COMPGROUPS.NET | Search | Post Question | Groups | Stream | About | Register

### DLI Image Compression v1.3

• Email
• Follow

Hello all,

DLI v1.3 has been completed and now available for download.  It has 2
major new/updated features.  First is an improved optimize for visual
quality mode.  It compresses up to 10% (avg 3%) more than v1.2 using
MS-SSIM as a measure.  The second major feature is a new "high speed"
visual mode that is fast enough for practical use.  It averages 9%
less compression than the optimum visual mode, while compressing and
decompressing over a thousand times faster.  More details and
comparison results for v1.3 are available on the DLI site.

Thank you,
Dennis

 0

See related articles to this posting

0 Replies
140 Views

Similar Articles

12/6/2013 4:28:38 PM
[PageSpeed]

Similar Artilces:

regular expression help #3 69794
; > 2.3.6 or 1.3.10 and nothing else. (2\.\3\.6|1\.3\.10) You may or may not want a "\b" before and/or after that. > I have tried various combinations but not getting much near > > This expr > [([:digit:]\.[:digit:]\.[:digit:])*]* This is wrong, but I think I understand what you were trying to do. First off, you said you wanted to match nothing but the above numbers. Did you mean that you wanted to match any triplet of dot-separated integers instead? That would be: \d+\.\d+\.\d+ Again, with or without "\b" around it for requiring word-boundaries... ^ 2 >is selecting this and everything else 1. Square parentheses do *not* do grouping; 2. * means match *zero* or more occurrencies of all this; 3. Consider using \d instead of [[:digit:]]. You may want something like \d+\.\d+\.\d+ or some improvement thereof: e.g. you may put limits on the length digit sequences and add checks for word boundaries. Michele -- {$_=pack'B8'x25,unpack'A8'x32,$a^=sub{pop^pop}->(map substr (($a||=join'',map--$|x\$_,(unpack'w',unpack'u','G^<R<Y]*YB=' ..'KYU;*EVH[.FHF2W+#"\Z*5TI/ER

Hello, As I wrote recently I've been on the lookout for old equiptment and decided to buy a C3750 from German ebay. However now I'm in need of a HP-UX 11i v1 media to be able to use it for anything useful. As this is a workstation there should be no problem to pick up another media since there is a license to use it. Do any of your have a spare media which could be re-sold or perhaps know where to find one? I'm currently bidding again at ebay however it's not like there is a lot of these out there. Thanks. -- Henrik

Pronunciation
How do you pronounce "Ada"? Is this pronunciation acceptable? <http://cougar.eb.com/sound/a/ada00001.wav> Perfect pronunciation (NOTE spelling correction, though). ----- Original Message ----- From: "Florian Weimer" <fw@deneb.enyo.de> Newsgroups: comp.lang.ada To: <comp.lang.ada@ada-france.org> Sent: Monday, June 20, 2005 2:54 PM Subject: Pronunciation > How do you pronounce "Ada"? Is this pronunciation acceptable? > > <http://cougar.eb.com/sound/a/ada00001.wav> > _______________________________________________ >

SWIPL + JPL multiple databases
Hi all, I'm trying to initialize multiple prolog engines or databases in JPL What I need is 2 different rule sets for my object in Java. Multiple prolog engines are not implemented is JPL, new consult delets the old one, and if I'm starting a java thread and loading a file from it (consult(rulesetX)) i got memory access errors. Probably there is another way to solve this problem. Any Ideas? Regards, Rolland Rolland, On 2006-11-28, roll x <1.gig.4.spam@gmail.com> wrote: > I'm trying to initialize multiple prolog engines or databases in JPL > What I need is 2 differen

