The DOI matches its Crossref record; 23 of the 27 references it cites were found in a registry or an index.
Road Extraction by Deep Residual U-Net
Open accessAbstract
Road extraction from aerial images has been a hot research topic in the field of remote sensing image analysis. In this letter, a semantic segmentation neural network, which combines the strengths of residual learning and U-Net, is proposed for road area extraction. The network is built with residual units and has similar architecture to that of U-Net. The benefits of this model are twofold: first, residual units ease training of deep networks. Second, the rich skip connections within the network could facilitate information propagation, allowing us to design networks with fewer parameters, however, better performance. We test our network on a public road data set and compare it with U-Net and other two state-of-the-art deep-learning-based road extraction methods. The proposed approach outperforms all the comparing methods, which demonstrates its superiority over recently developed state of the arts.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In this study, we opt for a deep residual U-Net [15], [44], which has already been successfully used for retinal OCT images.”
From Interactive Deep Learning-Based Retinal OCT Layer Segmentation Refinement by Regressing Translation Maps (2024) · cites this paper as [44]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 5×
“This somehow shares similar idea to that of residual neural network [21].”
BackgroundMethodResultInfluentialVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 1×
“Recently, lots of works have suggested that a deeper network would have better performance [19], [20].”
BackgroundU-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · cited 7×
“1) U-Net: In semantic segmentation, to get a finer result, it is very important to use low-level details, while retaining high-level semantic information [23], [24].”
BackgroundMethodInfluentialFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472 · cited 1×
“of-the-art performance on a variety of computer vision tasks, such as scene recognition [13] and object detection [14].”
BackgroundFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562 · cited 1×
“Recently, lots of works have suggested that a deeper network would have better performance [19], [20].”
BackgroundFully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834 · cited 3×
“One way to solve this problem is employing a pretrained network and then finetuning it on the target data set, as done in [23].”
BackgroundMethodIdentity mappings in deep residual networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_38 · Cited by 10,741 · cited 2×
“[22] presented a detailed discussion on impacts of different combinations and suggested a full preactivation design as shown in Fig.”
BackgroundMethodDeep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art
Zhang, L., Zhang, L., & Du, B. (2016) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2016.2540798 · Cited by 2,272
Learning to Detect Roads in High-Resolution Aerial Images
Mnih, V., & Hinton, G. E. (2010) · European Conference on Computer Vision · doi:10.1007/978-3-642-15567-3_16 · Cited by 681 · cited 11×
“To demonstrate the accuracy and efficiency of the proposed deep ResUnet, we test it on Massachusetts roads data set2 and compare it with three state of the art methods, including Mnih’s [2] method, Saito’s method [5], and U-Net [24].”
BackgroundMethodResultInfluential
Research works printed but not linked: 4
- zhou, 2014, Learning deep features for scene recognition using places database, Proc NIPS, 487
- mnih, 2012, Learning to label aerial images from noisy data, Proc ICML, 567
- ehrig, 2005, Relaxed precision and recall for ontology matching, Proc Workshop Integr Ontol, 25
- chollet, 2015, Keras
Cited by
Sorted by citations
Items 301–400 of 1,001
DAU-Net: Dual attention-aided U-Net for segmenting tumor in breast ultrasound images
Pramanik, P., Roy, A., Cuevas, E., Pérez‐Cisneros, M., & Sarkar, R. (2024) · PLoS ONE · doi:10.1371/journal.pone.0303670 · Cited by 48
“The residual U-Net architecture, introduced in [7], is a variant of the traditional U-Net model.”
Deep multi-feature learning architecture for water body segmentation from satellite images
Tambe, R. G., Talbar, S., & Chavan, S. (2021) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2021.103141 · Cited by 48
Silva-Rodríguez, J., Colomer, A., & Naranjo, V. (2021) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2020.101846 · Cited by 52
“This residual UNet configuration was proposed in [40], and showed to outperform other configurations for Gleason grading in [41].”
MethodLiu, S., Setio, A. A. A., Ghesu, F. C., Gibson, E., Grbić, S., Georgescu, B., & Comaniciu, D. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.3026261 · Cited by 53
“lse positive reduction (FPR) module as shown in Fig. 3. The candidate generation module is trained to achieve high sensitivity via over-detecting nodule candidates. We use three identical 3D ResUNets [42] as the CG backbone networks without weight sharing. The first CG network is firstly trained to output 3D heatmaps wi…”
MethodDing, B., Wang, H., Chen, P., Zhang, Y., Guo, Z., Feng, J., & Liang, R. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.3016829 · Cited by 60
“…with other CNN based segmentation methods, the BCL-U Net can leverage inter-slice spatial dependence with less parameter amount and better exploit anisotropy of volume data than [22], [25]; by residual learning and HDC, it can better deal with segmentation of objects with diverse sizes than [25].”
