Search results

1 – 10 of 58
Article
Publication date: 21 November 2022

Aslan Ahmet Haykir and Ilkay Oksuz

Data quality and data resolution are essential for computer vision tasks like medical image processing, object detection, pattern recognition and so on. Super-resolution is a way…

98

Abstract

Purpose

Data quality and data resolution are essential for computer vision tasks like medical image processing, object detection, pattern recognition and so on. Super-resolution is a way to increase the image resolution, and super-resolved images contain more information compared to their low-resolution counterparts. The purpose of this study is analyzing the effects of the super resolution models trained before on object detection for aerial images.

Design/methodology/approach

Two different models were trained using the Super-Resolution Generative Adversarial Network (SRGAN) architecture on two aerial image data sets, the xView and the Dataset for Object deTection in Aerial images (DOTA). This study uses these models to increase the resolution of aerial images for improving object detection performance. This study analyzes the effects of the model with the best perceptual index (PI) and the model with the best RMSE on object detection in detail.

Findings

Super-resolution increases the object detection quality as expected. But, the super-resolution model with better perceptual quality achieves lower mean average precision results compared to the model with better RMSE. It means that the model with a better PI is more meaningful to human perception but less meaningful to computer vision.

Originality/value

The contributions of the authors to the literature are threefold. First, they do a wide analysis of SRGAN results for aerial image super-resolution on the task of object detection. Second, they compare super-resolution models with best PI and best RMSE to showcase the differences on object detection performance as a downstream task first time in the literature. Finally, they use a transfer learning approach for super-resolution to improve the performance of object detection.

Details

Information Discovery and Delivery, vol. 51 no. 4
Type: Research Article
ISSN: 2398-6247

Keywords

Article
Publication date: 16 August 2019

Shuangshuang Liu and Xiaoling Li

Conventional image super-resolution reconstruction by the conventional deep learning architectures suffers from the problems of hard training and gradient disappearing. In order…

Abstract

Purpose

Conventional image super-resolution reconstruction by the conventional deep learning architectures suffers from the problems of hard training and gradient disappearing. In order to solve such problems, the purpose of this paper is to propose a novel image super-resolution algorithm based on improved generative adversarial networks (GANs) with Wasserstein distance and gradient penalty.

Design/methodology/approach

The proposed algorithm first introduces the conventional GANs architecture, the Wasserstein distance and the gradient penalty for the task of image super-resolution reconstruction (SRWGANs-GP). In addition, a novel perceptual loss function is designed for the SRWGANs-GP to meet the task of image super-resolution reconstruction. The content loss is extracted from the deep model’s feature maps, and such features are introduced to calculate mean square error (MSE) for the loss calculation of generators.

Findings

To validate the effectiveness and feasibility of the proposed algorithm, a lot of compared experiments are applied on three common data sets, i.e. Set5, Set14 and BSD100. Experimental results have shown that the proposed SRWGANs-GP architecture has a stable error gradient and iteratively convergence. Compared with the baseline deep models, the proposed GANs models have a significant improvement on performance and efficiency for image super-resolution reconstruction. The MSE calculated by the deep model’s feature maps gives more advantages for constructing contour and texture.

Originality/value

Compared with the state-of-the-art algorithms, the proposed algorithm obtains a better performance on image super-resolution and better reconstruction results on contour and texture.

Details

International Journal of Intelligent Computing and Cybernetics, vol. 12 no. 3
Type: Research Article
ISSN: 1756-378X

Keywords

Article
Publication date: 6 August 2021

A. Valli Bhasha and B.D. Venkatramana Reddy

The problems of Super resolution are broadly discussed in diverse fields. Rather than the progression toward the super resolution models for real-time images, operating…

Abstract

Purpose

The problems of Super resolution are broadly discussed in diverse fields. Rather than the progression toward the super resolution models for real-time images, operating hyperspectral images still remains a challenging problem.

