sensors-logo

Journal Browser

Journal Browser

Advances in Image Segmentation: Theory and Applications

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Sensing and Imaging".

Deadline for manuscript submissions: closed (20 February 2022) | Viewed by 25308

Special Issue Editors

Faculty of Electronics, Telecommunication and Informatics, Gdańsk University of Technology, 80-233 Gdańsk, Poland
Interests: image processing; computer vision; image segmentation; object detection; remote sensing image processing; synthetic aperture radar image processing; active contours; mathematical morphology; morphological image processing; pattern recognition algorithms
Institute of Applied Sciences and Intelligent Systems “ScienceApp", Consiglio Nazionale delle Ricerche, c/o Dhitech Campus Universitario Ecotekne, Via Monteroni s/n, 73100 Lecce, Italy
Interests: computer vision; pattern recognition; video surveillance; object tracking; deep learning; audience measurements; visual interaction; human–robot interaction
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Segmentation is aimed at identifying the borders of objects in the analysed digital image or at splitting the image into various, non-overlapping regions. Segmentation is one of the most important steps in digital image processing systems as its result directly impacts the results of subsequent processing methods, e.g., 3D reconstruction and visualisation, distinguishing of features or classification. Practice shows that it is extremely difficult to arrive at a universal method that would produce high results for different segmentation problems that are being solved. This is all the more so as digital images are acquired with scanners/sensors of different types and with different characteristics.

Image segmentation has become a major topic of interest in various domains, including medical imaging, environmental remote sensing field, land cover applications, etc. In medical image analysis, segmentation can be defined as a method allowing, e.g., the precise shape of the potential lesions to be determined or the shape of the organ to be determined. Very different imaging techniques are available for medical use today (for example, ultrasonography (USG), computed tomography (CT), magnetic resonance imaging (MR)), while segmentation methods should be as closely matched to source images processed as possible.

In remote sensing, segmentation allows assigning labels to image pixels so that pixels in the same region or object are associated with the same label. It can be said that high spatial resolution (HSR) images acquired from planes, satellites or from unmanned aerial vehicles (UAVs) as well as from other platforms are increasingly available. HSR images come from different sensor types, such as hyperspectral, multispectral, synthetic aperture radar (SAR) or thermal infrared sensors.

This Special Issue aims to publish original papers, as well as review articles addressing emerging trends in image segmentation. The main topics include but are not limited to the following:

- Biomedical Image Segmentation (different modalities, e.g., CT, MRI, USG);

- SAR image segmentation;

- Segmentation methods for multisensor data analysis;

- Segmentation methods for time-series data analysis (e.g., agricultural crop areas, urban areas growing or drought/flood monitoring);

- Machine learning;

- Deep learning;

- Review of segmentation methods.

Dr. Marcin Ciecholewski
Dr. Cosimo Distante
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. 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

  • biomedical image segmentation (different modalities, e.g., CT, MRI, USG)
  • SAR image segmentation
  • segmentation methods for multisensor data analysis
  • segmentation methods for time-series data analysis (e.g., agricultural crop areas, urban areas growing or drought/flood monitoring)
  • machine learning
  • deep learning
  • review of segmentation methods

Published Papers (8 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Jump to: Review

14 pages, 13028 KiB  
Article
Multi-Path U-Net Architecture for Cell and Colony-Forming Unit Image Segmentation
by Vilen Jumutc, Dmitrijs Bļizņuks and Alexey Lihachev
Sensors 2022, 22(3), 990; https://0-doi-org.brum.beds.ac.uk/10.3390/s22030990 - 27 Jan 2022
Cited by 7 | Viewed by 2833
Abstract
U-Net is the most cited and widely-used deep learning model for biomedical image segmentation. In this paper, we propose a new enhanced version of a ubiquitous U-Net architecture, which improves upon the original one in terms of generalization capabilities, while addressing several immanent [...] Read more.
U-Net is the most cited and widely-used deep learning model for biomedical image segmentation. In this paper, we propose a new enhanced version of a ubiquitous U-Net architecture, which improves upon the original one in terms of generalization capabilities, while addressing several immanent shortcomings, such as constrained resolution and non-resilient receptive fields of the main pathway. Our novel multi-path architecture introduces a notion of an individual receptive field pathway, which is merged with other pathways at the bottom-most layer by concatenation and subsequent application of Layer Normalization and Spatial Dropout, which can improve generalization performance for small datasets. In general, our experiments show that the proposed multi-path architecture outperforms other state-of-the-art approaches that embark on similar ideas of pyramid structures, skip-connections, and encoder–decoder pathways. A significant improvement of the Dice similarity coefficient is attained at our proprietary colony-forming unit dataset, where a score of 0.809 was achieved for the foreground class. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

