Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.5
Journals
Applied Sciences
Special Issues
AI-Based Image Processing: 2nd Edition
applsci-logo
Submit to Applied Sciences Review for Applied Sciences Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

AI-Based Image Processing: 2nd Edition

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (30 April 2024) | Viewed by 25685

Special Issue Editors

Dr. Jianping Luo

E-Mail Website
Guest Editor
Guangdong Key Laboratory of Intelligent Information Processing, College of Electronics and Information Engineering, Shenzhen University, Shenzhen 518118, China
Interests: multi-objective evolutionary algorithm; evolutionary multiobjective optimization; pareto front
Prof. Dr. Hongying Meng

E-Mail Website
Guest Editor
Electronic and Electrical Engineering, Brunel University London, Uxbridge UB8 3PH, UK
Interests: image processing; artificial intelligence; signal processing; affective computing
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

With the rapid development of artificial intelligence, a series of artificial intelligence algorithms have been proposed, and intelligent applications have started to play an increasingly important role in industrial production and our social lives. The intersections of AI in medical image processing, image processing for intelligent transportation systems, satellite image processing, face recognition, object recognition, etc. are still challenging.

Therefore, a Special Issue on “AI-based Image Processing: 2nd Edition”, focusing on tackling the most pressing problems faced by the world is organized, and more specifically, in the topics specified below:

  • Image processing algorithms;
  • Image analytics;
  • Medial image processing;
  • Biomedical image analysis;
  • Image generation;
  • Image restoration and enhancement;
  • Image compression;
  • Edge detection;
  • Image segmentation;
  • Semantic segmentation;
  • Image classification;
  • Image inpainting;
  • Image captioning;
  • Feature detection and extraction;
  • Content-based image retrieval;
  • Optical character recognition;
  • Face recognition;
  • Emotion recognition;
  • Gesture recognition;
  • Object recognition and tracking.

Dr. Jianping Luo
Dr. Hongying Meng
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • image processing
  • feature detection and extraction
  • object recognition and tracking

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • Reprint: MDPI Books provides the opportunity to republish successful Special Issues in book format, both online and in print.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (15 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

22 pages, 15185 KiB  
Article
Low Tensor Rank Constrained Image Inpainting Using a Novel Arrangement Scheme
by Shuli Ma, Youchen Fan, Shengliang Fang, Weichao Yang and Li Li
Appl. Sci. 2025, 15(1), 322; https://doi.org/10.3390/app15010322 - 31 Dec 2024
Viewed by 722
Abstract
Employing low tensor rank decomposition in image inpainting has attracted increasing attention. This study exploited novel tensor arrangement schemes to transform an image (a low-order tensor) to a higher-order tensor without changing the total number of pixels. The developed arrangement schemes enhanced the [...] Read more.
Employing low tensor rank decomposition in image inpainting has attracted increasing attention. This study exploited novel tensor arrangement schemes to transform an image (a low-order tensor) to a higher-order tensor without changing the total number of pixels. The developed arrangement schemes enhanced the low rankness of images under three tensor decomposition methods: matrix SVD, tensor train (TT) decomposition, and tensor singular value decomposition (t-SVD). By exploiting the schemes, we solved the image inpainting problem with three low-rank constrained models that use the matrix rank, TT rank, and tubal rank as constrained priors. The tensor tubal rank and tensor train multi-rank were developed from t-SVD and TT decomposition, respectively. Then, ADMM algorithms were efficiently exploited for solving the three models. Experimental results demonstrate that our methods are effective for image inpainting and superior to numerous close methods. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

16 pages, 5589 KiB  
Article
Context-Encoder-Based Image Inpainting for Ancient Chinese Silk
by Quan Wang, Shanshan He, Miao Su and Feng Zhao
Appl. Sci. 2024, 14(15), 6607; https://doi.org/10.3390/app14156607 - 28 Jul 2024
Cited by 1 | Viewed by 1530
Abstract
The rapid advancement of deep learning technologies presents novel opportunities for restoring damaged patterns in ancient silk, which is pivotal for the preservation and propagation of ancient silk culture. This study systematically scrutinizes the evolutionary trajectory of image inpainting algorithms, with a particular [...] Read more.
The rapid advancement of deep learning technologies presents novel opportunities for restoring damaged patterns in ancient silk, which is pivotal for the preservation and propagation of ancient silk culture. This study systematically scrutinizes the evolutionary trajectory of image inpainting algorithms, with a particular emphasis on those firmly rooted in the Context-Encoder structure. To achieve this study’s objectives, a meticulously curated dataset comprising 6996 samples of ancient Chinese silk (256 × 256 pixels) was employed. Context-Encoder-based image inpainting models—LISK, MADF, and MEDFE—were employed to inpaint damaged patterns. The ensuing restoration effects underwent rigorous evaluation, providing a comprehensive analysis of the inherent strengths and limitations of each model. This study not only provides a theoretical foundation for adopting image restoration algorithms grounded in the Context-Encoder structure but also offers ample scope for exploration in achieving more effective restorations of ancient damaged silk. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

