Opening Up the Next Generation of Convolutional Neural Networks (CNN) for Palm Recognition
Dilip Sisodia, Charu Vaibhav Verma, Garima Joshi, P. Kaliyamoorthi, Ankur Gupta, Sabyasachi Pramanik, Faiz Mohammad · Advances in computational intelligence and robotics book series · 2024
A subset of biometrics called palm recognition has drawn a lot of interest because of its potential applications in a variety of fields, including as security systems and human-computer interaction. Convolutional neural networks (CNNs) are being used in this effort to look at the creation of a palm detecting system. In order to identify and classify palm orientation in digital photos, deep learning will be used. An essential part of the study is data collection, which comprises compiling a large dataset of accurately labeled palm photos for supervised learning. The authors handle data preparation and use methods like image scaling, pixel normalization, and data augmentation to get datasets ready for model training. The system is based on a CNN model architecture, which creates a neural network with the ability to intelligently classify and automatically gather data from palm photos. Convolutional layers are used in this architecture to extract features, while fully connected layers are used for classification.