• Home
  • Archive
  • Tools
  • Contact Us

The Customize Windows

Technology Journal

  • Cloud Computing
  • Computer
  • Digital Photography
  • Windows 7
  • Archive
  • Cloud Computing
  • Virtualization
  • Computer and Internet
  • Digital Photography
  • Android
  • Sysadmin
  • Electronics
  • Big Data
  • Virtualization
  • Downloads
  • Web Development
  • Apple
  • Android
Advertisement
You are here: Home » Guetzli : A New JPEG Encoder For Higher Compression

By Abhishek Ghosh March 27, 2017 9:31 am Updated on March 27, 2017

Guetzli : A New JPEG Encoder For Higher Compression

Advertisement

Google’s new algorithm promises to reduce the size of JPEG images by 35% without losing quality
Guetzli uses the JPEG format. So, softwares which can currently read standard JPEGs will also be able to read Guetzli JPEGs without any other modification. Guetzli Is A New JPEG Encoder For Higher Compression Developed By Google. Released free software generates ~35% smaller images than libjpeg. We are not going in to in depth technical details about the technology as readers can read the official Google blog with visual examples :

Vim
1
https://research.googleblog.com/2017/03/announcing-guetzli-new-open-source-jpeg.html

Guetzli is an algorithm very similar to Zopfli, which is another project also developed by Mountain View but for the compression of PNG and gzip file format.

Guetzli uses a large amount of memory, around 300MB of memory per 1 megapixel of the input image and a significant amount of CPU time – about 1 minute of CPU per 1 megapixel of input image.

Advertisement

---

Guetzli - A New JPEG Encoder For Higher Compression

 

Guetzli : A New JPEG Encoder For Higher Compression

 

As developer, you will be interested to check the Github project and try :

Vim
1
https://github.com/google/guetzli

On Mac, unless you want the build by hand, with Homebrew you can easily install by running the command :

Vim
1
brew install guetzli

which will return this output :

Vim
1
2
3
4
5
6
7
8
9
10
11
==> Installing dependencies for guetzli: libpng
==> Installing guetzli dependency: libpng
==> Downloading https://homebrew.bintray.com/bottles/libpng-1.6.29.sierra.bottle
######################################################################## 100.0%
==> Pouring libpng-1.6.29.sierra.bottle.tar.gz
?  /usr/local/Cellar/libpng/1.6.29: 26 files, 1.2MB
==> Installing guetzli
==> Downloading https://homebrew.bintray.com/bottles/guetzli-1.0.1.sierra.bottle
######################################################################## 100.0%
==> Pouring guetzli-1.0.1.sierra.bottle.tar.gz
?  /usr/local/Cellar/guetzli/1.0.1: 5 files, 229.2KB

Example command to convert image is :

Vim
1
guetzli [--quality Q] [--verbose] original.png output.jpg

There is no manual entry for guetzli. As example, I ran :

Vim
1
guetzli --quality 85 --verbose 14.jpg 14guetzli.jpg

which returned this output :

Vim
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
riginal Out[ 158254] BA[100.00%] D[0.0000] Score[158254.0000] (*)
Iter  1: f111111 quantization matrix:
  1  1  1  1  1  1  1  1     1  1  1  2  2  3  3  3     1  1  1  2  2  3  3  3
  1  1  1  1  1  1  1  1     1  1  1  2  3  3  3  3     1  1  1  2  3  3  3  3
  1  1  1  1  1  1  1  2     1  1  1  3  3  3  3  3     1  1  1  3  3  3  3  3
  1  1  1  1  1  1  2  2     2  2  3  3  3  3  3  3     2  2  3  3  3  3  3  3
  1  1  1  1  1  2  2  3     2  3  3  3  3  3  3  3     2  3  3  3  3  3  3  3
  1  1  1  1  2  2  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  1  1  1  2  2  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  1  1  2  2  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
Iter  1: f111111 GQ[ 7.08] Out[ 158254] BA[100.00%] D[0.0000] Score[158254.0000]
Iter  2: f111111 quantization matrix:
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
  3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3     3  3  3  3  3  3  3  3
Iter  2: f111111 GQ[22.56] Out[ 137554] BA[100.00%] D[0.3788] Score[137554.0000] (*)
Iter  3: f111111 quantization matrix:
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
  5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5     5  5  5  5  5  5  5  5
Iter  3: f111111 GQ[45.11] Out[ 125457] BA[100.00%] D[0.7007] Score[125457.0000] (*)
Iter  4: f111111 quantization matrix:
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
  7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7     7  7  7  7  7  7  7  7
Iter  4: f111111 GQ[67.67] Out[ 115489] BA[100.00%] D[0.9011] Score[115489.0000] (*)
Iter  5: f111111 quantization matrix:
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9     9  9  9  9  9  9  9  9
  9  9  9  9  9  9  9 11     9  9  9  9  9  9  9 11     9  9  9  9  9  9  9 11
Iter  5: f111111 GQ[90.31] Out[ 109336] BA[100.00%] D[1.1391] Score[109336.0000] (*)
Iter  6: f111111 quantization matrix:
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11    11 11 11 11 11 11 11 11
11 11 11 11 11 11 11 13    11 11 11 11 11 11 11 13    11 11 11 11 11 11 11 13
Iter  6: f111111 GQ[112.87] Out[ 105255] BA[100.00%] D[1.3803] Score[105255.0000] (*)
Iter  7: f111111 quantization matrix:
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13    13 13 13 13 13 13 13 13
13 13 13 13 13 13 15 15    13 13 13 13 13 13 15 15    13 13 13 13 13 13 15 15
Iter  7: f111111 GQ[135.52] Out[  99795] BA[100.00%] D[1.6504] Score[99795.0000] (*)
Iter  8: f111111 quantization matrix:
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 17    15 15 15 15 15 15 15 17    15 15 15 15 15 15 15 17
15 15 15 15 15 15 17 17    15 15 15 15 15 15 17 17    15 15 15 15 15 15 17 17
Iter  8: f111111 GQ[158.17] Out[  96721] BA[100.00%] D[2.1032] Score[4472752592435967084701623882088448.0000]
Iter  9: f111111 quantization matrix:
13 13 13 13 15 15 15 15    13 13 13 13 15 15 15 15    13 13 13 13 15 15 15 15
13 13 13 15 15 15 15 15    13 13 13 15 15 15 15 15    13 13 13 15 15 15 15 15
13 15 15 15 15 15 15 15    13 15 15 15 15 15 15 15    13 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
Iter  9: f111111 GQ[146.92] Out[  97700] BA[100.00%] D[1.8280] Score[97700.0000] (*)
 