18 pages, 9908 KiB  
Article
The Prediction of Incremental Damage on Optics from the Final Optic Assembly in an ICF High-Power Laser Facility
by Xueyan Hu, Wei Zhou, Huaiwen Guo, Xiaoxia Huang, Bowang Zhao, Wei Zhong, Qihua Zhu and Zhifei Chen
Appl. Sci. 2024, 14(12), 5226; https://doi.org/10.3390/app14125226 - 17 Jun 2024
Cited by 1 | Viewed by 940
Abstract
High-power laser facilities necessitate predicting incremental damage to final optics to identify evolving damage trends. In this study, we propose a surface damage detection method utilizing image segmentation employing ResNet-18 and a damage area estimation network employing U-Net++. Paired sets of online and [...] Read more.
High-power laser facilities necessitate predicting incremental damage to final optics to identify evolving damage trends. In this study, we propose a surface damage detection method utilizing image segmentation employing ResNet-18 and a damage area estimation network employing U-Net++. Paired sets of online and offline images of optics obtained from a large laser facility are used to train the network. The trends of varying damage could be identified by incorporating additional experimental parameters. A key advantage of the proposed method is that the network can be trained end to end on small samples, eliminating the need for manual labeling or feature extraction. The software developed based on these models can facilitate the daily inspection and maintenance of optics in large laser facilities. By effectively applying deep learning techniques, we successfully addressed the challenges faced by traditional methods in handling complex environments, achieving the accurate identification and prediction of damages on optics. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

22 pages, 10900 KiB  
Article
Removal of Color-Document Image Show-Through Based on Self-Supervised Learning
by Mengying Ni, Zongbao Liang and Jindong Xu
Appl. Sci. 2024, 14(11), 4568; https://doi.org/10.3390/app14114568 - 26 May 2024
Cited by 1 | Viewed by 1446
Abstract
Show-through phenomena have always been a challenging issue in color-document image processing, which is widely used in various fields such as finance, education, and administration. Existing methods for processing color-document images face challenges, including dealing with double-sided documents with show-through effects, accurately distinguishing [...] Read more.
Show-through phenomena have always been a challenging issue in color-document image processing, which is widely used in various fields such as finance, education, and administration. Existing methods for processing color-document images face challenges, including dealing with double-sided documents with show-through effects, accurately distinguishing between foreground and show-through parts, and addressing the issue of insufficient real image data for supervised training. To overcome these challenges, this paper proposes a self-supervised-learning-based method for removing show-through effects in color-document images. The proposed method utilizes a two-stage-structured show-through-removal network that incorporates a double-cycle consistency loss and a pseudo-similarity loss to effectively constrain the process of show-through removal. Moreover, we constructed two datasets consisting of different show-through mixing ratios and conducted extensive experiments to verify the effectiveness of the proposed method. Experimental results demonstrate that the proposed method achieves competitive performance compared to state-of-the-art methods and can effectively perform show-through removal without the need for paired datasets. Specifically, the proposed method achieves an average PSNR of 33.85 dB on our datasets, outperforming comparable methods by a margin of 0.89 dB. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

