Deep Semantic Segmentation Network Research Articles

Identification of sweetpotato virus disease-infected leaves from field images using deep learning

IntroductionSweetpotato virus disease (SPVD) is widespread and causes significant economic losses. Current diagnostic methods are either costly or labor-intensive, limiting both efficiency and scalability.MethodsThe segmentation algorithm proposed in this study can rapidly and accurately identify SPVD lesions from field-captured photos of sweetpotato leaves. Two custom datasets, DS-1 and DS-2, are utilized, containing meticulously annotated images of sweetpotato leaves affected by SPVD. DS-1 is used for training, validation, and testing the model, while DS-2 is exclusively employed to validate the model’s reliability. This study employs a deep learning-based semantic segmentation network, DeepLabV3+, integrated with an Attention Pyramid Fusion (APF) module. The APF module combines a channel attention mechanism with multi-scale feature fusion to enhance the model’s performance in disease pixel segmentation. Additionally, a novel data augmentation technique is utilized to improve recognition accuracy in the edge background areas of real large images, addressing issues of poor segmentation precision in these regions. Transfer learning is applied to enhance the model’s generalization capabilities.ResultsThe experimental results indicate that the model, with 62.57M parameters and 253.92 Giga Floating Point Operations Per Second (GFLOPs), achieves a mean Intersection over Union (mIoU) of 94.63% and a mean accuracy (mAcc) of 96.99% on the DS-1 test set, and an mIoU of 78.59% and an mAcc of 79.47% on the DS-2 dataset.DiscussionAblation studies confirm the effectiveness of the proposed data augmentation and APF methods, while comparative experiments demonstrate the model’s superiority across various metrics. The proposed method also exhibits excellent detection results in simulated scenarios. In summary, this study successfully deploys a deep learning framework to segment SPVD lesions from field images of sweetpotato foliage, which will contribute to the rapid and intelligent detection of sweetpotato diseases.

LSL-SS-Net: level set loss-guided semantic segmentation networks for landslide extraction

ABSTRACT A landslide inventory is essential for numerous applications, including disaster prevention and mitigation. The emergence of advanced satellite technologies and the proliferation of extensive satellite imagery have greatly enhanced the role of deep learning-based semantic segmentation networks in the development of comprehensive landslide inventories. In deep learning, especially in semantic segmentation tasks, the loss function is crucial to guide the training of neural networks. However, typical loss functions have several challenges and shortcomings, including difficulty in handling objects of different scales and neglecting spatial correlation. These drawbacks hinder semantic segmentation networks from solely concentrating on accurately predicting landslide categories with rich semantic information. The spatial details (e.g. small landslides and precise landslide boundary information) in the segmentation results may be lost, reducing the accuracy of landslide extraction. Most studies only focus on training and detection in the local region. Consequently, detecting new unexplored region and detrimental to emergency response is challenging. This study presents a level set loss-guided semantic segmentation network, which can be integrated with different semantic segmentation networks, to address these issues. Three out of five study areas (e.g. Bijie, Jiuzhaigou, and Taitung) are used for model training and accuracy evaluation, while the rest, which are newly unexplored regions (e.g. Hokkaido and Haiti), are used to test the transferability of the proposed loss function in different regions. The study results validate the effectiveness of the designed level set loss in various semantic segmentation networks and in different study areas for landslide extraction.

A machine vision‐based intelligent segmentation method for dam underwater cracks using swarm optimization algorithm and deep learning

AbstractEnsuring the safety of water networks is a research hotspot in the current water conservancy industry, and dams are an important part. However, over time, the dam is prone to varying degrees of aging and disease, most of which are structural cracks. If they cannot be discovered and repaired in time, the normal operation of the dam will be affected, and even catastrophic accidents such as dam failure will occur. However, complex backgrounds and blurred images can easily lead to misjudgments by machine vision detection models, and high‐efficiency and accurate detection and evaluation technology are urgently needed. This paper combines the deep semantic segmentation network and the model hyperparameters optimization algorithm to propose a data‐intelligent perception method of dam underwater cracks driven by knowledge coupling. Taking the underwater detection of a concrete face rockfill dam as an example, the effectiveness of the model is verified by using the underwater vehicle as the carrier. Experimental results indicate that the developed method achieves an intersection‐union ratio of 0.9301, a precision rate of 0.9678, a precision rate of 0.9472, and a recall rate of 0.9577 in the test set. This shows that the constructed method has a high crack fine detection performance. In addition, the developed method has better segmentation performance in different complex underwater crack scenes, which further illustrates the high performance of the developed method.