13 pages, 1273 KiB  
Article
Influence of Pitch Angle Errors in 3D Scene Reconstruction Based on U-V Disparity: A Sensitivity Study
by Jonatán Felipe, Marta Sigut and Leopoldo Acosta
Sensors 2022, 22(1), 79; https://0-doi-org.brum.beds.ac.uk/10.3390/s22010079 - 23 Dec 2021
Cited by 1 | Viewed by 1988
Abstract
U-V disparity is a technique that is commonly used to detect obstacles in 3D scenes, modeling them as a set of vertical planes. In this paper, the authors describe the general lines of a method based on this technique for fully reconstructing 3D [...] Read more.
U-V disparity is a technique that is commonly used to detect obstacles in 3D scenes, modeling them as a set of vertical planes. In this paper, the authors describe the general lines of a method based on this technique for fully reconstructing 3D scenes, and conduct an analytical study of its performance and sensitivity to errors in the pitch angle of the stereoscopic vision system. The equations of the planes calculated for a given error in this angle yield the deviation with respect to the ideal planes (with a zero error in the angle) for a large test set consisting of planes with different orientations, which is represented graphically to analyze the method’s qualitative and quantitative performance. The relationship between the deviation of the planes and the error in the pitch angle is observed to be linear. Two major conclusions are drawn from this study: first, that the deviation between the calculated and ideal planes is always less than or equal to the error considered in the pitch angle; and second, that even though in some cases the deviation of the plane is zero or very small, the probability that a plane of the scene deviates from the ideal by the greatest amount possible, which matches the error in the pitch angle, is very high. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

24 pages, 81345 KiB  
Article
Improving Automatic Renal Segmentation in Clinically Normal and Abnormal Paediatric DCE-MRI via Contrast Maximisation and Convolutional Networks for Computing Markers of Kidney Function
by Hykoush Asaturyan, Barbara Villarini, Karen Sarao, Jeanne S. Chow, Onur Afacan and Sila Kurugol
Sensors 2021, 21(23), 7942; https://0-doi-org.brum.beds.ac.uk/10.3390/s21237942 - 28 Nov 2021
Cited by 2 | Viewed by 2294
Abstract
There is a growing demand for fast, accurate computation of clinical markers to improve renal function and anatomy assessment with a single study. However, conventional techniques have limitations leading to overestimations of kidney function or failure to provide sufficient spatial resolution to target [...] Read more.
There is a growing demand for fast, accurate computation of clinical markers to improve renal function and anatomy assessment with a single study. However, conventional techniques have limitations leading to overestimations of kidney function or failure to provide sufficient spatial resolution to target the disease location. In contrast, the computer-aided analysis of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) could generate significant markers, including the glomerular filtration rate (GFR) and time–intensity curves of the cortex and medulla for determining obstruction in the urinary tract. This paper presents a dual-stage fully modular framework for automatic renal compartment segmentation in 4D DCE-MRI volumes. (1) Memory-efficient 3D deep learning is integrated to localise each kidney by harnessing residual convolutional neural networks for improved convergence; segmentation is performed by efficiently learning spatial–temporal information coupled with boundary-preserving fully convolutional dense nets. (2) Renal contextual information is enhanced via non-linear transformation to segment the cortex and medulla. The proposed framework is evaluated on a paediatric dataset containing 60 4D DCE-MRI volumes exhibiting varying conditions affecting kidney function. Our technique outperforms a state-of-the-art approach based on a GrabCut and support vector machine classifier in mean dice similarity (DSC) by 3.8% and demonstrates higher statistical stability with lower standard deviation by 12.4% and 15.7% for cortex and medulla segmentation, respectively. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Graphical abstract

