• Home
  • Search
  • Incorporating Deep Features into GEOBIA Paradigm for Remote Sensing Imagery Classification: A Patch-Based Approach
  • Open Access IconOpen Access
  • Cite Icon20
  • https://doi.org/10.3390/rs12183007Copy DOI Icon

Incorporating Deep Features into GEOBIA Paradigm for Remote Sensing Imagery Classification: A Patch-Based Approach

Show More
  • Abstract
  • Highlights & Summary
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

The fast and accurate creation of land use/land cover maps from very-high-resolution (VHR) remote sensing imagery is crucial for urban planning and environmental monitoring. Geographic object-based image analysis methods (GEOBIA) provide an effective solution using image objects instead of individual pixels in VHR remote sensing imagery analysis. Simultaneously, convolutional neural networks (CNN) have been widely used in the image processing field because of their powerful feature extraction capabilities. This study presents a patch-based strategy for integrating deep features into GEOBIA for VHR remote sensing imagery classification. To extract deep features from irregular image objects through CNN, a patch-based approach is proposed for representing image objects and learning patch-based deep features, and a deep features aggregation method is proposed for aggregating patch-based deep features into object-based deep features. Finally, both object and deep features are integrated into a GEOBIA paradigm for classifying image objects. We explored the influences of segmentation scales and patch sizes in our method and explored the effectiveness of deep and object features in classification. Moreover, we performed 5-fold stratified cross validations 50 times to explore the uncertainty of our method. Additionally, we explored the importance of deep feature aggregation, and we evaluated our method by comparing it with three state-of-the-art methods in a Beijing dataset and Zurich dataset. The results indicate that smaller segmentation scales were more conducive to VHR remote sensing imagery classification, and it was not appropriate to select too large or too small patches as the patch size should be determined by imagery and its resolution. Moreover, we found that deep features are more effective than object features, while object features still matter for image classification, and deep feature aggregation is a critical step in our method. Finally, our method can achieve the highest overall accuracies compared with the state-of-the-art methods, and the overall accuracies are 91.21% for the Beijing dataset and 99.05% for the Zurich dataset.

Similar Papers
  • Research Article
  • Citations19

Object-oriented multiscale deep features for hyperspectral image classification

  • May 03, 2020
  • International Journal of Remote Sensing
  • Liang Hong +1
  • Research Article
  • Citations5

Deep Learning Features Can Improve Radiomics-Based Prostate Cancer Aggressiveness Prediction.

  • Sep 01, 2024
  • JCO clinical cancer informatics
  • Nuno M Rodrigues +8
  • Research Article
  • Citations44

Deep Feature Aggregation and Image Re-Ranking With Heat Diffusion for Image Retrieval

  • Jun 01, 2019
  • IEEE Transactions on Multimedia
  • Shanmin Pang +4
  • Research Article
  • Citations52

Deep Convolutional Neural Network-Assisted Feature Extraction for Diagnostic Discrimination and Feature Visualization in Pancreatic Ductal Adenocarcinoma (PDAC) versus Autoimmune Pancreatitis (AIP)

  • Dec 11, 2020
  • Journal of Clinical Medicine
  • Sebastian Ziegelmayer +6
  • Research Article
  • Citations6

A customized ConvNeXt‐XL network with fusion of deep and handcrafted features for colposcopy image classification

  • Feb 19, 2024
  • International Journal of Imaging Systems and Technology
  • Thendral Natarajan +3
  • Research Article

Abstract GS5-04: Prediction of occult invasive disease in ductal carcinoma in situ using deep learning features

  • Feb 14, 2018
  • Cancer Research
  • Jy Lo +8
  • Research Article
  • Citations16

Image retrieval based on aggregated deep features weighted by regional significance and channel sensitivity

  • Jun 21, 2021
  • Information Sciences
  • Juxiang Zhou +3
  • Supplementary Content
  • Citations9

Convolutional Neural Network based Age Estimation from Facial Image and Depth Prediction from Single Image

  • Jan 01, 2016
  • ANU Open Research (Australian National University)
  • Jiayan Qiu
  • Book Chapter
  • Citations12

Chapter 13 - A user independent hand gesture recognition system using deep CNN feature fusion and machine learning technique

  • Jan 01, 2021
  • New Paradigms in Computational Modeling and Its Applications
  • Jaya Prakash Sahoo +2
  • PDF
  • Research Article
  • Citations3

Multimodal Deep Feature Fusion (MMDFF) for RGB-D Tracking

  • Nov 28, 2018
  • Complexity
  • Ming-Xin Jiang +4
  • Research Article
  • Citations52

LQGDNet: A Local Quaternion and Global Deep Network for Facial Depression Recognition

  • Jul 01, 2023
  • IEEE Transactions on Affective Computing
  • Yuanyuan Shang +6
  • Conference Article
  • Citations12

Object Tracking Based on Deep CNN Feature and Color Feature

  • Aug 01, 2018
  • Yujuan Qi +2
  • Research Article
  • Citations28

Enhancing deep discriminative feature maps via perturbation for face presentation attack detection

  • Dec 02, 2019
  • Image and Vision Computing
  • Yasar Abbas Ur Rehman +2
  • Research Article
  • Citations18

Evaluation of hidden Markov models using deep CNN features in isolated sign recognition

  • Feb 24, 2021
  • Multimedia Tools and Applications
  • Anil Osman Tur +1
  • Research Article
  • Citations24

Deep Ensemble Feature Network for Gastric Section Classification

  • Jan 01, 2021
  • IEEE Journal of Biomedical and Health Informatics
  • Ting-Hsuan Lin +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.