Extraction of water bodies from high-resolution remote sensing imagery based on a deep semantic segmentation network

The precise delineation of urban aquatic features is of paramount importance in scrutinizing water resources, monitoring floods, and devising water management strategies. Addressing the challenge of indistinct boundaries and the erroneous classification of shadowed regions as water in high-resolution remote sensing imagery, we introduce WaterDeep, which is a novel deep learning framework inspired by the DeepLabV3 + architecture and an innovative fusion mechanism for high- and low-level features. This methodology first creates a comprehensive dataset of high-resolution remote sensing images, then progresses through the Xception baseline network for low-level feature extraction, and harnesses densely connected Atrous Spatial Pyramid Pooling (ASPP) modules to assimilate multi-scale data into sophisticated high-level features. Subsequently, the network decoder amalgamates the elemental and intricate features and applies dual-line interpolation to the amalgamated dataset to extract aqueous formations from the remote images. Experimental evidence substantiates that WaterDeep outperforms its existing deep learning counterparts, achieving a stellar overall accuracy of 99.284%, FWIoU of 95.58%, precision of 97.562%, recall of 95.486%, and F1 score of 96.513%. It also excels in the precise demarcation of edges and the discernment of shadows cast by urban infrastructure. The superior efficacy of the proposed method in differentiating water bodies in complex urban environments has significant practical applications in real-world contexts.

Enhancing visual autonomous navigation in row-based crops with effective synthetic data generation

Abstract Introduction Service robotics is recently enhancing precision agriculture enabling many automated processes based on efficient autonomous navigation solutions. However, data generation and in-field validation campaigns hinder the progress of large-scale autonomous platforms. Simulated environments and deep visual perception are spreading as successful tools to speed up the development of robust navigation with low-cost RGB-D cameras. Materials and methods In this context, the contribution of this work resides in a complete framework to fully exploit synthetic data for a robust visual control of mobile robots. A wide realistic multi-crops dataset is accurately generated to train deep semantic segmentation networks and enabling robust performance in challenging real-world conditions. An automatic parametric approach enables an easy customization of virtual field geometry and features for a fast reliable evaluation of navigation algorithms. Results and conclusion The high quality of the generated synthetic dataset is demonstrated by an extensive experimentation with real crops images and benchmarking the resulting robot navigation both in virtual and real fields with relevant metrics.

A Deep Learning Method for Arctic Sea Ice Type Classification Based on Active-Passive Microwave Data

The Arctic-scale classification of sea ice type is important in the fields of macro-monitoring in the Arctic, climate change assessment, and long time-series interannual variability of sea ice. To date, a lack of deep learning-based algorithms for detecting Arctic sea ice types from microwave sensors is comparatively rare. In this study, a sea ice type classification algorithm based on the U-Net-CBAM deep learning semantic segmentation network was developed by carrying out research on sea ice type classification algorithms in the Arctic through the integrated use of ASCAT scatterometer and AMSR2 radiometer data. The results of the study show that the proposed method achieves an impressive average accuracy of 94.1% in extracting sea ice categories using AARI ice maps. The results of the algorithm show a high degree of agreement with the OSI-SAF sea ice type daily products.

Sea ice detection network for icebreakers in polar environments with attention-based deeplabv3+ architecture

Shipborne sea ice detection aboard icebreakers plays a paramount role in polar navigation. The continuous evolution of deep learning semantic segmentation networks has promoted the advancement of sea ice detection tasks. At this stage, there are relatively few studies on shipboard sea ice detection, and the accuracy of polar sea ice detection will be reduced due to problems such as blurred sea fog and indistinct boundaries. In this study, a shipboard sea ice detection dataset is constructed, and a sea ice detection method that combines multi-branch attention feature alignment and multi-scale feature extraction is proposed. The heterogeneous receptive field enhancement atrous spatial convolution pooling pyramid module is designed, and the feature alignment module based on the attention mechanism is constructed, which strengthens the model’s extraction of sea ice features and elevates representation performance. Experimental results underscore the heightened precision of our approach in sea ice detection, to some extent alleviating the issue of missed detections in new ice. It constitutes a positive contribution towards advancing shipborne sea ice detection in polar environments.

