ПАТЕНТНЫЙ ПОИСК В РФ
НОВЫЕ ПАТЕНТЫ, ЗАЯВКИ НА ПАТЕНТ
БИБЛИОТЕКА ПАТЕНТОВ НА ИЗОБРЕТЕНИЯ

Method and system for personal identification using 3D palmprint imaging

Патентный поиск по классам МПК-8:

Класс G06K9/62 способы и устройства для распознавания с использованием электронных средств


Классы МПК:G06K9/62 способы и устройства для распознавания с использованием электронных средств
Автор(ы): Zhang, David (Hong Kong, HK)
Lu, Guangming (Hong Kong, HK)
Luo, Nan (Hong Kong, HK)
Li, Wei (Hong Kong, HK)
Zhang, Lei (Hong Kong, HK)
Kanhangad, Vivek (Hong Kong, HK)
Патентообладатель(и): The Hong Kong Polytechnic University (Kowloon, HK)
Приоритеты:
подача заявки
22.04.2009
публикация патента
11.09.2012

РЕФЕРАТ (Abstract)

A biometric identification system (30) for identifying a person, the system (30) comprising: an image acquisition module (31) to capture a three-dimensional (3D) image of a palm of the person; a region of interest (ROI) extraction module (34) to extract a 3D subimage from the captured image; and a 3D features extraction module (36) to extract 3D palmprint features from the 3D subimage; wherein the extracted 3D palmprint features are compared to reference 3D palmprint features to verify the identity of the person.
Полный текст Патента US 8265347 + PDF


ФОРМУЛА ИЗОБРЕТЕНИЯ (CLAIMS)

We claim:

1. A biometric identification system for identifying a person, the system comprising: an image acquisition module to capture a three-dimensional (3D) image of a palm of the person; a region of interest (ROI) extraction module to extract a 3D subimage from the captured image; and a 3D features extraction module to extract 3D palmprint features from the 3D subimage; wherein the extracted 3D palmprint features are compared to reference 3D palmprint features to verify the identity of the person, and wherein the ROI extraction module extracts a two-dimensional (2D) subimage from the captured image, and further comprises: a 2D features extraction module to extract 2D palmprint features from the 2D subimage and to generate competitive code maps using the extracted 2D palmprint features, and wherein angular distances between the competitive code maps and reference competitive code maps are calculated in order to identify the person.

2. The system according to claim 1, wherein the 3D palmprint features include surface curvature of major palm lines.

3. The system according to claim 1, further comprising an infrared sensor to detect the presence of a palm to initiate image capture of the palm of the person.

4. The system according to claim 1, further comprising a Liquid Crystal Display (LCD) projector to generate arbitrary stripe patterns onto the surface of the palm to enable acquisition of depth information using active triangulation.

5. The system according to claim 1, wherein the image acquisition module is a charge coupled device (CCD) camera.

6. The system according to claim 1, wherein a matching score is calculated based on the comparison and the matching score is compared to a decision threshold, and if the matching score is greater than the decision threshold of a first decision module, the person is rejected as a fake palm or an impostor, and the identification process is terminated prior to a 2D Gabor feature extraction module extracting 2D palmprint features from the 2D subimage.

7. The system according to claim 1, wherein the ROI is a coordinate system which uses the gaps between the fingers as reference points, and the 3D subimage is a fixed size located at a central part of the palm.

8. The system according to claim 1, wherein the 3D features extraction module generates a curvature map using the extracted 3D palmprint features, and the curvature map is compared to a reference curvature map to determine whether a high correlation exists between the curvature maps in order to verify the identity of the person.

9. A biometric identification system for identifying a person, the system comprising: an image acquisition module to capture a three-dimensional (3D) image of a palm of the person; a region of interest (ROI) extraction module to extract a 3D subimage from the captured image; and a 3D features extraction module to extract 3D palmprint features from the 3D subimage; wherein the extracted 3D palmprint features are compared to reference 3D palmprint features to verify the identity of the person, wherein the 3D features extraction module generates a curvature map using the extracted 3D palmprint features, and the curvature map is compared to a reference curvature map to determine whether a high correlation exists between the curvature maps in order to verify the identity of the person, and wherein the 3D features extraction module uses a normalized local correlation method is used to compare the curvature map to the reference curvature map, using the expression: C=i=-NNj=-NN(Pij-P_)(Qij-Q_)[i=-NNj=-NN(Pij-P_)2][i=-NNj=-NN(Qij-Q_)2] where Pij and Qij are curvature values in the neighborhood of the points being matched in the curvature map and reference curvature map, respectively, and P and Q are the mean curvature values in those neighborhoods.


Предыдущий   патент US   Следующий

Наверх