Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions
Efficient parking space management in urban settings represents a growing challenge owing to the sustained increase in the vehicle fleet. This study presents a comparative evaluation of five object detection architectures —YOLOv8s, YOLOv11s, YOLOv12s, RT-DETR-L and Faster R-CNN—applied to real-time...
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Artículo |
| Lenguaje: | inglés |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | Enlace permanente (DOI o Handle) |
| Etiquetas: |
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1881700517897306112 |
|---|---|
| author | Fernando G. Yunganina Mamani Guver Leon Cori Coarite Jhon Alexander Chambi Vilca Ángel Rosendo Condori-Coaquira David Mamani-Pari Milton Edward Humpiri Flores Esteban Tocto-Cano |
| author_browse | David Mamani-Pari Esteban Tocto-Cano Fernando G. Yunganina Mamani Guver Leon Cori Coarite Jhon Alexander Chambi Vilca Milton Edward Humpiri Flores Ángel Rosendo Condori-Coaquira |
| author_facet | Fernando G. Yunganina Mamani Guver Leon Cori Coarite Jhon Alexander Chambi Vilca Ángel Rosendo Condori-Coaquira David Mamani-Pari Milton Edward Humpiri Flores Esteban Tocto-Cano |
| author_sort | Fernando G. Yunganina Mamani |
| collection | Producción científica |
| description | Efficient parking space management in urban settings represents a growing challenge owing to the sustained increase in the vehicle fleet. This study presents a comparative evaluation of five object detection architectures —YOLOv8s, YOLOv11s, YOLOv12s, RT-DETR-L and Faster R-CNN—applied to real-time intelligent vehicle occupancy monitoring under variable lighting conditions. The models were trained via transfer learning on a custom dataset of 1463 source images (21,944 annotated instances; expanded to 3511 files and 52,664 instances through offline augmentation of the training subset; three classes: free, occupied and unavailable) captured on a university campus located in Juliaca (Puno region), Peru, at 3824 m a.s.l. under daytime and nighttime clear-sky conditions from a single fixed-camera viewpoint. Each architecture was evaluated in ten independent experiments. Six dataset partitioning schemes of increasing strictness—a random control (R0) plus five leakage-controlled partitions—were evaluated. Under the strictest scheme (D3), simultaneously disjoint in acquisition date and camera viewpoint and therefore the most rigorous generalization estimate obtained in this study, accuracy ranges from mAP@0.5:0.95 of 0.9325 for Faster R-CNN to 0.8763 for YOLOv11s. Under the random partitioning conventionally applied to fixed-camera datasets, the same five architectures fell within 0.0055 of one another, all above 0.985, and their ranking was essentially inverted (Spearman ρ=−0.80). The differences in computational efficiency across architectures were statistically significant (H=47.06, p<0.001). YOLOv8s was the fastest of the four non-dominated architectures under the disjoint partition and was selected in 73.3% of weightings, although it ranked fourth in accuracy; its recommendation therefore rests on computational efficiency under a real-time constraint, whereas deployments that prioritize accuracy are better served by Faster R-CNN. The integrated system YOLOv8s + ByteTrack + FastAPI + Next.js 14 achieved per-slot accuracies of 87.5% and 91.8% under daytime and nighttime clear-sky conditions, respectively, using 1395 observations collected in a single university parking lot. For YOLOv8s, the transition from random to disjoint partitioning costs 0.1085 in mAP@0.5:0.95 (0.9913 to 0.8828), indicating that the near-saturated performance obtained under random partitioning substantially reflects the memorization of a fixed spatial configuration rather than generalization. The results support the feasibility of single-stage CNN architectures for intelligent parking monitoring in high-altitude Andean university environments under the evaluated acquisition conditions. |
| format | Article |
| id | openalex:W7204123370 |
| institution | UPeU |
| language | English |
| publishDate | 2026 |
| publishDateRange | 2026 |
| publishDateSort | 2026 |
| record_format | dspace |
| spelling | openalex:W7204123370Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting ConditionsFernando G. Yunganina MamaniGuver Leon Cori CoariteJhon Alexander Chambi VilcaÁngel Rosendo Condori-CoaquiraDavid Mamani-PariMilton Edward Humpiri FloresEsteban Tocto-Canoparking occupancy detectionreal-time object detectionYOLOv8YOLOv11Faster R-CNNRT-DETRtransfer learningvariable lighting conditionsnighttime detectiondata partitioningcomputational efficiencyparking space managementByteTrackuniversity parking lotsAdvanced Neural Network ApplicationsSmart Parking Systems ResearchVideo Surveillance and Tracking MethodsEfficient parking space management in urban settings represents a growing challenge owing to the sustained increase in the vehicle fleet. This study presents a comparative evaluation of five object detection architectures —YOLOv8s, YOLOv11s, YOLOv12s, RT-DETR-L and Faster R-CNN—applied to real-time intelligent vehicle occupancy