20 pages, 11168 KiB  
Article
Syn2Real Detection in the Sky: Generation and Adaptation of Synthetic Aerial Ship Images
by Yaoyuan Wu, Weijie Guo, Zhuoyue Tan, Yifei Zhao, Quanxing Zhu, Liaoni Wu and Zhiming Guo
Appl. Sci. 2024, 14(11), 4558; https://doi.org/10.3390/app14114558 - 25 May 2024
Viewed by 1473
Abstract
Object detection in computer vision requires a sufficient amount of training data to produce an accurate and general model. However, aerial images are difficult to acquire, so the collection of aerial image datasets is a priority issue. Building on the existing research on [...] Read more.
Object detection in computer vision requires a sufficient amount of training data to produce an accurate and general model. However, aerial images are difficult to acquire, so the collection of aerial image datasets is a priority issue. Building on the existing research on image generation, the goal of this work is to create synthetic aerial image datasets that can be used to solve the problem of insufficient data. We generated three independent datasets for ship detection using engine and generative model. These synthetic datasets are rich in virtual scenes, ship categories, weather conditions, and other features. Moreover, we implemented domain-adaptive algorithms to address the issue of domain shift from synthetic data to real data. To investigate the application of synthetic datasets, we validated the synthetic data using six different object detection algorithms and three existing real-world, ship detection datasets. The experimental results demonstrate that the methods for generating synthetic aerial image datasets can complete the insufficient data in aerial remote sensing. Additionally, domain-adaptive algorithms could further mitigate the discrepancy from synthetic data to real data, highlighting the potential and value of synthetic data in aerial image recognition and comprehension tasks in the real world. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

17 pages, 14564 KiB  
Article
P D N: A Priori Dictionary Network for Fashion Parsing
by Jue Hou, Yinwen Lu, Yang Yang and Zheng Liu
Appl. Sci. 2024, 14(8), 3509; https://doi.org/10.3390/app14083509 - 22 Apr 2024
Viewed by 1302
Abstract
The task of fashion parsing aims to assign pixel-level labels to clothing targets; thereby, parsing models are required to have good contextual recognition ability. However, the shapes of clothing components are complex, and the types are difficult to distinguish. Recent solutions focus on [...] Read more.
The task of fashion parsing aims to assign pixel-level labels to clothing targets; thereby, parsing models are required to have good contextual recognition ability. However, the shapes of clothing components are complex, and the types are difficult to distinguish. Recent solutions focus on improving datasets and supplying abundant priori information, but the utilization of features by more efficient methods is rarely explored. In this paper, we propose a multi-scale fashion parsing model called the Priori Dictionary Network (PDN), which includes a priori attention module and a multi-scale backbone. The priori attention module extracts high dimensional features from our designed clothing average template as a priori information dictionary (priori dictionary, PD), and the PD is utilized to activate the feature maps of a CNN from a multi-scale attention mechanism. The backbone is derived from classical models, and five side paths are designed to leverage the richer features of local and global contextual representations. To measure the performance of our method, we evaluated the model on four public datasets, the CFPD, UTFR-SBD3, ModaNet and LIP, and the experimental results show that our model stands out from other State of the Art in all four datasets. This method can assist with the labeling problem of clothing datasets. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

22 pages, 33607 KiB  
Article
Holoscopic Elemental-Image-Based Disparity Estimation Using Multi-Scale, Multi-Window Semi-Global Block Matching
by Bodor Almatrouk, Hongying Meng and Mohammad Rafiq Swash
Appl. Sci. 2024, 14(8), 3335; https://doi.org/10.3390/app14083335 - 15 Apr 2024
Cited by 1 | Viewed by 1302
Abstract
In Holoscopic imaging, a single aperture is used to acquire full-colour spatial images like a fly’s eye by gently altering angles between nearby lenses with a micro-lens array. Due to its simple data collection and visualisation methods, which provide robust and scalable spatial [...] Read more.
In Holoscopic imaging, a single aperture is used to acquire full-colour spatial images like a fly’s eye by gently altering angles between nearby lenses with a micro-lens array. Due to its simple data collection and visualisation methods, which provide robust and scalable spatial information, and its motion parallax, binocular disparity, and convergence, this technique may be able to overcome traditional 2D imaging issues like depth, scalability, and multi-perspective problems. A novel disparity-map-generating method uses angular information from a single Holoscopic image’s micro-images, or Elemental Images (EIs), to create a scene’s disparity map. Not much research has used EIs instead of Viewpoint Images (VPIs) for disparity estimation. This study investigates whether angular perspective data may replace spatial orthographic data. Using noise reduction and contrast enhancement, EIs with a low resolution and lack of texture are pre-processed to calculate the disparity. The Semi-Global Block Matching (SGBM) technique is used to calculate the disparity between EI pixels. A multi-resolution approach overcomes EIs’ resolution constraints, and a content-aware analysis dynamically modifies the SGBM window size settings to generate disparities across different texture and complexity levels. A background mask and nearby EIs with accurate backgrounds detect and rectify EIs with erroneous backgrounds. Our method generates disparity maps that outperform two state-of-the-art deep learning algorithms and VPIs in real images. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