YUV444 selected quantization matrix:
13 13 13 13 15 15 15 15    13 13 13 13 15 15 15 15    13 13 13 13 15 15 15 15
13 13 13 15 15 15 15 15    13 13 13 15 15 15 15 15    13 13 13 15 15 15 15 15
13 15 15 15 15 15 15 15    13 15 15 15 15 15 15 15    13 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15    15 15 15 15 15 15 15 15
Iter 10: f111111(7) up Coeffs[7511/9355] Blocks[1188/1517/34100] ValThres[1.7390] Out[  92821] EstErr[0.17%] BA[100.00%] D[1.8828] Score[92821.0000] (*)
Iter 11: f111111(7) up Coeffs[1844/1844] Blocks[988/988/34100] ValThres[0.1611] Out[  91712] EstErr[0.19%] BA[100.00%] D[1.8829] Score[91712.0000] (*)

Processed image nicely opened in Mac Preview.

Tagged With paperuri:(0ee59c2b833ed8da2904d2ce30274c88) , guetzli , jpeg encoder

This Article Has Been Shared 879 Times!

Facebook Twitter Pinterest
Abhishek Ghosh

About Abhishek Ghosh

Abhishek Ghosh is a Businessman, Orthopaedic Surgeon, Author and Blogger. You can keep touch with him on Twitter - @AbhishekCTRL.

Here’s what we’ve got for you which might like :

Articles Related to Guetzli : A New JPEG Encoder For Higher Compression

  • Google Plus is losing Traffic while FaceBook, Twitter touches record

    Google Plus is losing both the number of unique visits and time spend on the site, while FaceBook, Twitter touches record.

  • Google April Fools with a Bunch of Pranks

    Google April Fools with a Bunch of Pranks this year and involves kind of easter eggs in their most serious products like every year, here is our analysis.

  • OwnCloud Adds API on Community Edition For Easier Integration

    OwnCloud adds API on community edition for easier integration for the developers to create web applications and mobile applications in easier way.

  • Capabilities Required by the Professionals in the Era of Cloud Computing

    Capabilities required by the professionals in the era of Cloud Computing is not limited to IT sector but it includes the Health, Law and other professionals.

  • Secure Hash Algorithm 1 (SHA-1) Circumvented by Google

    SHA 1 is a Cryptographic Hashing Function. Secure Hash Algorithm 1 (SHA-1) is Circumvented by Google Research For Testing Vulnerability.

Additionally, performing a search on this website can help you. Also, we have YouTube Videos.

Take The Conversation Further ...

We'd love to know your thoughts on this article.
Meet the Author over on Twitter to join the conversation right now!

If you want to Advertise on our Article or want a Sponsored Article, you are invited to Contact us.

Contact Us

Subscribe To Our Free Newsletter

You can subscribe to our Free Once a Day, Regular Newsletter by clicking the subscribe button below.

Click To Subscribe

Please Confirm the Subscription When Approval Email Will Arrive in Your Email Inbox as Second Step.

Search this website…

 

Popular Articles

Our Homepage is best place to find popular articles!

Here Are Some Good to Read Articles :

  • Cloud Computing Service Models
  • What is Cloud Computing?
  • Cloud Computing and Social Networks in Mobile Space
  • ARM Processor Architecture
  • What Camera Mode to Choose
  • Indispensable MySQL queries for custom fields in WordPress
  • Windows 7 Speech Recognition Scripting Related Tutorials

Social Networks

  • Pinterest (21K Followers)
  • Twitter (5.3k Followers)
  • Facebook (5.7k Followers)
  • LinkedIn (3.7k Followers)
  • YouTube (1.3k Followers)
  • GitHub (Repository)
  • GitHub (Gists)
Looking to publish sponsored article on our website?

Contact us

Recent Posts

  • What is Standard Software February 28, 2021
  • WordPress Link to text Fragment February 27, 2021
  • How to Protect IP Cameras From Hackers February 25, 2021
  • 6 Sectors That Have Undergone Revamps in Digital Landscape February 24, 2021
  • What You Can Control with a Smartwatch and ESP32 February 23, 2021

 

About This Article

Cite this article as: Abhishek Ghosh, "Guetzli : A New JPEG Encoder For Higher Compression," in The Customize Windows, March 27, 2017, March 1, 2021, https://thecustomizewindows.com/2017/03/guetzli-new-jpeg-encoder-higher-compression/.

Source:The Customize Windows, JiMA.in

 

This website uses cookies. If you do not want to allow us to use cookies and/or non-personalized Ads, kindly clear browser cookies after closing this webpage.

Read Cookie Policy.

PC users can consult Corrine Chorney for Security.

Want to know more about us? Read Notability and Mentions & Our Setup.

Copyright © 2021 - The Customize Windows | dESIGNed by The Customize Windows

Copyright  · Privacy Policy  · Advertising Policy  · Terms of Service  · Refund Policy