ARCHIVES

Abstract: The quick Rise of various online Platforms has resulted in a massive influx of data, large in the form of photographes and vidéos. Uncompressed MultiMedia data, such as graphics, audio, and video, requiers a significant amount of Storage space and bandwidth. transmission. The need for appropriate image compression to combat This excessive data transmission is critical. It has become unavoidable to use approches. Image compression is the process of reducing the size of an image in order to Save space. Storage and transmission that Is useful Various approaches have emerged to overcome this challenge, but none of them had been successful. The rebuilt image suffers severe data loss, which is a big disadvantage. To deal with the situation This study promotes the value of deep Learning by establishing a Framework. Convolutional Autoencoder is a type of autoencoder that uses a Convolutional To develop a better picture compression model, a convolutional autoencoder model with 20 different layers and filters was created. This unsupervised machine learning technique compresses images using back propagation and reconstructs the input image with the least amount of loss. A new instance has been added to the architecture to do denoising with the least amount of dataloss possible. The architecture has demonstrated its success in image compression and denoising, but It has also cleared the way for additional research into the model's improvement in terms of compression factor and data loss reduction in higher- dimensional images.