author={D. Muhammad and Noorul Mubarak and M. Mohammed Sathik and S. Zulaikha Beevi},
  journal={International Journal of Computer Science and Information Technology},
In this paper, we have made improvements in region growing image segmentation. The First one is seeds select method, we use Harris corner detect theory to auto find growing seeds. Through this method, we can improve the segmentation speed. In this method, we use the Improved Harris corner detect theory for maintaining the distance vector between the seed pixel and maintain minimum distance between the seed pixels. The homogeneity criterion usually depends on image formation properties that are… 

Figures and Tables from this paper

Improved Region Growing Method for Magnetic Resonance Images (MRIs) Segmentation

An automatic homogeneity criterion based on estimating probability of pixel intensities of a given image is described and is applied to challenging applications: gray matter/white matter segmentation in magnetic resonance image (MRI) datasets.

Overview of automatic seed selection methods for biomedical images segmentation

An extensive survey on works carried out in the area of automatic seed point selection for biomedical images segmentation by seeded region growing algorithm is presented.

An Overview of Segmentation Algorithms for the Analysis of Anomalies on Medical Images

This paper classifies some of the widely used medical image segmentation algorithms based on their evolution, and the features of each generation are also discussed.

Region Growing and K-Means Hard Clustering Technique to Extract Brain Abnormalities in MRI Images

In this work, two adaptive region growing techniques were proposed to isolate and extract any abnormal tissues like tumors, edema and others to aid in detect diseases in MRI images of brain. Six T1

Improvement of Segmentation Efficiency in Mammogram Images Using Dual-ROI Method

The experimental outcome demonstrated that the proposed approach enhanced the segmentation efficiency by methods of statistical parameters contrasted with the existing operating procedures.

Adaptive Technique Depending on Region Growing and Soft Clustering to Detect Tumors in Different Modalities of MRI Brain Images

An adaptive technique based on the fuzzy clustering scheme showed that Region Growing segmentation improved its performance and it could separate the consists of the tumor regions, and the elapsed time of implementation is clearly reduced.


This paper presents an objective comparision of region bas e d segmentation techniques and classification of tumor in MR image using Artificial neural network (ANN) and Rad i a l Basis Function (RBF) and feed forward back propagation (FFBP) are trained the f e atu re vector for classification.

Locating seed points for automatic multi-organ segmentation using non-rigid registration and organ annotations

A method to automatically calculate the seed points for segmenting organs in three-dimensional (3D), non-annotated Computed Tomography and Magnetic Resonance volumes from the VISCERAL dataset is presented in this paper.

An Adaptive Fuzzy C-Means Algorithm for Improving MRI Segmentation

The proposed method is capable to avoid various problems of existing fuzzy clustering methods that solve the defect of noise and overcomes the coincident clusters problem of PCM.

Segmentation des régions d'intérêts dans des images médicales

The main objective of this thesis is the extraction of relevant regions of cardiac images by proposing an automatic segmentation of the right ventricle based on an active shape model (ASM) combined with a distance transform and an automatic initialization by the generalized Hough transform method.



Symmetric region growing

A set of theoretical criteria for a subclass of region-growing algorithms that are insensitive to the selection of the initial growing points are defined, which leads to a single-pass region- growing approach applicable to any dimensionality of images.


To segment the images, from segmentation techniques edge detection, thresholding, region growing and clustering are taken for this study.

Automated extraction of bronchus from 3D CT images of lung based on genetic algorithm and 3D region growing

A method to automate the segmentation of airway tree structures in lung from a stack of gray-scale computed tomography images and the final extracted bronchus area with the optimal threshold value is reconstructed and visualized by 3D texture mapping method.

Adaptive image region-growing

To decide if two regions should be merged, instead of comparing the difference of region feature means with a predefined threshold, the authors adaptively assess region homogeneity from region feature distributions, resulting in an algorithm that is robust with respect to various image characteristics.

Seeded Region Growing

This correspondence presents a new algorithm for segmentation of intensity images which is robust, rapid, and free of tuning parameters, and suggests two ways in which it can be employed, namely, by using manual seed selection or by automated procedures.

Dynamic Region Growing

A novel segmentation algorithm based on the region growing paradigm is presented, which assigns a stability value to each extracted region that reeects the robust-ness of that region.

A Computational Approach to Edge Detection

  • J. Canny
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 1986
There is a natural uncertainty principle between detection and localization performance, which are the two main goals, and with this principle a single operator shape is derived which is optimal at any scale.

Performance Evaluation of Edge Detection Techniques for Images in Spatial Domain

The comparative analysis of various Image Edge Detection methods is presented and it has been shown that the Canny's edge detection algorithm performs better than all these operators under almost all scenarios.

Adaptive region growing technique using polynomial functions for image approximation

Scale & Affine Invariant Interest Point Detectors

A comparative evaluation of different detectors is presented and it is shown that the proposed approach for detecting interest points invariant to scale and affine transformations provides better results than existing methods.