BackgroundMethodHong, Z., Ming, D., Zhou, K., Guo, Y., & Lu, T. (2018) · IEEE Access · doi:10.1109/access.2018.2867210 · Cited by 48
“In this experiment, the parameter ρ is set to 3, which is consistent with previous studies [52], [62].”
ResultUM-Net: Rethinking ICGNet for polyp segmentation with uncertainty modeling
Du, X.-Q., Xu, X., Chen, J.-J., Zhang, X., Li, L., Liu, H., & Li, S. (2024) · Medical Image Anal. · doi:10.1016/j.media.2024.103347 · Cited by 47
Mehmood, A., Gulzar, Y., Ilyas, Q. M., Jabbari, A., Ahmad, M., & Iqbal, S. (2023) · Cancers · doi:10.3390/cancers15143604 · Cited by 58
“However, studies have found that once the network has reached a certain depth, adding further depth either makes the network harder to train with insignificant performance gains or, occasionally, causes its performance to degrade [39].”
BackgroundEfficient automatic segmentation for multi-level pulmonary arteries: The PARSE challenge
Luo, G., Wang, K.-Q., Liu, J., Li, S., Liang, X., Li, X., Gan, S., Wang, W., Dong, S., Wang, W.-Y., et al. (2023) · arXiv.org · doi:10.48550/arxiv.2304.03708 · Cited by 47
“Therefore, PKU appended a skeleton decoder to the Res-Unet(Zhang et al., 2018) (Fig.”
MethodSelf-citationLiu, X.-Z., Wang, Z.-Y., Wan, J., Zhang, J.-L., Xi, Y., Liu, R.-Y., & Miao, Q. (2023) · Remote Sensing · doi:10.3390/rs15041049 · Cited by 47
“However, these methods are usually limited by fixed parameters and have been proven to underperform when applied to low-contrast images [6–8].”
BackgroundMethodBrain tumour segmentation based on an improved U-Net
Zheng, P., Zhu, X.-F., & Guo, W.-B. (2022) · BMC Medical Imaging · doi:10.1186/s12880-022-00931-1 · Cited by 47
“used the U-Net extended path to design residual connections and proposed a depth residual U-Net for image segmentation [13].”
MethodQuantization-aware Deep Optics for Diffractive Snapshot Hyperspectral Imaging
Li, L.-G., Wang, L., Song, W., Zhang, L., Xiong, Z.-W., & Huang, H. (2022) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr52688.2022.01916 · Cited by 47
“We adopt the ResUNet [49], a variant of the U-Net [35], as the decoder.”
MethodCascaded Residual Attention Enhanced Road Extraction from Remote Sensing Images
Li, S.-F., Liao, C., Ding, Y., Hu, H., Jia, Y., Chen, M., Xu, B., Ge, X., Liu, T.-Y., & Wu, D. (2021) · ISPRS Int. J. Geo Inf. · doi:10.3390/ijgi11010009 · Cited by 47
“To verify the proposed method, we conducted comparative experiments on the Massachusetts dataset to compare the performance of our method with other classical semantic segmentation methods, including the U-Net [27], SegNet [6], Res-U-Net [28], DeepLabv3+ [7], and D-LinkNet [54] series methods.”
BackgroundMethodReview on Active and Passive Remote Sensing Techniques for Road Extraction
Jia, J., Sun, H., Jiang, C., Karila, K., Karjalainen, M., Ahokas, E., Khoramshahi, E., Hu, P.-L., Chen, C., Xue, T., et al. (2021) · Remote Sensing · doi:10.3390/rs13214235 · Cited by 47
“A variant of FCN—DenseNets—including SegNet, DeepLab, RCFs, Y-net and U-Net decoder [49,51,52,54] can extract hierarchical features from images, especially high-resolution images.”
BackgroundJi, S., Zhang, Z.-L., Zhang, C., Wei, S., Lü, M., & Duan, Y. (2019) · International Journal of Remote Sensing · doi:10.1080/01431161.2019.1699973 · Cited by 59
“Although the encoder-decoder FCN structure has been widely used in semantic segmentation (Ronneberger, Fischer, and Brox 2015; Zhang, Liu, and Wang 2018; Badrinarayanan, Kendall, and Cipolla 2017; Yu et al. 2018b; Tian et al. 2019), its limitation as far as representing the global contextual…”
MethodSelf-citationVehicle Detection From High-Resolution Remote Sensing Imagery Using Convolutional Capsule Networks
Yu, Y., Gu, T., Guan, H., Li, D., & Jin, S. (2019) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2019.2912582 · Cited by 56
“Consequently, extensive studies have also been conducted for information extraction and interpretation from remote sensing images, such as road extraction [1], road feature extraction [2], vehicle detection [3], and traffic monitoring [4].”
