سامانه همایش های دانشگاه آزاد اسلامی واحد نجف آباد

صفحه اصلی / International Conference on Artificial Intelligence; City, Industry and Health

A Novel Adaptive Fuzzy-Based AI Approach for High-Density Salt-and-Pepper Noise Removal in MRI Images: Applications in Digital Health and Clinical Diagnostics

نویسندگان :

Alireza Naghsh¹ Mohammad Ebadi²

1- Department of Electrical Engineering, Na.C., Islamic Azad University, Najafabad, Iran 2- Department of Electrical Engineering, Na.C., Islamic Azad University, Najafabad, Iran

کلمات کلیدی :

Artificial intelligence،Digital MRI image،Fuzzy logic،Healthcare،Watermarking

چکیده :

In this study, we present an adaptive weighted fuzzy-based smart method for the removal of salt-and-pepper (SP) noise from digital MRI images, enhancing diagnostic reliability in healthcare applications. The proposed method begins with a watermarking strategy that accurately distinguishes noisy pixels from healthy ones. For each noisy pixel, a neighborhood window is constructed, and the weight of each healthy pixel within the neighborhood is determined using adaptive fuzzy logic system based on the pixel characteristics. Within this window, the noisy pixel is replaced by the weighted average of the healthy pixel values. In high density noise scenarios, where no healthy pixels are present within the window, subsequent noise removal iteration utilize the mean values of previously replaced pixels to determine an appropriate replacement. Evaluations demonstrate the superiority of this method in effectively reducing noise while preserving critical image details, making it particularly suitable for medical imaging applications where precision and clarity are essential. This approach represents a promising advancement in SP noise reduction techniques, paving the way for improved visualization and analysis under high-density noise conditions. The evaluation of the algorithm demonstrates an average of SSIM as 0.9197 and PSNR as 35.88.