Visual Computing and Understanding: New Developments and Trends

Journal Name	Impact Factor	CiteScore	Launched Year	First Decision (median)	APC
Applied Sciences applsci	2.5	5.3	2011	18.4 Days	CHF 2400	Submit
Computers computers	2.6	5.4	2012	15.5 Days	CHF 1800	Submit
Electronics electronics	2.6	5.3	2012	16.4 Days	CHF 2400	Submit
Information information	2.4	6.9	2010	16.4 Days	CHF 1600	Submit
Journal of Imaging jimaging	2.7	5.9	2015	18.3 Days	CHF 1800	Submit

33 pages, 36906 KiB

Open AccessArticle

Making Images Speak: Human-Inspired Image Description Generation

by Chifaa Sebbane, Ikram Belhajem and Mohammed Rziza

Information 2025, 16(5), 356; https://doi.org/10.3390/info16050356 - 28 Apr 2025

Despite significant advances in deep learning-based image captioning, many state-of-the-art models still struggle to balance visual grounding (i.e., accurate object and scene descriptions) with linguistic coherence (i.e., grammatical fluency and appropriate use of non-visual tokens such as articles and prepositions). To address these [...] Read more.

Despite significant advances in deep learning-based image captioning, many state-of-the-art models still struggle to balance visual grounding (i.e., accurate object and scene descriptions) with linguistic coherence (i.e., grammatical fluency and appropriate use of non-visual tokens such as articles and prepositions). To address these limitations, we propose a hybrid image captioning framework that integrates handcrafted and deep visual features. Specifically, we combine local descriptors—Scale-Invariant Feature Transform (SIFT) and Bag of Features (BoF)—with high-level semantic features extracted using ResNet50. This dual representation captures both fine-grained spatial details and contextual semantics. The decoder employs Bahdanau attention refined with an Attention-on-Attention (AoA) mechanism to optimize visual-textual alignment, while GloVe embeddings and a GRU-based sequence model ensure fluent language generation. The proposed system is trained on 200,000 image-caption pairs from the MS COCO train2014 dataset and evaluated on 50,000 held-out MS COCO pairs plus the Flickr8K benchmark. Our model achieves a CIDEr score of 128.3 and a SPICE score of 29.24, reflecting clear improvements over baselines in both semantic precision—particularly for spatial relationships—and grammatical fluency. These results validate that combining classical computer vision techniques with modern attention mechanisms yields more interpretable and linguistically precise captions, addressing key limitations in neural caption generation. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

17 pages, 39878 KiB

Open AccessArticle

Real-Time Volume-Rendering Image Denoising Based on Spatiotemporal Weighted Kernel Prediction

by Xinran Xu, Chunxiao Xu and Lingxiao Zhao

J. Imaging 2025, 11(4), 126; https://doi.org/10.3390/jimaging11040126 - 21 Apr 2025

Abstract

Volumetric Path Tracing (VPT) based on Monte Carlo (MC) sampling often requires numerous samples for high-quality images, but real-time applications limit samples to maintain interaction rates, leading to significant noise. Traditional real-time denoising methods use radiance and geometric features as neural network inputs, [...] Read more.

Volumetric Path Tracing (VPT) based on Monte Carlo (MC) sampling often requires numerous samples for high-quality images, but real-time applications limit samples to maintain interaction rates, leading to significant noise. Traditional real-time denoising methods use radiance and geometric features as neural network inputs, but lightweight networks struggle with temporal stability and complex mapping relationships, causing blurry results. To address these issues, a spatiotemporal lightweight neural network is proposed to enhance the denoising performance of VPT-rendered images with low samples per pixel. First, the reprojection technique was employed to obtain features from historical frames. Next, a dual-input convolutional neural network architecture was designed to predict filtering kernels. Radiance and geometric features were encoded independently. The encoding of geometric features guided the pixel-wise fitting of radiance feature filters. Finally, learned weight filtering kernels were applied to images’ spatiotemporal filtering to produce denoised results. The experimental results across multiple denoising datasets demonstrate that this approach outperformed the baseline models in terms of feature extraction and detail representation capabilities while effectively suppressing noise with superior performance and enhanced temporal stability. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

