Super Resolution for Noisy Images Using Convolutional Neural Networks

The images in high resolution contain more useful information than the low resolution. Thus, high-resolution digital are preferred over low-resolution images. Image super-resolution is one of principal techniques for generating major advantages methods that they economical, independent image capture devices, and can be statically used. In this paper, a single-image network model based on convolutional neural networks proposed by combining conventional autoencoder residual approaches. A network-based dictionary method used to train input addition, linear refined unit thresholds output provide better dictionary. Autoencoders aid removal noise from enhancement their quality. Secondly, processes it further create image. experimental results demonstrate outstanding performance our compared other traditional methods. produces clearer detailed images, as important real-life applications. Moreover, has advantage learning, enhancement, removal. Furthermore, training with improved preprocessing creates an efficient versatile network.