This project focuses on improving the accuracy of Convolutional Neural Network (CNN) models through hyperparameter optimization using Particle Swarm Optimization (PSO) and Genetic Algorithms (GA). By leveraging these optimization techniques, the study aims to find the best combination of hyperparameters that enhance the performance of CNN models. PSO and GA are employed to explore the hyperparameter space efficiently, adjusting parameters such as learning rate, batch size, number of layers, and other critical settings. The project involves extensive experimentation and analysis to compare the effectiveness of PSO and GA in achieving optimal CNN accuracy. The findings contribute to the field of deep learning by providing insights into advanced optimization strategies for neural networks..