16 pages, 16551 KiB

Open AccessArticle

Camera Pose Generation Based on Unity3D

by Hao Luo, Wenjie Luo and Wenzhu Yang

Information 2025, 16(4), 315; https://doi.org/10.3390/info16040315 - 16 Apr 2025

Viewed by 142

Abstract

Deep learning models performing complex tasks require the support of datasets. With the advancement of virtual reality technology, the use of virtual datasets in deep learning models is becoming more and more widespread. Indoor scenes represents a significant area of interest for the [...] Read more.

Deep learning models performing complex tasks require the support of datasets. With the advancement of virtual reality technology, the use of virtual datasets in deep learning models is becoming more and more widespread. Indoor scenes represents a significant area of interest for the application of machine vision technologies. Existing virtual indoor datasets exhibit deficiencies with regard to camera poses, resulting in problems such as occlusion, object omission, and objects having too small of a proportion of the image, and perform poorly in the training for object detection and simultaneous localization and mapping (SLAM) tasks. Aiming at the problems regarding the capacity of cameras to comprehensively capture scene objects, this study presents an enhanced algorithm based on rapidly exploring random tree star (RRT*) for the generation of camera poses in a 3D indoor scene. Meanwhile, in order to generate multimodal data for various deep learning tasks, this study designs an automatic image acquisition module under the Unity3D platform. The experimental results from running the model on several mainstream virtual indoor datasets—such as 3D-FRONT and Hypersim—indicate that the image sequences generated in this study show enhancements in terms of object capture rate and efficiency. Even in cluttered environments such as those in SceneNet RGB-D, the object capture rate remains stable at around 75%. Compared with the image sequences from the original datasets, those generated in this study achieve improvements in the object detection and SLAM tasks, with increases of up to approximately 30% in mAP for the YOLOv10 object detection task and up to approximately 10% in SR for the ORB-SLAM algorithm. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

21 pages, 1718 KiB

Open AccessArticle

Gaze Estimation Based on a Multi-Stream Adaptive Feature Fusion Network

by Changli Li, Enrui Tong, Kao Zhang, Nenglun Cheng, Zhongyuan Lai and Zhigeng Pan

Appl. Sci. 2025, 15(7), 3684; https://doi.org/10.3390/app15073684 - 27 Mar 2025

Viewed by 253

Abstract

Recently, with the widespread application of deep learning networks, appearance-based gaze estimation has made breakthrough progress. However, most methods focus on feature extraction from the facial region while neglecting the critical role of the eye region in gaze estimation, leading to insufficient eye [...] Read more.

Recently, with the widespread application of deep learning networks, appearance-based gaze estimation has made breakthrough progress. However, most methods focus on feature extraction from the facial region while neglecting the critical role of the eye region in gaze estimation, leading to insufficient eye detail representation. To address this issue, this paper proposes a multi-stream multi-input network architecture (MSMI-Net) based on appearance. The model consists of two independent streams designed to extract high-dimensional eye features and low-dimensional features, integrating both eye and facial information. A parallel channel and spatial attention mechanism is employed to fuse low-dimensional eye and facial features, while an adaptive weight adjustment mechanism (AWAM) dynamically determines the contribution ratio of eye and facial features. The concatenated high-dimensional and fused low-dimensional features are processed through fully connected layers to predict the final gaze direction. Extensive experiments on the EYEDIAP, MPIIFaceGaze, and Gaze360 datasets validate the superiority of the proposed method. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

18 pages, 9766 KiB

Open AccessArticle

MSG-YOLO: A Multi-Scale Dynamically Enhanced Network for the Real-Time Detection of Small Impurities in Large-Volume Parenterals

by Ziqi Li, Dongyao Jia, Zihao He and Nengkai Wu

Electronics 2025, 14(6), 1149; https://doi.org/10.3390/electronics14061149 - 14 Mar 2025

Cited by 1 | Viewed by 381

Abstract

The detection of small targets holds significant application value in the identification of small foreign objects within large-volume parenterals. However, existing methods often face challenges such as inadequate feature expression capabilities, the loss of detailed information, and difficulties in suppressing background interference. To [...] Read more.

