Building Instance Change Detection from High Spatial Resolution Remote Sensing Images with Improved Instance Segmentation Architecture
Bibliographic record
Abstract
The detection of ground surface changes can provide essential and valuable information for experts in the fields of geomatics, emergency management, and urban management. Many deep learning models have been proposed for detecting the change from remote sensing images. However, the majority of studies have not been able to simultaneously accomplish both building change detection and instance segmentation of changed buildings from high spatial resolution images. The information of the building instance change in high spatial resolution remote sensing images is crucial for disaster assessment and urban building management. To address these issues, we proposed a novel building instance change detection architecture based on the Cascade Mask R-CNN framework and designed a swin transformer siamese (STS) backbone to input the bi-temporal images and to improve the detection accuracy, called STS-ConcCMR. Due to the lack of remote sensing mask datasets, two building change detection datasets are transferred into the building instance change detection task. Our method achieves the best performance in F1 of90.7 and 94.5 points on the LEVIER-CD and WHU-CD datasets, respectively. And the performance of the proposed architecture can be effectively improved by substituting the siamese backbone for the non-siamese backbone. The AP and F1 are improved by 0.5 and 0.4 points on the LEVIR-CD test set, respectively, and by 2.6 points and 4.0 points on the WHU-CD test set. Experimental results demonstrated the superiority of the proposed architecture.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".