PENERAPAN ALGORITME KOMPRESI JPEG DAN METODE FUZZY C-MEANS PADA KOMPRESI CITRA BERBASIS ENTROPI
This entropy Advances in technology which rapidly grows demand a lot of information to be processed, stored and delivered. With the amount of the information grows bigger means the needs for memory to represent the information will also grows bigger. A digital image is an information in the form of...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Theses and Dissertations NonPeerReviewed |
| Published: |
[Yogyakarta] : Universitas Gadjah Mada
2013
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/126599/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=66826 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universitas Gadjah Mada |
| Summary: | This entropy
Advances in technology which rapidly grows demand a lot of information
to be processed, stored and delivered. With the amount of the information grows
bigger means the needs for memory to represent the information will also grows
bigger. A digital image is an information in the form of 2D (two dimension) which
is processed through the visual interpretation by human eye. To save more
information contained in the image therefore image compression is necessary.
JPEG is one of many lossy compression methods used in image
compression. JPEG is also widely known and commonly used nowadays. In this
final project, the author will use the JPEG's compression algorithm to generate
an entropy-based image compression. -based image compression will
be tested on image with 2 different types of initial compression format, .jpg and
.png. Before sample image is compressed by JPEG, it will be grouped into
clusters based on its information with the help of Fuzzy C-Means (FCM). To
speed up the iteration of FCM, the author will do an initialization to determine
cluster's first centroid. Contained information on each cluster's can be determined
based on its entropy, and then JPEG will apply compression with different quality
scale on each cluster.
Final results obtained is complete image formed from all clusters. Where
cluster with low entropy will not be compressed and cluster with high entropy will
be compressed. Based on assumption that a cluster with low entropy contains
important information and a cluster with high entropy contains less important
information. Sample image with .png compression format give out better output
than sample image with .jpg compression format, whether in terms size,
qualitative-based comparison (visual) and quantitative-based comparison
(mathematical calculation). |
|---|