Iterative Framework For Data Augmentation Of Segmented Fingerprints
Researchers have developed a new data augmentation technique to address the scarcity of infant fingerprint data for biometric systems. The iterative framework uses a convolutional neural network to intentionally introduce variations in segmented fingerprints, creating diverse examples for training. Experiments show this method effectively expands fingerprint variability while maintaining visual resemblance to original prints, paving the way for improved infant biometric matching systems. AI
IMPACT Enhances AI's ability to process and match biometric data from limited infant datasets.