Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic
Roads are critical for economic growth and trade but are constantly degraded by heavy traffic and adverse weather, leading to potholes that compromise safety. Traditional detection methods, like manual inspections, are labor-intensive, costly, and prone to errors. Existing automated systems also str...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Case Studies in Construction Materials |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214509525002384 |
| _version_ | 1826554947043328000 |
|---|---|
| author | Mario Roman-Garay Hector Rodriguez-Rangel Carlos Beltran Hernandez-Beltran Peter Lepej José Eleazar Arreygue-Rocha Luis Alberto Morales-Rosales |
| author_facet | Mario Roman-Garay Hector Rodriguez-Rangel Carlos Beltran Hernandez-Beltran Peter Lepej José Eleazar Arreygue-Rocha Luis Alberto Morales-Rosales |
| author_sort | Mario Roman-Garay |
| collection | DOAJ |
| description | Roads are critical for economic growth and trade but are constantly degraded by heavy traffic and adverse weather, leading to potholes that compromise safety. Traditional detection methods, like manual inspections, are labor-intensive, costly, and prone to errors. Existing automated systems also struggle with false positives, particularly in challenging conditions involving shadows, stains, or other environmental interferences. This research presents an architecture for detecting, measuring, and evaluating potholes, as well as generating maintenance recommendations. We integrate 2D images and 3D point clouds captured using the Intel RealSense D435i camera, generating a dataset of 583 images—299 containing potholes and 234 depicting environmental noise on various pavements. Each image is labeled through semantic segmentation and paired with corresponding point clouds. The architecture utilizes transfer learning with a Segformer network, achieving high detection performance with a Recall of 90.87 %, Precision of 90.01 %, Accuracy of 86.8 % F1 Score of 90.433 %, and a loss of 0.0431. The method achieves an IOU of 85.872 %, ensuring accurate diameter estimation, which contrasts with studies using lower IOU values where pothole dimensions are often underestimated due to incomplete detection. Our architecture provides reliable contour detection, facilitating the integration of image data and point clouds to estimate pothole dimensions with a depth estimation error of 5.94 mm. A fuzzy logic system processes these measurements to assess repair urgency and recommend appropriate repair techniques. |
| first_indexed | 2025-03-14T07:49:00Z |
| format | Article |
| id | doaj.art-7e0be9348cd54317b0d8ccd41dcf61df |
| institution | Directory Open Access Journal |
| issn | 2214-5095 |
| language | English |
| last_indexed | 2025-03-14T07:49:00Z |
| publishDate | 2025-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Case Studies in Construction Materials |
| spelling | doaj.art-7e0be9348cd54317b0d8ccd41dcf61df2025-03-03T04:23:44ZengElsevierCase Studies in Construction Materials2214-50952025-07-0122e04440Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logicMario Roman-Garay0Hector Rodriguez-Rangel1Carlos Beltran Hernandez-Beltran2Peter Lepej3José Eleazar Arreygue-Rocha4Luis Alberto Morales-Rosales5División de Estudios de Posgrado e Investigación, Instituto Tecnológico de Culiacán, Juan de Dios Batiz 310 pte, Culiacán, Sinaloa 80220, MéxicoDivisión de Estudios de Posgrado e Investigación, Instituto Tecnológico de Culiacán, Juan de Dios Batiz 310 pte, Culiacán, Sinaloa 80220, MéxicoDivisión de Estudios de Posgrado e Investigación, Instituto Tecnológico de Culiacán, Juan de Dios Batiz 310 pte, Culiacán, Sinaloa 80220, MéxicoFaculty of Agriculture and Life Sciences, University of Maribor, Pivola 10 2311 Hoče, Slovenia, Maribor, Maribor 2000, SloveniaMaestría en Infraestructura del Transporte en la Rama de las Vías Terrestres, Universidad Michoacana de San Nicolás de Hidalgo, Ciudad Universitaria, Avenida Francisco J. Múgica S/N, Morelia, Michoacán 58060, MéxicoFacultad de ingeniería civil, CONAHCYT-Universidad Michoacana de San Nicolás de Hidalgo, Ciudad Universitaria, Avenida Francisco J. Múgica S/N, Morelia, Michoacán 58060, México; Corresponding author.Roads are critical for economic growth and trade but are constantly degraded by heavy traffic and adverse weather, leading to potholes that compromise safety. Traditional detection methods, like manual inspections, are labor-intensive, costly, and prone to errors. Existing automated systems also struggle with false positives, particularly in challenging conditions involving shadows, stains, or other environmental interferences. This research presents an architecture for detecting, measuring, and evaluating potholes, as well as generating maintenance recommendations. We integrate 2D images and 3D point clouds captured using the Intel RealSense D435i camera, generating a dataset of 583 images—299 containing potholes and 234 depicting environmental noise on various pavements. Each image is labeled through semantic segmentation and paired with corresponding point clouds. The architecture utilizes transfer learning with a Segformer network, achieving high detection performance with a Recall of 90.87 %, Precision of 90.01 %, Accuracy of 86.8 % F1 Score of 90.433 %, and a loss of 0.0431. The method achieves an IOU of 85.872 %, ensuring accurate diameter estimation, which contrasts with studies using lower IOU values where pothole dimensions are often underestimated due to incomplete detection. Our architecture provides reliable contour detection, facilitating the integration of image data and point clouds to estimate pothole dimensions with a depth estimation error of 5.94 mm. A fuzzy logic system processes these measurements to assess repair urgency and recommend appropriate repair techniques.http://www.sciencedirect.com/science/article/pii/S2214509525002384Pothole detectionDepth estimationDeep learningPavement assessmentMaintenance recommendations |
| spellingShingle | Mario Roman-Garay Hector Rodriguez-Rangel Carlos Beltran Hernandez-Beltran Peter Lepej José Eleazar Arreygue-Rocha Luis Alberto Morales-Rosales Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic Case Studies in Construction Materials Pothole detection Depth estimation Deep learning Pavement assessment Maintenance recommendations |
| title | Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic |
| title_full | Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic |
| title_fullStr | Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic |
| title_full_unstemmed | Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic |
| title_short | Architecture for pavement pothole evaluation using deep learning, machine vision, and fuzzy logic |
| title_sort | architecture for pavement pothole evaluation using deep learning machine vision and fuzzy logic |
| topic | Pothole detection Depth estimation Deep learning Pavement assessment Maintenance recommendations |
| url | http://www.sciencedirect.com/science/article/pii/S2214509525002384 |
| work_keys_str_mv | AT marioromangaray architectureforpavementpotholeevaluationusingdeeplearningmachinevisionandfuzzylogic AT hectorrodriguezrangel architectureforpavementpotholeevaluationusingdeeplearningmachinevisionandfuzzylogic AT carlosbeltranhernandezbeltran architectureforpavementpotholeevaluationusingdeeplearningmachinevisionandfuzzylogic AT peterlepej architectureforpavementpotholeevaluationusingdeeplearningmachinevisionandfuzzylogic AT joseeleazararreyguerocha architectureforpavementpotholeevaluationusingdeeplearningmachinevisionandfuzzylogic AT luisalbertomoralesrosales architectureforpavementpotholeevaluationusingdeeplearningmachinevisionandfuzzylogic |