Image compression and decompression tool using SVD
Project description
Image compression and decompression using SVD
Installation
Install CLI using pipx:
pipx install git+https://github.com/toadharvard/svdi-cli.git
Or
pipx install svdi
Usage example
svdi compress --in-file=images/rafiq.bmp --out-file=images/liquidated.svdi --compression=3 --method=numpy
svdi decompress --in-file=images/liquidated.svdi --out-file=images/rafiq2.bmp
ls -l ./images
Available SVD functions
rsvd
— N Halko, P. G Martinsson, and J. A Tropp. Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions. Siam Review, 53(2):217-288, 2011.numpy
— NumPy'snp.linalg.svd
function.pcafast
— H. Li, G. C. Linderman, A. Szlam, K. P. Stanton, Y. Kluger, and M. Tygert. Algorithm 971: An implementation of a randomized algorithm for principal component analysis. Acm Transactions on Mathematical Software, 43(3):1-14, 2017.pi
— Power iterations method. URL: http://www.cs.yale.edu/homes/el327/datamining2013aFiles/07_singular_value_decomposition.pdf
Available commands
svdi --help
svdi compress --help
svdi decompress --help
SVDI Format
The CLI uses a simple binary format for storing SVD results in .svdi
files.
Each .svdi file consists of a header followed by the SVD results for each channel:
- Header
- Signature: A fixed 4-byte sequence b'SVDI' to identify the file format.
- Dimensions: Three 4-byte unsigned integers representing the shape of the matrices: m, n, and k.
- m: The number of rows in each U matrix.
- n: The number of columns in each Vh matrix.
- k: The size of each S vector, and the number of columns in U and rows in Vh.
- Channel Data
Repeated for each channel (NUMBER_OF_CHANNELS):
- U Matrix: A matrix of size m * k, with elements stored as 32-bit floats in row-major order.
- S Vector: A vector of length k, with elements stored as 32-bit floats.
- Vh Matrix: A matrix of size k * n, with elements stored as 32-bit floats in row-major order.
Licence
See the details of the license in the LICENCE file.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
svdi-1.0.1.tar.gz
(6.1 kB
view details)
Built Distribution
svdi-1.0.1-py3-none-any.whl
(8.8 kB
view details)
File details
Details for the file svdi-1.0.1.tar.gz
.
File metadata
- Download URL: svdi-1.0.1.tar.gz
- Upload date:
- Size: 6.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.0.0 CPython/3.12.3
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | ce6b23f3e1b22692ef7271782a86f9d0cac6bf5fb4e3b66473974dbc75813935 |
|
MD5 | c0fa9a9c65b527a288434aaa7fd9c4bb |
|
BLAKE2b-256 | 86adbd82d23ab39950c9e6d1c58363a4a9dd29a86a4bfd95eae0ba4d45571ed5 |
File details
Details for the file svdi-1.0.1-py3-none-any.whl
.
File metadata
- Download URL: svdi-1.0.1-py3-none-any.whl
- Upload date:
- Size: 8.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/5.0.0 CPython/3.12.3
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | fc125c23d87f803a48f82986dba7e1b56fb87fe2c197bf7d6b633fa474a46903 |
|
MD5 | 4af50c70f92e8ca9da12369735bf2f9b |
|
BLAKE2b-256 | 329c42cec105bff045e6666c5906449384c30ac43cc0cadf650e328f6bdb1bb9 |