Advanced image segmentation methods using partial differential equations: A concise comparison

Sliz, J.; Mikulka, Jan

doi:10.1109/piers.2016.7734800

Cited by 9 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcome the problem of surface topology of objects in the image, a function that represents the curve or surface as a set of zero levels for surfaces with higher dimensions is developed. In this way it turns out that a more accurate numerical implementation is obtained, so as to be able to solve topological problems better [46] [47]. The Level-Set method is quite simple and easy to adapt to calculate and analyze the interface changes of an image, either two or three dimensions [48].…”

Section: F) Pde Based Methodsmentioning

confidence: 99%

Image Segmentation Based on Color Dissimilarity

Karma¹,

Putra

Sudarma

et al. 2022

IJCAI

View full text Add to dashboard Cite

This study aims to develop a segmentation technique that can be used to detect objects in an image. The concept used is to imitate the human concept of recognizing an object based on its color difference. A color is considered different if it has different R, G or B values. Humans can only distinguish two colors clearly if they both have a DeltaE value of at least 8. The difference in DeltaE values is obtained from differences in the values of R, G or B, either alone or in pairs. The application of this concept to the segmentation technique has shown good results.This technique is able to give the same results and is good for images with JPG, PNG and BMP file types. To provide confidence in the performance of this method, a comparison is made with several segmentation methods. By measuring the similarity index using the Jaccard index, Sørensen-Dice index and BF Score, this color dissimilarity-based segmentation method has the best performance. Meanwhile, when measuring the processing time, this method has the second best performance. Thus, it can be concluded that this method is very suitable for detecting objects in an image. Povzetek: Tehnika segmentacije je bila razvita z imitacijo človeškega koncepta prepoznavanja predmeta, in sicer na podlagi barvnih razlik. Barve se štejejo za različne, če imajo vrednost DeltaE najmanj 8, ki jo dobimo iz razlik v vrednostih R, G ali B barv samih ali v parih. Izvajanje te tehnike kaže enake rezultate za datoteke JPG, PNG in BMP in je v veliko pomoč pri postopku identifikacije predmeta.

show abstract

Section: F) Pde Based Methodsmentioning

confidence: 99%

Image Segmentation Based on Color Dissimilarity

Karma¹,

Putra

Sudarma

et al. 2022

IJCAI

View full text Add to dashboard Cite

show abstract

“…Existing conventional post-reconstruction denoising techniques include: Gaussian filtering, Non-Local Mean filtering (NLM) [8], [9], anisotropic diffusion [10], [11] and blockmatching 3D (BM3D) [12], [13]. Variational PDE (partial differential equation) [14], one of the most recent techniques, have also gained popularity in many applications, such as image denoising [15] and segmentation [16]. However, deep learning could be a more effective technique to tackle these problems.…”

Section: Introductionmentioning

confidence: 99%

Learning to Denoise Gated Cardiac PET Images Using Convolutional Neural Networks

Gambin¹,

Tadi²,

Teuho³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Noise and motion artifacts in Positron emission tomography (PET) scans can interfere in diagnosis and result in inaccurate interpretations. PET gating techniques effectively reduce motion blurring, but at the cost of increasing noise, as only a subset of the data is used to reconstruct the image. Deep convolutional neural networks (DCNNs) could complement gating techniques by correcting such noise. However, there is little research on the specific application of DCNNs to gated datasets, which present additional challenges that are not considered in these studies yet, such as the varying level of noise depending on the gate, and performance pitfalls due to changes in the noise properties between non-gated and gated scans. To extend the current status of artificial intelligence (AI) in gated-PET imaging, we present a postreconstruction denoising approach based on U-Net architectures on cardiac dual-gated PET images obtained from 40 patients. To this end, we first evaluate the denoising performance of four different variants of the U-Net architecture (2D, semi-3D, 3D, Hybrid) on non-gated data to better understand the advantages of each type of model, and to shed more light on the factors to take in consideration when selecting a denoising architecture. Then, we tackle the denoising of gated-PET reconstructions, revising challenges and limitations, and propose two training approaches, which overcome the need for gated targets. Quantification results show that the proposed deep learning (DL) frameworks can successfully reduce noise levels while correctly preserving the original motionless resolution of the gates.

show abstract

“…The result is then processed through a series of image processing procedures such as Thresholding, Erosion and Dilation are performed in order to eliminate noise and separate the moving automobiles from the background. The number of vehicles is then counted conveniently and also useful for future planning and management of traffic (Abubakar, 2012;Bezdek, Ehrlich, & Full, 1984;Blumenstein & Verma, 1998;Hasan, Saha, Hoque, & Majumder, 2014;Indira & Ramesh, 2011;Kaur & Kaur, 2014;Lim & Isa, 2012;Lindeberg & Li, 1997;Sathya & Manavalan, 2011;Seal, Das, & Sen, 2015;Sliz & Mikulka, 2016;Srisha & Khan, 2013;Wagstaff, Cardie, Rogers, & Schroedl, 2001;Zhang, 2006).…”

Section: Introductionmentioning

confidence: 99%

A Framework to Reduce Road Congestion by means of Data Analytics

Yousaf

Saher

Shahzad

et al. 2021

REAL

View full text Add to dashboard Cite

The density of vehicles on the road especially in urban areas keeps on increasing to large amount day by day. Especially during the peak hours of the day, large amount of people wastes much of their time in traffic signals. Not only they waste energy by burning excess fuel and releasing CO2 emissions in the environment as well as their time and money. An idea has been proposed to monitor the traffic congestion by means of data analytics on image data and solve the critical traffic congestion issue. The CCTV or surveillance cameras installed at the top points on the roads acts as a medium to provide image data as an input to analyze road traffic congestion by counting the number of vehicles under specified interval of time. Monitoring of traffic congestion using image processing techniques is very useful for the future urban road planning such as: 1) if there is a need to make the road wider, 2) if there is a need to add more lanes on the road, 3) if there is need to make flyover or a bridge to control the traffic on the roads. It will help municipalities to structure and expansion of the roads.

show abstract

Advanced image segmentation methods using partial differential equations: A concise comparison

Cited by 9 publications

References 13 publications

Image Segmentation Based on Color Dissimilarity

Image Segmentation Based on Color Dissimilarity

Learning to Denoise Gated Cardiac PET Images Using Convolutional Neural Networks

A Framework to Reduce Road Congestion by means of Data Analytics

Contact Info

Product

Resources

About