Pointer Meter Reading Recognition by Joint Detection and Segmentation

To handle the task of pointer meter reading recognition, in this paper, we propose a deep network model that can accurately detect the pointer meter dial and segment the pointer as well as the reference points from the located meter dial. Specifically, our proposed model is composed of three stages: meter dial location, reference point segmentation, and dial number reading recognition. In the first stage, we translate the task of meter dial location into a regression task, which aims to separate bounding boxes by an object detection network. This results in the accurate and fast detection of meter dials. In the second stage, the dial region image determined by the bounding box is further processed by using a deep semantic segmentation network. After that, the segmented output is used to calculate the relative position between the pointer and reference points in the third stage, which results in the final output of reading recognition. Some experiments were conducted on our collected dataset, and the experimental results show the effectiveness of our method, with a lower computational burden compared to some existing works.

ACDSSNet: Atrous Convolution-Based Deep Semantic Segmentation Network for Efficient Detection of Sickle Cell Anemia.

In medical image processing, semantic segmentation plays an important role since, in most applications, it is required to find the exact location of the anomaly. It is tough than the segmentation or classification task since in this task class-belongingness of each pixel is predicted. The presence of noise, and variations of viewpoint, shape, and size of cells make it more challenging. In this work, two novel Atrous Convolution-based Deep Semantic Segmentation Networks: ACDSSNet-I, ACDSSNet-II are proposed for more accurate Sickle Cell Anemia (SCA) detection, which can mitigate these issues. The main contributions are: 1) Improvement of feature extraction performance by employing Atrous convolution-based dense prediction, which yields varying field-view with adaptive resolution; 2) Employment of Atrous spatial pyramid-based pooling resulting in more robust segmentation; 3) Upgrading the segmentation performance by adding an efficient decoder module to finetune the segmentation, particularly at object boundaries; 4) Design of modified DeepLabV3+ architectures (MDA) by introducing computationally efficient MobileNetV2 or ResNet50 as a base classifier; 5) Further performance improvement has been accomplished by hybridizing MDA-1 with MDA-2 by integrating the benefits of MobileNetV2 models and ADAM and SGDM optimizers; 6) Improvement of overall performance by efficiently utilizing the input image's saturation information only to minimize the false positive. Furthermore, the optimal selection of threshold value makes the hybridization of MDA-1 with MDA-2 efficient resulting in more accurate semantic segmentation. The experimental results illustrate the proposed model outperforms others with the best semantic segmentation performances: 98.21% accuracy, 99.00% specificity, and 0.9547 DSC value.

A Pre-Procession Module for Point-Based Deep Learning in Dense Point Clouds in the Ship Engineering Field

Recently, point cloud technology has been applied in the ship engineering field. However, the dense point cloud acquired by terrestrial laser scanning (TLS) technology in ship engineering applications brings an obstacle to some powerful and advanced point-based deep learning point cloud processing methods. This paper presents a deep learning pre-procession module to ensure the feasibility of processing dense point clouds on commonly available computer devices. The pre-procession module is designed according to the traditional point cloud processing methods and the PointNet++ paradigm, and is evaluated on two ship structure datasets and two popular point cloud datasets. Experimental results illustrate that (i) the proposed module improves the performance of point-based deep learning semantic segmentation networks, and (ii) the proposed module empowers the existing point-based deep learning networks with the capability to process dense input point clouds. The proposed module may provide a useful semantic segmentation tool for realistic dense point clouds in various industrial applications.

A deep model towards accurate boundary location and strong generalization for medical image segmentation

Accurate medical image segmentation plays a crucial role in computer-assisted diagnosis and monitoring. However, due to the complexity of medical images and the limitations of image acquisition, most of the current segmentation models still face the challenges of imprecise boundary positioning and poor generalization ability, which greatly affect their application in clinical practices. To address these issues, we propose a new deep learning-based medical image semantic segmentation network (BLG-Net). To improve the accuracy of object boundary location, a Local–global module (Logo) is proposed. It uses the combination of a self-attention mechanism and a cross-attention mechanism to model intra-sample and inter-sample long-range dependencies from local and global perspectives, respectively, to help recover the spatial structures. To further enhance the generalization ability of the BLG-Net, we develop a generalization enhancement module (GEM), which can adaptively adjust the classifier weights of the network according to the features of the test images extracted during the test process to alleviate the distribution shift issues. Besides, we design a multi-granularity loss (MGL) function, which can guide the network to learn the object region and its boundary details at the pixel, patch, and map levels, respectively. Extensive experiments on five medical image segmentation tasks demonstrate the improvements of our method. We further verify the model’s generalization ability to unseen data on two other popular datasets with different modalities.

