Comparative Analysis of Spatial Filters for Noise Reduction in Eye Images using PSNR and SSIM Values

Abstract- Ophthalmic imaging using mobile phones has develop universily because of its affordability and accessibility.However, the noise present in the smartphone-captured images has an effect on accuracy of clinical analysis . This studycalculates the effectiveness of Eighteen spatial filtering algorithms including the classic smoothing filters such as Gaussian,Median, and Bilateral as well as advanced methods like Anisotropic Diffusion, Non-Local Means for noise reduction insmartphone-captured photos displaying placido-disc reflections. Each filter was applied on the image to improve imagequality while maintaining important features in the image. Image quality was objectively assessed using algorithms suchas the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). The result of comparativeanalysis shows that Guided filter offered superior performance in pre-processing for Keratoconus grayscale image,achieving a PSNR of 62.29 dB and an SSIM of 0.9997, demonstrating the functioning in noise reduction while preserving image features , ultimately enhancing the accuracy of keratoconus detection algorithms by providing quality image forclearer, more reliable feature extraction. By applying spatial filtering techniques on smartphone-based ophthalmicimaging , this study has the capacity to improve future telemedicine applications, especially in areas with limited access tospecialised equipment for the same treatment .