Reconstruction of two dimensional images using deep learning

Super-resolution technologies are one of the tools used in image restoration, which aims to obtain highresolution content from low-resolution images. Super-resolution technology aims to increase the quality of low-resolution image by reconstructing it. Super-resolution applications are used in areas such as face recognition, medical imaging, satellite imaging and video resolution enhancement. Deep neural network models used for single-image super-resolution are quite successful in terms of computational performance. In these models, low-resolution images are converted to high-resolution using methods such as bicubic interpolation, but since the super-resolution process is performed in the high-resolution domain, it adds memory cost and computational complexity. In our proposed model, low-resolution image is given as input to convolutional neural network to reduce computational complexity. In our proposed model, subpixel convolution layer is presented that learns a series of filters to enhance low-resolution feature maps to highresolution images. In our proposed model, we add convolution layers to the efficient subpixel convolutional neural network (ESPCN) model and transfer the feature information of the current layer from the previous layer to the next upper layer to prevent the lost gradient value. In this study, the proposed efficient subpixel convolutional neural network (R-ESPCN) model is remodeled to reduce the time required for the real-time subpixel convolutional neural network to perform super-resolution operations on images. The results show that our method is significantly improved in terms of accuracy and can be applied to deep learning methods in the field of image data processing.