An AI-Driven Deep Autoencoder-Based Image Compression Framework for Optimizing Storage and Retrieval of Digital Content in Information Systems
Main Article Content
Abstract
In today's information systems, digital content has grown so much that storing data, transferring it and accessing multimedia resources in a timely fashion have become a problem. To meet these challenges, image compression is a key component in minimizing storage space needed while maintaining reasonable image quality. The study has introduced an AI-based Deep Autoencoder Based Image Compression Framework that can optimize storage and retrieval of digital content in information systems. The framework is based on a deep autoencoder network, which learns compact and meaningful representations of image data, allowing for efficient compression and reconstruction. The proposed method differs from the existing image compression methods as it uses artificial intelligence to automatically extract salient image features and reduce redundant information to gain benefits in terms of compression without sacrificing much in terms of visual quality. Typical test image sets and common performance measures, such as Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), Structural Similarity Index Measure (SSIM), Compression Ratio (CR), and Bits Per Pixel (BPP), are used for evaluating the effectiveness of the framework. The experimental results show that the proposed deep autoencoder model outperforms the traditional compression schemes in terms of compression efficiency with good image quality and structural similarity. Besides that, the structure can improve the storage efficiency and provide quicker access to digital information, thus applicable to large scale information systems handling large multimedia repositories. Incorporating AI into image compression enhances the scalability, resource utilization, and accessibility of content within large data sets. The proposed framework can be effectively used in a digital library, educational repository, cloud-based storage system, enterprise information systems, and multimedia content management system. The results show that using deep learning for image compression is a promising approach to meet the challenge of efficiently managing digital information, and is also important for the sustainable and intelligent operation of information systems in the age of massive digital data. 1