Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
6.8
3.9
Journals
Sensors
Sections
Visual Sensors
sensors-logo
Submit to Topical Collection Review for Sensors

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Visual Sensors

A topical collection in Sensors (ISSN 1424-8220). This collection belongs to the section "Physical Sensors".

Viewed by 10144

Editors

Prof. Dr. Oscar Reinoso Garcia

E-Mail Website
Collection Editor
Department of Systems and Automation Engineering, Universidad Miguel Hernández, Avinguda de la Universitat d'Elx, s/n, 03202 Elche, Alicante, Spain
Interests: computer vision; robotics; cooperative robotics
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Luis Payá

E-Mail Website
Collection Editor
System Engineering and Automation Department, Miguel Hernandez University, 03202 Elche, Spain
Interests: computer vision; omnidirectional imaging; appearance descriptors; image processing; mobile robotics; environment modeling; visual localization
Special Issues, Collections and Topics in MDPI journals

Topical Collection Information

Dear Colleagues,

Visual sensors are able to capture a large quantity of information from the environment around them. Nowadays, a wide variety of visual systems can be found, from the classical monocular systems to omnidirectional, RGB-D and more sophisticated 3D systems. Every configuration presents some specific characteristics that make them useful to solve different problems, with a unique vision system or a network of them, or even fusing the visual information with other sources of information. The range of applications of visual sensors is wide and varied. Amongst them, we can find robotics, industry, security, medicine, agriculture, quality control, visual inspection, surveillance, autonomous driving and navigation aid systems.

The visual information can be processed using a variety of approaches to extract relevant and useful information. Among them, machine learning and deep learning have experienced a great development over the past few years and have provided robust solutions to complex problems. The aim of this collection is to collect new research works and developments that present some of the possibilities that vision systems offer, focusing on the different configurations that can be used, new image processing algorithms and novel applications in any field. Furthermore, reviews presenting a deep analysis of a specific problem and the use of vision systems to address it would also be appropriate.

This Topical Collection invites contributions in the following topics (but is not limited to them):

  • Image processing;
  • Visual pattern recognition;
  • Object recognition by visual sensors;
  • Movement estimation or registration from images;
  • Visual sensors in robotics;
  • Visual sensors in industrial applications;
  • Computer vision for quality evaluation;
  • Visual sensors in agriculture;
  • Computer vision in medical applications;
  • Computer vision in autonomous or computer-aided driving;
  • Environment modeling and reconstruction from images;
  • Visual Localization;
  • Visual SLAM;
  • Deep learning from images.

Prof. Dr. Oscar Reinoso Garcia
Prof. Dr. Luis Payá
Collection 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 collection 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. Sensors 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 2600 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

  • 3D imaging
  • Stereo visual systems
  • Omnidirectional visual systems
  • Quality assessment
  • Pattern recognition
  • Visual registration
  • Visual navigation
  • Visual mapping
  • LiDAR/vision system
  • Multi-visual sensors
  • RGB-D cameras
  • Fusion of visual information
  • Fusion with other sources of information
  • Networks of visual sensors
  • Machine learning
  • Deep learning

Published Papers (4 papers)

Download All Papers

2023

Jump to: 2022

17 pages, 16040 KiB  
Article
GammaGAN: Gamma-Scaled Class Embeddings for Conditional Video Generation
by Minjae Kang and Yong Seok Heo
Sensors 2023, 23(19), 8103; https://doi.org/10.3390/s23198103 - 27 Sep 2023
Viewed by 849
Abstract
In this paper, we propose a new model for conditional video generation (GammaGAN). Generally, it is challenging to generate a plausible video from a single image with a class label as a condition. Traditional methods based on conditional generative adversarial networks (cGANs) often [...] Read more.
In this paper, we propose a new model for conditional video generation (GammaGAN). Generally, it is challenging to generate a plausible video from a single image with a class label as a condition. Traditional methods based on conditional generative adversarial networks (cGANs) often encounter difficulties in effectively utilizing a class label, typically by concatenating a class label to the input or hidden layer. In contrast, the proposed GammaGAN adopts the projection method to effectively utilize a class label and proposes scaling class embeddings and normalizing outputs. Concretely, our proposed architecture consists of two streams: a class embedding stream and a data stream. In the class embedding stream, class embeddings are scaled to effectively emphasize class-specific differences. Meanwhile, the outputs in the data stream are normalized. Our normalization technique balances the outputs of both streams, ensuring a balance between the importance of feature vectors and class embeddings during training. This results in enhanced video quality. We evaluated the proposed method using the MUG facial expression dataset, which consists of six facial expressions. Compared with the prior conditional video generation model, ImaGINator, our model yielded relative improvements of 1.61%, 1.66%, and 0.36% in terms of PSNR, SSIM, and LPIPS, respectively. These results suggest potential for further advancements in conditional video generation. Full article
►▼ Show Figures

