CHEYMA Nadir

Aymen AbdElKader CHERIF , Mohammed Salim BENSALEM

A Feature Extraction Method for Iris Recognition System Based on CNN(Transfer Learning)

Bentahar Adel , Djouilem Aboubakar

Deep feature extraction and classification for finger vein images

KHERFI AHMED AYOUB , KOADRI HICHEM

DEEP TRANSFER LEARNING FOR EAR RECOGNITION

Biometrics has emerged as a pervasive technology for secure and non-invasive identity verification. Palmprint recognition, in particular, offers a rich set of creases and texture features that enhance robustness against spoofing. However, despite advances in deep learning, it remains unclear whether off-the-shelf transfer learning yields superior performance on palmprint datasets. In this work, we conduct a comparative study on the Birjand University Mobile Palmprint Database (BMPD) by applying transfer learning with pre-trained convolutional neural networks (VGG16, ResNeXt50, MobileNetV2, and DenseNet121), optimizing hyperparameters via grid search, and evaluating using accuracy, precision, recall, and F1 score. Our experiments show that DenseNet121 with the Adam optimizer achieved the highest accuracy of 99.38% compared to the other models used. However, with the RMSprop optimizer, MobileNetV2 achieved the best accuracy, also with a value of 99.38%. Our study demonstrates that transfer learning techniques can be a good choice for palmprint recognition.

Citation

KHEIRA FAIROUZ Bedjekina , BILAL Attallah , BRIK Youcef , Cheyma nadir , ,(2025-06-11), Palmprint recognition systems using transfer learning techniques: A comparative study,Third national Conference on Materials Sciences And Engineering, (MSE’25),University of Hassiba Benbouali of Chlef (online presentation)

The majority of previous research relied on palm veins, fingerprints, and other biometrics. However, because finger veins are located behind the skin and are both more secure than fingerprint systems and uniquely different for each person, they cannot be used for falsification. This presentation discusses the use of the vgg16 algorithm in finger vein recognition systems. The extensive set of experiments shows that the accuracy achievable with the proposed approach can go beyond the 94% correct identification rate for SDUMLA HMT data.

Citation

Cheyma nadir , ,(2022), Finger vein recognition system using transfer learning,Journée doctorale en électronique 2022,Msila

Finger vein based cnn for human recognition

Citation

Cheyma nadir , BILAL Attallah , BRIK Youcef , ,(2022), Finger vein based cnn for human recognition,ICATEEE2022,M'sila-Algeria

Ear Recognition using Ensemble of Deep Features and Machine Learning Classifiers

Citation

Cheyma nadir , BILAL Attallah , BRIK Youcef , ,(2022), Ear Recognition using Ensemble of Deep Features and Machine Learning Classifiers,ICCTA 2022,Alexandria, Egypt

Recognition System using ear based deep learning features

Citation

Cheyma nadir , ,(2022), Recognition System using ear based deep learning features,ICATEEE2022,M'sila-Algeria