17 pages, 6014 KiB  
Article
Rectification for Stitched Images with Deformable Meshes and Residual Networks
by Yingbo Fan, Shanjun Mao, Mei Li, Zheng Wu, Jitong Kang and Ben Li
Appl. Sci. 2024, 14(7), 2821; https://doi.org/10.3390/app14072821 - 27 Mar 2024
Viewed by 1319
Abstract
Image stitching is an important method for digital image processing, which is often prone to the problem of the irregularity of stitched images after stitching. And the traditional image cropping or complementation methods usually lead to a large number of information loss. Therefore, [...] Read more.
Image stitching is an important method for digital image processing, which is often prone to the problem of the irregularity of stitched images after stitching. And the traditional image cropping or complementation methods usually lead to a large number of information loss. Therefore, this paper proposes an image rectification method based on deformable mesh and residual network. The method aims to minimize the information loss at the edges of the spliced image and the information loss inside the image. Specifically, the method can select the most suitable mesh shape for residual network regression according to different images. Its loss function includes global loss and local loss, aiming to minimize the loss of image information within the grid and global target. The method in this paper not only greatly reduces the information loss caused by irregular shapes after image stitching, but also adapts to different images with various rigid structures. Meanwhile, its validation on the DIR-D dataset shows that the method outperforms the state-of-the-art methods in image rectification. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

12 pages, 3775 KiB  
Article
Multi-Identity Recognition of Darknet Vendors Based on Metric Learning
by Yilei Wang, Yuelin Hu, Wenliang Xu and Futai Zou
Appl. Sci. 2024, 14(4), 1619; https://doi.org/10.3390/app14041619 - 17 Feb 2024
Viewed by 1660
Abstract
Dark web vendor identification can be seen as an authorship aliasing problem, aiming to determine whether different accounts on different markets belong to the same real-world vendor, in order to locate cybercriminals involved in dark web market transactions. Existing open-source datasets for dark [...] Read more.
Dark web vendor identification can be seen as an authorship aliasing problem, aiming to determine whether different accounts on different markets belong to the same real-world vendor, in order to locate cybercriminals involved in dark web market transactions. Existing open-source datasets for dark web marketplaces are outdated and cannot simulate real-world situations, while data labeling methods are difficult and suffer from issues such as inaccurate labeling and limited cross-market research. The problem of identifying vendors’ multiple identities on the dark web involves a large number of categories and a limited number of samples, making it difficult to use traditional multiclass classification models. To address these issues, this paper proposes a metric learning-based method for dark web vendor identification, collecting product data from 21 currently active English dark web marketplaces and using a multi-dimensional feature extraction method based on product titles, descriptions, and images. Using pseudo-labeling technology combined with manual labeling improves data labeling accuracy compared to previous labeling methods. The proposed method uses a Siamese neural network with metric learning to learn the similarity between vendors and achieve the recognition of vendors’ multiple identities. This method achieved better performance with an average F1-score of 0.889 and an accuracy rate of 97.535% on the constructed dataset. The contributions of this paper lie in the proposed method for collecting and labeling data for dark web marketplaces and overcoming the limitations of traditional multiclass classifiers to achieve effective recognition of vendors’ multiple identities. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

20 pages, 24378 KiB  
Article
Diffusion-Denoising Process with Gated U-Net for High-Quality Document Binarization
by Sangkwon Han, Seungbin Ji and Jongtae Rhee
Appl. Sci. 2023, 13(20), 11141; https://doi.org/10.3390/app132011141 - 10 Oct 2023
Cited by 1 | Viewed by 2469
Abstract
The binarization of degraded documents represents a crucial preprocessing task for various document analyses, including optical character recognition and historical document analysis. Various convolutional neural network models and generative models have been used for document binarization. However, these models often struggle to deliver [...] Read more.
The binarization of degraded documents represents a crucial preprocessing task for various document analyses, including optical character recognition and historical document analysis. Various convolutional neural network models and generative models have been used for document binarization. However, these models often struggle to deliver generalized performance on noise types the model has not encountered during training and may have difficulty extracting intricate text strokes. We herein propose a novel approach to address these challenges by introducing the use of the latent diffusion model, a well-known high-quality image-generation model, into the realm of document binarization for the first time. By leveraging an iterative diffusion-denoising process within the latent space, our approach excels at producing high-quality, clean, binarized images and demonstrates excellent generalization using both data distribution and time steps during training. Furthermore, we enhance our model’s ability to preserve text strokes by incorporating a gated U-Net into the backbone network. The gated convolution mechanism allows the model to focus on the text region by combining gating values and features, facilitating the extraction of intricate text strokes. To maximize the effectiveness of our proposed model, we use a combination of the latent diffusion model loss and pixel-level loss, which aligns with the model’s structure. The experimental results on the Handwritten Document Image Binarization Contest and Document Image Binarization Contest benchmark datasets showcase the superior performance of our proposed model compared to existing methods. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

