Finger Vein Identification based on Feature Extraction using Line Tracking Method over Edge Detection for Improved Accuracy

The aim of the study is to implement finger vein recognition for authorized person identification in security systems for smart homes, industries, and banks. The chosen machine learning techniques are Line Tracking Algorithm and Edge Detection. This phase involves selection of collection of data, training and testing of selected data with suggested classifiers Line Tracking and Edge Detection. For SPSS analysis, the outcomes of the two classifiers are categorized into two groups, each consisting of 20 samples. A G-power pre-test score of 80% and a 95% confidence interval (CI) are used for analysis. Line Tracking Algorithm achieved an accuracy of 93.7220%. Edge Detection achieved an accuracy of 92.1620%. The selected Line Tracking Algorithm showed a statistically significant improvement in digital security compared to the Edge Detection model, with a p-value of 0.004 (p<0.05) in the SPSS statistical analysis. The Line Tracking Algorithm demonstrated a higher accuracy rate (93.7220%) compared to the Edge Detection model (92.1620%). Overall, the study suggests that the Line Tracking Algorithm is more effective for finger vein recognition in security applications compared to the Edge Detection method.