Design/methodology/approach

This paper aims to develop the enhanced image super-resolution model using “optimized Non-negative Structured Sparse Representation (NSSR), Adaptive Discrete Wavelet Transform (ADWT), and Optimized Deep Convolutional Neural Network”. Once after converting the HR images into LR images, the NSSR images are generated by the optimized NSSR. Then the ADWT is used for generating the subbands of both NSSR and HRSB images. The residual image with this information is obtained by the optimized Deep CNN. All the improvements on the algorithms are done by the Opposition-based Barnacles Mating Optimization (O-BMO), with the objective of attaining the multi-objective function concerning the “Peak Signal-to-Noise Ratio (PSNR), and Structural similarity (SSIM) index”. Extensive analysis on benchmark hyperspectral image datasets shows that the proposed model achieves superior performance over typical other existing super-resolution models.

Findings

From the analysis, the overall analysis of the suggested and the conventional super resolution models relies that the PSNR of the improved O-BMO-(NSSR+DWT+CNN) was 38.8% better than bicubic, 11% better than NSSR, 16.7% better than DWT+CNN, 1.3% better than NSSR+DWT+CNN, and 0.5% better than NSSR+FF-SHO-(DWT+CNN). Hence, it has been confirmed that the developed O-BMO-(NSSR+DWT+CNN) is performing well in converting LR images to HR images.

Originality/value

This paper adopts a latest optimization algorithm called O-BMO with optimized Non-negative Structured Sparse Representation (NSSR), Adaptive Discrete Wavelet Transform (ADWT) and Optimized Deep Convolutional Neural Network for developing the enhanced image super-resolution model. This is the first work that uses O-BMO-based Deep CNN for image super-resolution model enhancement.

Article
Publication date: 1 March 2013

Lei Zeng, Xiaofeng Li and Jin Xu

The purpose of this paper is to introduce an improved method for joint training of low‐ and high‐resolution dictionaries in the single image super resolution. With simulations…

Abstract

Purpose

The purpose of this paper is to introduce an improved method for joint training of low‐ and high‐resolution dictionaries in the single image super resolution. With simulations, the proposed method is thereafter evaluated.

Design/methodology/approach

Sparse representations of low‐resolution image patches are used to reconstruct the high‐resolution image patches with high resolution dictionary. By using different factors, the scheme weights the two dictionaries in the high‐ and low‐resolution spaces in the training process. It is reasonable to achieve better reconstructed images with more emphasis on the high‐resolution spaces.

Findings

An improved joint training algorithm based on K‐SVD is developed with flexible weight factors on dictionaries in the high‐ and low‐resolution spaces. From the experiment results, the proposed scheme outperforms the classic bicubic interpolation and neighbor‐embedding learning based method.

Originality/value

By using flexible weight factors in joint training of the dictionaries for super resolution, better reconstruction results can be achieved.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 32 no. 2
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 23 August 2019

Haiqing He, Ting Chen, Minqiang Chen, Dajun Li and Penggen Cheng

This paper aims to present a novel approach of image super-resolution based on deep–shallow cascaded convolutional neural networks for reconstructing a clear and high-resolution…

Abstract

Purpose

This paper aims to present a novel approach of image super-resolution based on deep–shallow cascaded convolutional neural networks for reconstructing a clear and high-resolution (HR) remote sensing image from a low-resolution (LR) input.

Design/methodology/approach

The proposed approach directly learns the residuals and mapping between simulated LR and their corresponding HR remote sensing images based on deep and shallow end-to-end convolutional networks instead of assuming any specific restored models. Extra max-pooling and up-sampling are used to achieve a multiscale space by concatenating low- and high-level feature maps, and an HR image is generated by combining LR input and the residual image. This model ensures a strong response to spatially local input patterns by using a large filter and cascaded small filters. The authors adopt a strategy based on epochs to update the learning rate for boosting convergence speed.

Findings

The proposed deep network is trained to reconstruct high-quality images for low-quality inputs through a simulated dataset, which is generated with Set5, Set14, Berkeley Segmentation Data set and remote sensing images. Experimental results demonstrate that this model considerably enhances remote sensing images in terms of spatial detail and spectral fidelity and outperforms state-of-the-art SR methods in terms of peak signal-to-noise ratio, structural similarity and visual assessment.

