AI-BASED CONCRETE SURFACE POROSITY AND DEFECT DETECTION USING QUANTUM DEEP LEARNING IMAGE ANALYSIS AND ARDUINO UNO

Concrete surface porosity, bugholes, cracks, honeycombing, and local spalling influence durability, appearance, and theperformance of protective coatings and repair systems. Conventional visual inspection is dependent on access, lighting,inspector experience, and subjective interpretation, while image-based deep learning can provide repeatable localization ofvisible surface anomalies. This paper formulates a proposed hybrid quantum–classical image-analysis framework for identifyingvisible concrete-surface porosity and related defects, with an Arduino Uno used only as a low-level status and indicationinterface. The proposed workflow acquires RGB surface images, applies region-of-interest selection and photometricnormalization, extracts compact visual descriptors through a classical convolutional encoder, and maps the reduced featurevector into a parameterized quantum circuit. Measurement expectation values are passed to a classical classification head toestimate defect class and confidence. A segmentation-oriented branch estimates the visible defect mask and computes an imagebased area fraction. The Arduino Uno is not assigned image processing or quantum computation because its ATmega328Pplatform has limited memory and computational resources; inference is instead assigned to a host computer or suitable edgegateway, with a proposed serial status link to the controller. Since no project dataset, camera specification, circuitimplementation, or measured observations were supplied, the experimental graphs and numerical examples in this manuscriptare explicitly illustrative and are not reported as measured performance. The contribution is a reproducible research design,mathematical evaluation framework, and hardware–software integration concept that requires validation on annotated concreteimages and physical specimens before engineering deployment

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-21
DOI
https://doi.org/10.5281/zenodo.24423363
Primary Topic
Infrastructure Maintenance and Monitoring
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

AI-BASED CONCRETE SURFACE POROSITY AND DEFECT DETECTION USING QUANTUM DEEP LEARNING IMAGE ANALYSIS AND ARDUINO UNO

KETHAVATH AMINA, G.SUDHAKAR, AJMEERA GANESH, UGLAVATH NILESH
Zenodo (CERN European Organization for Nuclear Research)
Infrastructure Maintenance and Monitoring
article

AI-BASED CONCRETE SURFACE POROSITY AND DEFECT DETECTION USING QUANTUM DEEP LEARNING IMAGE ANALYSIS AND ARDUINO UNO

KETHAVATH AMINA, G.SUDHAKAR, AJMEERA GANESH, UGLAVATH NILESH
article en

Abstract

Concrete surface porosity, bugholes, cracks, honeycombing, and local spalling influence durability, appearance, and theperformance of protective coatings and repair systems. Conventional visual inspection is dependent on access, lighting,inspector experience, and subjective interpretation, while image-based deep learning can provide repeatable localization ofvisible surface anomalies. This paper formulates a proposed hybrid quantum–classical image-analysis framework for identifyingvisible concrete-surface porosity and related defects, with an Arduino Uno used only as a low-level status and indicationinterface. The proposed workflow acquires RGB surface images, applies region-of-interest selection and photometricnormalization, extracts compact visual descriptors through a classical convolutional encoder, and maps the reduced featurevector into a parameterized quantum circuit. Measurement expectation values are passed to a classical classification head toestimate defect class and confidence. A segmentation-oriented branch estimates the visible defect mask and computes an imagebased area fraction. The Arduino Uno is not assigned image processing or quantum computation because its ATmega328Pplatform has limited memory and computational resources; inference is instead assigned to a host computer or suitable edgegateway, with a proposed serial status link to the controller. Since no project dataset, camera specification, circuitimplementation, or measured observations were supplied, the experimental graphs and numerical examples in this manuscriptare explicitly illustrative and are not reported as measured performance. The contribution is a reproducible research design,mathematical evaluation framework, and hardware–software integration concept that requires validation on annotated concreteimages and physical specimens before engineering deployment

Zenodo (CERN European Organization for Nuclear Research)
Infrastructure Maintenance and Monitoring
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon