Place: Large Lecture Room – CVC
Affiliation: Instituto Universitario para el Desarrollo Tecnológico y la Innovación en Comunicaciones, ULPG, Spain
The aim of this seminar is to introduce two biometrics devices researched in our research institute (Instituto Universitario para el Desarrollo Tecnológico y la Innovación en Comunicaciones, Universidad de Las Palmas de Gran Canaria).
The contactless hand based biometric identification system uses geometric and palm features. Hand images are acquired using two commercial webcams with 1200×1600 pixel resolution. One of the webcams has been modified to operate in the near infrared band (NIR). Images acquired in the NIR and visible band are used for obtaining the geometric and palm features respectively.
Different ways of correlate both images will be discussed:
- Acquiring the image using a cold mirror or using active shape models.
- The palm features were obtained combining a Orthogonal Line Ordinal filter and Scalar Invariant Feature Transform (SIFT).
- Different classifiers are used for each feature: Support Vector Machine, Minimum Hamming distance. Score combination strategies are also discussed.
- Evaluation methods will be also commented using more than 8000 hand images from three different databases.
- Refering to the off-line handwritten signature verifier, we propose a method for taking into account the gray level of the signature strokes using statistical texture features.
- The co-occurrence matrix and local binary pattern are analysed and used as features.
- Genuine samples and random forgeries have been used to train an SVM model and random and skilled forgeries have been used for testing it.