Efficient Edge-Based Image Interpolation Method Using Neighboring Slope Information

Khan, Sajid Ali; Lee, Dong‐Ho; Khan, Muhammad Asif; Gilal, Abdul Rehman; Mujtaba, Ghulam

doi:10.1109/access.2019.2942004

Cited by 17 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the aspect of objective evaluation, as shown in Table 2, a comparison in terms of PSNR and SSIM is made between the ADTV-SRGAN, and the Bicubic [1], A+ [2], SRCNN [7], VDSR [13], LapSRN [14], GuideAE [15], and SRGAN [20]. The parameters of these methods remain unchanged and can be referred in the corresponding study.…”

Section: B Comparison With Other Methodsmentioning

confidence: 99%

“…The first consists of interpolation methods [1] that estimate the pixel values of interpolation points by using information on the neighborhood of known pixel points. The cons of interpolation methods is that although they have very less complexity, however, poor selection of neighboring pixels for interpolation often leads to the creation of artifacts [2]. The second consists of twostep upscaling methods [3] that apply simple interpolation methods in first stage to upscale the image followed by edge refinement to reduce the artifacts created due to the first stage interpolation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Diagonal Total-Variation Generative Adversarial Network for Super-Resolution Imaging

et al. 2020

View full text Add to dashboard Cite

To address problems that the loss function does not correlate well with perceptual vision in super-resolution methods based on the convolutional neural network(CNN), a novel model called the ADTV-SRGAN is designed based on the adaptive diagonal total-variation generative adversarial network. Combined with global perception and the local structure adaptive method, spatial loss based on the diagonal variation model is proposed to make the loss function can be adjusted according to the spatial features. Pixel loss and characteristic loss are in combination with the spatial loss for the fusing optimization of the total loss function such that high-frequency details of the images are maintained to improve their quality. The results of experiment show that the proposed method can obtain competitive results in objective evaluations. In subjective assessment, images reconstructed by it are clear, delicate, and natural, and it preserved edgeand texture-related details.

show abstract

Section: B Comparison With Other Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adaptive Diagonal Total-Variation Generative Adversarial Network for Super-Resolution Imaging

et al. 2020

View full text Add to dashboard Cite

show abstract

“…These methods make use of the correlation of the data of the damaged points or missing points collected in different time. They can realize efficient interpolation, but it is easy to produce artifacts [21]. The slope is integral and composed of numerous points.…”

Section: Introductionmentioning

confidence: 99%

An Improved Spatio-Temporal Kriging Interpolation Algorithm and Its Application in Slope

et al. 2020

View full text Add to dashboard Cite

Slope stability analysis based on the deformation monitoring data has been commonly used to predict and warn slope disasters. However, due to breakdown of the monitoring equipment or restrictions and interferences of internal and external factors in the area, the loss of data is unavoidable during the process of the slope monitoring. This problem can be solved by the spatio-temporal Kriging interpolation algorithm. However, the subjective factors and theoretical semi-variogram of variogram models, and many parameters estimation may lead to low interpolation precision and poor calculation efficiency. In this paper, a hybrid spatio-temporal interpolation algorithm was presented by combining the improved adaptive genetic algorithm (IAGA) with the spatio-temporal Kriging interpolation method, and it was applied to monitor the deformation of HP1 slope in Yuebao open-pit mine. The results showed that the interpolation precision of the proposed method was about 1 times higher than that of the traditional spatio-temporal Kriging method and the spatio-temporal method of Gaussian process regression method. In addition, after the interpolation analysis of the missing part of the monitoring data, the maximum cumulative displacement of all monitoring points was around 35.8 mm, and the rate of the displacement deformation didn't exceed 2 mm/d for three consecutive days. The deformation trend was basically consistent with the actual slope. It shows that the improved spatio-temporal Kriging interpolation algorithm (ISTKIA) has certain feasibility and reliability, and could provide a new idea for the research of related problems in related fields. INDEX TERMS Slope deformation monitoring, spatio-temporal Kriging interpolation, ISTKIA, stability analysis.

show abstract

“…Getting any information from scenery images is not simple task, it involves deep feature extraction. Many approaches [6]- [11] to this type of computer vision problems have been proposed. The research [12]- [15] in this area is mostly about end-to-end text recognition systems consisting two stages including text detection and text recognition.…”

Section: Introductionmentioning

confidence: 99%

Deep Neural Networks Combined with STN for Multi-Oriented Text Detection and Recognition

Katper¹,

Rehman²,

Alshanqiti³

et al. 2020

IJACSA

View full text Add to dashboard Cite

Developing systems for interpreting visuals, such as images, videos is really challenging but important task to be developed and applied on benchmark datasets. This study solves the very challenge by using STN-OCR model consisting of deep neural networks (DNN) and Spatial Transformer Networks (STNs). The network architecture of this study consists of two stages: localization network and recognition network. In the localization network it finds and localizes text regions and generates sampling grid. Whereas, in the recognition network, text regions will be input and then this network learns to recognize text including low resolution, curved and multioriented text. Deep learning-based approaches require a lot of data for training effectively, therefore, this study has used two benchmark datasets, Street View House Numbers (SVHN) and International Conference on Document Analysis and Recognition (ICDAR) 2015 to evaluate the system. The STN-OCR model achieves better results than literature on these datasets.

show abstract

Efficient Edge-Based Image Interpolation Method Using Neighboring Slope Information

Cited by 17 publications

References 17 publications

Adaptive Diagonal Total-Variation Generative Adversarial Network for Super-Resolution Imaging

Adaptive Diagonal Total-Variation Generative Adversarial Network for Super-Resolution Imaging

An Improved Spatio-Temporal Kriging Interpolation Algorithm and Its Application in Slope

Deep Neural Networks Combined with STN for Multi-Oriented Text Detection and Recognition

Contact Info

Product

Resources

About