Enhancing glomeruli segmentation through cross-species pre-training

The importance of kidney biopsy, a medical procedure in which a small tissue sample is extracted from the kidney for examination, is increasing due to the rising incidence of kidney disorders. This procedure helps diagnosing several kidney diseases which are cause of kidney function changes, as well as guiding treatment decisions, and evaluating the suitability of potential donor kidneys for transplantation. In this work, a deep learning system for the automatic segmentation of glomeruli in biopsy kidney images is presented. A novel cross-species transfer learning approach, in which a semantic segmentation network is trained on mouse kidney tissue images and then fine-tuned on human data, is proposed to boost the segmentation performance. The experiments conducted using two deep semantic segmentation networks, MobileNet and SegNeXt, demonstrated the effectiveness of the cross-species pre-training approach leading to an increased generalization ability of both models.

Towards evaluating the robustness of deep neural semantic segmentation networks with Feature-Guided Method

Although deep neural networks (DNNs) demonstrated its superior performance in computer vision, recent works proved that the vulnerability of deep neural task systems to carefully crafted human-imperceptible perturbations. We observe that the majority of adversarial attack methods for generating perturbations are based on rotation or modification of the input-diagnostic. Regardless of the characteristics completed by additional sources, such approaches yield comparable properties to enrich initial details. It motivates us to consider views about using various information apart from original inputs to deliver adversarial features. For such needs, we induce a simple yet adaptable adversarial attack strategy with a Feature-Guided Method (FGM) for crafting adversarial examples (AEs) in segmentation domain. FGM first induces the multi-source patterns that are apart from the original inputs. Then, producing feature diversities generated with the original data to deliver the perturbed components. Finally, FGM blends original inputs with created features in a defined norm constraint to form the adversarial examples. In such way, it preserves the original positive class-general characteristics and enriches the new positive class-specific diversities when performing adversarial attacks. Moreover, FGM employs the adaptive gradient-based strategy on such generated information, which lowers the risk of falling into the local optimum when searching for the decision boundary of the source and target models in latent space. We conduct detailed experiments to evaluate the performance of the proposed method compared to baselines on public segmentation models. The experimental results reveal better performance of FGMs in fooling source and target segmentation systems leading large margins over 5% on mIoU, mRec, and mAcc. We also deploy the adversarial training with proposed work and PGD on widely used models. Our approach improves the robust quality of adversarially trained models on FCN, PSPNet, and DeepLabv3 with various backbones by significant margins with over 13% improvement on mIoU and 12% on mRec, which indicates a better impact of deploying such mechanisms on robust deep neural segmentation models.

UAV Remote-Sensing Image Semantic Segmentation Strategy Based on Thermal Infrared and Multispectral Image Features

The availability of high-resolution imagery resources for semantic segmentation research has expanded significantly due to the rapid development of remote sensing technology utilizing Unmanned Aerial Vehicles (UAVs). These images provide researchers with a more accurate view of the region of interest and allow for more detailed analysis and interpretation of the images. However, semantic segmentation based on UAV remote sensing imagery still faces new challenges in deriving ground objects. In contrast to the commonly used multispectral (MS) imagery, thermal infrared (TIR) imagery can record the emission of ground objects, making the temperature characteristics of TIR imagery and the color characteristics of MS imagery complementary. These two approaches can be used synergistically to provide more comprehensive image information. On this basis, we propose a strategy for semantic segmentation of UAV images by utilizing both TIR and MS image features. The approach combines principal component analysis (PCA) transformation with a deep learning semantic segmentation network, namely Deeplv3. The effectiveness of the proposed strategy is evaluated by comparing it with both traditional supervised classification algorithms and deep learning algorithms. According to the results, the proposed strategy exhibits greater robustness, achieving a mean pixel accuracy (MPA) of 92.8% and a mean intersection over union (MIOU) of 73.5%. These results outperform several classical deep learning semantic segmentation algorithms that were also evaluated. The proposed strategy would be beneficial to promote the development of semantic segmentation technology for UAV remote sensing images.

Relationships between quantitative retinal microvascular characteristics and cognitive function based on automated artificial intelligence measurements.

