Financial forecasting using computational intelligence nowadays remains a hot topic. Recent improvements in deep neural networks allow us to predict financial market behavior. In our work we first implement a novel approach of , which converts financial time-series data to 2-D images and then feeds the generated images to a convolutional neural network as an input. We then hypothesize that the performance of the model can be improved using different techniques. Specifically, in our work, we improve the computational and financial performance of the previous approach by 1) fine-tuning the neural network hyperparameters, 2) creating images with 5 channels corresponding to indicator clusters, 3) improving financial evaluation using take profit and stop loss techniques, 4) evolutionary optimized parameters for trading strategy. The results of this study show that the above-mentioned strategies improve the model considerably. We conclude with future work that can be done in order to further improve the computational and financial performance of the model.