15 pages, 2969 KiB  
Article
Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation
by Yun Jiang, Huixia Yao, Shengxin Tao and Jing Liang
Sensors 2021, 21(18), 6177; https://0-doi-org.brum.beds.ac.uk/10.3390/s21186177 - 15 Sep 2021
Cited by 6 | Viewed by 1722
Abstract
Segmentation of retinal vessels is a critical step for the diagnosis of some fundus diseases. Methods: To further enhance the performance of vessel segmentation, we propose a method based on a gated skip-connection network with adaptive upsampling (GSAU-Net). In GSAU-Net, a novel skip-connection [...] Read more.
Segmentation of retinal vessels is a critical step for the diagnosis of some fundus diseases. Methods: To further enhance the performance of vessel segmentation, we propose a method based on a gated skip-connection network with adaptive upsampling (GSAU-Net). In GSAU-Net, a novel skip-connection with gating is first utilized in the extension path, which facilitates the flow of information from the encoder to the decoder. Specifically, we used the gated skip-connection between the encoder and decoder to gate the lower-level information from the encoder. In the decoding phase, we used an adaptive upsampling to replace the bilinear interpolation, which recovers feature maps from the decoder to obtain the pixelwise prediction. Finally, we validated our method on the DRIVE, CHASE, and STARE datasets. Results: The experimental results showed that our proposed method outperformed some existing methods, such as DeepVessel, AG-Net, and IterNet, in terms of accuracy, F-measure, and AUCROC. The proposed method achieved a vessel segmentation F-measure of 83.13%, 81.40%, and 84.84% on the DRIVE, CHASE, and STARE datasets, respectively. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

12 pages, 12444 KiB  
Communication
Generation of the NIR Spectral Band for Satellite Images with Convolutional Neural Networks
by Svetlana Illarionova, Dmitrii Shadrin, Alexey Trekin, Vladimir Ignatiev and Ivan Oseledets
Sensors 2021, 21(16), 5646; https://0-doi-org.brum.beds.ac.uk/10.3390/s21165646 - 21 Aug 2021
Cited by 11 | Viewed by 3874
Abstract
The near-infrared (NIR) spectral range (from 780 to 2500 nm) of the multispectral remote sensing imagery provides vital information for landcover classification, especially concerning vegetation assessment. Despite the usefulness of NIR, it does not always accomplish common RGB. Modern achievements in image processing [...] Read more.
The near-infrared (NIR) spectral range (from 780 to 2500 nm) of the multispectral remote sensing imagery provides vital information for landcover classification, especially concerning vegetation assessment. Despite the usefulness of NIR, it does not always accomplish common RGB. Modern achievements in image processing via deep neural networks make it possible to generate artificial spectral information, for example, to solve the image colorization problem. In this research, we aim to investigate whether this approach can produce not only visually similar images but also an artificial spectral band that can improve the performance of computer vision algorithms for solving remote sensing tasks. We study the use of a generative adversarial network (GAN) approach in the task of the NIR band generation using only RGB channels of high-resolution satellite imagery. We evaluate the impact of a generated channel on the model performance to solve the forest segmentation task. Our results show an increase in model accuracy when using generated NIR compared to the baseline model, which uses only RGB (0.947 and 0.914 F1-scores, respectively). The presented study shows the advantages of generating the extra band such as the opportunity to reduce the required amount of labeled data. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

17 pages, 584 KiB  
Article
Fully Parallel Implementation of Otsu Automatic Image Thresholding Algorithm on FPGA
by Wysterlânya K. P. Barros, Leonardo A. Dias and Marcelo A. C. Fernandes
Sensors 2021, 21(12), 4151; https://doi.org/10.3390/s21124151 - 17 Jun 2021
Cited by 13 | Viewed by 2789
Abstract
This work proposes a high-throughput implementation of the Otsu automatic image thresholding algorithm on Field Programmable Gate Array (FPGA), aiming to process high-resolution images in real-time. The Otsu method is a widely used global thresholding algorithm to define an optimal threshold between two [...] Read more.
This work proposes a high-throughput implementation of the Otsu automatic image thresholding algorithm on Field Programmable Gate Array (FPGA), aiming to process high-resolution images in real-time. The Otsu method is a widely used global thresholding algorithm to define an optimal threshold between two classes. However, this technique has a high computational cost, making it difficult to use in real-time applications. Thus, this paper proposes a hardware design exploiting parallelization to optimize the system’s processing time. The implementation details and an analysis of the synthesis results concerning the hardware area occupation, throughput, and dynamic power consumption, are presented. Results have shown that the proposed hardware achieved a high speedup compared to similar works in the literature. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