The detection of small targets holds significant application value in the identification of small foreign objects within large-volume parenterals. However, existing methods often face challenges such as inadequate feature expression capabilities, the loss of detailed information, and difficulties in suppressing background interference. To tackle the task of the high-speed and high-precision detection of tiny foreign objects in production scenarios involving large infusions, this paper introduces a multi-scale dynamic enhancement network (MSG-YOLO) based on an improved YOLO framework. The primary innovation is the design of a multi-scale dynamic grouped channel enhancement convolution module (MSG-CECM). This module captures multi-scale contextual features through parallel dilated convolutions, enhances the response of critical areas by integrating channel-space joint attention mechanisms, and employs a dynamic grouping strategy for adaptive feature reorganization. In the channel dimension, cross-scale feature fusion and a squeeze-excitation mechanism optimize feature weight distribution; in the spatial dimension, local maximum responses and spatial attention enhance edge details. Furthermore, the module features a lightweight design that reduces computational costs through grouped convolutions. The experiments conducted on our custom large infusion dataset (LVPD) demonstrate that our method improves the mean Average Precision (mAP) by 2.2% compared to the baseline YOLOv9 and increases small target detection accuracy (AP_small) by 3.1% while maintaining a real-time inference speed of 58 FPS. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

25 pages, 104688 KiB

Open AccessArticle

No-Reference Quality Assessment of Infrared Image Colorization with Color–Spatial Features

by Dian Sheng, Weiqi Jin, Xia Wang and Li Li

Electronics 2025, 14(6), 1126; https://doi.org/10.3390/electronics14061126 - 12 Mar 2025

Viewed by 444

Abstract

LDANet represents an innovative no-reference quality assessment model specifically engineered to evaluate colorized infrared images. This is a crucial task for various applications, and existing methods often fail to capture color-specific distortions. The proposed model distinguishes itself by uniquely combining color feature extraction [...] Read more.

LDANet represents an innovative no-reference quality assessment model specifically engineered to evaluate colorized infrared images. This is a crucial task for various applications, and existing methods often fail to capture color-specific distortions. The proposed model distinguishes itself by uniquely combining color feature extraction through latent Dirichlet allocation (LDA) with spatial feature extraction enhanced by multichannel and spatial attention mechanisms. It employs a dual-feature approach that facilitates thorough assessment of both color fidelity and detail preservation in colorized images. The architecture of LDANet encompasses two critical components: an LDA-based color feature extraction module which meticulously analyzes and learns color distribution patterns, and a spatial feature extraction module that leverages an inception network bolstered by attention mechanisms to effectively capture multiscale spatial characteristics. Rigorous experimental validation conducted on a specialized dataset of colorized infrared images demonstrates that LDANet significantly outperforms existing leading no-reference image quality assessment methods. This study reports the effectiveness of integrating color-specific features within a quality assessment framework tailored for infrared image colorization, representing a meaningful advancement in this domain. These findings emphasize the essential role of color feature integration in the evaluation of colorized infrared images, providing a robust tool for optimizing colorization algorithms and enhancing their practical applications. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

18 pages, 8347 KiB

Open AccessArticle

Shallow Subsurface Wavefield Data Interpolation Method Based on Transfer Learning

by Danfeng Zang, Jian Li, Chuankun Li, Hengran Zhang, Zhipeng Pei and Yixiang Ma

Appl. Sci. 2025, 15(4), 1964; https://doi.org/10.3390/app15041964 - 13 Feb 2025

Viewed by 453

Abstract

The deployment density of surface sensors can significantly impact the accuracy of subsurface shallow seismic field energy inversion. With finite budget constraints, it is often not feasible to deploy a large number of sensors, resulting in limited seismic signal acquisition that hinders accurate [...] Read more.