Introduction: The purpose of this study is to assess the relationship between retinal vascular characteristics and cognitive function using artificial intelligence techniques to obtain fully automated quantitative measurements of retinal vascular morphological parameters. Methods: A deep learning-based semantic segmentation network ResNet101-UNet was used to construct a vascular segmentation model for fully automated quantitative measurement of retinal vascular parameters on fundus photographs. Retinal photographs centered on the optic disc of 3107 participants (aged 50-93years) from the Beijing Eye Study 2011, a population-based cross-sectional study, were analyzed. The main parameters included the retinal vascular branching angle, vascular fractal dimension, vascular diameter, vascular tortuosity, and vascular density. Cognitive function was assessed using the Mini-Mental State Examination (MMSE). Results: The results showed that the mean MMSE score was 26.34 ± 3.64 (median: 27; range: 2-30). Among the participants, 414 (13.3%) were classified as having cognitive impairment (MMSE score < 24), 296 (9.5%) were classified as mild cognitive impairment (MMSE: 19-23), 98 (3.2%) were classified as moderate cognitive impairment (MMSE: 10-18), and 20 (0.6%) were classified as severe cognitive impairment (MMSE < 10). Compared with the normal cognitive function group, the retinal venular average diameter was significantly larger (p = 0.013), and the retinal vascular fractal dimension and vascular density were significantly smaller (both p < 0.001) in the mild cognitive impairment group. The retinal arteriole-to-venular ratio (p = 0.003) and vascular fractal dimension (p = 0.033) were significantly decreased in the severe cognitive impairment group compared to the mild cognitive impairment group. In the multivariate analysis, better cognition (i.e., higher MMSE score) was significantly associated with higher retinal vascular fractal dimension (b = 0.134, p = 0.043) and higher retinal vascular density (b = 0.152, p = 0.023) after adjustment for age, best corrected visual acuity (BCVA) (logMAR) and education level. Discussion: In conclusion, our findings derived from an artificial intelligence-based fully automated retinal vascular parameter measurement method showed that several retinal vascular morphological parameters were correlated with cognitive impairment. The decrease in retinal vascular fractal dimension and decreased vascular density may serve as candidate biomarkers for early identification of cognitive impairment. The observed reduction in the retinal arteriole-to-venular ratio occurs in the late stages of cognitive impairment.

Study on Single-Tree Extraction Method for Complex RGB Point Cloud Scenes

With the development of sensor technology and point cloud generation techniques, there has been an increasing amount of high-quality forest RGB point cloud data. However, popular clustering-based point cloud segmentation methods are usually only suitable for pure forest scenes and not ideal for scenes with multiple ground features or complex terrain. Therefore, this study proposes a single-tree point cloud extraction method that combines deep semantic segmentation and clustering. This method first uses a deep semantic segmentation network, Improved-RandLA-Net, which is developed based on RandLA-Net, to extract point clouds of specified tree species by adding an attention chain to improve the model’s ability to extract channel and spatial features. Subsequently, clustering is employed to extract single-tree point clouds from the segmented point clouds. The feasibility of the proposed method was verified in the Gingko site, the Lin’an Pecan site, and a Fraxinus excelsior site in a conference center. Finally, semantic segmentation was performed on three sample areas using pre- and postimproved RandLA-Net. The experiments demonstrate that Improved-RandLA-Net had significant improvements in Accuracy, Precision, Recall, and F1 score. At the same time, based on the semantic segmentation results of Improved-RandLA-Net, single-tree point clouds of three sample areas were extracted, and the final single-tree recognition rates for each sample area were 89.80%, 75.00%, and 95.39%, respectively. The results demonstrate that our proposed method can effectively extract single-tree point clouds in complex scenes.

Distinguishing Ulva prolifera and Sargassum horneri by using multi-feature-based ResUnet algorithm

In recent years, two types of macroalgae, namely, Ulva prolifera and Sargassum horneri, have appeared occasionally together in the Yellow Sea and the East China Sea. Remote sensing enables timely and cost-effective observation of macroalgae across large areas. In the available studies, the recognition and classification of the two macroalgae are primarily based on spectral difference analysis. In this study, the spectral features, indices and textural feature parameters of the macroalgae targets were extracted and a preliminary multi-feature dataset was constructed based on Sentinel-2 images. Feature selection was performed using SHAP-based importance analysis and Bhattacharyya distance. From this, a multi-feature dataset was created and used as an input to a deep semantic segmentation network of improved ResUnet. The experiments of intelligent recognition and classification of U. prolifera and S. horneri were carried out using the proposed multi-feature-based ResUnet algorithm, with specific F1-scores of 96.7% and 96.8%, respectively. The proposed multi-feature-based ResUnet algorithm can obtain efficient and high-accuracy results for the recognition and classification of marine floating macroalgae. It achieves accurate remote sensing monitoring of the two types of marine floating macroalgae and has significant theoretical research significance and practical application value.

Adaptive Local Cross-Channel Vector Pooling Attention Module for Semantic Segmentation of Remote Sensing Imagery

Adding an attention module to the deep convolution semantic segmentation network has significantly enhanced the network performance. However, the existing channel attention module focusing on the channel dimension neglects the spatial relationship, causing location noise to transmit to the decoder. In addition, the spatial attention module exemplified by self-attention has a high training cost and challenges in execution efficiency, making it unsuitable to handle large-scale remote sensing data. We propose an efficient vector pooling attention (VPA) module for building the channel and spatial location relationship. The module can locate spatial information better by performing a unique vector average pooling in the vertical and horizontal dimensions of the feature maps. Furthermore, it can also learn the weights directly by using the adaptive local cross-channel interaction. Multiple weight learning ablation studies and comparison experiments with the classical attention modules were conducted by connecting the VPA module to a modified DeepLabV3 network using ResNet50 as the encoder. The results show that the mIoU of our network with the addition of an adaptive local cross-channel interaction VPA module increases by 3% compared to the standard network on the MO-CSSSD. The VPA-based semantic segmentation network can significantly improve precision efficiency compared with other conventional attention networks. Furthermore, the results on the WHU Building dataset present an improvement in IoU and F1-score by 1.69% and 0.97%, respectively. Our network raises the mIoU by 1.24% on the ISPRS Vaihingen dataset. The VPA module can also significantly improve the network’s performance on small target segmentation.

Fault-Seg-Net: A method for seismic fault segmentation based on multi-scale feature fusion with imbalanced classification

Fault identification has important geological significance and practical production value. Due to the effects of earth filtering and environmental noise, it is difficult to identify minor faults, and manual fault identification is inefficient. In this study, an end-to-end deep learning semantic segmentation network Fault-Seg-Net is proposed to identify fault on seismic images, which simultaneously learns global semantic features and local detailed features. In Fault-Seg-Net, a multi-scale residual module is designed to expand the receptive field to mine fine-grained fault features from the low-dimensional feature space. Fault-Seg-Attention module is designed to model long-distance dependencies of pixel spatial location to compensate for the spatial continuity loss. In addition, a compound loss is used to guide the model training to handle imbalanced seismic image segmentation tasks. Experimental results on synthetic datasets have verified that Fault-Seg-Net can achieve high Precision (88.6%), Recall (89.2%), Dice (88.8%) and mIoU (81.5%) simultaneously, which is significantly better than traditional image processing methods and deep learning semantic segmentation networks. Experimental results on real large-scale field datasets have verified that Fault-Seg-Net has important practical value and strong robustness. This study provides an effective solution for intelligent seismic fault identification under complex geological environment.

DU-Net-Cloud: a smart cloud-edge application with an attention mechanism and U-Net for remote sensing images and processing

In recent ages, the use of deep learning approaches to extract ground object information from remote sensing high-resolution images has attracted extensive attention in many fields. Nevertheless, due to the high similarity of features between roads, prevailing deep learning semantic segmentation networks commonly demonstrate reduced continuity in road segmentation. Besides this, the role of advanced computing technologies including cloud and edge infrastructures has also become very important due to large number of images and their storage requirements. In order to better study the road details in images related to remote sensing, this paper suggests a road extraction technique which is basically founded on Dimensional U-Net (DU-Net) network. At the deepening level of the U-Net network, a parallel attention mechanism, known as ProCBAM, is added and implemented to the feature transmission step of the classical U-Net network. Moreover, we use and implement the edge-cloud architecture to develop and construct a unique remote sensing image service system that integrates several datacenters and their related edge infrastructure. In the proposed system, the edge network is primarily used for caching and distributing the processed remote sensing images, while the remote datacenter serves as the cloud platform and is responsible for the storage and processing of original remote sensing images. The results show that the proposed cloud enabled DU-Net model has achieved good performance in road segmentation. We observed that it can achieve improved road segmentation and resolve the issue of reduced continuity of road segmentation when compared with other state-of-the-art learning networks. Moreover, our empirical evaluations suggest that the proposed system not only distributes the workload of processing tasks across the edges but also achieves data efficiency among them, which enhances image processing efficiency and reduces data transmission costs.