An image compression method based on Ramanujan Sums and measures of central dispersion

This paper introduces a simple lossy image compression method based on Ramanujan Sums cq(n) and the statistical measures of numerical data such as mean and standard deviation. The Ramanujan Sum cq(n) has been used in digital signal processing for a variety of applications nowadays. Some of them incl...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Ratio mathematica Ročník 41; s. 53 - 63
Hlavní autoři: Sajikumar, S, Dasan, J, Hema, V
Médium: Journal Article
Jazyk:angličtina
Vydáno: Pescara Accademia Piceno Aprutina dei Velati in Teramo (APAV) 01.12.2021
Accademia Piceno Aprutina dei Velati
Témata:
Algorithms
Compression tests
Digital signal processing
Edge detection
Image compression
Kernels
lossy image compression
Mean
ramanujan sum cq(n)
Standard deviation
Sums
ISSN:1592-7415, 2282-8214
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper introduces a simple lossy image compression method based on Ramanujan Sums cq(n) and the statistical measures of numerical data such as mean and standard deviation. The Ramanujan Sum cq(n) has been used in digital signal processing for a variety of applications nowadays. Some of them include the recently developed image kernels for edge detection, extraction of periodicity from signals, etc. The presented compression algorithm is an extension of the edge detection algorithm using an integer image kernel based on Ramanujan Sums. We propose a block-based compression algorithm that detects edges in the images using this image kernel and then compresses the image by storing kernel operation values, the mean and standard deviation for each block instead of pixel values. The proposed method has the advantage of low computational complexity and shows its ability in fast reconstruction and high compression that can be achieved for different block sizes.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1592-7415
2282-8214
DOI:10.23755/rm.v41i0.683