Title:	FEASIBILITY STUDY AND EMPIRICAL ANALYSIS OF A LOW-COST FINGERPRINT RECOGNITION FOR IMMUNIZATION TRACING
Author(s):	Esther Mukoya, Richard Rimiru and Michael Kimwele
ISBN:	978-989-8704-34-4
Editors:	Pedro Isaías and Hans Weghorn
Year:	2021
Edition:	Single
Type:	Full
First Page:	117
Last Page:	125
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Many Fingerprint recognition systems mostly use minutiae features. Feature extraction module, a critical stage in fingerprint recognition, is mainly dependent on the image quality. Practically, the process of obtaining a good quality fingerprint image can be challenging. This is so especially with children fingerprint images collected using standard fingerprint devices. Children fingerprint recognition can be applied in immunization clinics to trace children attending the immunization. This will help to create effective follow-ups especially for children who are likely to miss their immunizations. Given that children fingerprints are of lesser quality, especially when collected via standard equipment, a system that can improve the quality and hence make the image recognizable is ideal. In this work, the authors focus on children fingerprint images which comprise of well-known distortions such as wetness, small area and blurriness. An accurate and efficient fingerprint feature extraction from children's fingerprints which are usually of poor-quality fingerprint image is a difficult task. Fingerprint enhancement technique is introduced using residual dense network to perform image Super resolution. Local features are extracted using dense connected convolutional layers. This is then followed by image filtering. The output of the processed image has a black background with the bifurcations shown in white. In this research, the image was inverted to obtain the final output with white background and black lines. Quality scores were obtained using the National Institute of Standards and Technology fingerprint Image Quality algorithm to determine the quality improvement after the image enhancement and the method provides better results. Experimental results showed that the enhancement improved the visual clarity of the low-quality images. Thus, with ideal enhancement, children's fingerprints can be applied in follow-ups of immunization clinics.
	Go Back