Research on the natural image super-resolution reconstruction algorithm based on compressive perception theory and deep learning model

Duan, Ganglong; Hu, Wang; Wang, Jianren

doi:10.1016/j.neucom.2015.12.125

Cited by 21 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where, t start and t end indicate the start time and end time, respectively. In addition, the sparse basis and the random measurement matrices use discrete cosine orthogonal basis and orthogonal symmetric Toeplitz matrices [36,37], respectively.…”

Section: Experiments and Results Analysismentioning

confidence: 99%

Adaptive Algorithm on Block-Compressive Sensing and Noisy Data Estimation

2019

View full text Add to dashboard Cite

In this paper, an altered adaptive algorithm on block-compressive sensing (BCS) is developed by using saliency and error analysis. A phenomenon has been observed that the performance of BCS can be improved by means of rational block and uneven sampling ratio as well as adopting error analysis in the process of reconstruction. The weighted mean information entropy is adopted as the basis for partitioning of BCS which results in a flexible block group. Furthermore, the synthetic feature (SF) based on local saliency and variance is introduced to step-less adaptive sampling that works well in distinguishing and sampling between smooth blocks and detail blocks. The error analysis method is used to estimate the optimal number of iterations in sparse reconstruction. Based on the above points, an altered adaptive block-compressive sensing algorithm with flexible partitioning and error analysis is proposed in the article. On the one hand, it provides a feasible solution for the partitioning and sampling of an image, on the other hand, it also changes the iteration stop condition of reconstruction, and then improves the quality of the reconstructed image. The experimental results verify the effectiveness of the proposed algorithm and illustrate a good improvement in the indexes of the Peak Signal to Noise Ratio (PSNR), Structural Similarity (SSIM), Gradient Magnitude Similarity Deviation (GMSD), and Block Effect Index (BEI).

show abstract

Section: Experiments and Results Analysismentioning

confidence: 99%

Adaptive Algorithm on Block-Compressive Sensing and Noisy Data Estimation

2019

View full text Add to dashboard Cite

show abstract

“…Here, DNN classifier [20] [22] is employed for the classification process of breast cancer using the generated feature vector from a segmented blood cell. The DNN classifier includes a number of hidden layers linked among the input layer and the output layer.…”

Section: Structure Of Dnnmentioning

confidence: 99%

ALOA: Ant Lion Optimization Algorithm-based Deep Learning for Breast Cancer Classification

Fernandis¹

2021

View full text Add to dashboard Cite

Normally, breast cancer is the most familiar disease, and it is the second leading type of cancer death among women. Breast cancer disease begins in malignant stage and it slowly develops from breast cell. Periodic clinical checks and self-tests assist for early prediction of breast cancer, thus the survival rate can increase significantly. In this paper, Ant Lion Optimization (ALO)-based Deep Neural Network (DNN) technique is developed for breast cancer classification. Here, the Type 2 Fuzzy and Cuckoo Search (T2FCS) method is utilized for pre-processing the input image. The color-based thresholding scheme is employed for segmenting the blood cells from the input breast cancer image. Furthermore, the statistical features are extracted from an input image for predicting breast cancer. Besides, the cancer classification is performed using a DNN classifier. The ALO algorithm is developed in this method to train the classifier for better performance. In addition, the developed breast cancer classification approach achieves a better classification performance based on several parameters, such as specificity, accuracy, and sensitivity. Thus, the developed ALO-based DNN method obtained enhanced results with high accuracy of 95.34%, sensitivity of 97.54%, and specificity of 95.34%.

show abstract

“…19 Some authors have evaluated the combined uses of NNs and CS. 20,21 The idea is to first recover images from compressive or low-resolution (LR) samples and then map the result into HR images by applying properly trained superresolution convolutional neural networks (SRCNNs). 22 Such concepts have been used in a large number of computer vision problems, including image enhancement, such as denoising 23 and deblurring.…”

Section: Introductionmentioning

confidence: 99%

Deep learning for enhancing wavefield image quality in fast non-contact inspections

Esfandabadi

Bilodeau

Masson

et al. 2019

Structural Health Monitoring

View full text Add to dashboard Cite

Ultrasonic wavefield imaging with a non-contact technology can provide detailed information about the health status of an inspected structure. However, high spatial resolution, often necessary for accurate damage quantification, typically demands a long scanning time. In this work, we investigate a novel methodology to acquire high-resolution wavefields with a reduced number of measurement points to minimize the acquisition time. Such methodology is based on the combination of compressive sensing and convolutional neural networks to recover high spatial frequency information from low-resolution images. A data set was built from 652 wavefield images acquired with a laser Doppler vibrometer describing guided ultrasonic wave propagation in eight different structures, with and without various simulated defects. Out of those 652 images, 326 cases without defect and 326 cases with defect were used as a training database for the convolutional neural network. In addition, 273 wavefield images were used as a testing database to validate the proposed methodology. For quantitative evaluation, two image quality metrics were calculated and compared to those achieved with different recovery methods or by training the convolutional neural network with non-wavefield images data set. The results demonstrate the capability of the technique for enhancing image resolution and quality, as well as similarity to the wavefield acquired on the full high-resolution grid of scan points, while reducing the number of measurement points down to 10% of the number of scan points for a full grid.

show abstract

Research on the natural image super-resolution reconstruction algorithm based on compressive perception theory and deep learning model

Cited by 21 publications

References 44 publications

Adaptive Algorithm on Block-Compressive Sensing and Noisy Data Estimation

Adaptive Algorithm on Block-Compressive Sensing and Noisy Data Estimation

ALOA: Ant Lion Optimization Algorithm-based Deep Learning for Breast Cancer Classification

Deep learning for enhancing wavefield image quality in fast non-contact inspections

Contact Info

Product

Resources

About