Originality/value

The proposed method can reconstruct an HR remote sensing image from an LR input and significantly improve the quality of remote sensing images in terms of spatial detail and fidelity.

Details

Sensor Review, vol. 39 no. 5
Type: Research Article
ISSN: 0260-2288

Keywords

Article
Publication date: 3 August 2023

Yandong Hou, Zhengbo Wu, Xinghua Ren, Kaiwen Liu and Zhengquan Chen

High-resolution remote sensing images possess a wealth of semantic information. However, these images often contain objects of different sizes and distributions, which make the…

Abstract

Purpose

High-resolution remote sensing images possess a wealth of semantic information. However, these images often contain objects of different sizes and distributions, which make the semantic segmentation task challenging. In this paper, a bidirectional feature fusion network (BFFNet) is designed to address this challenge, which aims at increasing the accurate recognition of surface objects in order to effectively classify special features.

Design/methodology/approach

There are two main crucial elements in BFFNet. Firstly, the mean-weighted module (MWM) is used to obtain the key features in the main network. Secondly, the proposed polarization enhanced branch network performs feature extraction simultaneously with the main network to obtain different feature information. The authors then fuse these two features in both directions while applying a cross-entropy loss function to monitor the network training process. Finally, BFFNet is validated on two publicly available datasets, Potsdam and Vaihingen.

Findings

In this paper, a quantitative analysis method is used to illustrate that the proposed network achieves superior performance of 2–6%, respectively, compared to other mainstream segmentation networks from experimental results on two datasets. Complete ablation experiments are also conducted to demonstrate the effectiveness of the elements in the network. In summary, BFFNet has proven to be effective in achieving accurate identification of small objects and in reducing the effect of shadows on the segmentation process.

Originality/value

The originality of the paper is the proposal of a BFFNet based on multi-scale and multi-attention strategies to improve the ability to accurately segment high-resolution and complex remote sensing images, especially for small objects and shadow-obscured objects.

Details

International Journal of Intelligent Computing and Cybernetics, vol. 17 no. 1
Type: Research Article
ISSN: 1756-378X

Keywords

Article
Publication date: 8 January 2018

Shafinaz Mohd Basir, Idnin Pasya, Tajmalludin Yaakob, Nur Emileen Abd Rashid and Takehiko Kobayashi

This paper aims to present an approach of utilizing multiple-input multiple-output (MIMO) radar concept to enhance pedestrian classification in automotive sensors. In a practical…

Abstract

Purpose

This paper aims to present an approach of utilizing multiple-input multiple-output (MIMO) radar concept to enhance pedestrian classification in automotive sensors. In a practical environment, radar signals reflected from pedestrians and slow-moving vehicles are similar in terms of reflecting angle and Doppler returns, inducing difficulty for target discrimination. An efficient discrimination between the two targets depends on the ability of the sensor to extract unique characteristics from each target, for example, by exploiting Doppler signatures. This study describes the utilization of MIMO radar for Doppler measurement and demonstrates its application to improve pedestrian classification through actual laboratory measurements.

Design/methodology/approach

Multiple non-modulated sinusoidal signals are transmitted orthogonally over a MIMO array using time division scheme, illuminating human and non-human targets. The reflected signal entering each of the receiving antenna are combined at the radar receiver prior to Doppler processing. Doppler histogram was formulated based on a series of measurements, and the Doppler spread of the targets was determined from the histograms. Results were compared between MIMO and conventional single antenna systems.

Findings

Measurement results indicated that the MIMO configuration provides able to capture more Doppler information compared to conventional single antenna systems, enabling a more precise discrimination between pedestrian and other slow-moving objects on the road.

Originality/value

The study demonstrated the effectiveness of using MIMO configuration in radar-based automotive sensor to enhance the accuracy of Doppler estimation, which is seldom highlighted in literature of MIMO radars. The result also indicated its usefulness in improving target discrimination capability of the radar, through actual measurement.

