A coloring fuzzy graph approach for image classification

Gómez, Daniel; Montero, Javier; Yáñez, Javier

doi:10.1016/j.ins.2006.01.006

Cited by 55 publications

(19 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since people are often unable to extract useful knowledge from such huge datasets, data mining [16] has become a research focus in recent years. Among the several functions of data mining, classification is crucially important and has been applied successfully to several areas such as automatic text summarization and categorization [17,38], image classification [15], and virus detection of new malicious emails [31]. Although real-word data mining tasks often involve continuous attributes, some classification algorithms such as AQ [18,26], CLIP [6,7] and CN2 [8] can only handle categorical attribute, while others can handle continuous attributes but would perform better on categorical attributes [36].…”

Section: Introductionmentioning

confidence: 99%

A discretization algorithm based on Class-Attribute Contingency Coefficient

Tsai

Lee

Yang

2008

Information Sciences

185

100

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

A discretization algorithm based on Class-Attribute Contingency Coefficient

Tsai

Lee

Yang

2008

Information Sciences

185

100

View full text Add to dashboard Cite

“…The second binary coloring is then applied separately to the sub graph generated by those pixels colored as "0", to obtain the color classes "00" and "01", and to the sub graph generated by those pixels colored as "1", to obtain the color classes "10" and "11". This hierarchical process of binary coloring procedures is repeated a number of iterations, until a significative segmentation is obtained (see Gómez et al 2007Gómez et al , 2006. Possible homogeneous regions will be associated to connected pixels with the same color.…”

Section: Searching For Fuzzy Classes In Remote Sensingmentioning

confidence: 99%

“…2 (see http://www.mat.ucm. es/fcs and Gómez et al (2006) for computational results). Such a coloring algorithm offers several possible pictures of the image.…”

Section: Final Commentsmentioning

confidence: 99%

Fuzzy sets in remote sensing classification

Gómez

Montero

2007

Soft Comput

View full text Add to dashboard Cite

Supervised classification in remote sensing is a very complex problem and involves steps of different nature, including a serious data preprocessing. The final objective can be stated in terms of a classification of isolated pixels between classes, which can be either previously known or not (for example, different land uses), but with no particular shape nither well defined borders. Hence, a fuzzy approach seems natural in order to capture the structure of the image. In this paper we stress that some useful tools for a fuzzy classification can be derived from fuzzy coloring procedures, to be extended in a second stage to the complete non visible spectrum. In fact, the image is considered here as a fuzzy graph defined on the set of pixels, taking advantage of fuzzy numbers in order to summarize information. A fuzzy model is then presented, to be considered as a decision making aid tool. In this way we generalize the classical definition of fuzzy partition due to Ruspini, allowing in addition a first evaluation of the quality of the classification in this way obtained, in terms of three basic indexes (measuring covering, relevance and overlapping of our family of classes).

show abstract

“…In the past, the probability statistics (PS) method [11,[22][23][24][25] was usually used for target control. Along with the development of information and industry technology, at present, people are becoming more and more interested in fuzzy control [1,[3][4][5][6][7][8][9][10]12,13,15,[17][18][19]21,26] and rough sets for target control [2,20,27,28]; again, by the combination of fuzzy sets (FS) theory and rough sets (RS) theory, i.e., fuzzy rough sets (FRS) theory [14,16,29], we can obtain a new control algorithm, which here is called the fuzzy rough (FR) control algorithm. However, what differences are there between the new FR algorithm and the PS algorithm on target control?…”

Section: Introductionmentioning

confidence: 99%

Research on two different mathematical theories on control

Wang

Pang

et al. 2008

Journal of Computational and Applied Mathematics

View full text Add to dashboard Cite

With a brand-new theory, this paper not only provides the differences of attributes in concept, formula expression and function type between fuzzy rough sets and probability statistics, but also introduces their differences in algorithms on target control for better solving the control problem. Some new definitions and theorems concerning fuzzy rough sets and probability statistics are given, but this paper mainly makes a comparison of two control algorithms for the target tracking. The simulation results show that the comprehensive performance of the fuzzy rough sets algorithm is better than that of the probability statistics algorithm, but its control effect is not as good as that of the latter on multisensor target control. Finally, some problems concerning the combination of fuzzy rough sets and the probability statistics phenomenon to be solved and development trends are discussed. By these investigations, we can choose the optimal control algorithms for accomplishing better target control.

show abstract

A coloring fuzzy graph approach for image classification

Cited by 55 publications

References 22 publications

A discretization algorithm based on Class-Attribute Contingency Coefficient

A discretization algorithm based on Class-Attribute Contingency Coefficient

Fuzzy sets in remote sensing classification

Research on two different mathematical theories on control

Contact Info

Product

Resources

About