The Trace transform and its applications

Kadyrov, A. A.; Petrou, Maria

doi:10.1109/34.946986

Cited by 219 publications

(174 citation statements)

References 9 publications

Supporting

Mentioning

166

Contrasting

Unclassified

Order By: Relevance

“…GCM methodology is a tool that identifies features like coarseness, smoothness and regularity [16]. This methodology based on the intensity levels of digital image and grey levels of SAR images.…”

Section: A Analysis With Gcm Methodologymentioning

confidence: 99%

“…SAR images plays significant role in non military and military applications such as positioning the systems and guided application systems. SAR image analysis is used in variety applications like mining [11], oil pollution monitoring [12,13,14], oceanography [15,16,17,18]. Images collected from different SAR image databases is extensively used in the classification of space borne and airborne images [19,20,21].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Classification of Synthetic Aperture Radar Images using Fuzzy SVMs

Sekar¹

2017

IJRASET

View full text Add to dashboard Cite

In this study, the classification of synthetic aperture radar images are analyses using fuzzy support vector machines. The issues in the classification of SAR images are not addressed properly and remains an unsolved problem in few instances. This study provides one of the best methodology to classify SAR based image databases with fuzzy support vector machines. All the concepts which are derived and applied are based machine learning algorithms. The designed and developed methodology has got many applications in the area like weather forecasting, flood impact monitoring, designing crop calenders, early detection of drought and many more. All the algorithms applied in this study are pixel oriented. This study is carried out on the SAR images collected from the publicly available AXA EORC data set. We designed a model based mathematics and statistics which is developed in the python programming language. Several algorithms are designed to analyze SAR images in which fuzzy support methodology is exhibited high performance. Reports discussed in this study shown that the efficiency of fuzzy support vector machines to classify synthetic aperture radar image is 95.3% without any heuristic information.

show abstract

Section: A Analysis With Gcm Methodologymentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Classification of Synthetic Aperture Radar Images using Fuzzy SVMs

Sekar¹

2017

IJRASET

View full text Add to dashboard Cite

show abstract

“…Trace Transform has been successfully applied to many image processing tasks such as image database retrieval [2], texture classification [4] and characters recognition [5]. Evolutionary Trace Transform (ETT) [6] has recently been developed to construct efficient Trace transform triple features to represent an image.…”

Section: Introductionmentioning

confidence: 99%

“…1) should considerably differ from each other. Trace transform (TT) [2] calculates functionals on image pixels along straight lines projected in different orientations. For example, Fig.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Study of Multi-objective Evolutionary Trace Transform Methods for Robust Feature Extraction

Albukhanajer

Jin

Briffa

et al. 2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Recently, Evolutionary Trace Transform (ETT) has been developed to extract efficient features (called triple features) for invariant image identification using multi-objective evolutionary algorithms. This paper compares two methods of Evolutionary Trace Transform (method I and II) evolved through similar objectives by minimizing the withinclass variance (Sw) and maximizing the between-class variance (S b ) of image features. However, each solution on the Pareto front of method I represents one triple features (i.e. 1D) to be combined with another solution to construct 2D feature space, whereas each solution on the Pareto front of method II represents a complete pair of triple features (i.e. 2D). Experimental results show that both methods are able to produce stable and consistent features. Moreover, method II has denser solutions distributed in the convex region of the Pareto front than in method I. Nevertheless, method II takes longer time to evolve than method I. Although the Trace transforms are evolved offline on one set of low resolution (64 × 64) images, they can be applied to extract features from various standard 256 × 256 images.

show abstract

“…Restricting ourselves, therefore, in computer vision to features that we can consciously identify as characterising our cognition excludes the vast number of features that our sub-conscious uses and which we cannot usually identify. We may, however, replace the mechanism of our sub-conscious with a Mathematical tool that allows us to construct thousands of features that do not have physical or other meaning: we may use the Trace transform which is an alternative image representation and from which we can construct the so called triple features [2]. In [2] it was shown how one can construct such features invariant to rotation, translation and scaling, while in [3] it was shown how to construct object signatures invariant to affine transforms.…”

Section: Introductionmentioning

confidence: 99%

Texture classification with thousands of features

Kadyrov¹,

Talepbour²,

Petrou³

2002

Procedings of the British Machine Vision Conference 2002

Self Cite

View full text Add to dashboard Cite

The Trace transform is a generalisation of the Radon transform that allows one to construct image features that do not necessarily have meaning in terms of human perception, but they measure different image characteristics. The ability of producing thousands of features from an image allows one to be selective as to which are appropriate for a particular task. In this paper we propose the use of such an approach to the problem of texture discrimination and compare its results with the classical co-occurrence matrix approach where usually the features used are fewer than ten.

show abstract

The Trace transform and its applications

Cited by 219 publications

References 9 publications

Classification of Synthetic Aperture Radar Images using Fuzzy SVMs

Classification of Synthetic Aperture Radar Images using Fuzzy SVMs

A Comparative Study of Multi-objective Evolutionary Trace Transform Methods for Robust Feature Extraction

Texture classification with thousands of features

Contact Info

Product

Resources

About