At present, deep-learning methods have been widely used in road extraction from remote-sensing images and have effectively improved the accuracy of road extraction. However, these methods are still affected by the loss of spatial features and the lack of global context information. To solve these problems, we propose a new network for road extraction, the coord-dense-global (CDG) model, built on three parts: a coordconv module by putting coordinate information into feature maps aimed at reducing the loss of spatial information and strengthening road boundaries, an improved dense convolutional network (DenseNet) that could make full use of multiple features through own dense blocks, and a global attention module designed to highlight high-level information and improve category classification by using pooling operation to introduce global information. When tested on a complex road dataset from Massachusetts, USA, CDG achieved clearly superior performance to contemporary networks such as DeepLabV3+, U-net, and D-LinkNet. For example, its mean (intersection of the prediction and ground truth regions over their union) and mean score (evaluation metric for the harmonic mean of the and metrics) were 61.90% and 76.10%, respectively, which were 1.19% and 0.95% higher than the results of D-LinkNet (the winner of a road-extraction contest). In addition, CDG was also superior to the other three models in solving the problem of tree occlusion. Finally, in universality research with the Gaofen-2 satellite dataset, the CDG model also performed well at extracting the road network in the test maps of Hefei and Tianjin, China.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7180710PMC
http://dx.doi.org/10.3390/s20072064DOI Listing

Publication Analysis

Top Keywords

road extraction
16
remote-sensing images
8
loss spatial
8
cdg model
8
road
7
improved method
4
method road
4
extraction
4
extraction high-resolution
4
high-resolution remote-sensing
4

Similar Publications

Background: The prevalence of depression in COVID-19 patients is notably high, disrupting daily life routines and compounding the burden of other chronic health conditions. In addition, to elucidate the connection between COVID-19 and depression, we conducted an analysis of commonly differentially expressed genes [co-DEGs], uncovering potential biomarkers and therapeutic avenues specific to COVID-19-related depression.

Methods: We obtained gene expression profiles from the Gene Expression Omnibus [GEO] database with strategic keyword searches ["COVID-19", "depression," and "SARS"].

View Article and Find Full Text PDF

Antemortem diagnostic tests and treatment outcomes using ivermectin in 13 cases of presumptive feline neurocuterebriasis.

Can Vet J

January 2025

Department of Clinical Studies, Ontario Veterinary College, University of Guelph, 50 Stone Road East, Guelph, Ontario N1G 2W1.

Background: Limited reports exist on the antemortem presumptive diagnosis and treatment of feline neurocuterebriasis. A 3-day treatment protocol reported for 3 cats had no adverse effects. This protocol comprised ivermectin (0.

View Article and Find Full Text PDF

Dynamic EIT technology for real-time non-invasive monitoring of acute pulmonary embolism: a porcine model experiment.

Respir Res

January 2025

Shaanxi Provincial Key Laboratory of Bioelectromagnetic Detection and Intelligent Perception, Department of Biomedical Engineering, Air Force Medical University, Xi'an, 710032, China.

Background: Acute pulmonary embolism represents the third most prevalent cardiovascular pathology, following coronary heart disease and hypertension. Its untreated mortality rate is as high as 20-30%, which represents a significant threat to patient survival. In view of the current lack of real-time monitoring techniques for acute pulmonary embolism, this study primarily investigates the potential of the pulsatility electrical impedance tomography (EIT) technique for the detection and real-time monitoring of acute pulmonary embolism through the collection and imaging of the pulsatile signal of pulmonary blood flow.

View Article and Find Full Text PDF

Background: Gallbladder and biliary diseases (GABD) represent prevalent disorders of the digestive system.

Methods: Data on age-standardized incidence rate (ASIR), age-standardized mortality rate (ASMR), and age-standardized disability-adjusted life years (DALYs) rate (ASDR) were extracted from the Global Burden of Disease (GBD) 2021 study. The estimated annual percentage change (EAPC) was utilized to quantify temporal trends in GABD.

View Article and Find Full Text PDF

Silibinin alleviates acute liver failure by modulating AKT/GSK3β/Nrf2/GPX4 pathway.

Naunyn Schmiedebergs Arch Pharmacol

January 2025

Jiangsu Key Laboratory of TCM Evaluation and Translational Research, Department of Chinese Materia Medica, School of Traditional Chinese Pharmacy, China Pharmaceutical University, 639 Longmian Road, Nanjing, 211198, P. R. China.

Silibinin (Sil) is a major bioactive component of silymarin, extracted from the fruit and seeds of Silybum marianum. Silibinin meglumine (SM) is a water-soluble derivative of silibinin that has shown significant potential in liver fibrosis. However, the potential effects and underlying mechanisms of SM on acute liver failure (ALF) are still not fully understood.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!