Advanced real-time traffic sign segmentation and classification using hybrid deep convolutional architectures on gtsdb and benchmark datasets

Muqimova , Anaxon

Techscience.uz - техника фанлари долзарб масалалри · 2026-yil

Annotatsiya

Autonomous driving systems and Advanced Driver Assistance Systems (ADAS) heavily rely on precise visual recognition of traffic control infrastructure under varying real-world conditions. While traditional object detection methods utilize bounding boxes, pixel-level semantic and instance segmentation provide essential geometric detail required for motion planning and vehicle control. This paper presents a rigorous comparative analysis of deep learning models, specifically U-Net, Mask R-CNN, and YOLOv8-Seg, for real-time traffic sign segmentation. Training and evaluation were conducted on benchmark datasets including the German Traffic Sign Detection Benchmark (GTSDB) and TT100K. To address extreme pixel-level class imbalance inherent in roadside imagery, a hybrid compound loss function combining Focal Loss and Dice Loss was developed. Experimental results indicate that YOLOv8-Seg achieves an optimal trade-off with a mean Average Precision (mAP50) of 89.7% and Intersection over Union (IoU) of 81.3% while operating at 45.2 Frames Per Second (FPS) on standard embedded GPU hardware. U-Net demonstrated superior boundary reconstruction accuracy with an IoU of 82.1%, whereas Mask R-CNN achieved peak precision (91.2% mAP50) at reduced inference speeds (12.4 FPS). The source code implementation and optimized hyperparameter setup in PyTorch 2.0 are detailed.

Maqola ma’lumotlari
MualliflarMuqimova , Anaxon
JurnalTechscience.uz - техника фанлари долзарб масалалри
NashriyotЖиззах политехника институти
ISSN3030-3702, 3030-3702
Nashr sanasi2026-09-16
Jild4
Son9
Betlar19-24
TilIngliz
Ilmiy sohaTexnika
DOI10.47390/ts-v4i9y2026n03
Indeksga qo‘shilgan2026-09-17

DOI: 10.47390/ts-v4i9y2026n03 · Maqolaning asl sahifasi

Kalit so‘zlar

traffic sign segmentation, deep neural networks, U-Net, YOLOv8-Seg, GTSDB, Focal Loss, Dice Loss, autonomous vehicles, embedded vision., yo'l belgilarini segmentatsiyalash, chuqur neyron tarmoqlar, U-Net, YOLOv8-Seg, GTSDB, Focal Loss, Dice Loss, avtonom transport vositalari, o'rnatilgan ko'rish tizimlari (embedded vision).

Ilmiy soha

Texnika

Iqtibos keltirish

APAMuqimova , Anaxon (2026). ADVANCED REAL-TIME TRAFFIC SIGN SEGMENTATION AND CLASSIFICATION USING HYBRID DEEP CONVOLUTIONAL ARCHITECTURES ON GTSDB AND BENCHMARK DATASETS. Techscience.uz - техника фанлари долзарб масалалри, 4(9), 19-24. https://doi.org/10.47390/ts-v4i9y2026n03
GOST (OAK)Muqimova , Anaxon ADVANCED REAL-TIME TRAFFIC SIGN SEGMENTATION AND CLASSIFICATION USING HYBRID DEEP CONVOLUTIONAL ARCHITECTURES ON GTSDB AND BENCHMARK DATASETS // Techscience.uz - техника фанлари долзарб масалалри. – 2026. – T. 4, № 9. – P. 19-24. – DOI: 10.47390/ts-v4i9y2026n03.
MLAMuqimova , Anaxon. “ADVANCED REAL-TIME TRAFFIC SIGN SEGMENTATION AND CLASSIFICATION USING HYBRID DEEP CONVOLUTIONAL ARCHITECTURES ON GTSDB AND BENCHMARK DATASETS.” Techscience.uz - техника фанлари долзарб масалалри, 4(9), 19-24, 2026. https://doi.org/10.47390/ts-v4i9y2026n03
ChicagoMuqimova , Anaxon. “ADVANCED REAL-TIME TRAFFIC SIGN SEGMENTATION AND CLASSIFICATION USING HYBRID DEEP CONVOLUTIONAL ARCHITECTURES ON GTSDB AND BENCHMARK DATASETS.” Techscience.uz - техника фанлари долзарб масалалри, 4(9), 19-24 (2026). https://doi.org/10.47390/ts-v4i9y2026n03
HarvardMuqimova , Anaxon (2026) ‘ADVANCED REAL-TIME TRAFFIC SIGN SEGMENTATION AND CLASSIFICATION USING HYBRID DEEP CONVOLUTIONAL ARCHITECTURES ON GTSDB AND BENCHMARK DATASETS’, Techscience.uz - техника фанлари долзарб масалалри, 4(9), 19-24. https://doi.org/10.47390/ts-v4i9y2026n03
BibTeX
@article{muqimovaanaxon2026,
  title = {ADVANCED REAL-TIME TRAFFIC SIGN SEGMENTATION AND CLASSIFICATION USING HYBRID DEEP CONVOLUTIONAL ARCHITECTURES ON GTSDB AND BENCHMARK DATASETS},
  author = {Muqimova , Anaxon},
  journal = {Techscience.uz - техника фанлари долзарб масалалри},
  year = {2026},
  volume = {4},
  number = {9},
  pages = {19-24},
  doi = {10.47390/ts-v4i9y2026n03},
  url = {https://doi.org/10.47390/ts-v4i9y2026n03}
}

Boshqa maqola uchun iqtibos tayyorlash

Techscience.uz - техника фанлари долзарб масалалри jurnalidan boshqa maqolalar

Techscience.uz - техника фанлари долзарб масалалри — barcha maqolalar · 2026-yil arxivi

Texnika sohasidagi boshqa maqolalar

Texnika sohasidagi OAK jurnallari