BackgroundDong, A.-N., Dou, J., Li, C.-D., Chen, Z.-Q., Ji, J., Xing, K., Zhang, J., & Daud, H. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3424680 · Cited by 47
“…Experiment A. Specifically, we compared the performance of five different network models, including the popular semantic segmentation algorithms VggUnet [27], FCN [37], DeeplabV3 [38], ResUnet [39], and our constructed MFFLnet, in both the Hokkaido landslide scene and the Luding landslide scene.”
MethodHettihewa, K. G., Kobchaisawat, T., Tanpowpong, N., & Chalidabhongse, T. H. (2023) · Scientific Reports · doi:10.1038/s41598-023-46580-4 · Cited by 49
“Deep Residual U-Net (ResUNet) is developed by Zhang et al. 8 as a combined architecture by utilizing the strengths of U-Net 4 and deep residual learning 9 .”
BackgroundMethodRecent advances in deep-learning-enhanced photoacoustic imaging
Yang, J., Choi, S., Kim, J., Park, B., & Kim, C. (2023) · Advanced Photonics Nexus · doi:10.1117/1.apn.2.5.054001 · Cited by 52
“The models tested included U-Net, 69 FD U-Net, 89 Y-Net, FD Y-Net, 78 deep residual U-Net (deep ResU-Net), 133 and GAN.”
MethodLocal Adaptive Illumination-Driven Input-Level Fusion for Infrared and Visible Object Detection
Wu, J., Shen, T., Wang, Q.-W., Tao, Z., Zeng, K., & Song, J. (2023) · Remote Sensing · doi:10.3390/rs15030660 · Cited by 46
GA-Net: A geometry prior assisted neural network for road extraction
Chen, X., Sun, Q., Guo, W., Qiu, C., & Yu, A.-Z. (2022) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2022.103004 · Cited by 46
Survey of Road Extraction Methods in Remote Sensing Images Based on Deep Learning
Liu, P., Wang, Q., Yang, G.-C., Li, L., & Zhang, H. (2022) · PFG – Journal of Photogrammetry Remote Sensing and Geoinformation Science · doi:10.1007/s41064-022-00194-z · Cited by 46
“At present, some review papers (Abdollahi et al. 2020b; Dai et al. 2020; Lian et al. 2020; Zhang et al. 2018b) have summarized and analysed road extraction methods in remote sensing images.”
BackgroundMethodInfluentialAbdollahi, A., Pradhan, B., & Alamri, A. (2021) · GIScience & Remote Sensing · doi:10.1080/15481603.2021.1972713 · Cited by 46
“Zhang et al. (2018) combined the UNet model with residual learning to extract a road area from the Massachusetts road dataset.”
Infrared Image Segmentation for Photovoltaic Panels Based on Res-UNet
Zhang, H., Hong, X., Zhou, S., & Wang, Q. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-31654-9_52 · Cited by 58
City-Scale Road Extraction from Satellite Imagery v2: Road Speeds and Travel Times
Etten, A. V. (2019) · IEEE Workshop/Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093593 · Cited by 46
“Extracting road pixels in small image chips from aerial imagery has a rich history (e.g. [33], [19] [29], [33], [25], [15]).”
BackgroundA Lightweight Network for Building Extraction From Remote Sensing Images
Huang, H., Chen, Y., & Wang, R. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3131331 · Cited by 50
“Due to the rapid development of pixel level convolutional neural network with the excellent performance in various computer vision tasks, it has been gradually applied to remote sensing image interpretation [21-27], including the classification, extraction, change detection of vegetation and so on.”
MVSI-Net: Multi-view attention and multi-scale feature interaction for brain tumor segmentation
Sun, J.-D., Hu, M., Wu, X.-S., Tang, C.-S., Lahza, H., Wang, S.-H., & Zhang, Y.-D. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106484 · Cited by 45
Fan, J., Li, J.-J., Liu, Y.-P., & Zhang, F. (2024) · Engineering applications of artificial intelligence · doi:10.1016/j.engappai.2023.107638 · Cited by 45
CNN and Transformer Fusion for Remote Sensing Image Semantic Segmentation
Chen, X., Li, D.-F., Liu, M.-Y., & Jia, J. (2023) · Remote Sensing · doi:10.3390/rs15184455 · Cited by 45
Kalinaki, K., Malik, O., & Lai, D. (2023) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2023.103453 · Cited by 45
Road Extraction With Satellite Images and Partial Road Maps
Xu, Q.-X., Long, C., Yu, L., & Zhang, C. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3261332 · Cited by 45
“[13] designed a deep residual UNet model to extract road features”
MethodLi-SegPNet: Encoder-Decoder Mode Lightweight Segmentation Network for Colorectal Polyps Analysis
Sharma, P., Gautam, A., Maji, P., Pachori, R. B., & Balabantaray, B. K. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3216269 · Cited by 58
“Literaturerevealsthatthereisanopportunitytocontributeto theadvancementofpolypsegmentation,andamoreefficient networkwillgiveanedgetotheclinicalapplicabilityofthe automaticmodelinCRCanalysis.”
