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.

Highlights

  • Due to its real-time and low-cost, land use/land cover (LULC) mapping using remote sensing images has received widespread attention in recent decades [1]

  • We showed an effective method of integrating deep features into geographic object-based image analysis (GEOBIA) for VHR remote sensing imagery classification

  • We proposed a patch-based approach for representing image objects using patches and learning patch-based deep features and a deep feature aggregation method for aggregating patch-based deep features into object-based deep features, in order to extract deep features from irregular image objects through convolutional neural networks (CNN)

Read more

Summary

Introduction

Due to its real-time and low-cost, land use/land cover (LULC) mapping using remote sensing images has received widespread attention in recent decades [1]. Ground entities in very-high-resolution (VHR) imagery often appear to have complex structures, demonstrating the increased heterogeneities within ground entities, interclass similarities and intraclass differences, hindering LULC mapping [3,4]. Besides the complexity of urban ground entities and the large heterogeneity within ground entities, the intraclass heterogeneities and interclass similarities are very high in VHR images. All these make it difficult to classify image objects through only handcrafted features; more robust features are required for improving VHR imagery classification [12]

Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call