This manuscript presents, palmprint recognition by combining different texture extraction approac... more This manuscript presents, palmprint recognition by combining different texture extraction approaches with high accuracy. The Region of Interest (ROI) is decomposed into different frequencytime sub-bands by wavelet transform up-to two levels and only the approximate image of two levels is selected, which is known as Approximate Image ROI (AIROI). This AIROI has information of principal lines of the palm. The Competitive Index is used as the features of the palmprint, in which six Gabor filters of different orientations convolve with the palmprint image to extract the orientation information from the image. The winner-take-all strategy is used to select dominant orientation for each pixel, which is known as Competitive Index. Further, PCA is applied to select highly uncorrelated Competitive Index features, to reduce the dimensions of the feature vector, and to project the features on Eigen space. The similarity of two palmprints is measured by the Euclidean distance metrics. The algor...
This work proposes Block-wise uniform local binary pattern histogram (BULBPH) followed by kernel ... more This work proposes Block-wise uniform local binary pattern histogram (BULBPH) followed by kernel discrimination analysis (KDA) as descriptor for palmprint recognition. BULBPH provides distribution of uniform patterns (such as line and wrinkles) in local region and can be better used as palmprint features. KDA is applied on BULBPH to reduce dimension and enhance discriminative capability using chi-RBF kernel. The experiments are conducted on four palmprint databases and performance is compared with related descriptors. It is observed that KDA on BULBPH descriptor achieves more than 99% accuracy with 4.04 decidability index on four palmprint databases.
Rotation and noise invariant feature extraction is a challenge in palmprint recognition. This wor... more Rotation and noise invariant feature extraction is a challenge in palmprint recognition. This work presents a novel RDF descriptor based on Radon, Dual tree complex wavelet, and Fourier transforms. Combined properties of these transforms help to explore efficiency and robustness of RDF descriptor for palmprint identification. Radon transform can capture directional features of the palmprint and is robust to additive white Gaussian noise also. It converts rotation into translation. 1D Dual tree complex wavelet transform (DTCWT) applied on Radon coefficients in angle direction removes translation in Radon coefficients due to palmprint rotation. The magnitude of 2D Fourier transform performed on resultant coefficients helps to extract rotation and illumination invariant features. The performance of the proposed RDF descriptor is evaluated on noisy and rotated palmprints upto 10∘. Trained with normal palmprints only, the proposed system gives good results for rotated and noisy palmprints. Experiments are performed on PolyU 2D, CASIA, and IIITDMJ databases. Theoretical foundations and experimental results show the robustness of RDF descriptor against additive white noise and rotation.
2014 9th International Conference on Industrial and Information Systems (ICIIS), 2014
This paper presents a multispectral palmprint recognition approach based on palm line orientation... more This paper presents a multispectral palmprint recognition approach based on palm line orientation feature extracted with high order steerable filter. Gaussian function is used as isotropic window to design a high order steerable filter. The orientation features are selected as per dominant filter response for a particular orientation. Optimum values for parameters, i.e., standard deviation and number of orientations are found experimentally in order to obtain low equal error rate (EER) and high correct identification rate (CIR). Weighted score level fusion strategy is applied to combine the score of all spectral palmprints. A recognition rate of 99.97% is achieved with high decidability index (DI) and low EER. Further, the proposed approach is compared with traditional competitive code method for multispectral PolyU palmprint database.
2011 International Conference on Signal Processing, Communication, Computing and Networking Technologies, 2011
Abstract Among the various palmprint recognition techniques, competitive coding technique is very... more Abstract Among the various palmprint recognition techniques, competitive coding technique is very effective and gives efficient results. In the competitive coding method, gabor filters of six different orientation convolve with the palmprint image to extract the orientation ...
