Image Compression and Enhancement Using Deep Learning Algorithm

These days, applications like transmission and database storage require image compression. This essay examines and discusses picture compression, the need for compression, basic tenets, classifications of compression, and different algorithms a compression of images. This essay aims to provide a guide for choose one of the widely used image compression techniques constructed using Wavelet, JPEG/DCT, VQ, and fractal methods. We examine and debate the benefits and drawbacks of these techniques for grayscale picture compression provide an experimental comparison using a common image size of 256 x 25. Many highly tuned applications, like geophysics, navigation, nondestructive testing, and medical imaging, which demand precise recoveries of original pictures, require lossless compression. Deep learning-based techniques have increasingly been used in image compression with numerous encouraging outcomes. For addressing the traditional as well as the current issues of image compression, in this work lite module based neural network is proposed. Further, the performance of the algorithm is optimized using the slime mold optimization technique. The performance of the proposed method is compared with the traditional methods of image compression such as CNN and SVM in terms of PSNR, MSE, and SSIM metrics.