The deployment density of surface sensors can significantly impact the accuracy of subsurface shallow seismic field energy inversion. With finite budget constraints, it is often not feasible to deploy a large number of sensors, resulting in limited seismic signal acquisition that hinders accurate inversion of the shallow subsurface explosions. To address the challenge of insufficient sensor signals needed for inversion, we conducted a study on a subsurface shallow wavefield data interpolation method based on transfer learning. This method is designed to increase the overall signal acquisition by interpolating signals at target locations from limited measurement points. Our research employs neural networks to interpolate real seismic data, supplementing the sampled signals. Given the lack of extensive samples from actual data collection, we devised a training approach that combines numerically simulated signals with real collected signals. Initially, we performed conventional interpolation training using a deep interpolation network with complete synthetic gather images obtained from numerical simulations. Subsequently, the feature extraction part was frozen, and the interpolation network was transferred to real datasets, where it was trained using incomplete gather images. Finally, these incomplete gather images were re-input into the trained network to obtain interpolated results at the target locations. Our study demonstrates the superiority of our method by comparing it with two other interpolation networks and validating the effectiveness of transfer learning through four sets of ablation experiments in the actual test. This method can also be applied to other shallow geological structures to generate a large number of seismic signals for energy inversion. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

22 pages, 5498 KiB

Open AccessArticle

Small-Sample Target Detection Across Domains Based on Supervision and Distillation

by Fusheng Sun, Jianli Jia, Xie Han, Liqun Kuang and Huiyan Han

Electronics 2024, 13(24), 4975; https://doi.org/10.3390/electronics13244975 - 18 Dec 2024

Cited by 1 | Viewed by 679

Abstract

To address the issues of significant object discrepancies, low similarity, and image noise interference between source and target domains in object detection, we propose a supervised learning approach combined with knowledge distillation. Initially, student and teacher models are jointly trained through supervised and [...] Read more.

To address the issues of significant object discrepancies, low similarity, and image noise interference between source and target domains in object detection, we propose a supervised learning approach combined with knowledge distillation. Initially, student and teacher models are jointly trained through supervised and distillation-based approaches, iteratively refining the inter-model weights to mitigate the issue of model overfitting. Secondly, a combined convolutional module is integrated into the feature extraction network of the student model, to minimize redundant computational effort; an explicit visual center module is embedded within the feature pyramid network, to bolster feature representation; and a spatial grouping enhancement module is incorporated into the region proposal network, to mitigate the adverse effects of noise on the outcomes. Ultimately, the model undergoes a comprehensive optimization process that leverages the loss functions originating from both the supervised and knowledge distillation phases. The experimental results demonstrate that this strategy significantly boosts classification and identification accuracy on cross-domain datasets; when compared to the TFA (Task-agnostic Fine-tuning and Adapter), CD-FSOD (Cross-Domain Few-Shot Object Detection) and DeFRCN (Decoupled Faster R-CNN for Few-Shot Object Detection), with sample orders of magnitude 1 and 5, increased the detection accuracy by 1.67% and 1.87%, respectively. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

19 pages, 7944 KiB

Open AccessArticle

Method for Reconstructing Velocity Field Images of the Internal Structures of Bridges Based on Group Sparsity

by Jian Li, Jin Li, Chenli Guo, Hongtao Wu, Chuankun Li, Rui Liu and Lujun Wei

Electronics 2024, 13(22), 4574; https://doi.org/10.3390/electronics13224574 - 20 Nov 2024

Viewed by 682

Abstract

Non-destructive testing (NDT) enables the determination of internal defects and flaws in concrete structures without damaging them, making it a common application in current bridge concrete inspections. However, due to the complexity of the internal structure of this type of concrete, limitations regarding [...] Read more.