ABSTRACT The proposed work aims to improve the performance of palmprint recognition system by red... more ABSTRACT The proposed work aims to improve the performance of palmprint recognition system by reducing the size of templates. Discrete Wavelet Transform assists in scaling down the size of extracted Region of Interest of palmprint approximately by a factor of 16, while preserving the potential information used to discriminate among different palms. This work proposes XOR-SUM Code, which is based on the fusion of the real and the imaginary Palm Code images for different orientations (say, $N$ ). The fused images of all orientations are added together to get line features of palmprint. The resulting image is then coded into $\left\lceil (N+1)/2 \right\rceil $ bits to get line features of reduced size. Experiments are carried out on Hong Kong PolyU palmprint database that contains 8000 palmprint images of 400 different palms. XOR-SUM Code technique is compared with Palm Code, Competitive Code, and Contourlet Transform-based techniques in terms of Equal Error Rate (EER), Decidability Index (DI), and Genuine Acceptance Rate (GAR). The XOR-SUM Code technique results in EER, DI, and GAR of 1.83 %, 4.4433, and 98.17 % respectively.
Abstract. A method for palmprint verification and identification, which is robust to blurred and ... more Abstract. A method for palmprint verification and identification, which is robust to blurred and occluded palmprints is proposed. An algorithm is proposed for region of interest (ROI) extraction that aligns each palmprint ROI on the same position. First-level decomposition of the ROI using a Haar wavelet gives an approximation ROI (AROI) and reduces computational overhead. Local phase quantization (LPQ) based on quantization of phase information by local Fourier transform is applied on AROI to obtain blur invariant palmprint features. The LPQ image is divided into M×M nonoverlapping blocks, and histograms of these blocks are considered as features. The entropy of each block is used to identify it as occluded or nonoccluded. The average chi-square distance between corresponding nonoccluded blocks is calculated for measuring similarity. Experiments performed on PolyU 2D, CASIA, IITD, and IIITDMJ palmprint databases show that the proposed system is independent of acquisition device. Moreover, to justify the robustness of the proposed system, experiments are performed with variable amounts of blur and occlusion present in test palmprints. The performance of the proposed system is compared with several local feature-based palmprint recognition techniques on normal as well as occluded palmprints.
Abstract This paper presents an algorithm to extract the region of interest (ROI) from the palm p... more Abstract This paper presents an algorithm to extract the region of interest (ROI) from the palm print image of the Hong Kong PolyU large-scale palm print database (version 2). Competitive coding method is used for feature extraction. Coding based methods are ...
This manuscript presents, palmprint recognition by combining different texture extraction approac... more This manuscript presents, palmprint recognition by combining different texture extraction approaches with high accuracy. The Region of Interest (ROI) is decomposed into different frequencytime sub-bands by wavelet transform up-to two levels and only the approximate image of two levels is selected, which is known as Approximate Image ROI (AIROI). This AIROI has information of principal lines of the palm. The Competitive Index is used as the features of the palmprint, in which six Gabor filters of different orientations convolve with the palmprint image to extract the orientation information from the image. The winner-take-all strategy is used to select dominant orientation for each pixel, which is known as Competitive Index. Further, PCA is applied to select highly uncorrelated Competitive Index features, to reduce the dimensions of the feature vector, and to project the features on Eigen space. The similarity of two palmprints is measured by the Euclidean distance metrics. The algor...
This work proposes Block-wise uniform local binary pattern histogram (BULBPH) followed by kernel ... more This work proposes Block-wise uniform local binary pattern histogram (BULBPH) followed by kernel discrimination analysis (KDA) as descriptor for palmprint recognition. BULBPH provides distribution of uniform patterns (such as line and wrinkles) in local region and can be better used as palmprint features. KDA is applied on BULBPH to reduce dimension and enhance discriminative capability using chi-RBF kernel. The experiments are conducted on four palmprint databases and performance is compared with related descriptors. It is observed that KDA on BULBPH descriptor achieves more than 99% accuracy with 4.04 decidability index on four palmprint databases.
Rotation and noise invariant feature extraction is a challenge in palmprint recognition. This wor... more Rotation and noise invariant feature extraction is a challenge in palmprint recognition. This work presents a novel RDF descriptor based on Radon, Dual tree complex wavelet, and Fourier transforms. Combined properties of these transforms help to explore efficiency and robustness of RDF descriptor for palmprint identification. Radon transform can capture directional features of the palmprint and is robust to additive white Gaussian noise also. It converts rotation into translation. 1D Dual tree complex wavelet transform (DTCWT) applied on Radon coefficients in angle direction removes translation in Radon coefficients due to palmprint rotation. The magnitude of 2D Fourier transform performed on resultant coefficients helps to extract rotation and illumination invariant features. The performance of the proposed RDF descriptor is evaluated on noisy and rotated palmprints upto 10∘. Trained with normal palmprints only, the proposed system gives good results for rotated and noisy palmprints. Experiments are performed on PolyU 2D, CASIA, and IIITDMJ databases. Theoretical foundations and experimental results show the robustness of RDF descriptor against additive white noise and rotation.
