Regions adjacency graph applied to color image segmentation

Trémeau, Alain; Colantoni, Philippe

doi:10.1109/83.841950

Cited by 210 publications

(101 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Second, the region adjacency graph (RAG) was built to simplify the evaluation processing. The RAG is an undirected graph, which is used to express the spatial relationship between adjacent segments, with nodes representing segments and arcs representing their adjacency [40,41]. Third, the JM distance of each pair of segments was calculated based on the RAG as follows:…”

Section: Area-weighted Variance (Wv) and Jeffries-matusita (Jm) Distamentioning

confidence: 99%

Unsupervised Segmentation Evaluation Using Area-Weighted Variance and Jeffries-Matusita Distance for Remote Sensing Images

Wang

Liu

2018

Remote Sensing

View full text Add to dashboard Cite

Abstract:Image segmentation is an important process and a prerequisite for object-based image analysis. Thus, evaluating the performance of segmentation algorithms is essential to identify effective segmentation methods and to optimize the scale. In this paper, we propose an unsupervised evaluation (UE) method using the area-weighted variance (WV) and Jeffries-Matusita (JM) distance to compare two image partitions to evaluate segmentation quality. The two measures were calculated based on the local measure criteria, and the JM distance was improved by considering the contribution of the common border between adjacent segments and the area of each segment in the JM distance formula, which makes the heterogeneity measure more effective and objective. Then the two measures were presented as a curve when changing the scale from 8 to 20, which can reflect the segmentation quality in both over-and under-segmentation. Furthermore, the WV and JM distance measures were combined by using three different strategies. The effectiveness of the combined indicators was illustrated through supervised evaluation (SE) methods to clearly reveal the segmentation quality and capture the trade-off between the two measures. In these experiments, the multiresolution segmentation (MRS) method was adopted for evaluation. The proposed UE method was compared with two existing UE methods to further confirm their capabilities. The visual and quantitative SE results demonstrated that the proposed UE method can improve the segmentation quality.

show abstract

Section: Area-weighted Variance (Wv) and Jeffries-matusita (Jm) Distamentioning

confidence: 99%

Unsupervised Segmentation Evaluation Using Area-Weighted Variance and Jeffries-Matusita Distance for Remote Sensing Images

Wang

Liu

2018

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…• Region Adjacency Graphs (RAG) [19] which provide very useful and common ways of describing the structure of a picture: vertices represent regions and edges represent region adjacency relationship.…”

Section: Representation Of Images By Graphsmentioning

confidence: 99%

Graph-based tools for microscopic cellular image segmentation

Ta¹,

Lézoray²,

Elmoataz³

et al. 2009

Pattern Recognition

View full text Add to dashboard Cite

We propose a framework of graph based tools for the segmentation of microscopic cellular images. This framework is based on an object oriented analysis of imaging problems in pathology. Our graph tools rely on a general formulation of discrete functional regularization on weighted graphs of arbitrary topology. It leads to a set of useful tools which can be combined together to address various image segmentation problems in pathology. To provide fast image segmentation algorithms, we also propose an image simplification based on graphs as a pre processing step. The abilities of this set of image processing discrete tools are illustrated through automatic and interactive segmentation schemes for color cytological and histological images segmentation problems.

show abstract

“…Two approaches were developed to deal with this problem. The first one involves merging adjacent regions according to some criterion after the use of the watershed algorithm (Courses & Surveys 1999;Haris et al 1998;Tremeau & Colantoni 2000). The second one uses markers to reduce over-segmentation (Salembier & Pardas 1994;Salembier & Marques 1999).…”

Section: Magnetic Element Segmentation With the Marker-controlled Watmentioning

confidence: 99%

“…This method is generalized from watersheds, which is a popular segmentation method for separating touching or adjacent objects (Beucher & Meyer 1993;Courses & Surveys 1999;Haris et al 1998;Tremeau & Colantoni 2000). However, it has been reported that the watershed algorithm often leads to over-segmentation (Vincent 1993), meaning that the method may partition one magnetic element to a set of small magnetic elements.…”

Section: Introductionmentioning

confidence: 99%

Properties of magnetic elements in the quiet Sun using the marker-controlled watershed method

Xie

Zhang

et al. 2009

A&A

View full text Add to dashboard Cite

Context. The quiet Sun is an important part of understanding the global magnetic properties of the Sun. A recently launched observation system, named HINODE, provides a lot of high-resolution images for studying the quiet Sun. Obviously, it is time-consuming to analyze these images by hand. It is desirable to develop a technique for recognizing magnetic elements, thus automatically computing magnetic properties and the relationship between magnetic elements and granulation. Aims. We design an automatic method of recognizing magnetic elements based on the features of HINODE magnetograms and of measuring their properties. Then we study the relationship between magnetic elements and granulation. Methods. We used the magnetogram, continuum image, and Dopplergram on April 16, 2007, which were taken with the Solar Optical Telescope instrument aboard HINODE. The field of view is 147. 60 × 162. 30 in a quiet solar region, locating at disk center. We introduced the mark-controlled watershed method to detect magnetic elements automatically, because it is a popular image-segmentation method for dealing with overlapping objects. We took the centers that are the local maximum in all directions as the marks for restraining over-segmentation. We computed the properties of the detected magnetic elements and the relation among magnetic field strength, relative continuum intensity, and Doppler velocity at the same locations of magnetic elements. Results. We obtain the following results: (1) 34% of our observation region are covered by magnetic fields; (2) the magnetic flux distribution of all elements reaches a peak at 1.07 × 10 16 Mx for the whole region; (3) the relative continuum intensity distribution at the locations of magnetic elements reaches a peak at 0.97, which shows that the majority of magnetic elements located at the areas where the relative continuum intensity is less than its average. The relative continuum intensities in the areas with strong flux density are the median, meaning that the strong magnetic elements are usually located at the boundary of granulation; (4) the absolute Doppler velocity distribution at the locations of magnetic elements reaches a peak at 1.00 km s −1 , and the majority of weak magnetic elements located at the areas where the absolute velocity is greater than 1.00 km s −1 ; (5) strongly magnetized regions only have weak absolute Doppler velocities. The absolute velocity is lower than 1.00 km s −1 in the regions where the magnetic flux density of elements is higher than 100 G.

show abstract

Regions adjacency graph applied to color image segmentation

Cited by 210 publications

References 12 publications

Unsupervised Segmentation Evaluation Using Area-Weighted Variance and Jeffries-Matusita Distance for Remote Sensing Images

Unsupervised Segmentation Evaluation Using Area-Weighted Variance and Jeffries-Matusita Distance for Remote Sensing Images

Graph-based tools for microscopic cellular image segmentation

Properties of magnetic elements in the quiet Sun using the marker-controlled watershed method

Contact Info

Product

Resources

About