This study introduces a Convolutional Neural Network (CNN) approach for identifying finger-vein patterns within a personal recognition framework. Finger-vein pattern recognition stands out as a highly secure and convenient method for personal authentication. The model proposed in this paper, infrared LED and CCD cameras are utilized to extract the vein patterns of the fingers as they are not visible under visible light. The authentication model drafted consists of pre-processing the image and classification of patterns. In image pre-processing state, Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are utilized for dimensionality reduction and feature extraction, respectively. Whereas Support Vector Machine (SVM) and CNN are employed for pattern classification of images. PCA works to remove noise in discarded dimensions while LDA preserves the crucial features. These features ae then used for pattern grouping and recognition purposes. This CNN model attains a classification accuracy of 98% showcasing better performance as compared to the SVM approach of ANFIS in the existing system.