoder in Unet is fused with the first-stage segmentation results. Using multi-scale information fusion to enhance information expression, and applying the depth information extraction module to refine the results, effectively reducing the false positive rate. Finally, we introduce deep supervision as a mechanism to mitigate vanishing and exploding gradient problems caused by deep models. By conducting experiments on a benchmark dataset of coronary artery segmentation, our method indeed improves the continuity of small branch segmentation results while reducing the false positive rate. The proposed method has good segmentation performance and generalization ability, outperforming multiple state-of-the-art algorithms on various indicators.

MsDC-DEQ-Net: Deep Equilibrium Model (DEQ) with Multiscale Dilated Convolution for Image Compressive Sensing (CS)

Compressive sensing (CS) is a technique that enables the recovery of sparse signals using fewer measurements than traditional sampling methods. To address the computational challenges of CS reconstruction, our objective is to develop an interpretable and concise neural network model for reconstructing natural images using CS. We achieve this by mapping one step of the iterative shrinkage thresholding algorithm (ISTA) to a deep network block, representing one iteration of ISTA. To enhance learning ability and incorporate structural diversity, we integrate aggregated residual transformations (ResNeXt) and squeeze-and-excitation mechanisms into the ISTA block. This block serves as a deep equilibrium layer connected to a semi-tensor product network for convenient sampling and providing an initial reconstruction. The resulting model, called MsDC-DEQ-Net, exhibits competitive performance compared to state-of-the-art network-based methods. It significantly reduces storage requirements compared to deep unrolling methods, using only one iteration block instead of multiple iterations. Unlike deep unrolling models, MsDC-DEQ-Net can be iteratively used, gradually improving reconstruction accuracy while considering computation tradeoffs. Additionally, the model benefits from multiscale dilated convolutions, further enhancing performance.

CTFNet: Long-Sequence Time-Series Forecasting Based on Convolution and Time-Frequency Analysis.

Although current time-series forecasting methods have significantly improved the state-of-the-art (SOTA) results for long-sequence time-series forecasting (LSTF), they still have difficulty in capturing and extracting the features and dependencies of long-term sequences and suffer from information utilization bottlenecks and high-computational complexity. To address these issues, a lightweight single-hidden layer feedforward neural network (SLFN) combining convolution mapping and time-frequency decomposition called CTFNet is proposed with three distinctive characteristics. First, time-domain (TD) feature mining-in this article, a method for extracting the long-term correlation of horizontal TD features based on matrix factorization is proposed, which can effectively capture the interdependence among different sample points of a long time series. Second, multitask frequency-domain (FD) feature mining-this can effectively extract different frequency feature information of time-series data from the FD and minimize the loss of data features. Integrating multiscale dilated convolutions, simultaneously focusing on both global and local context feature dependencies at the sequence level, and mining the long-term dependencies of the multiscale frequency information and the spatial dependencies among the different scale frequency information, break the bottleneck of data utilization, and ensure the integrity of feature extraction. Third, highly efficient-the CTFNet model has a short training time and fast inference speed. Our empirical studies with nine benchmark datasets show that compared with state-of-the-art methods, CTFNet can reduce prediction error by 64.7% and 53.7% for multivariate and univariate time series, respectively.

Plant leaf infected spot segmentation using robust encoder-decoder cascaded deep learning model

ABSTRACT Leaf infection detection and diagnosis at an earlier stage can improve agricultural output and reduce monetary costs. An inaccurate segmentation may degrade the accuracy of disease classification due to some different and complex leaf diseases. Also, the disease’s adhesion and dimension can overlap, causing partial under-segmentation. Therefore, a novel robust Deep Encoder-Decoder Cascaded Network (DEDCNet) model is proposed in this manuscript for leaf image segmentation that precisely segments the diseased leaf spots and differentiates similar diseases. This model is comprised of an Infected Spot Recognition Network and an Infected Spot Segmentation Network. Initially, ISRN is designed by integrating cascaded CNN with a Feature Pyramid Pooling layer to identify the infected leaf spot and avoid an impact of background details. After that, the ISSN developed using an encoder-decoder network, which uses a multi-scale dilated convolution kernel to precisely segment the infected leaf spot. Moreover, the resultant leaf segments are provided to the pre-learned CNN models to learn texture features followed by the SVM algorithm to categorize leaf disease classes. The ODEDCNet delivers exceptional performance on both the Betel Leaf Image and PlantVillage datasets. On the Betel Leaf Image dataset, it achieves an accuracy of 94.89%, with high precision (94.35%), recall (94.77%), and F-score (94.56%), while maintaining low under-segmentation (6.2%) and over-segmentation rates (2.8%). It also achieves a remarkable Dice coefficient of 0.9822, all in just 0.10 seconds. On the PlantVillage dataset, the ODEDCNet outperforms other existing models with an accuracy of 96.5%, demonstrating high precision (96.61%), recall (96.5%), and F-score (96.56%). It excels in reducing under-segmentation to just 3.12% and over-segmentation to 2.56%. Furthermore, it achieves a Dice coefficient of 0.9834 in a mere 0.09 seconds. It evident for the greater efficiency on both segmentation and categorization of leaf diseases contrasted with the existing models.

