Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Dec 7, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Automatic Changes Detection between Outdated Building Maps and New VHR Images Based on Pre-Trained Fully

Yunsheng Zhang¹, Yaochen Zhu², Haifeng Li¹

¹School of Geoscience and Info-Physics, Central South University, Changsha 410083, China.

Sensors (Basel, Switzerland)

|September 30, 2020

Summary

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

UAV Vision-Based Method for Multi-Point Displacement Measurement of Bridges.

Sensors (Basel, Switzerland)·2026

Same author

Displaced Femoral Neck Fractures Treated with Percutaneous Compression Plates in Elderly Individuals: An Effect Analysis Based on Imaging.

Current medical imaging·2025

Same author

Helical dislocation-driven plasticity and flexible high-performance thermoelectric generator in α-Mg<sub>3</sub>Bi<sub>2</sub> single crystals.

Nature communications·2025

Same author

Diagnostic performance of metagenomic next-generation sequencing based on alveolar lavage fluid in unexplained lung shadows.

Diagnostic microbiology and infectious disease·2024

Same author

Sewage surveillance revealed the seasonality and prevalence of respiratory syncytial virus and its implications for seasonal immunization strategy in low and middle-income regions of China.

Water research·2024

Same author

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

Frontiers in plant science·2024

This summary is machine-generated.

This study introduces an efficient method for detecting building changes in high spatial resolution remotely sensed images. The approach utilizes a deep convolutional neural network (DCNN) and random forest (RF) for accurate feature extraction and classification, reducing manual effort.

Area of Science:

Remote Sensing
Geospatial Analysis
Computer Vision

Background:

Detecting changes in building basemaps using high spatial resolution remotely sensed (HRS) images is challenging due to manual data labeling and ineffective hand-crafted features.
Existing methods struggle with efficiency and accuracy in identifying urban structural modifications from aerial or satellite imagery.

Purpose of the Study:

To develop an efficient and accurate method for detecting changes between existing building basemaps and new HRS images.
To improve feature extraction and classification performance in remote sensing change detection tasks.

Main Methods:

A fully convolutional feature extractor, reconstructed from a deep convolutional neural network (DCNN) pre-trained on the Pascal VOC dataset, was employed for pixel-wise feature extraction.

Keywords:

VHR images changes detection fully convolutional feature maps outdated building map

Related Experiment Videos

Last Updated: Dec 7, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Salient features were selected using a random forest (RF) algorithm, followed by a data cleaning process involving cross-validation and label-uncertainty estimation for RF classifier training.

Initial pixel-wise classification results were refined using a superpixel-based graph cuts algorithm to generate the final change map.

Main Results:

The proposed method demonstrated high accuracy in detecting building changes across simulated and real datasets.
The approach achieved a low false alarm rate, indicating reliable change detection capabilities.
The integration of DCNN for feature extraction and RF for classification significantly improved efficiency and performance.

Conclusions:

The developed method offers an effective solution for automated building change detection in HRS imagery.
The combination of deep learning features and machine learning classifiers provides a robust framework for geospatial change analysis.
This research contributes to more efficient and accurate monitoring of urban development and infrastructure changes.