Method for Image Restoration from Out-of-Focus Light with Code Implementation

Method for Image Restoration from Out-of-Focus Light, where light diffraction causes image distortion, which is then computationally reversed. Achieving this requires implementing a series of complex algorithms and techniques. Initially, blurred image analysis and preprocessing (e.g., using MATLAB's imgaussfilt for noise reduction) determine the degree and type of distortion. Subsequently, diffraction theory and image restoration algorithms—such as Wiener deconvolution or Richardson-Lucy deconvolution—are applied to reconstruct the original image. These algorithms operate by modeling light propagation paths and diffraction effects through point spread function (PSF) estimation, often implemented with Python's scikit-image library or MATLAB's deconvblind function. Advanced techniques like deep learning and convolutional neural networks (CNNs) can further enhance accuracy, using architectures like U-Net trained on paired blurred/ sharp images with TensorFlow or PyTorch. Thus, image restoration from defocused light remains a challenging domain demanding continuous research into optimized PSF modeling and real-time processing feasibility.