Figure 1

22 pages, 11325 KiB  
Article
Automated Video-Based Capture of Crustacean Fisheries Data Using Low-Power Hardware
by Sebastian Gregory Dal Toé, Marie Neal, Natalie Hold, Charlotte Heney, Rebecca Turner, Emer McCoy, Muhammad Iftikhar and Bernard Tiddeman
Sensors 2023, 23(18), 7897; https://doi.org/10.3390/s23187897 - 15 Sep 2023
Cited by 1 | Viewed by 853
Abstract
This work investigates the application of Computer Vision to the problem of the automated counting and measuring of crabs and lobsters onboard fishing boats. The aim is to provide catch count and measurement data for these key commercial crustacean species. This can provide [...] Read more.
This work investigates the application of Computer Vision to the problem of the automated counting and measuring of crabs and lobsters onboard fishing boats. The aim is to provide catch count and measurement data for these key commercial crustacean species. This can provide vital input data for stock assessment models, to enable the sustainable management of these species. The hardware system is required to be low-cost, have low-power usage, be waterproof, available (given current chip shortages), and able to avoid over-heating. The selected hardware is based on a Raspberry Pi 3A+ contained in a custom waterproof housing. This hardware places challenging limitations on the options for processing the incoming video, with many popular deep learning frameworks (even light-weight versions) unable to load or run given the limited computational resources. The problem can be broken into several steps: (1) Identifying the portions of the video that contain each individual animal; (2) Selecting a set of representative frames for each animal, e.g, lobsters must be viewed from the top and underside; (3) Detecting the animal within the frame so that the image can be cropped to the region of interest; (4) Detecting keypoints on each animal; and (5) Inferring measurements from the keypoint data. In this work, we develop a pipeline that addresses these steps, including a key novel solution to frame selection in video streams that uses classification, temporal segmentation, smoothing techniques and frame quality estimation. The developed pipeline is able to operate on the target low-power hardware and the experiments show that, given sufficient training data, reasonable performance is achieved. Full article
►▼ Show Figures

Figure 1

2022

Jump to: 2023

23 pages, 12581 KiB  
Article
An Online Rail Track Fastener Classification System Based on YOLO Models
by Chen-Chiung Hsieh, Ti-Yun Hsu and Wei-Hsin Huang
Sensors 2022, 22(24), 9970; https://doi.org/10.3390/s22249970 - 17 Dec 2022
Cited by 7 | Viewed by 2874
Abstract
In order to save manpower on rail track inspection, computer vision-based methodologies are developed. We propose utilizing the YOLOv4-Tiny neural network to identify track defects in real time. There are ten defects covering fasteners, rail surfaces, and sleepers from the upward and six [...] Read more.
In order to save manpower on rail track inspection, computer vision-based methodologies are developed. We propose utilizing the YOLOv4-Tiny neural network to identify track defects in real time. There are ten defects covering fasteners, rail surfaces, and sleepers from the upward and six defects about the rail waist from the sideward. The proposed real-time inspection system includes a high-performance notebook, two sports cameras, and three parallel processes. The hardware is mounted on a flat cart running at 30 km/h. The inspection results about the abnormal track components could be queried by defective type, time, and the rail hectometer stake. In the experiments, data augmentation by a Cycle Generative Adversarial Network (GAN) is used to increase the dataset. The number of images is 3800 on the upward and 967 on the sideward. Five object detection neural network models—YOLOv4, YOLOv4-Tiny, YOLOX-Tiny, SSD512, and SSD300—were tested. The YOLOv4-Tiny model with 150 FPS is selected as the recognition kernel, as it achieved 91.7%, 92%, and 91% for the mAP, precision, and recall of the defective track components from the upward, respectively. The mAP, precision, and recall of the defective track components from the sideward are 99.16%, 96%, and 94%, respectively. Full article
►▼ Show Figures

Figure 1

26 pages, 10295 KiB  
Article
Interpretation and Transformation of Intrinsic Camera Parameters Used in Photogrammetry and Computer Vision
by Kuan-Ying Lin, Yi-Hsing Tseng and Kai-Wei Chiang
Sensors 2022, 22(24), 9602; https://doi.org/10.3390/s22249602 - 07 Dec 2022
Cited by 5 | Viewed by 4690
Abstract
The precision modelling of intrinsic camera geometry is a common issue in the fields of photogrammetry (PH) and computer vision (CV). However, in both fields, intrinsic camera geometry has been modelled differently, which has led researchers to adopt different definitions of intrinsic camera [...] Read more.
The precision modelling of intrinsic camera geometry is a common issue in the fields of photogrammetry (PH) and computer vision (CV). However, in both fields, intrinsic camera geometry has been modelled differently, which has led researchers to adopt different definitions of intrinsic camera parameters (ICPs), including focal length, principal point, radial distortion, decentring distortion, affinity and shear. These ICPs are indispensable for vision-based measurements. These differences can confuse researchers from one field when using ICPs obtained from a camera calibration software package developed in another field. This paper clarifies the ICP definitions used in each field and proposes an ICP transformation algorithm. The originality of this study lies in its use of least-squares adjustment, applying the image points involving ICPs defined in PH and CV image frames to convert a complete set of ICPs. This ICP transformation method is more rigorous than the simplified formulas used in conventional methods. Selecting suitable image points can increase the accuracy of the generated adjustment model. In addition, the proposed ICP transformation method enables users to apply mixed software in the fields of PH and CV. To validate the transformation algorithm, two cameras with different view angles were calibrated using typical camera calibration software packages applied in each field to obtain ICPs. Experimental results demonstrate that our proposed transformation algorithm can be used to convert ICPs derived from different software packages. Both the PH-to-CV and CV-to-PH transformation processes were executed using complete mathematical camera models. We also compared the rectified images and distortion plots generated using different ICPs. Furthermore, by comparing our method with the state of art method, we confirm the performance improvement of ICP conversions between PH and CV models. Full article
►▼ Show Figures

Figure 1

Back to TopTop