Organizations: Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, Zhejiang 315200, P.R. China · Zhejiang Key Laboratory of Industrial Intelligence and Digital Twin, Eastern Institute of Technology, Ningbo, Zhejiang 315200, P.R. China · Department of Building Environment and Energy Engineering, The Hong Kong Polytechnic University, Kowloon, Hong Kong, China · School of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, China · School of Urban Planning and Design, Peking University, Shenzhen, 518055, China
Component-level information supports rooftop photovoltaic (PV) inspection, but recognizing individual components across complex scenes remains challenging. We present PVIS-10K, a dataset of 10,065 drone images with component-level annotations covering 63,632 physical PV modules across 22 rooftop installations. The high-resolution red-green-blue (RGB) images were acquired by unmanned aerial vehicles (UAVs, also known as drones) in Ningbo, Shaoxing and Shenzhen, China. Multi-angle oblique observations capture installation-height and scale variation, dense arrangements, occlusion, shadows and reflections. Three-dimensional delineation, reprojection and image-by-image review produced 3,516,675 image-level polygon annotations, stored in Common Objects in Context (COCO) format. Station-disjoint partitions support evaluation on unseen stations within and across cities. Four supervised models and text-prompted zero-shot Segment Anything Model 3 (SAM 3) provide full-resolution benchmarks of instance matching and foreground coverage. PVIS-10K supports component localization, delineation and image-based counting for drone-based rooftop PV inspection research.
Figures & tables
Figure 1: Challenges in drone-based rooftop PV component recognition. Panels a – d show installation geometry and scale variation, dense layouts, partial occlusion and shadows, and illumination and reflectance variation, respectively. Insets b1 and b2 in b and the lower-left inset in c show enlarged regions. Yellow outlines indicate selected ground-truth instances, including projected occluded portions.
Dataset
Image count
Annotation count
GSD
Target level
Annotation type
Annotation workflow
View representation
Bradbury 9 , 16
601
19,433
30
Non- component
Semantic segmentation
Manual 2D annotation
Orthographic view
BDAPPV 10
46,132
—
10; 20
Non- component
Semantic segmentation
Crowdsourced 2D annotation
Overhead views
Queens 11
667
∼ 14,000
15.2
Non- component
Semantic segmentation
Manual 2D annotation
Orthographic view
PV01 13
645
5,252
10
Non- component
Semantic segmentation
—
Orthographic view
Emmen 15 , 17
5,327
4,389
7.5
Non- component
Semantic segmentation
Manual 2D annotation
Orthographic view
PVIS-10K (ours)
10,065
3,516,675
∼ 0.5–1.1
Component
Instance segmentation
3D annotation and 2D reprojection
Multi-angle oblique views
Table 1: Representative open-access datasets for rooftop PV mapping.
Figure 2: Dataset construction workflow. Left, image acquisition, reconstruction and three-dimensional component annotation. Centre, projection of component polygons into selected images. Right, manual review and correction, annotation export and station-level partitioning. Dashed lines indicate projection rays; solid arrows indicate workflow progression.
City
Stations
Height
Assignment
Ningbo
18
20 m
12 training stations; 6 independent test stations
Shaoxing
3
40 m
Cross-city testing
Shenzhen
1
20 m
Cross-city testing
Table 2: Study areas, flight heights and partition assignments.
Training and test image records, component polygons and category definitions
station_splits.csv
CSV
Station, city, partition, role and relative paths
README.md README_zh-CN.md
Markdown
Dataset description and usage instructions
LICENSE.md
Text
CC BY 4.0 licence notice
Table 3: Core files in the dataset.
City
Split
Stations
Images
2D instances
Physical
Instances
components
per image
Ningbo
Training
12
5,266
1,578,541
28,731
299.8
Test
6
2,439
723,943
13,176
296.8
Shaoxing
Test
3
1,906
1,096,688
18,091
575.4
Shenzhen
Test
1
454
117,503
3,634
258.8
Total
22
10,065
3,516,675
63,632
349.4
Table 4: Dataset size and partition composition.
Figure 3: Station-level dataset composition. a , number of images; b , two-dimensional instance annotations, in thousands; c , mean number of instances per image. Colours indicate the city and training or test subset.
Parameter
Mask R-CNN
HTC
SOLOv2
Mask2Former
Backbone
ResNet-50 28
ResNet-50
ResNet-50
ResNet-50
Feature processing
FPN 29
FPN
FPN
Multi-scale deformable-attention pixel decoder
Backbone initialization
Torchvision pretrained
Torchvision pretrained
Torchvision pretrained
Torchvision pretrained
Optimizer
SGD
SGD
SGD
AdamW
Base learning rate
0.02
0.02
0.01
0.0001
Backbone learning-rate multiplier
1
1
1
0.1
Table 5: Supervised instance segmentation training configurations.