21 pages, 5457 KiB  
Article
Research on Students’ Action Behavior Recognition Method Based on Classroom Time-Series Images
by Zhaoyu Shou, Mingbang Yan, Hui Wen, Jinghua Liu, Jianwen Mo and Huibing Zhang
Appl. Sci. 2023, 13(18), 10426; https://doi.org/10.3390/app131810426 - 18 Sep 2023
Cited by 5 | Viewed by 2301
Abstract
Students’ action behavior performance is an important part of classroom teaching evaluation. To detect the action behavior of students in classroom teaching videos, and based on the detection results, the action behavior sequence of individual students in the teaching time of knowledge points [...] Read more.
Students’ action behavior performance is an important part of classroom teaching evaluation. To detect the action behavior of students in classroom teaching videos, and based on the detection results, the action behavior sequence of individual students in the teaching time of knowledge points is obtained and analyzed. This paper proposes a method for recognizing students’ action behaviors based on classroom time-series images. First, we propose an improved Asynchronous Interaction Aggregation (AIA) network for student action behavior detection. By adding a Multi-scale Temporal Attention (MsTA) module and a Multi-scale Channel Spatial Attention (MsCSA) module to the fast pathway and slow pathway, respectively, the accuracy of student action behavior recognition is improved in SlowFast, which is the backbone network of the improved AIA network,. Second, the Equalized Focal Loss function is introduced to improve the category imbalance that exists in the student action behavior dataset. Experimental results on the student action behavior dataset show that the method proposed in this paper can detect different action behaviors of students in the classroom and has better detection performance compared to the original AIA network. Finally, based on the results of action behavior recognition, the seat number is used as the index to obtain the action behavior sequence of individual students during the teaching time of knowledge points and the performance of students in this period is analyzed. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

18 pages, 5975 KiB  
Article
Intelligent Simulation Technology Based on RCS Imaging
by Jiaxing Hao, Xuetian Wang, Sen Yang and Hongmin Gao
Appl. Sci. 2023, 13(18), 10119; https://doi.org/10.3390/app131810119 - 8 Sep 2023
Viewed by 1441
Abstract
The target simulation of airplanes is an important research topic. It is particularly important to find the right balance between high performance and low cost. In order to balance the contradictions between realistic target simulations and controllable costs, the scientific formulation of the [...] Read more.
The target simulation of airplanes is an important research topic. It is particularly important to find the right balance between high performance and low cost. In order to balance the contradictions between realistic target simulations and controllable costs, the scientific formulation of the performance parameters of target simulation is the key to achieving high performance. This paper proposes an intelligent simulation technology based on RCS imaging simulation through the combination of 60° variation corner reflector and a Luneberg lens reflector. It is designed to simulate several important RCS characteristics of the aircraft. At the same time, the different RCS images are automatically shifted to the corresponding gear position to achieve the purpose of simulation, and the price is low and the performance is good. It can be used for the training of radar target searching. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

22 pages, 4219 KiB  
Article
MixerNet-SAGA A Novel Deep Learning Architecture for Superior Road Extraction in High-Resolution Remote Sensing Imagery
by Wei Wu, Chao Ren, Anchao Yin and Xudong Zhang
Appl. Sci. 2023, 13(18), 10067; https://doi.org/10.3390/app131810067 - 6 Sep 2023
Cited by 4 | Viewed by 1533
Abstract
In this study, we address the limitations of current deep learning models in road extraction tasks from remote sensing imagery. We introduce MixerNet-SAGA, a novel deep learning model that incorporates the strengths of U-Net, integrates a ConvMixer block for enhanced feature extraction, and [...] Read more.
In this study, we address the limitations of current deep learning models in road extraction tasks from remote sensing imagery. We introduce MixerNet-SAGA, a novel deep learning model that incorporates the strengths of U-Net, integrates a ConvMixer block for enhanced feature extraction, and includes a Scaled Attention Gate (SAG) for augmented spatial attention. Experimental validation on the Massachusetts road dataset and the DeepGlobe road dataset demonstrates that MixerNet-SAGA achieves a 10% improvement in precision, 8% in recall, and 12% in IoU compared to leading models such as U-Net, ResNet, and SDUNet. Furthermore, our model excels in computational efficiency, being 20% faster, and has a smaller model size. Notably, MixerNet-SAGA shows exceptional robustness against challenges such as same-spectrum–different-object and different-spectrum–same-object phenomena. Ablation studies further reveal the critical roles of the ConvMixer block and SAG. Despite its strengths, the model’s scalability to extremely large datasets remains an area for future investigation. Collectively, MixerNet-SAGA offers an efficient and accurate solution for road extraction in remote sensing imagery and presents significant potential for broader applications. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

