RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM

Satellite remote sensing images contain complex and diverse ground object information and the images exhibit spatial multi-scale characteristics, making the panoptic segmentation of satellite remote sensing images a highly challenging task. Due to the lack of large-scale annotated datasets for panor...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhuoran Liu, Zizhen Li, Ying Liang, Claudio Persello, Bo Sun, Guangjun He, Lei Ma
Format: Article
Language:English
Published: MDPI AG 2024-10-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/21/4002
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850192957476438016
author Zhuoran Liu
Zizhen Li
Ying Liang
Claudio Persello
Bo Sun
Guangjun He
Lei Ma
author_facet Zhuoran Liu
Zizhen Li
Ying Liang
Claudio Persello
Bo Sun
Guangjun He
Lei Ma
author_sort Zhuoran Liu
collection DOAJ
description Satellite remote sensing images contain complex and diverse ground object information and the images exhibit spatial multi-scale characteristics, making the panoptic segmentation of satellite remote sensing images a highly challenging task. Due to the lack of large-scale annotated datasets for panoramic segmentation, existing methods still suffer from weak model generalization capabilities. To mitigate this issue, this paper leverages the advantages of the Segment Anything Model (SAM), which can segment any object in remote sensing images without requiring any annotations and proposes a high-resolution remote sensing image panoptic segmentation method called Remote Sensing Panoptic Segmentation SAM (RSPS-SAM). Firstly, to address the problem of global information loss caused by cropping large remote sensing images for training, a Batch Attention Pyramid was designed to extract multi-scale features from remote sensing images and capture long-range contextual information between cropped patches, thereby enhancing the semantic understanding of remote sensing images. Secondly, we constructed a Mask Decoder to address the limitation of SAM requiring manual input prompts and its inability to output category information. This decoder utilized mask-based attention for mask segmentation, enabling automatic prompt generation and category prediction of segmented objects. Finally, the effectiveness of the proposed method was validated on the high-resolution remote sensing image airport scene dataset RSAPS-ASD. The results demonstrate that the proposed method achieves segmentation and recognition of foreground instances and background regions in high-resolution remote sensing images without the need for prompt input, while providing smooth segmentation boundaries with a panoptic segmentation quality (PQ) of 57.2, outperforming current mainstream methods.
format Article
id doaj-art-611237c19eb84915aa63d6054aa2d3c6
institution OA Journals
issn 2072-4292
language English
publishDate 2024-10-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-611237c19eb84915aa63d6054aa2d3c62025-08-20T02:14:23ZengMDPI AGRemote Sensing2072-42922024-10-011621400210.3390/rs16214002RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAMZhuoran Liu0Zizhen Li1Ying Liang2Claudio Persello3Bo Sun4Guangjun He5Lei Ma6School of Geodesy and Geomatics, Beijing University of Civil Engineering and Architecture, Beijing 102627, ChinaState Key Laboratory of Space-Earth Integrated Information Technology, Beijing Institute of Satellite Information Engineering, Beijing 100095, ChinaState Key Laboratory of Space-Earth Integrated Information Technology, Beijing Institute of Satellite Information Engineering, Beijing 100095, ChinaDepartment of Earth Observation Science, Faculty ITC, University of Twente, 7500AE Enschede, The NetherlandsCollege of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, ChinaState Key Laboratory of Space-Earth Integrated Information Technology, Beijing Institute of Satellite Information Engineering, Beijing 100095, ChinaSchool of Geography and Ocean Science, Nanjing University, Nanjing 210023, ChinaSatellite remote sensing images contain complex and diverse ground object information and the images exhibit spatial multi-scale characteristics, making the panoptic segmentation of satellite remote sensing images a highly challenging task. Due to the lack of large-scale annotated datasets for panoramic segmentation, existing methods still suffer from weak model generalization capabilities. To mitigate this issue, this paper leverages the advantages of the Segment Anything Model (SAM), which can segment any object in remote sensing images without requiring any annotations and proposes a high-resolution remote sensing image panoptic segmentation method called Remote Sensing Panoptic Segmentation SAM (RSPS-SAM). Firstly, to address the problem of global information loss caused by cropping large remote sensing images for training, a Batch Attention Pyramid was designed to extract multi-scale features from remote sensing images and capture long-range contextual information between cropped patches, thereby enhancing the semantic understanding of remote sensing images. Secondly, we constructed a Mask Decoder to address the limitation of SAM requiring manual input prompts and its inability to output category information. This decoder utilized mask-based attention for mask segmentation, enabling automatic prompt generation and category prediction of segmented objects. Finally, the effectiveness of the proposed method was validated on the high-resolution remote sensing image airport scene dataset RSAPS-ASD. The results demonstrate that the proposed method achieves segmentation and recognition of foreground instances and background regions in high-resolution remote sensing images without the need for prompt input, while providing smooth segmentation boundaries with a panoptic segmentation quality (PQ) of 57.2, outperforming current mainstream methods.https://www.mdpi.com/2072-4292/16/21/4002panoptic segmentationsegment anything modelremote sensingdeep learning
spellingShingle Zhuoran Liu
Zizhen Li
Ying Liang
Claudio Persello
Bo Sun
Guangjun He
Lei Ma
RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM
Remote Sensing
panoptic segmentation
segment anything model
remote sensing
deep learning
title RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM
title_full RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM
title_fullStr RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM
title_full_unstemmed RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM
title_short RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM
title_sort rsps sam a remote sensing image panoptic segmentation method based on sam
topic panoptic segmentation
segment anything model
remote sensing
deep learning
url https://www.mdpi.com/2072-4292/16/21/4002
work_keys_str_mv AT zhuoranliu rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam
AT zizhenli rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam
AT yingliang rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam
AT claudiopersello rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam
AT bosun rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam
AT guangjunhe rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam
AT leima rspssamaremotesensingimagepanopticsegmentationmethodbasedonsam