2014 9th International Conference on Industrial and Information Systems (ICIIS), 2014
This paper presents a multispectral palmprint recognition approach based on palm line orientation... more This paper presents a multispectral palmprint recognition approach based on palm line orientation feature extracted with high order steerable filter. Gaussian function is used as isotropic window to design a high order steerable filter. The orientation features are selected as per dominant filter response for a particular orientation. Optimum values for parameters, i.e., standard deviation and number of orientations are found experimentally in order to obtain low equal error rate (EER) and high correct identification rate (CIR). Weighted score level fusion strategy is applied to combine the score of all spectral palmprints. A recognition rate of 99.97% is achieved with high decidability index (DI) and low EER. Further, the proposed approach is compared with traditional competitive code method for multispectral PolyU palmprint database.
2011 International Conference on Signal Processing, Communication, Computing and Networking Technologies, 2011
Abstract Among the various palmprint recognition techniques, competitive coding technique is very... more Abstract Among the various palmprint recognition techniques, competitive coding technique is very effective and gives efficient results. In the competitive coding method, gabor filters of six different orientation convolve with the palmprint image to extract the orientation ...
ABSTRACT The proposed work aims to improve the performance of palmprint recognition system by red... more ABSTRACT The proposed work aims to improve the performance of palmprint recognition system by reducing the size of templates. Discrete Wavelet Transform assists in scaling down the size of extracted Region of Interest of palmprint approximately by a factor of 16, while preserving the potential information used to discriminate among different palms. This work proposes XOR-SUM Code, which is based on the fusion of the real and the imaginary Palm Code images for different orientations (say, $N$ ). The fused images of all orientations are added together to get line features of palmprint. The resulting image is then coded into $\left\lceil (N+1)/2 \right\rceil $ bits to get line features of reduced size. Experiments are carried out on Hong Kong PolyU palmprint database that contains 8000 palmprint images of 400 different palms. XOR-SUM Code technique is compared with Palm Code, Competitive Code, and Contourlet Transform-based techniques in terms of Equal Error Rate (EER), Decidability Index (DI), and Genuine Acceptance Rate (GAR). The XOR-SUM Code technique results in EER, DI, and GAR of 1.83 %, 4.4433, and 98.17 % respectively.
Abstract. A method for palmprint verification and identification, which is robust to blurred and ... more Abstract. A method for palmprint verification and identification, which is robust to blurred and occluded palmprints is proposed. An algorithm is proposed for region of interest (ROI) extraction that aligns each palmprint ROI on the same position. First-level decomposition of the ROI using a Haar wavelet gives an approximation ROI (AROI) and reduces computational overhead. Local phase quantization (LPQ) based on quantization of phase information by local Fourier transform is applied on AROI to obtain blur invariant palmprint features. The LPQ image is divided into M×M nonoverlapping blocks, and histograms of these blocks are considered as features. The entropy of each block is used to identify it as occluded or nonoccluded. The average chi-square distance between corresponding nonoccluded blocks is calculated for measuring similarity. Experiments performed on PolyU 2D, CASIA, IITD, and IIITDMJ palmprint databases show that the proposed system is independent of acquisition device. Moreover, to justify the robustness of the proposed system, experiments are performed with variable amounts of blur and occlusion present in test palmprints. The performance of the proposed system is compared with several local feature-based palmprint recognition techniques on normal as well as occluded palmprints.
Abstract This paper presents an algorithm to extract the region of interest (ROI) from the palm p... more Abstract This paper presents an algorithm to extract the region of interest (ROI) from the palm print image of the Hong Kong PolyU large-scale palm print database (version 2). Competitive coding method is used for feature extraction. Coding based methods are ...
Uploads
Papers by Deepti Tamrakar