17 pages, 5445 KiB  
Article
RepNet: A Lightweight Human Pose Regression Network Based on Re-Parameterization
by Xinjing Zhang and Qixun Zhou
Appl. Sci. 2023, 13(16), 9475; https://doi.org/10.3390/app13169475 - 21 Aug 2023
Viewed by 1915
Abstract
Human pose estimation, as the basis of advanced computer vision, has a wide application perspective. In existing studies, the high-capacity model based on the heatmap method can achieve accurate recognition results, but it encounters many difficulties when used in real-world scenarios. To solve [...] Read more.
Human pose estimation, as the basis of advanced computer vision, has a wide application perspective. In existing studies, the high-capacity model based on the heatmap method can achieve accurate recognition results, but it encounters many difficulties when used in real-world scenarios. To solve this problem, we propose a lightweight pose regression algorithm (RepNet) that introduces a multi-parameter network structure, fuses multi-level features, and combines the idea of residual likelihood estimation. A well-designed convolutional architecture is used for training. By reconstructing the parameters of each level, the network model is simplified, and the computation time and efficiency of the detection task are optimized. The prediction performance is also improved by the output of the maximum likelihood model and the reversible transformation of the underlying distribution learned by the flow generation model. RepNet achieves a recognition accuracy of 66.1 AP on the COCO dataset, at a computational speed of 15 ms on GPU and 40 ms on CPU. This resolves the contradiction between prediction accuracy and computational complexity and contributes to research in lightweight pose estimation. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

19 pages, 35842 KiB  
Article
Real-Time Intelligent Detection System for Illegal Wearing of On-Site Power Construction Worker Based on Edge-YOLO and Low-Cost Edge Devices
by Rong Chang, Bangyuan Li, Junpeng Dang, Chuanxu Yang, Anning Pan and Yang Yang
Appl. Sci. 2023, 13(14), 8287; https://doi.org/10.3390/app13148287 - 18 Jul 2023
Cited by 4 | Viewed by 2168
Abstract
Ensuring personal safety and preventing accidents are critical aspects of power construction safety supervision. However, current monitoring methods are inefficient and unreliable as most of them rely on manual monitoring and transmission, which results in slow detection and delayed warnings regarding violations. To [...] Read more.
Ensuring personal safety and preventing accidents are critical aspects of power construction safety supervision. However, current monitoring methods are inefficient and unreliable as most of them rely on manual monitoring and transmission, which results in slow detection and delayed warnings regarding violations. To overcome these challenges, we propose an intelligent detection system that can accurately identify instances of illegal wearing of power construction workers in real-time. Firstly, we integrated the squeeze-and-excitation (SE) module into our convolutional neural network to enhance detection accuracy. This module effectively prioritizes informative features while suppressing less relevant ones, resulting in improved overall performance. Secondly, we present an embedded real-time detection system that utilizes Jetson Xavier NX and Edge-YOLO. This system promptly detects and alerts power construction workers of instances of illegal wearing behavior. To ensure a lightweight implementation, we design appropriate detection heads based on target size and distribution, reducing model parameters while enhancing detection speed and minimizing accuracy loss. Additionally, we employed data augmentation to enhance the system’s robustness. Our experimental results demonstrate that our improved Edge-YOLO model achieves high detection precision and recall rates of 0.964 and 0.966, respectively, with a frame rate of 35.36 frames per second when deployed on Jetson Xavier NX. Therefore, Edge-YOLO proves to be an ideal choice for intelligent real-time detection systems, providing superior accuracy and speed performance compared to the original YOLOv5s model and other models in the YOLO series for safety monitoring at construction sites. Full article
(This article belongs to the Special Issue AI-Based Image Processing: 2nd Edition)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-15
Back to TopTop