Image Segmentation Based on Level Set Method

Jiang, Xin; Zhang, Ren‐Jie; Nie, Shengdong

doi:10.1016/j.phpro.2012.05.143

Cited by 51 publications

(30 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(a) The heatmap representation of the Q-table produced from several episodes of the RL algorithm applied to connect the hand drawn segments in (b). The connected path is [3,7,2,6,0,8,1].…”

Section: Classification Via Region Mergingmentioning

confidence: 99%

See 1 more Smart Citation

Subcellular structure segmentation from cryo-electron tomograms via machine learning

Yang

Zhou

Gao

et al. 2020

Preprint

View full text Add to dashboard Cite

We describe how to use several machine learning techniques organized in a learning pipeline to segment and identify subcellular structures from cryo electron tomograms. These tomograms are difficult to analyze with traditional segmentation tools. The learning pipeline in our approach starts from supervised learning via a special convolutional neural network trained with simulated data. It continues with semi-supervised reinforcement learning and/or a region merging techniques that try to piece together disconnected components that should belong to the same subcellular structure. A parametric or non-parametric fitting procedure is then used to enhance the segmentation results and quantify uncertainties in the fitting. Domain knowledge is used in generating the training data for the neural network and in guiding the fitting procedure through the use of appropriately chosen priors and constraints. We demonstrate that the approach proposed here work well for extracting membrane surfaces of protein reconstituted liposomes in a cellular environment that contains other artifacts.April

show abstract

“…(a) The heatmap representation of the Q-table produced from several episodes of the RL algorithm applied to connect the hand drawn segments in (b). The connected path is [3,7,2,6,0,8,1].…”

Section: Classification Via Region Mergingmentioning

confidence: 99%

“…Although a number of automated segmentation algorithms and tools have been developed in the last few decades for high contrast medical 3D imaging [2][3][4][5][6][7][8], most of them perform poorly on cryo-ET datasets.…”

Section: Introductionmentioning

confidence: 99%

Subcellular structure segmentation from cryo-electron tomograms via machine learning

Yang

Zhou

Gao

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Level iSet iBased iSegmentation: iThe ilevel iset imethod ifor icapturing idynamic iinterfaces iand ishapes iwas ifirstly iintroduced iby iOsher iand iSethian i [27]. iThe iprimary iidea iof ithe ilevel iset imethod iis ito irepresent icontours ias ia izero ilevel iset iof ian iimplicit ifunction idefined iin ia ihigher idimension, inormally ireferred ito ias ithe ilevel iset ifunction iand ito iderive ithe ilevel iset ifunction iaccording ito ia ipartial idifferential iequation i(PDE) i [28]. iFor imedical iprocessing ipurpose, iit iis ibeen ilinked ito icomputer iapplications.…”

Section: Segmentation Techniques Used In Brain Tumor Segmentationmentioning

confidence: 99%

Assessment on Brain Tumor Detection Techniques in Hyperintense Mr Images

Jefferson*,

Shunmugasundaram

2020

IJRTE

View full text Add to dashboard Cite

Brain tumors have different characteristics such as shape, size, location, and image intensities. Magnetic-resonance images (MRIs) typically have a degree of noise and randomness associated with the natural random nature of brain structure. MRI is a profoundly created medical imaging strategy giving a range of data about the individual’s delicate tissue structure. Even though it gives a rich data, the complex dynamics of the tumor evolution cannot be captured perfectly because of the uncertainty in the tumor segmentations. Different methods are available to identify and segment a brain tumor. Stages of medical image processing in brain tumor detection are discussed in this paper and overview of the analogous papers is quoted by analyzing several research papers. This paper provides delving of technologies which can be used to prognosticate brain tumor.

show abstract

“…This algorithm calculated a proxy statistic for each textural feature, called the feature weight (W i=1:q ), which was used to estimate the relevance of the textural feature to the target variable or metallurgical phase [45]. To determine the feature weight W i , the reliefF algorithm employed an iterative updating scheme for weights, which was executed z number of times (z was the user-defined parameter, z ∈ Z + ) in the following three steps: (1) an instance x t=1:p was sampled at random from the dataset D without replacement in the first step; (2) k (a user-defined parameter, k ∈ Z + ) number of nearest instances of x t were determined from each class in the second step; and (3) weights of each textural feature were estimated and updated using Equation (18) in the third step. The set of k nearest instances that were identified in the second step were referred to as nearest-hit instances, h j , if the class label of k instances were the same as that of the class label of x t , and they were referred to as nearest-miss instances, m j , if the class label of the k instances differed from that of the class label of x t , where j took the values from 1 to k. Based on the nearest-hit and nearest-miss instances identified in the second step, this algorithm rewarded or penalized the textural features by updating their weights using Equation (18), which assigned higher weights to features that were strongly correlated to the target variable when compared to the irrelevant features.…”

Section: Feature Rankingmentioning

confidence: 99%

“…Techniques of discontinuity approach include Sobel operator [13], Laplacian of Gaussian (LoG) operator [14], Laplacian operator [15], and canny operator [15]. Techniques of similarity approach include histogram-based thresholding [16], region splitting and merging [17], level-set [18], clustering, and water shedding [11]. Among these techniques, histogram-based thresholding (or Otsu's method [16]) is extensively used for image analysis or segmentation of microstructures [19].…”

Section: Introductionmentioning

confidence: 99%

Texture-Based Metallurgical Phase Identification in Structural Steels: A Supervised Machine Learning Approach

Naik

Sajid

Kiran

2019

Metals

View full text Add to dashboard Cite

Automatic identification of metallurgical phases based on thresholding methods in microstructural images may not be possible when the pixel intensities associated with the metallurgical phases overlap and, hence, are indistinguishable. To circumvent this problem, additional visual information about the metallurgical phases, referred to as textural features, are considered in this study. Mathematically, textural features are the second order statistics of an image domain and can be distinct for each metallurgical phase. Textural features are evaluated from the gray level co-occurrence matrix (GLCM) of each metallurgical phase (ferrite, pearlite, and martensite) present in heat-treated ASTM A36 steels in this study. The dataset of textural features and pixel intensities generated for the metallurgical phases is used to train supervised machine learning classifiers, which are subsequently employed to predict the metallurgical phases in the microstructure. Naïve Bayes (NB), k-nearest neighbor (K-NN), linear discriminant analysis (LDA), and decision tree (DT) classifiers are the four classifiers employed in this study. The performances of all four classifiers were assessed prior to their deployment, and the classification accuracy was found to be >97%. The proposed technique has two unique advantages: (1) unlike pixel intensity-based methods, the proposed method does not misclassify the grain boundaries as a metallurgical phase, and (2) the proposed method does not require the end-user to input the number of phases present in the microstructure.

show abstract

Image Segmentation Based on Level Set Method

Cited by 51 publications

References 14 publications

Subcellular structure segmentation from cryo-electron tomograms via machine learning

Subcellular structure segmentation from cryo-electron tomograms via machine learning

Assessment on Brain Tumor Detection Techniques in Hyperintense Mr Images

Texture-Based Metallurgical Phase Identification in Structural Steels: A Supervised Machine Learning Approach

Contact Info

Product

Resources

About