Usman, M., Rehman, A., Saleem, A. M., Jawaid, R., Byon, S.-S., Kim, S.-H., Lee, B.-D., Heo, M.-S., & Shin, Y.-G. (2022) · Italian National Conference on Sensors · doi:10.3390/s22249877 · Cited by 45
“The ResUNet or Deep Residual UNet architecture was utilized for 3D mandibular canal segmentation [16], an architecture that relies on deep residual learning and U-Net.”
MethodKhan, S. D., Alarabi, L., & Basalamah, S. M. (2022) · The Arabian journal for science and engineering · doi:10.1007/s13369-022-07082-z · Cited by 45
“[18] proposed a deep residual network based on UNet for road extraction in satellite images.”
BackgroundMethodLiu, B., Zhang, B., Feng, H., Wu, S., Yang, J., Zou, Y., & Siddique, K. H. M. (2021) · International Soil and Water Conservation Research · doi:10.1016/j.iswcr.2021.10.004 · Cited by 56
Deep Learning for Real-Time Aerodynamic Evaluations of Arbitrary Vehicle Shapes
Jacob, S. J., Mrosek, M., Othmer, C., & Köstler, H. (2022) · SAE International Journal of Passenger Vehicle Systems · doi:10.4271/15-15-02-0006 · Cited by 37
“There have been multiple variations and improvements for various applications [24, 25], including CFD [6].”
MethodM2H-Net: A Reconstruction Method For Hyperspectral Remotely Sensed Imagery
Deng, L., Sun, J., Chen, Y., Lu, H., Duan, F., Zhu, L., & Fan, T.-X. (2021) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2021.01.019 · Cited by 45
Ahmed, H., La, H. M., & Tran, K. T. (2020) · Automation in Construction · doi:10.1016/j.autcon.2020.103393 · Cited by 47
“Earlier studies employing Residual Networks and their variants have attested to their superior performance towards tackling a vast range of research problems [72] [73] [74].”
BackgroundAutomated crater detection with human level performance
Lee, C., & Hogan, J. (2020) · Computational Geosciences · doi:10.1016/j.cageo.2020.104645 · Cited by 45
“In stage 1 of the CDA, we use a ResUNET(Zhang et al., 2018) neural network to find circular crater features in global imagery or terrain model datasets (section 2.”
MethodKumar, S., Tzou, J.-C., & Wu, S.-C. (2020) · Energies · doi:10.3390/en13071772 · Cited by 52
“The work in [62] proposed a deep residual U-net architecture (ResUnet) for extracting road from aerial images.”
BackgroundImproving the Efficiency of Encoder-Decoder Architecture for Pixel-Level Crack Detection
Chen, H., Lin, H., & Yao, M. (2019) · IEEE Access · doi:10.1109/access.2019.2961375 · Cited by 49
“ResUnet [49] is a shallower UNet-like model that is built with residual units [50].”
BackgroundTransSea: Hybrid CNN–Transformer With Semantic Awareness for 3-D Brain Tumor Segmentation
Liu, Y., Ma, Y., Zhu, Z.-Q., Cheng, J., & Chen, X. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3413130 · Cited by 45
“Variants of U-Net, such as UNet ++ [32] and Res-UNet [33], have significantly enhanced performance.”
BackgroundRadhakrishnan, K., Scott, K. A., & Clausi, D. A. (2021) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2021.3076109 · Cited by 45
“segmentation problems [12], [13], but have also been used for regression problems such as pansharpening, and have achieved good results [14].”
BackgroundMachine learning approaches for improving atomic force microscopy instrumentation and data analytics
Masud, N., Rade, J., Hasib, M. H. H., Krishnamurthy, A., & Sarkar, A. (2024) · Frontiers of Physics · doi:10.3389/fphy.2024.1347648 · Cited by 44
An Accurate and Efficient Neural Network for OCTA Vessel Segmentation and a New Dataset
Ning, H., Wang, C.-L., Chen, X.-R., & Li, S. (2023) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp48485.2024.10447708 · Cited by 44
“We compare our proposed network with UNet[11], UNet++[18], ResUNet[19], OVS-Net[8] and FARGO[5].”
