Digital image processing, computer science

Wavelets,sub and coding, multi-resolution expansions.

1. Digital image processing Submitted by k.Abisha II MSc cs Nadar Saraswathi college of arts and science

2. Wavelets Wavelets play a crucial role in digital image processing, offering a powerful tool for image analysis, compression, and reconstruction. Here are some key aspects of wavelets in digital image processing: Image Compression: Wavelet transforms compress images by representing them in a more compact form, reducing the amount of data required to store or transmit them. Image Denoising : Wavelets help remove noise from images by separating the image into different frequency subbands and eliminating noise in the high-frequency subbands Image Enhancement: Wavelets enhance images by adjusting the contrast and sharpness in specific frequency subbands .

3. Image Segmentation: Wavelets aid in image segmentation by identifying edges and boundaries in image. Image Fusion: Wavelets combine multiple images into a single image, useful in applications like medical imaging or satellite imaging. Image Watermarking: Wavelets embed watermarks into images for copyright protection and authentication. Some common wavelet transforms used in digital image processing include: Discrete Wavelet Transform (DWT) Fast Wavelet Transform (FWT) Continuous Wavelet Transform (CWT) Stationary Wavelet Transform (SWT)

4. Wavelet-based techniques offer advantages like: Multi-resolution analysis Spatial and frequency localization Robustness to noise and compression However, wavelet-based methods also have limitations, such as: Computational complexity Selection of optimal wavelet basis **Interpretation of wavelet coefficients** By understanding wavelets and their applications in digital image processing, you can develop more efficient and effective image processing algorithms.

5. Sub and coding In digital image processing, subband coding and wavelet coding are related techniques used for image compression and analysis, Here’s a brief overview: Subband Coding: Divide the image into different frequency subbands using filters. Each subband represents a specific range of frequencies. Code each subband separately using techniques like quantization and entropy coding. Combine the coded subbands to form the compressed image.

6. Wavelet coding: Apply the wavelet transform to the image to decompose it into different frequency subbands . The wavelet transform uses scaling and translation parameters to represent the image at multiple resolutions. Code the wavelet coefficients in each subband using techniques like quantization and entropy coding. Store or transmit the coded wavelet coefficients.

7. Multi-resolution expansions There are several types of multi-resolution expansions used in digital image processing: Wavelet Transform: Decomposes an image into different frequency subbands , representing the image at multiple resolutions. Pyramid Transform: A multi-resolution representation of an image, with each level being a filtered and downsampled version of the previous level. Laplacian Pyramid: A pyramid representation where each level is the difference between the original image and a filtered version. Gaussian pyramid: A pyramid representation where each level is a filtered and downsampled version of the previous level, using a Gaussian filter.

8. Mallat Algorithm: An efficient algorithm for computing the wavelet transform, using a multi-resolution approach. Fast Wavelet Transform (FWT): An efficient algorithm for computing the wavelet transform, using a multi-resolution approach. Discrete Wavelet Transform (DWT): A multi-resolution representation of an image, using a wavelet basis. Multi-resolution expansions offer several benefits, including: Efficient representation: compact representation of images. Multi-scale analysis: Analyze images at multiple scales. Robustness to noise: Reduce noise in images. Flexibility: Allow for adjustable resolution and quality levels.

9. Thank you!!!