Image Dehazing using Fourier Transform and Bilateral Filtering
Vidhu Chaudhary, Ashish Tiwari · 2024
Image dehazing is a crucial technique in enhancing visual quality and restoring image clarity, particularly in outdoor scenes where atmospheric haze can obscure details. This paper presents a novel approach that integrates Fourier Transform and Bilateral Filtering to effectively remove haze from images. The Fourier Transform is employed to analyze the image in the frequency domain, enabling the identification and manipulation of haze-related frequency components. By selectively filtering these components, we can enhance the visibility of the underlying scene while preserving essential details. Subsequently, Bilateral Filtering is applied to smooth the image while maintaining edge sharpness, resulting in a natural-looking output free from artifacts. The proposed method demonstrates significant improvements in image contrast and visibility compared to traditional dehazing techniques, making it suitable for various applications in photography, remote sensing, and autonomous systems. Experimental results showcase the effectiveness of this hybrid approach, with quantitative metrics and visual comparisons illustrating enhanced performance in both indoor and outdoor environments. In this research paper, we explore the evaluation of a dehazing algorithm's efficacy through quantitative metrics like Peak Signal-To-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). The goal is to assess the algorithm's capability in enhancing image quality by reducing haze. A dataset comprising hazy images and their corresponding dehazed versions is utilized. PSNR quantifies image fidelity by measuring noise introduced during dehazing, while SSIM evaluates structural similarity and perceptual quality. Through comprehensive metric calculations and result analysis, the project provides insights into the algorithm's performance, highlighting its strengths and potential areas for improvement. This evaluation framework serves as a foundation for advancing dehazing techniques aimed at improving visibility and image clarity in various applications.