@article{CRSIJ26000106,
                  author = {Udeagbala Remigius Ndidika , Egbonwonu Emmanuel Livinus, Ezema D. C., Ogbodo Ikechukwu Ogbodo, Okika Stephen S.},
                  title = {Performance Analysis of Coiflet Wavelet and Wavelet Packet Transform for Biometric Fingerprint Image Compression },
                  journal = {Cosmo Research and Science International Journal},
                  year = {2025},
                  volume = {1},
                  number = {5},
                  pages = {252-270},
                  issn = {3108-1584},
                  url = {https://cosmorsij.com/published/CRSIJ26000106.pdf},
                  abstract = {Fingerprint image is one of the most acceptable means of biometric capturing, due to its unique features and convenient capturing process. The rapid increase in world population has led to the increase in population of people needed to be captured. Similarly, substantial storage, transmission and computation costs are obvious, thus their compression is advantageous to reduce these requirements. Coiflet wavelet is one of the best signal transformation and filtering technique. This work presents a carefully designed approach to biometric fingerprint image compression and performance analysis in wavelet and wavelet packet transform. This was achieved through the determination of the percentage Retain Energy (RE) and Number of Zeros (NZ) for different levels Coiflet-type wavelets at different threshold values of 235, 245 and 255, at fixed decomposition level 3 using wavelet and wavelet packet transform. 8-bit grayscale right thumb digitized image of size 380×400 was used for the experiment. The result shows that, at the first threshold value, Wavelet Transform (WT), RE (%) values increased from 96.75% to 98.87% while that of NZ, decreases from 97.53% to 96.67% across the five levels of coiflet wavelet. At the second threshold value, RE increased from 96.64% to 98.83% while NZ decrease from 97.58% to 96.70% across the five levels of the coiflet. At the third threshold value, RE increases from 96.53% to 98.79%, while NZ deceases from 97.63% to 96.73%. Then for WPT, at the first threshold value, RE increased from 97.02% to 98.80% and Nz also increased from 97.69% to 98.01% across the five levels of the coiflet. At the second threshold value, RE increased from 96.90% to 98.76%, also NZ increased from 97.75% to 98.03%. At the third threshold value, RE increased from 96.76% to 98.72%, also NZ increased from 97.80% to 98.06% across the five levels of the coiflet-type wavelet. This means that the Wavelet Packet Transform (WPT) has more compression rate than Wavelet Transform (WT), thereby requiring less storage space and overall cost. },
                  keywords = {Wavelet Transform, Wavelet Packet Transform, Retained Energy, Number of Zero, Biometric system, fingerprint image compression, coiflet wavelet.},
                  month = {March}
        }