In the recent scenario, understanding how the brain performs the high-level visual tasks, based on fMRI signals. In the ancient study, the image or stimuli is decoded with noise in fMRI data and it has high computational complexity. we present a novel image reconstruction method to reconstruct the images from the fMRI signals based on the computational model of the Recurrent Neural Network. RNN not only allowed the feed forward and backward is allowed, but also allowed its self-feedback.initially, we extract the vector representation of the images in each layer and second, transform this vector representation to its original image. To visualize perceptual content from the recorded brain activity in the form of voxel / EEG to pixel mapping and to reduce high computational complexity using parallel algorithms.