Retraction Notice: Shape Based Feature Extraction for Currency recognition and compare the accuracy and sensitivity with ORB algorithm
S.Koti Niteesh Reddy, S. Prem Kumar, R. Karthikeyan · 2022 International Conference on Cyber Resilience (ICCR) · 2022
Aim: The aim of this research is to recognize the currency using image processing techniques. In this research novel shape based feature extraction method is proposed and the accuracy is compared with (Oriented FAST and Rotated BRIEF) ORB algorithm. Materials and Methods: The performance of the proposed system has been evaluated using a challenging dataset Github with a wide variety of conditions. The research includes two groups. Total sample size of 40 is taken for analysis. The Sample size was calculated using clinical.com by keeping alpha error-threshold by 0.05, 95 % confidence interval, power 80 % for control group (Shape based feature extraction$\mathrm{N}=20$) and study group (ORB algorithm$\mathrm{N}=20$). The analysis shows that the Shape based feature extraction method has an accuracy of 97.48 % and the accuracy of ORB algorithm is 89.85 %. The sensitivity of the Shape based feature extraction method is 97.20 % and the sensitivity of the ORB algorithm is 88.81 %. Significance value is 0.000 (p ¡ 0.05, 2-tailed). Conclusion: In this work it is found that the novel shape based feature extraction method performs significantly better than ORB algorithm in terms of accuracy and sensitivity.