The data’s dimensionality reduction improves in the model’s construction with less manual labor and speeds up the learning and generalization phases of the machine learning process. LDA strives to reduce spreading inside the class while improving the distance between each class’s means. LDA assesses both within and between classes, LDA performs well by maximizing the distance between the means of each class, which results in improved classification outcomes. The objective of the research paper is to implement the dimensionality reduction technique to digital image processing by performing feature extraction in 2D, deploying Linear discriminant analysis. Initially feature extraction is executed which contributes to gather large set of data value. Consequently, the obtained vector values are stored in the database and the dimension of the same is reduced using linear discriminant analysis.