City
Model
Mask AP
AP50
AP75
IoU
F1
Ningbo
Mask R-CNN
65.37
79.88
68.17
90.77
95.16
HTC
72.72
85.87
75.57
90.42
94.97
SOLOv2
68.93
83.59
73.60
91.87
95.76
Mask2Former
65.18
79.59
66.89
92.65
96.18
SAM 3
4.75
7.68
4.95
89.49
94.45
Shaoxing
Mask R-CNN
55.30
68.75
60.86
90.14
94.81
Table 6: City-level instance segmentation and foreground coverage results (%).
Figure 4: Qualitative comparison of component-level instance segmentation. Columns show the original image, ground truth, HTC and SAM 3 predictions. Rows show a , densely packed components (station 022); b , inter-row occlusion (032); c , structural occlusion (028); and d , brightness variation (022). Crop and display scale are matched within each row. Green denotes ground-truth masks in the reference column and correct predictions in the model columns; yellow denotes split instances, where one PV component is divided into multiple predicted instances; purple denotes merged instances, where multiple PV components are combined into one predicted instance; and blue denotes missed instances, where a ground-truth component is not detected. Blue overlays mark missed ground-truth instances, not predicted masks. Colours illustrate the error types shown in these examples.
High-fidelity photovoltaic (PV) module mapping requires complete instance recovery and accurate delineation. Vision foundation models offer zero-shot segmentation, but converting incomplete and inconsistent predictions across views into a complete, geometrically accurate map of individual modules remains challenging. We propose the Multi-View PV Module Mapper (MV-PVMapper), a training-free framework coupling vision foundation models with unmanned aerial vehicle (UAV) photogrammetry. First, Segment Anything Model 3 (SAM 3) generates candidates from original images using text prompts. Their contours are then transferred into object space through per-module slope correction to limit distortion from surface-model noise, and verified against representative module dimensions to guide cross-view instance selection. Finally, accepted instances return to all covering images as box prompts for SAM~3, supplementing text prompts with spatial cues for iterative segmentation, verification and fusion. On five-direction surveys of three operational rooftop PV plants at approximately 1 cm nadir ground sampling distance, two feedback iterations recover 7,710 of 7,711 reference modules with no false positives at an intersection-over-union (IoU) matching threshold of 0.5. Macro-averaged area precision and recall reach 99.00 % and 98.52 %, respectively. Averaged across sites, the aggregated Jaccard index reaches 97.55 %, compared with 60.73 % and 41.24 % for orthomosaic segmentation and multi-view pixel-voting baselines, respectively; average precision at the stricter IoU threshold of 0.95 reaches 95.67 %. Ablation studies show that module-size constraints are critical for instance separation, while slope correction and feedback to all covering views improve boundary accuracy. These results are achieved without task-specific image annotation or segmentation-model weight updates.
WenHao Duan, Qi Chen, Penglei Zhang +4
School of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, China · Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, Zhejiang 315200, China · Zhejiang Key Laboratory of Industrial Intelligence and Digital Twin, Eastern Institute of Technology, Ningbo, Zhejiang 315200, China +2
Rooftop photovoltaic (PV) systems account for the vast majority of PV grid connections, yet no open, comprehensive, installation-level dataset of these systems exists: public registries aggregate data only above a capacity threshold, and remote sensing-based detection efforts, while extensive, are typically confined to a single method, a limited geographic scope, or a single point in time. We introduce OpenPVMapper, a nationwide, multi-source database of rooftop PV installations in mainland France, built by aggregating and reconciling complementary sources: a deep learning-based detection pipeline deployed on nationwide aerial imagery, OpenStreetMap and a probabilistic building-level detection dataset. The resulting database contains 1,135,850 installations, totaling approximately 15.01~GWp of installed capacity, each documented with its provenance, detection method, and, where available, a manual validation flag. Manual review of a stratified sample of 1,862 installations places the database's overall precision at approximately 74--75%, with corroboration across independent sources bringing a substantial, quantified precision gain. By aggregating independent sources rather than relying on any single detection method, OpenPVMapper reaches a level of confidence beyond what any one source could provide on its own, while remaining extensible to further sources as they become available. It is released under an open (CC-BY) license alongside the full source code used to build it.
Utility-scale photovoltaic (PV) inspection is dominated by aerial surveys, which are effective for identifying module-level hot cells but provide limited visibility of critical components located under the array, including connectors, cabling, fasteners and structural elements. These assets are difficult to inspect consistently at scale, and their defects are often spatially small, partially occluded and difficult to localise from isolated images. We present an automated 3D visual-thermal inspection system (Tricorder) for under-panel PV asset inspection. The approach combines LiDAR-inertial Simultaneous Localization and Mapping (SLAM) with geometrically calibrated RGB and thermal imagery and machine-learning-based object detection and segmentation. Detected assets are associated with the reconstructed 3D environment, allowing visual and thermal observations from multiple viewpoints to be spatially fused and localised. We demonstrate the approach on utility-scale PV data for MC4 connector detection and connection-state recognition, thermal anomaly identification, and detection of structural components including bolts, nuts and torque tubes. The results demonstrate a pathway from image-based inspection toward spatially grounded autonomous inspection, where defects are detected and linked to specific assets and their 3D locations within the solar farm.