CNN-Based Image Quality Classification Considering Quality Degradation in Bridge Inspection Using an Unmanned Aerial Vehicle

Gwon, Gi-Hun; Lee, Jin Hwan; Kim, In‐Ho; Jung, Hyung‐Jo

doi:10.1109/access.2023.3238204

Cited by 13 publications

(3 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When a reference image is available [15], the performance of the merged image is assessed using Mutual Information (MI) [16], Correlation Coefficient (CC) [17], Structural Content (SC), and Normalized Cross Correlation (NCC) [14]. However, if the reference image is not available, the performance of the merged image is assessed using metrics such as Structural Blind/Reference less Image Spatial Quality Evaluator (BRISQUE) [18], Naturalness Image Quality Evaluator (NIQE) [19], Perceptionbased Image Quality Evaluator (PIQE) [15].…”

Section: Color Transform Methodsmentioning

confidence: 99%

Quantitative Analysis of Blurry Color Image Fusion Techniques using Color Transform

Mohammed,

Mohamad,

Abbas

et al. 2023

Al-Mustansiriyah Journal of Science

View full text Add to dashboard Cite

This work focuses on fused color images resulting from motion blur (left and right) with a blur block size of 11 pixels. The color conversion process was performed from RGB color space (Red, Green, Blue) to HSV (Hue, Saturation, Value), L*a*b*, and Ycbcr (Luminance, Chrominance) color space. The traditional (addition, multiplication) and proposed fusion techniques (absolute real standard deviation) were used for this purpose. The data was examined by quality criteria with reference (Mutual Information, Correlation Coefficient, Structural Content, Normalized Cross Correlation) and without reference (Blind Reference less Image Spatial Quality Evaluator, Naturalness Image Quality Evaluator, and Perception-based Image Quality Evaluator). In results and depending on the criteria, the best fusion method is the proposed real standard deviation.

show abstract

Section: Color Transform Methodsmentioning

confidence: 99%

Quantitative Analysis of Blurry Color Image Fusion Techniques using Color Transform

Mohammed,

Mohamad,

Abbas

et al. 2023

Al-Mustansiriyah Journal of Science

View full text Add to dashboard Cite

show abstract

“…Wu et al [125] used UAVs and remote sensing to monitor earth-moving excavators at construction sites where monitoring cameras are not available for installation. Gwon et al [126] propose an image quality assessment method using a CNN for UAV-based bridge inspection considering various degradation factors. Xing et al [127] improve the YOLOv5 DL model for UAV pavement crack detection, achieving real-time pixellevel detection with higher accuracy and faster speed.…”

Section: ) Infrastructurementioning

confidence: 99%

A Survey on Applications of Unmanned Aerial Vehicles Using Machine Learning

Teixeira,

Miguel,

Silva

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Unmanned Aerial Vehicles (UAVs) play an important role in many applications, including health, transport, telecommunications and safe and rescue operations. Their adoption can improve the speed and precision of applications when compared to traditional solutions based on handwork. The use of UAVs brings scientific and technological challenges. In this context, Machine Learning (ML) techniques provide solutions to several problems concerning the use of UAVs in civil and military applications. An increasing number of scientific papers on the use of ML in UAVs context have been published in academic journals. In this work, we present a literature review on the use of ML techniques in UAVs, outlining the most recurrent areas and the most commonly used ML techniques in UAV applications. The results reveal that applications in the areas of environment, communication and security are among the main research topics.INDEX TERMS unmanned aerial vehicle, machine learning, literature review, UAV applications, neural networks.

show abstract

“…In recent years, the application of UAVs equipped with advanced image sensors for infrastructure inspection has garnered significant attention in the field of maintenance and structural health monitoring [1][2][3]. This emerging research area aims to exploit the advantages offered by robot and image sensor-based inspections, notably their remote control capabilities and enhanced accessibility, surpassing the limitations of conventional visual inspection methods [4,5]. By enabling more efficient and cost-effective inspections of large-scale structures within reduced time frames, these advanced techniques hold the potential to revolutionize inspection practices.…”

Section: Introductionmentioning

confidence: 99%

Image-to-Image Translation-Based Structural Damage Data Augmentation for Infrastructure Inspection Using Unmanned Aerial Vehicle

Gwon,

Lee,

Kim

et al. 2023

Drones

Self Cite

View full text Add to dashboard Cite

As technology advances, the use of unmanned aerial vehicles (UAVs) and image sensors for structural monitoring and diagnostics is becoming increasingly critical. This approach enables the efficient inspection and assessment of structural conditions. Furthermore, the integration of deep learning techniques has been proven to be highly effective in detecting damage from structural images, as demonstrated in our study. To enable effective learning by deep learning models, a substantial volume of data is crucial, but collecting appropriate instances of structural damage from real-world scenarios poses challenges and demands specialized knowledge, as well as significant time and resources for labeling. In this study, we propose a methodology that utilizes a generative adversarial network (GAN) for image-to-image translation, with the objective of generating synthetic structural damage data to augment the dataset. Initially, a GAN-based image generation model was trained using paired datasets. When provided with a mask image, this model generated an RGB image based on the annotations. The subsequent step generated domain-specific mask images, a critical task that improved the data augmentation process. These mask images were designed based on prior knowledge to suit the specific characteristics and requirements of the structural damage dataset. These generated masks were then used by the GAN model to produce new RGB image data incorporating various types of damage. In the experimental validation conducted across the three datasets to assess the image generation for data augmentation, our results demonstrated that the generated images closely resembled actual images while effectively conveying information about the newly introduced damage. Furthermore, the experimental validation of damage detection with augmented data entailed a comparative analysis between the performance achieved solely with the original dataset and that attained with the incorporation of additional augmented data. The results for damage detection consistently demonstrated that the utilization of augmented data enhanced performance when compared to relying solely on the original images.

show abstract

CNN-Based Image Quality Classification Considering Quality Degradation in Bridge Inspection Using an Unmanned Aerial Vehicle

Cited by 13 publications

References 38 publications

Quantitative Analysis of Blurry Color Image Fusion Techniques using Color Transform

Quantitative Analysis of Blurry Color Image Fusion Techniques using Color Transform

A Survey on Applications of Unmanned Aerial Vehicles Using Machine Learning

Image-to-Image Translation-Based Structural Damage Data Augmentation for Infrastructure Inspection Using Unmanned Aerial Vehicle

Contact Info

Product

Resources

About