An Approach for the Pan Sharpening of Very High Resolution Satellite Images Using a CIELab Color Based Component Substitution Algorithm

Rahımzadeganasl, Alıreza; Algancı, Uğur; Göksel, Çiğdem

doi:10.3390/app9235234

Cited by 10 publications

(7 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A conversion of images RGB color space to the CIE 1976 was done using standard illuminant D65, [0.9504, 1.0000, 1.0888], which simulates noon daylight with a color temperature of 6504 K. The L*a*b* (Lab) colors space was conducted using the Hering's theory for its effectiveness in measuring small color changes [28]. The theory shows that only red (R), green (G), blue (B), and yellow (Y) can characterize the hue component [29]. The LAB color space is a 3-dimensional color system with dimension L* for lightness, and a* and b* for color, where L* is the luminance presented in Eq.…”

Section: Pre-processing Stagementioning

confidence: 99%

Malignant skin melanoma detection using image augmentation by oversampling in nonlinear lower-dimensional embedding manifold

Abayomi‐Alli¹,

Damaševičius²,

Misra³

et al. 2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

The continuous rise in skin cancer cases, especially in malignant melanoma, has resulted in a high mortality rate of the affected patients due to late detection. Some challenges affecting the success of skin cancer detection include small datasets or data scarcity problem, noisy data, imbalanced data, inconsistency in image sizes and resolutions, unavailability of data, reliability of labeled data (ground truth), and imbalance of skin cancer datasets. This study presents a novel data augmentation technique based on covariant Synthetic Minority Oversampling TEchnique (SMOTE) to address the data scarcity and class imbalance problem. We propose an improved data augmentation model for effective detection of melanoma skin cancer. Our method is based on data oversampling in a non-linear lower-dimensional embedding manifold for creating synthetic melanoma images. The proposed data augmentation technique is used to generate a new skin melanoma dataset using dermoscopic images from the publicly available P H 2 dataset. The augmented images were used to train the SqueezeNet deep learning model. The experimental results in binary classification scenario show a significant improvement in detection of melanoma with respect to accuracy (92.18%), sensitivity (80.77%), specificity (95.1%), and F1-score (80.84%). We also improved the multi-class classification results in melanoma detection to 89.2% (sensitivity), 96.2% (specificity) for atypical nevus detection, 65.4% (sensitivity), 72.2% (specificity), and for common nevus detection 66% (sensitivity), 77.2% (specificity). The proposed classification framework outperforms some of the state-of-the-art methods in detecting skin melanoma.

show abstract

Section: Pre-processing Stagementioning

confidence: 99%

Malignant skin melanoma detection using image augmentation by oversampling in nonlinear lower-dimensional embedding manifold

Abayomi‐Alli¹,

Damaševičius²,

Misra³

et al. 2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…(1) Correlation Coefficient (CC) measures the correlation between the original multispectral (MS k ) and fused images (MS are correlated [75,76].…”

Section: Quality Indicesmentioning

confidence: 99%

Automation of Pan-Sharpening Methods for Pléiades Images Using GIS Basic Functions

2021

View full text Add to dashboard Cite

Pan-sharpening methods allow the transfer of higher resolution panchromatic images to multispectral ones concerning the same scene. Different approaches are available in the literature, and only a part of these approaches is included in remote sensing software for automatic application. In addition, the quality of the results supplied by a specific method varies according to the characteristics of the scene; for consequence, different algorithms must be compared to find the best performing one. Nevertheless, pan-sharpening methods can be applied using GIS basic functions in the absence of specific pan-sharpening tools, but this operation is expensive and time-consuming. This paper aims to explain the approach implemented in Quantum GIS (QGIS) for automatic pan-sharpening of Pléiades images. The experiments are carried out on data concerning the Greek island named Lesbo. In total, 14 different pan-sharpening methods are applied to reduce pixel dimensions of the four multispectral bands from 2 m to 0.5 m. The automatic procedure involves basic functions already included in GIS software; it also permits the evaluation of the quality of the resulting images supplying the values of appropriate indices. The results demonstrate that the approach provides the user with the highest performing method every time, so the best possible fused products are obtained with minimal effort in a reduced timeframe.

show abstract

“…1 A PAN band captures a broad range of wavelengths with a high spatial detail content, whereas MS bands capture narrower ranges of wavelengths, with lower spatial resolution. 2 The pansharpened image has both the high spatial detail content of the PAN data and the spectral features of the MS data. By combining the source images, pansharpening can enhance the visual interpretation of the data and improve the accuracy of image analysis.…”

Section: Introductionmentioning

confidence: 99%

Comparative analysis of deep learning-based pansharpening methods for improved image classification accuracy

Yilmaz,

Asikoglu

2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

.Pansharpened images are frequently utilized as base data in image classification applications. Nonetheless, the accuracy of image classification heavily relies on the efficiency of the pansharpening strategy applied. With numerous existing pansharpening approaches available, it becomes challenging for analysts to select the one that yields the best outcome. Recently, the deep learning (DL)-based pansharpening approaches have become popular due to their capabilities. Thus, this study aims at examining the image classification performance of pansharpened images generated by several commonly used DL-based pansharpening algorithms that rely on pre-trained models and comparing them with those of several traditional pansharpening algorithms. The experiments that were conducted in two test sites indicated that the DL-based pansharpening algorithms could be used in various circumstances. It can also be inferred that the DL-based pansharpening algorithms enhanced the image classification accuracy more effectively than many other traditional algorithms. The use of pre-trained models led to robust pansharpening results and, therefore, superior classification accuracies in most cases.

show abstract

An Approach for the Pan Sharpening of Very High Resolution Satellite Images Using a CIELab Color Based Component Substitution Algorithm

Cited by 10 publications

References 42 publications

Malignant skin melanoma detection using image augmentation by oversampling in nonlinear lower-dimensional embedding manifold

Malignant skin melanoma detection using image augmentation by oversampling in nonlinear lower-dimensional embedding manifold

Automation of Pan-Sharpening Methods for Pléiades Images Using GIS Basic Functions

Comparative analysis of deep learning-based pansharpening methods for improved image classification accuracy

Contact Info

Product

Resources

About