ResultAn Improved U-Net Image Segmentation Method and Its Application for Metallic Grain Size Statistics
Shi, P., Duan, M., Yang, L., Feng, W., Ding, L., & Jiang, L.-W. (2022) · Materials · doi:10.3390/ma15134417 · Cited by 44
“ResU-Net [39] is a semantic segmentation neural network which combines the strengths of residual learning, and U-Net is proposed for road area extraction.”
BackgroundDCS-TransUperNet: Road Segmentation Network Based on CSwin Transformer with Dual Resolution
Zhang, Z., Miao, C., Liu, C., & Tian, Q. (2022) · Applied Sciences · doi:10.3390/app12073511 · Cited by 44
“Super performance [22–24] about CNN has been shown in remote sensing image segmentation, especially U-Net with an encoder–decoder structure, e.g., in order to enhance the effectiveness of road feature representation, a variant of U-Net with residual was presented by Zhang et al. [25].”
BackgroundMethodSelf-citationMoldovanu, S., Damian, F. A., Biswas, K. C., Florescu, A., & Moraru, L. (2021) · Cancers · doi:10.3390/cancers13215256 · Cited by 48
Radar-Based Human Target Detection using Deep Residual U-Net for Smart Home Applications
Stephan, M., & Santra, A. (2019) · International Conference on Machine Learning and Applications · doi:10.1109/icmla.2019.00035 · Cited by 50
“Recently, deep residual U-net like network architectures have been proposed for identifying road lines from SAR images [18].”
BackgroundLi, Y., Guo, L., Rao, J., Xu, L., & Jin, S. (2019) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2018.2878771 · Cited by 44
“The popular CNN-based methods contain VGG [13], fully convolutional network (FCN) [14], [15], cascaded CNN [16], and U-Net [17].”
MethodZhao, H., Wu, B., Zhang, M., Jiang, L., Tian, F., Xie, Y., Zeng, H., Zheng, Z., Ma, Z., Wang, M., et al. (2025) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2025.01.034 · Cited by 45
Deng, L., Zhang, A., Guo, J.-J., & Liu, Y. (2023) · Remote Sensing · doi:10.3390/rs15061530 · Cited by 43
TransBTSV2: Towards Better and More Efficient Volumetric Segmentation of Medical ImagesPreprint
Li, J., Wang, W., Chen, C., Zhang, T., Zha, S., Wang, J., & Yu, H. Z. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2201.12785 · Cited by 32
“In the following research, U-Net variants such as U-Net++ [8] and Res-UNet [9] further improve the segmentation performance.”
BackgroundMethodHyperspectral change detection based on modification of UNet neural networks
Moustafa, M. S., Mohamed, S. A., Ahmed, S., & Nasr, A. H. (2021) · Journal of Applied Remote Sensing · doi:10.1117/1.jrs.15.028505 · Cited by 50
“The first variant of UNet is Residual UNet, 38 which was introduced as an extension to benefit from the residual learning as shown in Fig.”
Methodde Bem, P. P., Júnior, O. A. D. C., Júnior, O. A. D. C., Gomes, R. A. T., & Guimarães, R. F. (2020) · Remote Sensing · doi:10.3390/rs12162576 · Cited by 44
“A further enhancement of the architecture has been proposed through the insertion of residual connections within the architecture’s blocks, resulting in what has been called ResUnet [51,64].”
BackgroundZheng, J., Shao, A., Yan, Y., Wu, J., & Zhang, M. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3292112 · Cited by 46
“From these images, the objects of interest can be automatically extracted using approaches, such as semantic segmentation, which pixel-wisely assigns class labels and aids in road prospecting [1], urban planning [2], and land detection [3].”
MethodSeismic Random Noise Attenuation by Applying Multiscale Denoising Convolutional Neural Network
Zhong, T., Cheng, M., Dong, X.-T., & Wu, N. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3095922 · Cited by 43
“Meanwhile, many multiscale networks, such as U-Net [46] and the encoder–decoder network (ED-Net) [47], have been proposed, mainly concentrating on face identification and semantic segmentation.”
BackgroundInfluentialMapping Roads in the Brazilian Amazon with Artificial Intelligence and Sentinel-2
Botelho, J., Costa, S. C. P., Ribeiro, J. G., & Souza, C. M. (2022) · Remote Sensing · doi:10.3390/rs14153625 · Cited by 43
“Mapping roads from satellite imagery is an important image processing research topic to improve urban planning, transportation systems, and agricultural organization [1,2].”
BackgroundLu, X., & Weng, Q.-H. (2025) · Isprs Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2025.07.013 · Cited by 42
A Multi-Scale Context Aware Attention Model for Medical Image Segmentation
Alam, M. S., Wang, D., Liao, Q., & Sowmya, A. (2022) · IEEE journal of biomedical and health informatics · doi:10.1109/jbhi.2022.3227540 · Cited by 42
“Jha et al. [18] developed ResUNet ++ that was an improved version of ResUNet [28].”
BackgroundWang, S., Mu, X., Yang, D., He, H., & Zhao, P. (2021) · Remote Sensing · doi:10.3390/rs13030465 · Cited by 50
“Moreover, Zhang et al. [31] developed a deep residual U-Net for road extraction, which combined the U-Net architecture and residual units.”
BackgroundInfluentialChen, Z.-Y., Fan, W., Zhong, B., Li, J., Du, J., & Wang, C. (2020) · IEEE transactions on intelligent transportation systems (Print) · doi:10.1109/tits.2019.2939536 · Cited by 42
“2) Quantitation Comparison: To quantitatively illustrate the performance of our method, we compare our method with three other state-of-the-art methods [6], [7], [31], [46], [49], [50] on Shaoshan dataset.”
BackgroundMethodResultInfluentialBehera, T., Bakshi, S., Nappi, M., & Sa, P. K. (2023) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2023.3239119 · Cited by 42
“Several common RS tasks in this domain comprise buildings extraction [34], [35], road networks extraction [36], [37], [38], and vegetation extraction [39].”
BackgroundRoad Extraction from Remote Sensing Imagery with Spatial Attention Based on Swin Transformer
Zhu, X., Huang, X.-H., Cao, W., Yang, X.-F., Zhou, Y.-F., & Wang, S. (2024) · Remote Sensing · doi:10.3390/rs16071183 · Cited by 41
Conv-trans dual network for landslide detection of multi-channel optical remote sensing images
Chen, X., Liu, M.-Z., Li, D.-F., Jia, J., Yang, A., Zheng, W.-F., & Yin, L. (2023) · Frontiers in Earth Science · doi:10.3389/feart.2023.1182145 · Cited by 41
Learning Context-Based Nonlocal Entropy Modeling for Image Compression
Li, M., Zhang, K., Li, J.-X., Zuo, W., Timofte, R., & Zhang, D. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3104974 · Cited by 41
“Besides, the skip connection could facilitate information propagation and ease the training of the network [41].”
BackgroundUrban road mapping based on an end-to-end road vectorization mapping network framework
Chen, D., Zhong, Y., Zheng, Z., Ma, A., & Lu, X. (2021) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2021.05.016 · Cited by 46
Deep Learning-Based Defect Detection System in Steel Sheet Surfaces
Amin, D., & Akhter, S. (2020) · 2020 IEEE Region 10 Symposium (TENSYMP) · doi:10.1109/tensymp50017.2020.9230863 · Cited by 41
MimickNet, Mimicking Clinical Image Post-Processing Under Black-Box Constraints
Huang, O., Long, W., Bottenus, N., Lerendegui, M., Trahey, G., Farsiu, S., & Palmeri, M. (2019) · IUS · doi:10.1109/ultsym.2019.8925597 · Cited by 41
“Other encoder-decoder Unet flavours exist which exploit residual learning [15], [16], wavelet transforms [17], and dense blocks [18], [19].”
BackgroundHekal, A. A., Elnakib, A., Moustafa, H., & Amer, H. M. (2024) · IEEE Access · doi:10.1109/access.2024.3351564 · Cited by 41
“Residual learning, introduced by He et al. [41], addresses the degradation problem caused by increasing network depth during training [42].”
BackgroundZhou, F., Wen, G., Ma, Y., Geng, H., Huang, R., Pei, L., Yu, W., Chu, L., & Qiu, R. (2022) · Applied Sciences · doi:10.3390/app12115336 · Cited by 41
“The residual U-Net architecture from [79] is utilized for vegetation segmentation.”
MethodA skin lesion segmentation network with edge and body fusion
Ma, Q.-S., Wang, G., Li, Y., Mao, K.-M., Xu, L., & Zhao, Y.-H. (2025) · Applied Soft Computing · doi:10.1016/j.asoc.2024.112683 · Cited by 40
Deep learning in cropland field identification: A review
Xu, F., Yao, X., Zhang, K., Yang, H. F., Feng, Q., Li, Y., Yan, S., Gao, B., Li, S., Yang, J., et al. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.109042 · Cited by 48
MRI brain tumor segmentation using residual Spatial Pyramid Pooling-powered 3D U-Net
Vijay, S., Guhan, T., Srinivasan, K., Vincent, P., & Chang, C.-Y. (2023) · Frontiers in Public Health · doi:10.3389/fpubh.2023.1091850 · Cited by 40
“The performance for image segmentation is greatly improved by many U-Net variants, including U-Net ++ (11), two-stage cascaded U-Net (12), and Res-U-Net (13).”
