Radon and discrete cosine transforms based feature extraction and dimensionality reduction approach for face recognition

Jadhav, Dattatray V.; Holambe, Raghunath S.

doi:10.1016/j.sigpro.2008.04.017

Cited by 40 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the work of Magli et al [6], a postprocessing method was presented based on wavelet and Radon transforms for detecting four squares. In other research such as [7][8][9], the Radon transform was used for face recognition and compared to pattern recognition methods.…”

Section: Introductionmentioning

confidence: 99%

Estimating facial angles using Radon transform

Bakhshali¹,

Shamsi²

2015

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Analysis of human facial images has become increasingly important due to its numerous applications. In this regard, extracting facial parameters is vital and various studies have been done in this field. Using these methods, different facial parameters such as the location of eyebrows, eye length, and nose angles are extracted. In this article, a robust and automatic method is introduced for determining facial angles from profile view images using the Radon transform. The Radon transform is a type of linear integration along a specific direction and angles play an important role in its performance. The proposed algorithm not only has good precision, but also efficient performance. The precision of this method in angle measurements is 98.51% compared with manual measurement. Compared to various other angle estimation algorithms, the proposed method has higher efficiency.

show abstract

Section: Introductionmentioning

confidence: 99%

Estimating facial angles using Radon transform

Bakhshali¹,

Shamsi²

2015

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…They are used especially in speech and image transform coding [3,4], DCT-based subband decomposition in speech and image compression [5], or video transcoding [6]. Other important applications are: block filtering, feature extraction [7], digital signal interpolation [8], image resizing [9], transform-adaptive filtering [10,11], and filter banks [12].…”

Section: Introductionmentioning

confidence: 99%

Novel VLSI Algorithm and Architecture with Good Quantization Properties for a High-Throughput Area Efficient Systolic Array Implementation of DCT

Chiper

Ungureanu

2010

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

Using a specific input-restructuring sequence, a new VLSI algorithm and architecture have been derived for a high throughput memory-based systolic array VLSI implementation of a discrete cosine transform. The proposed restructuring technique transforms the DCT algorithm into a cycle-convolution and a pseudo-cycle convolution structure as basic computational forms. The proposed solution has been specially designed to have good fixed-point error performances that have been exploited to further reduce the hardware complexity and power consumption. It leads to a ROM based VLSI kernel with good quantization properties. A parallel VLSI algorithm and architecture with a good fixed point implementation appropriate for a memory-based implementation have been obtained. The proposed algorithm can be mapped onto two linear systolic arrays with similar length and form. They can be further efficiently merged into a single array using an appropriate hardware sharing technique. A highly efficient VLSI chip can be thus obtained with appealing features as good architectural topology, processing speed, hardware complexity and I/O costs. Moreover, the proposed solution substantially reduces the hardware overhead involved by the pre-processing stage that for short length DCT consumes an important percentage of the chip area.

show abstract

“…For example, a method which combine RT and singular value decomposition (SVD) was proposed for image recognition in [8], a feature extraction and dimensionality reduction approach for face recognition based on RT and DCT was presented in [9]. But the precision of RT is dependent on the number of projection in different direction, and RT is redundant and difficult to be perfectly reconstructed.…”

Section: Introductionmentioning

confidence: 99%

Study of Finite Radon Transform in Face Recognition

Zhang

Wang

2010

2010 Second International Conference on Computer Modeling and Simulation

View full text Add to dashboard Cite

A novel feature extraction method based on finite Radon transform (FRAT), soft threshold (ST), and two dimensional maximum margin criterion (2DMMC) was proposed. The framework of the proposed method consists of three steps: First, each face image was projected into FRAT domain by FRAT; then ST was used to select main FRAT coefficients; finally, 2DMMC was used to extract features for face classification from the main FRAT coefficients. The experimental results on Yale and ORL face databases show that the proposed method is feasible and efficient.

show abstract

Radon and discrete cosine transforms based feature extraction and dimensionality reduction approach for face recognition

Cited by 40 publications

References 15 publications

Estimating facial angles using Radon transform

Estimating facial angles using Radon transform

Novel VLSI Algorithm and Architecture with Good Quantization Properties for a High-Throughput Area Efficient Systolic Array Implementation of DCT

Study of Finite Radon Transform in Face Recognition

Contact Info

Product

Resources

About