VSTNet: Robust watermarking scheme based on voxel space transformation for diffusion tensor imaging images

Deep watermarking has been successfully applied to real-value domain (e.g., CT, MRI, DWI) medical images for copyright protection; however, manifold-value domain images, such as diffusion tensor imaging (DTI), have received less academic attention. The non-Euclidean nature of DTI images hinders deep models from generating plausible DTI images because it is difficult to guarantee that each generated DTI image voxel lies on a 3 × 3 symmetric positive definite manifold. In this paper, we propose a robust watermarking scheme based on voxel space transformation for DTI images. First, a voxel space transformation is proposed, which effectively solves the problem of fiber orientation estimation in Riemann networks for DTI images and obtains the eigenvalues and eigenvectors of each voxel using singular value decomposition (SVD) to transform the DTI voxels from the 3 × 3 symmetric positive definite manifold space to Euclidean space. Second, SVD is combined with a deep neural network to extract the high-level features of each voxel eigenvalue in the DTI images and embedding watermarking messages, avoiding the false positive errors problem in traditional watermarking algorithms based on an SVD transform. Finally, using a deep neural network combining multiscale dilated convolution, dense residual connection and channel attention, the algorithm demonstrates high robustness against various intentional or unintentional attacks on DTI images while ensuring the good visual quality and diffusion characteristics of the DTI images embedded with watermarking messages. Experiments show that the watermarking message extraction accuracy reaches 97.8% for Crop (p = 0.7) attacks and 95.4% for Gaussian Blur (k = 7) attacks.

E2LNet: An Efficient and Effective Lightweight Network for Panoramic Depth Estimation.

Monocular panoramic depth estimation has various applications in robotics and autonomous driving due to its ability to perceive the entire field of view. However, panoramic depth estimation faces two significant challenges: global context capturing and distortion awareness. In this paper, we propose a new framework for panoramic depth estimation that can simultaneously address panoramic distortion and extract global context information, thereby improving the performance of panoramic depth estimation. Specifically, we introduce an attention mechanism into the multi-scale dilated convolution and adaptively adjust the receptive field size between different spatial positions, designing the adaptive attention dilated convolution module, which effectively perceives distortion. At the same time, we design the global scene understanding module to integrate global context information into the feature maps generated using the feature extractor. Finally, we trained and evaluated our model on three benchmark datasets which contains the virtual and real-world RGB-D panorama datasets. The experimental results show that the proposed method achieves competitive performance, comparable to existing techniques in both quantitative and qualitative evaluations. Furthermore, our method has fewer parameters and more flexibility, making it a scalable solution in mobile AR.

Generative adversarial networks with multi-scale and attention mechanisms for underwater image enhancement

The images captured underwater are usually degraded due to the effects of light absorption and scattering. Degraded underwater images exhibit color distortion, low contrast, and blurred details, which in turn reduce the accuracy of marine biological monitoring and underwater object detection. To address this issue, a generative adversarial network with multi-scale and an attention mechanism is proposed to improve the quality of underwater images. To extract more effective features within the generative network, several modules are introduced: a multi-scale dilated convolution module, a novel attention module, and a residual module. These modules are utilized to design a generative network with a U-shaped structure. The multi-scale dilated convolution module is designed to extract features at multiple scales and expand the receptive field to capture more global information. The attention module directs the network’s focus towards important features, thereby reducing the interference from redundant feature information. To improve the discriminative power of the adversarial network, a multi-scale discriminator is designed. It has two output feature maps with different scales. Additionally, an improved loss function for the generative adversarial network is proposed. This improvement involves incorporating the total variation loss into the traditional loss function. The performance of different methods for enhancing underwater images is evaluated using the EUVP dataset and UIEB dataset. The experimental results demonstrate that the enhanced underwater images exhibit better quality and visual effects compared to other methods.

Dual Parallel Branch Fusion Network for Road Segmentation in High-Resolution Optical Remote Sensing Imagery

Road segmentation from high-resolution (HR) remote sensing images plays a core role in a wide range of applications. Due to the complex background of HR images, most of the current methods struggle to extract a road network correctly and completely. Furthermore, they suffer from either the loss of context information or high redundancy of details information. To alleviate these problems, we employ a dual branch dilated pyramid network (DPBFN), which enables dual-branch feature passing between two parallel paths when it is merged to a typical road extraction structure. A DPBFN consists of three parts: a residual multi-scaled dilated convolutional network branch, a transformer branch, and a fusion module. Constructing pyramid features through parallel multi-scale dilated convolution operations with multi-head attention block can enhance road features while suppressing redundant information. Both branches after fusing can solve shadow or vision occlusions and maintain the continuity of the road network, especially on a complex background. Experiments were carried out on three datasets of HR images to showcase the stable performance of the proposed method, and the results are compared with those of other methods. The OA in the three data sets of Massachusetts, Deep Globe, and GF-2 can reach more than 98.26%, 95.25%, and 95.66%, respectively, which has a significant improvement compared with the traditional CNN network. The results and explanation analysis via Grad-CAMs showcase the effective performance in accurately extracting road segments from a complex scene.

Multi-scale feature fusion network with local attention for lung segmentation

Computer-assisted medical care can benefit from the lung region segmentation method. Numerous methods provide end-to-end solutions, these methods employ convolution neural networks to segment lung regions from images. The low contrast, unpredictable appearance, and other problems in medical images have an effect on the accuracy of existing methods. In order to overcome the aforementioned issues, the MSDC (multi-scale dilated convolution) module is added to the short-cut connection, so as to fuse multi-scale features with various receptive fields to obtain more global information of lung area. Moreover, a local attention module which includes channel attention and spatial attention is suggested to give more weight to the lung area to lower the influence of background. Several lung segmentation datasets are employed to evaluate the segmentation performance of images qualitatively and quantitatively. From the experimental results, we can see that the segmentation accuracy of our model outperforms many recent image segmentation methods.