BackgroundDe-Noising of Photoacoustic Microscopy Images by Attentive Generative Adversarial Network
He, D., Zhou, J., Shang, X., Tang, X., Luo, J., & Chen, S. (2022) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2022.3227105 · Cited by 60
“U-Net U-Net [50] architecture has been widely and effectively applied in various biomedical imaging applications, while Residual U-Net [44] is one of the variants of U-Net.”
BackgroundMethodMaksum, Y., Amirli, A., Amangeldi, A., Inkarbekov, M., Ding, Y., Romagnoli, A., Rustamov, S., & Akhmetov, B. (2022) · Journal of Industrial Information Integration · doi:10.1016/j.jii.2022.100352 · Cited by 38
Zhang, J., Hu, Q., Li, J., & Ai, M. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.3003425 · Cited by 45
“[7] proposed a semantic segmentation road extraction neural network, which is constructed using the units and has a similar architecture to U-Net, which make the model use fewer parameters but achieve better performance.”
MethodSAT-Net: a side attention network for retinal image segmentation
Tong, H., Fang, Z.-J., Wei, Z.-R., Cai, Q., & Gao, Y.-B. (2021) · Applied intelligence (Boston) · doi:10.1007/s10489-020-01966-z · Cited by 40
Zhang, J., & Sheng, G. (2020) · Journal of Petroleum Science and Engineering · doi:10.1016/j.petrol.2020.107527 · Cited by 40
Automatic Gully Detection: Neural Networks and Computer Vision
Gafurov, A., & Yermolayev, O. (2020) · Remote Sensing · doi:10.3390/rs12111743 · Cited by 40
Wang, S., Yang, H., Wu, Q., Zheng, Z., Wu, Y., & Li, J. (2020) · Italian National Conference on Sensors · doi:10.3390/s20072064 · Cited by 40
“complexity of road networks and the interference of other objects (such as parking lots, building roofs and trees), it is difficult to extract roads effectively [5,6].”
BackgroundMethodAlam, F., Mehmood, R., Katib, I. A., Altowaijri, S. M., & Albeshri, A. (2019) · Journal on spesial topics in mobile networks and applications · doi:10.1007/s11036-019-01319-2 · Cited by 40
“…in literature such as map-supervised based on CNN [17], U-Net which is a deep CNN with residual units which ease training with fewer parameters without sacrificing performance [18], CNN method for road detection particularly in extreme weather conditions [19] and Bayesian network based RBNet [20].”
MethodUsing the wide-range attention U-Net for road segmentation
Yuan, M., Liu, Z., & Wang, F. (2019) · Remote Sensing Letters · doi:10.1080/2150704x.2019.1574990 · Cited by 40
Towards Accurate Medical Image Segmentation With Gradient-Optimized Dice Loss
Qi, M., & Xiao, X. (2024) · IEEE Signal Processing Letters · doi:10.1109/lsp.2023.3329437 · Cited by 40
“The results in Table I show that with the application of GODC, the Dice index of Residual U-Net [26] is improved by 0.9 points in the pancreatic tumor segmentation task.”
MethodInfluentialJiang, L., Li, M., & Wang, X. (2023) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2023.3262586 · Cited by 38
“Conventional encoder–decoder networks like ResUNet [28] use skip connections to aggregate low-and high-level semantics.”
An Enhanced Residual U-Net for Microaneurysms and Exudates Segmentation in Fundus Images
Kou, C., Li, W., Yu, Z.-K., & Yuan, L. (2020) · IEEE Access · doi:10.1109/access.2020.3029117 · Cited by 40
“In addition, we construct the ResU-Net with a 9-level architecture, a deeper network than the ResU-Net in [36].”
MethodmResU-Net: multi-scale residual U-Net-based brain tumor segmentation from multimodal MRI
Li, P., Li, Z., Wang, Z.-J., Li, C.-X., & Wang, M.-N. (2023) · Medical and Biological Engineering and Computing · doi:10.1007/s11517-023-02965-1 · Cited by 39
“Zhang et al. [18] suggested DeepResUNet and noted that it is very crucial to fuse low-level semantic information with high-level semantic information using skip connections.”
Edge-aware Feature Aggregation Network for Polyp Segmentation
Zhou, T., Zhang, Y., Chen, G., Zhou, Y., Wu, Y., & Fan, D.-P. (2023) · Machine Intelligence Research · doi:10.1007/s11633-023-1479-8 · Cited by 39
“However, the existing U-Net structures, such as ResU-Net [18] , H-DenseUNet [19] and Attention U-Net [20] , directly adopt a skip connection to fuse feature representations from the encoder to the decoder.”
