Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Automated Road Extraction from Satellite Imagery Integrating Dense Depthwise Dilated Separable Spatial Pyramid Pooling with DeepLabV3+

Arpan Mahara, Md Rezaul Karim Khan, Naphtali D. Rishe, Wenjia Wang, Seyed Masoud Sadjadi

TL;DR

This study hypothesizes that the integration of DenseDDSSPP, combined with an appropriately selected backbone network and a Squeeze-and-Excitation block, will generate an efficient dense feature map by focusing on relevant features, leading to more precise and accurate road extraction from Remote Sensing images.

Abstract

Road Extraction is a sub-domain of Remote Sensing applications; it is a subject of extensive and ongoing research. The procedure of automatically extracting roads from satellite imagery encounters significant challenges due to the multi-scale and diverse structures of roads; improvement in this field is needed. The DeepLab series, known for its proficiency in semantic segmentation due to its efficiency in interpreting multi-scale objects' features, addresses some of these challenges caused by the varying nature of roads. The present work proposes the utilization of DeepLabV3+, the latest version of the DeepLab series, by introducing an innovative Dense Depthwise Dilated Separable Spatial Pyramid Pooling (DenseDDSSPP) module and integrating it in place of the conventional Atrous Spatial Pyramid Pooling (ASPP) module. This modification enhances the extraction of complex road structures from satellite images. This study hypothesizes that the integration of DenseDDSSPP, combined with an appropriately selected backbone network and a Squeeze-and-Excitation block, will generate an efficient dense feature map by focusing on relevant features, leading to more precise and accurate road extraction from Remote Sensing images. The results section presents a comparison of our model's performance against state-of-the-art models, demonstrating better results that highlight the effectiveness and success of the proposed approach.

Automated Road Extraction from Satellite Imagery Integrating Dense Depthwise Dilated Separable Spatial Pyramid Pooling with DeepLabV3+

TL;DR

This study hypothesizes that the integration of DenseDDSSPP, combined with an appropriately selected backbone network and a Squeeze-and-Excitation block, will generate an efficient dense feature map by focusing on relevant features, leading to more precise and accurate road extraction from Remote Sensing images.

Abstract

Road Extraction is a sub-domain of Remote Sensing applications; it is a subject of extensive and ongoing research. The procedure of automatically extracting roads from satellite imagery encounters significant challenges due to the multi-scale and diverse structures of roads; improvement in this field is needed. The DeepLab series, known for its proficiency in semantic segmentation due to its efficiency in interpreting multi-scale objects' features, addresses some of these challenges caused by the varying nature of roads. The present work proposes the utilization of DeepLabV3+, the latest version of the DeepLab series, by introducing an innovative Dense Depthwise Dilated Separable Spatial Pyramid Pooling (DenseDDSSPP) module and integrating it in place of the conventional Atrous Spatial Pyramid Pooling (ASPP) module. This modification enhances the extraction of complex road structures from satellite images. This study hypothesizes that the integration of DenseDDSSPP, combined with an appropriately selected backbone network and a Squeeze-and-Excitation block, will generate an efficient dense feature map by focusing on relevant features, leading to more precise and accurate road extraction from Remote Sensing images. The results section presents a comparison of our model's performance against state-of-the-art models, demonstrating better results that highlight the effectiveness and success of the proposed approach.

Paper Structure

This paper contains 19 sections, 16 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Proposed Approach
  4. Dilated Convolution in Spatial Pyramid Pooling
  5. Depth-wise Dilated Separable Convolution
  6. ASPP Module
  7. DeepLabV3+
  8. Integration of DenseDDSSPP into the Network
  9. Selection of suitable backbone network and Incorporation of Squeeze-and-Excitation (SE) block:
  10. Experiments and Results
  11. Datasets
  12. Massachusetts Road Dataset:
  13. DeepGlobe Road Dataset:
  14. Evaluation Metrics
  15. Experimental Environment and Baselines
  16. ...and 4 more sections

Figures (5)

  • Figure 1: Depiction of Different Convolutions
  • Figure 2: Architecture of DeepLabV3+ with DenseDDSSPP module
  • Figure 3: Squeeze-and-Excitation Block
  • Figure 4: Comparative Results of Road Extraction from Massachusetts Dataset. The figure presents a side-by-side comparison of road extracted by various models, including the Proposed Model, against the Ground Truth, highlighting the effectiveness of each approach in synthesizing accurate road extraction.
  • Figure 5: Comparative Results of Road Extraction from DeepGlobe Road Dataset. The figure presents a side-by-side comparison of road extracted by various models, including the Proposed Model, against the Ground Truth.