With the rapid growth of modern technology, ensuring secure and reliable identification has become a major necessity. This thesis presents the design and implementation of a fingerprint recognition system using MATLAB for secure and reliable biometric authentication. Traditional security methods such as passwords, PINs, and ID cards are increasingly vulnerable to unauthorized access, duplication, and hacking. To address these limitations, biometric systems offer a more dependable solution by using unique human characteristics. Among various biometric techniques, fingerprint recognition is widely preferred because of its uniqueness, permanence, accuracy, and cost-effectiveness. The proposed system compares two fingerprint images to determine whether they belong to the same individual using digital image processing techniques. The process begins with image acquisition, where fingerprint images are imported into MATLAB for analysis. Since raw images may contain noise and poor contrast, several pre-processing techniques are applied to improve image quality. These include grayscale conversion, resizing, histogram equalization, and noise reduction. Segmentation is then performed to isolate the fingerprint area, followed by binarization to clearly distinguish ridges and valleys. Further processing involves thinning or skeletonization, which reduces ridge thickness to a single pixel for accurate feature extraction. Distinct fingerprint features are then identified and used in the matching process to calculate a similarity score between images. This score is compared with a predefined threshold to verify fingerprint authenticity. If the score exceeds the threshold, the system declares the fingerprint as valid; otherwise, it is marked invalid. Experimental results show that the system successfully differentiates between matching and non-matching fingerprints under normal conditions. Although performance may decline with poor image quality or incorrect finger placement, the system demonstrates the effectiveness of image processing techniques in biometric recognition. This work provides a strong foundation for future enhancements using advanced algorithms and real-time biometric applications.

This work presented the design and implementation of a fingerprint recognition system using MATLAB, focusing on image processing techniques and biometric authentication. The main objective of the project was to compare two fingerprint images and determine whether they belong to the same person. To achieve this, several processing stages were implemented, including grayscale conversion, resizing, image enhancement, noise removal, segmentation, binarization, and thinning. These steps improved image quality and enabled accurate extraction of fingerprint features such as ridge patterns and structural details. The extracted features were used in the matching stage to calculate similarity between fingerprint images. Based on a predefined threshold, the system classified fingerprints as either VALID or NOT VALID. Experimental results demonstrated that the system successfully identified matching and non-matching fingerprints with satisfactory accuracy under normal conditions. Matching fingerprints produced high similarity scores, while different fingerprints generated low scores, confirming the reliability of the system. The study also highlighted the importance of image enhancement and preprocessing in improving recognition performance. MATLAB proved to be an effective platform for implementing and visualizing the fingerprint recognition process. However, the system may face limitations when dealing with poor-quality images, distortion, noise, or improper finger placement, which can affect matching accuracy. Overall, this dissertation successfully developed a functional fingerprint recognition system and demonstrated the significance of image processing in biometric authentication. The system can serve as a useful prototype for applications such as security systems, attendance monitoring, and access control, with future scope for further improvements and real-time implementation.

[1] Deng et al., “Deep Learning-Based Face Recognition Using Transformer Architectures,” IEEE Transactions on Biometrics, Behavior, and Identity Science, 2026. [2] Kakade et al., “A Comprehensive Study on Multimodal Biometric Authentication Systems,” International Journal of Advanced Computer Science, 2026. [3] Rjab et al., “Blockchain-Based Biometric Authentication in Decentralized Systems,” Computers & Security Journal, 2026. [4] Ma et al., “Gait Recognition Using Deep Learning for Human Identification at a Distance,” Pattern Recognition Letters, 2026. [5] Gopal Varma et al., “Multimodal Biometric Authentication Using AI: Fingerprint, Face, and Voice Integration,” Journal of Artificial Intelligence Research, 2025. [6] Gimba et al., “CNN-Based Multimodal Biometric System for Face and Fingerprint Recognition,” Expert Systems with Applications, 2025. [7] Garg et al., “Multimodal Biometric System Using Fingerprint, Iris, and ECG Signals,” Biomedical Signal Processing and Control, 2025. [8] Al-Refai et al., “Evaluation of Biometric Modalities: Security and Performance Analysis,” Journal of Information Security, 2025. [9] Kumar and Zhou, “Deep Learning Approaches for Fingerprint Recognition,” IEEE Access, 2024. [10] Krivokuca Hahn et al., “iCarB: A Multimodal Biometric Dataset for Real-World Applications,” arXiv preprint arXiv:2411.xxxx, 2024. [11] Meiramkhanov et al., “Hybrid Fingerprint Recognition Using CNN and Gabor Filters,” Applied Sciences, 2024. [12] Al-Dabbas et al., “Face Recognition Using Machine Learning and Deep Learning Techniques,” Engineering, Technology & Applied Science Research, 2024. [13] Zhang et al., “Deep Learning-Based Iris Recognition System,” Pattern Recognition Journal, 2024. [14] Singh et al., “Fingerprint Recognition Using MATLAB-Based Image Processing Techniques,” International Journal of Image Processing, 2023. [15] Sharma et al., “Face Recognition System Using MATLAB and PCA-LDA Techniques,” International Journal of Computer Applications, 2023. [16] Patel et al., “Multimodal Biometric System Using MATLAB: Face and Fingerprint Recognition,” Procedia Computer Science, 2022. [17] Ahmed et al., “Voice Recognition Using Machine Learning for Biometric Authentication,” IEEE Access, 2022. [18] Roy et al., “Iris Recognition Using Image Processing Techniques,” International Journal of Biometrics, 2021.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close