qiong-hu / digital_image_processing

Practices of image deblurring, image inpainting, and image classification.

Digital Image Processing

Project 1: Image Deblur using Deconvolution Algorithm

Content

• Apply naive deconvolution, Wiener deconvolution on blurred in frequency domain
• Estimate frequency-dependent Signal-Noise Ratio (SNR) by analyzing the power spectral density of images
• Experiment with different approximation of SNR function for Wiener Filter algorithm
• Evaluate quantitatively the performance of different deconvolution algorithms

Project 2: Image Inpainting using Sparse Coding and Dictionary Learning

Content

• Use Orthogonal Matching Pursuit (OMP) to solve sparse coding problem
• Learn sparse dictionary using K-SVD algorithm
• Implement these two algorithms for an image inpainting task, which aims to remove extra text content on an image
• Evaluate quantitatively the reconstruction performance

Project 3: Image Classification using Convolutional Neural Networks (CNN)

Content

• Implement data augmentation methods when training a convolutional neural networks (CNN) on images
• Design a CNN pipeline for multi-class image classification task with 6 layers, which is only consisted of convolutional layers and fully-connected layers
• User rectified linear unit (ReLU) as the activation function and cross entropy loss as the loss function
• Evaluate the classification performance