Non-destructive testing (NDT) enables the determination of internal defects and flaws in concrete structures without damaging them, making it a common application in current bridge concrete inspections. However, due to the complexity of the internal structure of this type of concrete, limitations regarding measurement point placement, and the extensive detection area, accurate defect detection cannot be guaranteed. This paper proposes a method that combines the Simultaneous Algebraic Reconstruction Technique with Group Sparsity Regularization (SART-GSR) to achieve tomographic imaging of bridge concrete under sparse measurement conditions. Firstly, a mathematical model is established based on the principles of the tomographic imaging of bridge concrete; secondly, the SART algorithm is used to solve for its velocity values; thirdly, on the basis of the SART results, GSR is applied for optimized solution processing; finally, simulation experiments are conducted to verify the reconstruction effects of the SART-GSR algorithm compared with those of the SART and ART algorithms. The results show that the SART-GSR algorithm reduced the relative error to 1.5% and the root mean square error to 89.76 m/s compared to the SART and ART algorithms. This improvement in accuracy makes it valuable for the tomographic imaging of bridge concrete and provides a reference for defect detection in bridge concrete. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

19 pages, 5749 KiB

Open AccessArticle

Video Anomaly Detection Based on Global–Local Convolutional Autoencoder

by Fusheng Sun, Jiahao Zhang, Xiaodong Wu, Zhong Zheng and Xiaowen Yang

Electronics 2024, 13(22), 4415; https://doi.org/10.3390/electronics13224415 - 11 Nov 2024

Viewed by 1029

Abstract

Video anomaly detection (VAD) plays a crucial role in fields such as security, production, and transportation. To address the issue of overgeneralization in anomaly behavior prediction by deep neural networks, we propose a network called AMFCFBMem-Net (appearance and motion feature cross-fusion block memory [...] Read more.

Video anomaly detection (VAD) plays a crucial role in fields such as security, production, and transportation. To address the issue of overgeneralization in anomaly behavior prediction by deep neural networks, we propose a network called AMFCFBMem-Net (appearance and motion feature cross-fusion block memory network), which combines appearance and motion feature cross-fusion blocks. Firstly, dual encoders for appearance and motion are employed to separately extract these features, which are then integrated into the skip connection layer to mitigate the model’s tendency to predict abnormal behavior, ultimately enhancing the prediction accuracy for abnormal samples. Secondly, a motion foreground extraction module is integrated into the network to generate a foreground mask map based on speed differences, thereby widening the prediction error margin between normal and abnormal behaviors. To capture the latent features of various models for normal samples, a memory module is introduced at the bottleneck of the encoder and decoder structures. This further enhances the model’s anomaly detection capabilities and diminishes its predictive generalization towards abnormal samples. The experimental results on the UCSD Pedestrian dataset 2 (UCSD Ped2) and CUHK Avenue anomaly detection dataset (CUHK Avenue) demonstrate that, compared to current cutting-edge video anomaly detection algorithms, our proposed method achieves frame-level AUCs of 97.5% and 88.8%, respectively, effectively enhancing anomaly detection capabilities. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

16 pages, 6488 KiB

Open AccessArticle

3D-CNN Method for Drowsy Driving Detection Based on Driving Pattern Recognition

by Jimin Lee, Soomin Woo and Changjoo Moon

Electronics 2024, 13(17), 3388; https://doi.org/10.3390/electronics13173388 - 26 Aug 2024

Viewed by 1216

Abstract

Drowsiness impairs drivers’ concentration and reaction time, doubling the risk of car accidents. Various methods for detecting drowsy driving have been proposed that rely on facial changes. However, they have poor detection for drivers wearing a mask or sunglasses, and they do not [...] Read more.

Drowsiness impairs drivers’ concentration and reaction time, doubling the risk of car accidents. Various methods for detecting drowsy driving have been proposed that rely on facial changes. However, they have poor detection for drivers wearing a mask or sunglasses, and they do not reflect the driver’s drowsiness habits. Therefore, this paper proposes a novel method to detect drowsy driving even with facial detection obstructions, such as masks or sunglasses, and regardless of the driver’s different drowsiness habits, by recognizing behavioral patterns. We achieve this by constructing both normal driving and drowsy driving datasets and developing a 3D-CNN (3D Convolutional Neural Network) model reflecting the Inception structure of GoogleNet. This binary classification model classifies normal driving and drowsy driving videos. Using actual videos captured inside real vehicles, this model achieved a classification accuracy of 85% for detecting drowsy driving without facial obstructions and 75% for detecting drowsy driving when masks and sunglasses are worn. Our results demonstrate that the behavioral pattern recognition method is effective in detecting drowsy driving. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