Farooq, M. U., Ullah, Z., & Gwak, J. (2023) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2022.102173 · Cited by 35
Pixel-Level Anomaly Detection via Uncertainty-aware Prototypical Transformer
Huang, C., Liu, C., Zhang, Z., Wu, Z.-H., Wen, J., Jiang, Q., & Xu, Y. (2022) · ACM Multimedia · doi:10.1145/3503161.3548082 · Cited by 39
“Specifically, U-shaped networks [11, 19, 33, 46, 49] adopting an encoder-decoder structure with skip connections, have significantly improved the performance compared with early methods.”
ResultSelf-citationDing, Y., Yang, Q., Wang, Y., Chen, D., Qin, Z., & Zhang, J. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102511 · Cited by 33
Self-citationAutomatic void content assessment of composite laminates using a machine-learning approach
Machado, J., Tavares, J., Camanho, P., & Correia, N. (2022) · Composite structures · doi:10.1016/j.compstruct.2022.115383 · Cited by 39
Aerial Images Meet Crowdsourced Trajectories: A New Approach to Robust Road Extraction
Liu, L., Yang, Z., Li, G., Wang, K., Chen, T., & Lin, L. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3141821 · Cited by 44
“For instance, Cheng et al. [12] proposed a cascaded end-to-end CNN to cope with the road detection and centerline extraction simultaneously with two cascaded CNN. Zhang et al. [43] developed a semantic segmentation neural network, which combined the residual learning and U-Net to extract road areas.”
BackgroundMethodInfluentialLiver tumor segmentation using 2.5D UV-Net with multi-scale convolution
Zhang, C., Hua, Q., Chu, Y., & Wang, P. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104424 · Cited by 43
Deep Learning for Multilabel Remote Sensing Image Annotation With Dual-Level Semantic Concepts
Zhu, P., Tan, Y., Zhang, L., Wang, Y., Mei, J., Liu, H., & Wu, M. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2019.2960466 · Cited by 47
“To solve this problem, many semantic segmentation algorithms have been proposed, in which each pixel is assigned with an object label [7], [8].”
BackgroundSelf-citationAbdollahi, A., Pradhan, B., & Alamri, A. M. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3143855 · Cited by 53
Cho, Y., Misra, S., Managuli, R., Barr, R. G., Lee, J., & Kim, C. (2024) · Ultrasound in Medicine & Biology · doi:10.1016/j.ultrasmedbio.2024.11.020 · Cited by 37
Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report
Huijben, E. M. C., Terpstra, M., Galapon, A., Pai, S., Thummerer, A., Koopmans, P., Afonso, M., van Eijnatten, M., Gurney-Champion, O., Chen, Z.-X., et al. (2024) · doi:10.48550/arxiv.2403.08447 · Cited by 38
“Team Elekta only participated in task 1, where they employed a 2.5D pix2pix (Isola et al., 2017) model using a Re-sUnet (Zhang et al., 2018b) generator and a discriminator implemented similarly to the encoding part of the ResUnet.”
MethodSelf-citationBhuyan, K., Meena, S., Nava, L., van Westen, C. V., Floris, M., & Catani, F. (2023) · GIScience & Remote Sensing · doi:10.1080/15481603.2023.2182057 · Cited by 38
“, 2015) due to 163 its robust network structure and segmented pixels as outputs(Ghorbanzadeh, et al. , 2021; Prakash et 164 al. , 2021; Zhang et al., 2018).”
BackgroundMethodYu, F., He, B., Li, K., Yan, T.-H., Shen, Y., Wang, Q., & Wu, M. (2021) · Applied Ocean Research · doi:10.1016/j.apor.2021.102608 · Cited by 38
Metadata
- Authors
- Zhang, Z., Liu, Q., & Wang, Y.
- Affiliations
- State Key Laboratory of Virtual Reality Technology and Systems (CN)
- Venue
- IEEE Geoscience and Remote Sensing Letters
- Published
- 2018-05
- Volume
- 15
- Issue
- 5
- Pages
- 749-753
- ISSN
- 1545-598X (ISSN-L)
- 1558-0571
- OpenAlex
- W2774320778
- Cited by
- 3,179
- References
- 27
- Citations by year
- 2026: 331 · 2025: 526 · 2024: 616 · 2023: 510 · 2022: 441 · 2021: 342 · 2020: 241 · 2019: 147 · 2018: 25
- FWCI
- 151.19
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Geoscience and Remote Sensing Letters (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- Natural Science Foundation of China (61601011)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Automated Road and Building Extraction
- Physical Sciences › Engineering › Civil and Structural Engineering › Infrastructure Maintenance and Monitoring
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- road extraction, aerial images, semantic segmentation, residual learning, U-Net, skip connections
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
