A hierarchical classifier using new support vector machines for automatic target recognition

Casasent, David P.; Wang, Yu-Chiang

doi:10.1016/j.neunet.2005.06.033

Cited by 43 publications

(20 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 10 shows the flow chart of hierarchical recognition. In the last years, there were several literatures [4,10,12,13] on the similar topic to also discuss hierarchical target recognition. However, the target recognition methods discussed in these literatures were different from the hierarchical method discussed in this paper.…”

Section: Hierarchical Recognition Methodsmentioning

confidence: 99%

“…The literature [10] used some data mining, such as chemogenomic data mining, proteomic data mining, integrated text mining and integrated mining, to carry out the target discovery. The literature [12] proposed a binary hierarchical classifier that was called support vector representation and discrimination machine classifier (SVRDM) to achieve automatic target recognition. The SVRDM classifier was used at each node in the hierarchy.…”

Section: Hierarchical Recognition Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A hierarchical target recognition method based on image processing

Sun

Xiong³

et al. 2014

AIR

View full text Add to dashboard Cite

In order to provide an accurate and rapid target recognition method for some military affairs, public security, finance and other departments, this paper studied firstly a variety of fuzzy signal, analyzed the uncertainties classification and their influence, eliminated fuzziness processing, presents some methods and algorithms for fuzzy signal processing, and compared with other methods on image processing. Where, the fuzzy signal processing is that a blurred signal is dealt with by eliminating fuzziness. Moreover, this paper used the wavelet packet analysis to carry out feature extraction of target for the first time, extracted the coefficient feature and energy feature of wavelet transformation, gave the matching and recognition methods, compared with the existing target recognition methods by experiment, and presented the hierarchical recognition method. In target feature extraction process, the more detailed and rich texture feature of target can be obtained by wavelet packet to image decomposition to compare with the wavelet decomposition. In the process of matching and recognition, the hierarchical recognition method is presented to improve the recognition speed and accuracy. The wavelet packet transformation is used to carry out the image decomposition. Through experiment results, the proposed recognition method has the high precision, fast speed, and its correct recognition rate is improved by an average 6.13% than that of existing recognition methods. These researches development in this paper can provide an important theoretical reference and practical significance to improve the real-time and accuracy on fuzzy target recognition.

show abstract

Section: Hierarchical Recognition Methodsmentioning

confidence: 99%

Section: Hierarchical Recognition Methodsmentioning

confidence: 99%

A hierarchical target recognition method based on image processing

Sun

Xiong³

et al. 2014

AIR

View full text Add to dashboard Cite

show abstract

“…Common approaches are to simply compute the distances between class means [12,14] or to represent the classes in a high dimensional space using a Gaussian kernel [13,20].…”

Section: Existing Approachesmentioning

confidence: 99%

“…This determines the partitioning at a given level. When applied recursively, this simple yet popular [12,14,20] method allows to construct a class hierarchy. K-means clustering minimizes the distances to cluster centers, thus tries to find compact clusters.…”

Section: Top-down Recursive Partitioningmentioning

confidence: 99%

Constructing Category Hierarchies for Visual Recognition

Marszałek

Schmid

2008

Lecture Notes in Computer Science

107

View full text Add to dashboard Cite

Abstract. Class hierarchies are commonly used to reduce the complexity of the classification problem. This is crucial when dealing with a large number of categories. In this work, we evaluate class hierarchies currently constructed for visual recognition. We show that top-down as well as bottom-up approaches, which are commonly used to automatically construct hierarchies, incorporate assumptions about the separability of classes. Those assumptions do not hold for visual recognition of a large number of object categories. We therefore propose a modification which is appropriate for most top-down approaches. It allows to construct class hierarchies that postpone decisions in the presence of uncertainty and thus provide higher recognition accuracy. We also compare our method to a one-against-all approach and show how to control the speed-foraccuracy trade-off with our method. For the experimental evaluation, we use the Caltech-256 visual object classes dataset and compare to stateof-the-art methods.

show abstract

“…A preferable approach that we use is to use a binary hierarchical classification structure [2] as in Figure I. This structure makes a coarse separation between classes in the beginning (at upper levels) in the hierarchy, and finer decisions are made later at lower levels.…”

Section: Introductionmentioning

confidence: 99%

A support vector hierarchical method for multi-class classification and rejection

Wang

Casasent

2009

2009 International Joint Conference on Neural Networks

Self Cite

View full text Add to dashboard Cite

We address both recognition of true classes and rejection of unseen false classes inputs, as occurs in many realistic pattern recognition problems. We advance a hierarchical binary-decision classifier and produce analog outputs at each node, with values proportional to the class conditional probabilities. This yields a new soft-decision hierarchical classifier (hard decisions are not made at each node) . The hierarchy is designed by our new support vector clustering method, which selects the classes to be separated at each node in the hierarchy. Use of our SVRDM (support vector representation and discrimination machine) classifiers at each node provides generalization and rejection ability. The softdecision SVRDM output allows use of the confidence score for each class at each node; this is shown to improve classification (for true classes) and rejection (for false classes) performance. New aspects of this paper are that we provide remarks on our hierarchical design method, including our hierarchical clustering rule, and discus s the meaning and the use of probabilities in our soft-decision hierarchical SVRDM classifiers. We also provide initial tests results on a new database (COIL) that allows large class problem to be addressed. No prior work considered rejection of false classes on this database.

show abstract

A hierarchical classifier using new support vector machines for automatic target recognition

Cited by 43 publications

References 5 publications

A hierarchical target recognition method based on image processing

A hierarchical target recognition method based on image processing

Constructing Category Hierarchies for Visual Recognition

A support vector hierarchical method for multi-class classification and rejection

Contact Info

Product

Resources

About