17 pages, 6653 KiB

Open AccessArticle

Supervised-Learning-Based Method for Restoring Subsurface Shallow-Layer Q Factor Distribution

by Danfeng Zang, Jian Li, Chuankun Li, Mingxing Ma, Chenli Guo and Jiangang Wang

Electronics 2024, 13(11), 2145; https://doi.org/10.3390/electronics13112145 - 30 May 2024

Cited by 1 | Viewed by 697

Abstract

The distribution of shallow subsurface quality factors (Q) is a crucial indicator for assessing the integrity of subsurface structures and serves as a primary parameter for evaluating the attenuation characteristics of seismic waves propagating through subsurface media. As the complexity of underground spaces [...] Read more.

The distribution of shallow subsurface quality factors (Q) is a crucial indicator for assessing the integrity of subsurface structures and serves as a primary parameter for evaluating the attenuation characteristics of seismic waves propagating through subsurface media. As the complexity of underground spaces increases, regions expand, and testing environments diversify, the survivability of test nodes is compromised, resulting in sparse effective seismic data with a low signal-to-noise ratio (SNR). Within the confined area defined by the source and sensor placement, only the Q factor along the wave propagation path can be estimated with relative accuracy. Estimating the Q factor in other parts of the area is challenging. Additionally, in recent years, deep neural networks have been employed to address the issue of missing values in seismic data; however, these methods typically require large datasets to train networks that can effectively fit the data, making them less applicable to our specific problem. In response to this challenge, we have developed a supervised learning method for the restoration of shallow subsurface Q factor distributions. The process begins with the construction of an incomplete labeled data volume, followed by the application of a block-based data augmentation technique to enrich the training samples and train the network. The uniformly partitioned initial data are then fed into the trained network to obtain output data, which are subsequently combined to form a complete Q factor data volume. We have validated this training approach using various networks, all yielding favorable results. Additionally, we compared our method with a data augmentation approach that involves creating random masks, demonstrating that our method reduces the mean absolute percentage error (MAPE) by 5%. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

14 pages, 5118 KiB

Open AccessArticle

Domain Adaptive Subterranean 3D Pedestrian Detection via Instance Transfer and Confidence Guidance

by Zengyun Liu, Zexun Zheng, Tianyi Qin, Liying Xu and Xu Zhang

Electronics 2024, 13(5), 982; https://doi.org/10.3390/electronics13050982 - 4 Mar 2024

Cited by 3 | Viewed by 1269

Abstract

With the exploration of subterranean scenes, determining how to ensure the safety of subterranean pedestrians has gradually become a hot research topic. Considering the poor illumination and lack of annotated data in subterranean scenes, it is essential to explore the LiDAR-based domain adaptive [...] Read more.

With the exploration of subterranean scenes, determining how to ensure the safety of subterranean pedestrians has gradually become a hot research topic. Considering the poor illumination and lack of annotated data in subterranean scenes, it is essential to explore the LiDAR-based domain adaptive detectors for localizing the spatial location of pedestrians, thus providing instruction for evacuation and rescue. In this paper, a novel domain adaptive subterranean 3D pedestrian detection method is proposed to adapt pre-trained detectors from the annotated road scenes to the unannotated subterranean scenes. Specifically, an instance transfer-based scene updating strategy is designed to update the subterranean scenes by transferring instances from the road scenes to the subterranean scenes, aiming to create sufficient high-quality pseudo labels for fine-tuning the pre-trained detector. In addition, a pseudo label confidence-guided learning mechanism is constructed to fully utilize pseudo labels of different qualities under the guidance of confidence scores. Extensive experiments validate the superiority of our proposed domain adaptive subterranean 3D pedestrian detection method. Full article

(This article belongs to the Topic Visual Computing and Understanding: New Developments and Trends)

► Show Figures

Figure 1

Topic Menu

Topic Editors

Visual Computing and Understanding: New Developments and Trends

Topic Information

Keywords

Participating Journals

Published Papers (13 papers)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI