Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Point processes for unsupervised line network extraction in remote sensing.

Caroline Lacoste1, Xavier Descombes, Josiane Zerubia

  • 1CREATIS, INSA, 7 rue Jean Capelle, bat. Blaise Pascal, F-69621 Villeurbanne Cedex, France. caroline.lacoste@creatis.insa-lyon.fr

IEEE Transactions on Pattern Analysis and Machine Intelligence
|October 22, 2005
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Burst-Suppression EEG in Early Infantile Developmental and Epileptic Encephalopathies: Phenotype, Genotype, and Outcome.

Neurology·2026
Same author

Study protocol FRENCH-ARRIVE: labor induction in low-risk nulliparous women at 39 weeks of gestation to reduce cesarean births: a randomized trial of induction versus expectant management in France.

BMC pregnancy and childbirth·2026
Same author

Uncertainty-aware AI for tumor subtyping with histology and immunohistochemistry: A multi-center study in Renal Cell Carcinoma.

Computer methods and programs in biomedicine·2026
Same author

Prostaglandins versus transcervical balloon for induction of labor for fetal growth restriction: study protocol for a randomized, multicenter, open-label, superiority trial (PROBIN).

BMC pregnancy and childbirth·2026
Same author

Oral ivermectin versus 5% permethrin cream to treat children and adults with classic scabies: multicentre, assessor blinded, cluster randomised clinical trial.

BMJ (Clinical research ed.)·2026
Same author

Linear structure unfolding : application to mouse brain in spatial transcriptomics.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces the "Quality Candy" model for unsupervised line network extraction from images. It improves upon the "Candy" model by incorporating quality coefficients, enhancing topological and radiometric feature analysis for better results.

Area of Science:

  • Remote Sensing
  • Computer Vision
  • Geographic Information Systems

Background:

  • Automated extraction of linear features like roads and rivers from imagery is challenging.
  • Existing methods often struggle to balance topological accuracy with radiometric detail.

Purpose of the Study:

  • To develop an unsupervised method for extracting line networks from remotely sensed images.
  • To improve upon existing models by better utilizing network topology and radiometric properties.

Main Methods:

  • Modeling line networks as interacting line segments using an object process.
  • Employing the "Quality Candy" prior model focusing on topological properties.
  • Utilizing a data term based on statistical tests for radiometric properties, with accurate and efficient computation options.

Related Experiment Videos

  • Parameter calibration and optimization via Reversible Jump Markov Chain Monte Carlo (RJMCMC) with accelerated convergence using proposal kernels.

    • Main Results:

    • The "Quality Candy" model significantly outperforms the previous "Candy" model in unsupervised line network extraction.
    • Quantitative evaluation on satellite and aerial imagery confirms improved accuracy compared to manual extractions.
    • Demonstrated the benefit of quality coefficients in the prior density for interaction modeling.

    Conclusions:

    • The "Quality Candy" model offers a robust approach to unsupervised extraction of complex line networks.
    • The integration of topological quality coefficients enhances the model's performance in remote sensing applications.
    • Offline computation of data potential and its use in RJMCMC algorithms improve efficiency and relevance.