Details

Sensor Review, vol. 38 no. 2
Type: Research Article
ISSN: 0260-2288

Keywords

Article
Publication date: 28 March 2019

Jean Paul Simon

This paper aims to clarify the notion of artificial intelligence (AI), reviewing the present scope of the phenomenon through its main applications. It aims at describing the…

4827

Abstract

Purpose

This paper aims to clarify the notion of artificial intelligence (AI), reviewing the present scope of the phenomenon through its main applications. It aims at describing the various applications while assessing the markets, highlighting some of the leading industrial sectors in the field. Therefore, it identifies pioneering companies and the geographical distribution of AI companies.

Design/methodology/approach

The paper builds upon an in-depth investigation of public initiatives focusing mostly on the EU. It is based on desk research, a comprehensive review of the main grey and scientific literature in this field.

Findings

The paper notes that there is no real consensus on any definition for this umbrella term, that the definition does fluctuate over time but highlights some of the main changes and advances that took place over the past 60 years. It stresses that, in spite of the hype, on both the business and consumer sides, the demand appears uncertain. The scope of the announced disruptions is not easy to assess, technological innovation associated with AI may be modest or take some time to be fully deployed. However, some companies and regions are leading already in the field.

Research limitations/implications

The paper, based on desk research, does not consider any expert opinions. Besides, the scientific literature on the phenomenon is still scarce (but not the technical one in the specific research sectors of AI). Most of the data come from consultancies or government publications which may introduce some bias, although the paper gathered various, often conflicting viewpoints.

Originality/value

The paper gives a thorough review of the available literature (consultancies, governments) stressing the limitations of the available research on economic and social aspects. It aims at providing a comprehensive overview of the major trends in the field. It gives a global overview of companies and regions.

Details

Digital Policy, Regulation and Governance, vol. 21 no. 3
Type: Research Article
ISSN: 2398-5038

Keywords

Article
Publication date: 17 March 2016

suryanarayana gunnam and Ravindra Dhuli

The purpose of this paper is to present an improved wavelet based approach in single image super resolution (SISR). The proposed method generates high resolution (HR) images by…

Abstract

Purpose

The purpose of this paper is to present an improved wavelet based approach in single image super resolution (SISR). The proposed method generates high resolution (HR) images by preserving the image contrast and edges simultaneously.

Design/methodology/approach

Covariance based interpolation algorithm is employed to obtain an initial estimate of the unknown HR image. The proposed method preserves the image contrast, by applying singular value decomposition (SVD) based correction on the dual-tree complex wavelet transform (DT-CWT) coefficients. In addition, the dual operating mode diffusion based shock filter (DBSF) ensures noise mitigation and edge preservation.

Findings

Experimental results on various test images reveal superiority of the proposed method over the existing SISR techniques in terms of peak signal to noise ratio (PSNR), structural similarity index measure (SSIM) and visual quality.

Originality/value

With SVD based correction, the proposed method preserves the image contrast and also the DBSF operation helps to achieve sharper edges.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering , vol. 35 no. 3
Type: Research Article
ISSN: 0332-1649

Book part
Publication date: 14 March 2024

Mousumi Bose, Lilly Ye and Yiming Zhuang

Today's marketing is dominated by decision-making based on artificial intelligence and machine learning. This study focuses on one semi- and unsupervised machine learning…

Abstract

Today's marketing is dominated by decision-making based on artificial intelligence and machine learning. This study focuses on one semi- and unsupervised machine learning technique, generative adversarial networks (GANs). GANs are a type of deep learning architecture capable of generating new data similar to the training data that were used to train it, and thus, it is designed to learn a generative model that can produce new samples. GANs have been used in multiple marketing areas, especially in creating images and video and providing customized consumer contents. Through providing a holistic picture of GANs, including its advantage, disadvantage, ethical considerations, and its current application, the study attempts to provide business some strategical orientations, including formulating strong marketing positioning, creating consumer lifetime values, and delivering desired marketing tactics in product, promotion, pricing, and distribution channel. Through using GANs, marketers will create unique experiences for consumers, build strategic focus, and gain competitive advantages. This study is an original endeavor in discussing GANs in marketing, offering fresh insights in this research topic.

Details

The Impact of Digitalization on Current Marketing Strategies
Type: Book
ISBN: 978-1-83753-686-3

Keywords

1 – 10 of 58