monitoring under variable lighting conditions. The models were trained via transfer learning on a custom dataset of 1463 source images (21,944 annotated instances; expanded to 3511 files and 52,664 instances through offline augmentation of the training subset; three classes: free, occupied and unavailable) captured on a university campus located in Juliaca (Puno region), Peru, at 3824 m a.s.l. under daytime and nighttime clear-sky conditions from a single fixed-camera viewpoint. Each architecture was evaluated in ten independent experiments. Six dataset partitioning schemes of increasing strictness—a random control (R0) plus five leakage-controlled partitions—were evaluated. Under the strictest scheme (D3), simultaneously disjoint in acquisition date and camera viewpoint and therefore the most rigorous generalization estimate obtained in this study, accuracy ranges from mAP@0.5:0.95 of 0.9325 for Faster R-CNN to 0.8763 for YOLOv11s. Under the random partitioning conventionally applied to fixed-camera datasets, the same five architectures fell within 0.0055 of one another, all above 0.985, and their ranking was essentially inverted (Spearman ρ=−0.80). The differences in computational efficiency across architectures were statistically significant (H=47.06, p<0.001). YOLOv8s was the fastest of the four non-dominated architectures under the disjoint partition and was selected in 73.3% of weightings, although it ranked fourth in accuracy; its recommendation therefore rests on computational efficiency under a real-time constraint, whereas deployments that prioritize accuracy are better served by Faster R-CNN. The integrated system YOLOv8s + ByteTrack + FastAPI + Next.js 14 achieved per-slot accuracies of 87.5% and 91.8% under daytime and nighttime clear-sky conditions, respectively, using 1395 observations collected in a single university parking lot. For YOLOv8s, the transition from random to disjoint partitioning costs 0.1085 in mAP@0.5:0.95 (0.9913 to 0.8828), indicating that the near-saturated performance obtained under random partitioning substantially reflects the memorization of a fixed spatial configuration rather than generalization. The results support the feasibility of single-stage CNN architectures for intelligent parking monitoring in high-altitude Andean university environments under the evaluated acquisition conditions.Articlehttps://doi.org/10.3390/s26175329W7204123370enSensors1424-82202026-08-22 |
| spellingShingle | parking occupancy detection real-time object detection YOLOv8 YOLOv11 Faster R-CNN RT-DETR transfer learning variable lighting conditions nighttime detection data partitioning computational efficiency parking space management ByteTrack university parking lots Advanced Neural Network Applications Smart Parking Systems Research Video Surveillance and Tracking Methods Fernando G. Yunganina Mamani Guver Leon Cori Coarite Jhon Alexander Chambi Vilca Ángel Rosendo Condori-Coaquira David Mamani-Pari Milton Edward Humpiri Flores Esteban Tocto-Cano Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions |
| title | Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions |
| title_full | Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions |
| title_fullStr | Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions |
| title_full_unstemmed | Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions |
| title_short | Comparative Evaluation of Deep Learning Object Detectors for Real-Time Parking Occupancy Detection Under Variable Lighting Conditions |
| title_sort | comparative evaluation of deep learning object detectors for real time parking occupancy detection under variable lighting conditions |
| topic | parking occupancy detection real-time object detection YOLOv8 YOLOv11 Faster R-CNN RT-DETR transfer learning variable lighting conditions nighttime detection data partitioning computational efficiency parking space management ByteTrack university parking lots Advanced Neural Network Applications Smart Parking Systems Research Video Surveillance and Tracking Methods |
| topic_facet | parking occupancy detection real-time object detection YOLOv8 YOLOv11 Faster R-CNN RT-DETR transfer learning variable lighting conditions nighttime detection data partitioning computational efficiency parking space management ByteTrack university parking lots Advanced Neural Network Applications Smart Parking Systems Research Video Surveillance and Tracking Methods |
| url | https://doi.org/10.3390/s26175329 |
| work_keys_str_mv | AT fernandogyunganinamamani comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions AT guverleoncoricoarite comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions AT jhonalexanderchambivilca comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions AT angelrosendocondoricoaquira comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions AT davidmamanipari comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions AT miltonedwardhumpiriflores comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions AT estebantoctocano comparativeevaluationofdeeplearningobjectdetectorsforrealtimeparkingoccupancydetectionundervariablelightingconditions |