26 pages, 6956 KiB  
Article
Attention-Guided Network with Densely Connected Convolution for Skin Lesion Segmentation
by Shengxin Tao, Yun Jiang, Simin Cao, Chao Wu and Zeqi Ma
Sensors 2021, 21(10), 3462; https://0-doi-org.brum.beds.ac.uk/10.3390/s21103462 - 16 May 2021
Cited by 12 | Viewed by 2352
Abstract
The automatic segmentation of skin lesions is considered to be a key step in the diagnosis and treatment of skin lesions, which is essential to improve the survival rate of patients. However, due to the low contrast, the texture and boundary are difficult [...] Read more.
The automatic segmentation of skin lesions is considered to be a key step in the diagnosis and treatment of skin lesions, which is essential to improve the survival rate of patients. However, due to the low contrast, the texture and boundary are difficult to distinguish, which makes the accurate segmentation of skin lesions challenging. To cope with these challenges, this paper proposes an attention-guided network with densely connected convolution for skin lesion segmentation, called CSAG and DCCNet. In the last step of the encoding path, the model uses densely connected convolution to replace the ordinary convolutional layer. A novel attention-oriented filter module called Channel Spatial Fast Attention-guided Filter (CSFAG for short) was designed and embedded in the skip connection of the CSAG and DCCNet. On the ISIC-2017 data set, a large number of ablation experiments have verified the superiority and robustness of the CSFAG module and Densely Connected Convolution. The segmentation performance of CSAG and DCCNet is compared with other latest algorithms, and very competitive results have been achieved in all indicators. The robustness and cross-data set performance of our method was tested on another publicly available data set PH2, further verifying the effectiveness of the model. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

Review

Jump to: Research

21 pages, 32074 KiB  
Review
Computational Methods for Liver Vessel Segmentation in Medical Imaging: A Review
by Marcin Ciecholewski and Michał Kassjański
Sensors 2021, 21(6), 2027; https://0-doi-org.brum.beds.ac.uk/10.3390/s21062027 - 12 Mar 2021
Cited by 31 | Viewed by 5220
Abstract
The segmentation of liver blood vessels is of major importance as it is essential for formulating diagnoses, planning and delivering treatments, as well as evaluating the results of clinical procedures. Different imaging techniques are available for application in clinical practice, so the segmentation [...] Read more.
The segmentation of liver blood vessels is of major importance as it is essential for formulating diagnoses, planning and delivering treatments, as well as evaluating the results of clinical procedures. Different imaging techniques are available for application in clinical practice, so the segmentation methods should take into account the characteristics of the imaging technique. Based on the literature, this review paper presents the most advanced and effective methods of liver vessel segmentation, as well as their performance according to the metrics used. This paper includes results available for four imaging methods, namely: computed tomography (CT), computed tomography angiography (CTA), magnetic resonance (MR), and ultrasonography (USG). The publicly available datasets used in research are also presented. This paper may help researchers gain better insight into the available materials and methods, making it easier to develop new, more effective solutions, as well as to improve existing approaches. This article analyzes in detail various segmentation methods, which can be divided into three groups: active contours, tracking-based, and machine learning techniques. For each group of methods, their theoretical and practical characteristics are discussed, and the pros and cons are highlighted. The most advanced and promising approaches are also suggested. However, we conclude that liver vasculature segmentation is still an open problem, because of the various deficiencies and constraints researchers need to address and try to eliminate from the solutions used. Full article
(This article belongs to the Special Issue Advances in Image Segmentation: Theory and Applications)
Show Figures

Figure 1

Back to TopTop