enumerate in table
\setbox0\hbox{1.\enskip}% \def\item{\par\leftskip\wd0\advance\myenumc by 1% \noindent\llap{\the\myenumc.\enskip}}}% \def\endmyenum{\par\endgroup} \begin{document} \begin{center} \begin{tabular}{lp{1.0in}} \toprule Col header 1 & Col header 2 \\ \midrule Entry & \begin{myenum} \item Item 1 short \item Item 2 is very long \item Item 3 is very very very long \end{myenum} \\ Entry & \begin{enumerate} \item Item 1 short \item Item 2 is very long \item Item 3 is very very very long \end{enumerate...; \usepackage{array} > \usepackage{booktabs} > > \newcount\myenumc > \def\myenum{\begingroup% > \myenumc0\setbox0\hbox{1.\enskip}% > =A0 \def\item{\par\leftskip\wd0\advance\myenumc by 1% > =A0 =A0 \noindent\llap{\the\myenumc.\enskip}}}% > \def\endmyenum{\par\endgroup} > > \begin{document} > \begin{center} > =A0 \begin{tabular}{lp{1.0in}} \toprule > =A0 =A0 Col header 1 & Col header 2 \\ \midrule > =A0 =A0 Entry & > =A0 =A0 \begin{myenum} > =A0 =A0 \item Item 1 short > =A0 =A0 \item Item 2 is very long > =A0 =A0 \item Item 3 is very very very

None #3
program, be able to display=20 it with some kind of color shading (possibly the original image shining=20 through) and manipulate the image pixels within the ROI. I checked the=20 IDL help and the newsgroup archive and drew the following conclusions=20 (without doing much testing yet): 1. DEFROI is not for widget programs. 2. CW_DEFROI is meant to do what I want, but is kind of clunky.=20 According to the source code, it's been written 1993 and never updated. 3. XROI would actually work well, but (1) opens the image in its own=20 window, while I would prefer to define the ROI in my... to the source code, it's been written 1993 and never updated. > > 3. XROI would actually work well, but (1) opens the image in its own > window, while I would prefer to define the ROI in my existing draw > widget, and (2) offers tons of options which will confuse the user > (statistics, histogram, multiple ROIs, even loading a new image) and > which I would preferably not offer to him. > > 4. David recommends to write your own code to define ROIs. As I > understand it, I would include some code to draw a freehand or polygon > path and then use

Unicode #3
Is xHarbour Unicode ready? -- @~@ Remain silent. Nothing from soldiers is real! / v \ Simplicity is Beauty! May the Force and farces be with you! /( _ )\ (Fedora 15 i686) Linux 3.3.4 ^ ^ 21:27:02 up 1 day 1:22 0 users load average: 0.08 0.03 0.05 不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA): http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa Hi Why you need this? Regards Luiz On 1 maio, 10:30, Man-wai Chang <toylet.toy...@gmail.com> wrote: > Is xHarbour Unicode ready? > > -- > =C2=A0 =C2=A0@~@ =C2=A0 Remain silent. Nothing from soldiers

Support #3
Dear All, What are the different protocols supported by MATLAB except TCP/IP ? Please reply, as it is on priority basis. Response awaited.

re #3
any possibility to measure the internal variable x? p2 and p3 can be estimated directly from the response to l.

simplification #3
Hello. ff = Pi*Cos[1/7*Pi]*Cos[2/7*Pi]* Cos[3/7*Pi]/Sin[Pi*Cos[1/7*Pi]*Cos[2/7*Pi]*Cos[3/7*Pi]]; Let's make an attempt to simplify ff. In[190]:= FullSimplify[ff] Out[190]= Pi*Cos[Pi/7]*Cos[(2*Pi)/7]*Cos[(3*Pi)/7]*Csc[Pi*Cos[Pi/7]*Cos[(2*Pi)/ 7]*Cos[(3*Pi)/7]] Let's try harder In[194]:= o1=FullSimplify[Together[TrigToExp[ff]]] Out[194]= (1/4)*Sqrt[1 + 1/Sqrt[2]]*Pi or as an another way take In[199]:= o2=FullSimplify[TrigFactor //@ ff] Out[199]= (1/8)*Pi*Csc[Pi/8] My first question is how someone can arrive from o1 to o2 and vice versa. Secondly, o1 was obtained by FullSimplify[Together[TrigToExp[ff]]]. Why doesn't In[206]:= FullSimplify[ff, TransformationFunctions -> {Automatic, TrigToExp, Together}] Out[206]= Pi*Cos[Pi/7]*Cos[(2*Pi)/7]*Cos[(3*Pi)/7]*Csc[Pi*Cos[Pi/7]*Cos[(2*Pi)/ 7]*Cos[(3*Pi)/7]] do the same thing? What I miss here? Thirdly, why the following does suceed? In[3]:= FullSimplify[ff, TransformationFunctions -> {Automatic, FullSimplify}] Out[3]= (1/8)*Pi*Csc[Pi/8] Thank you very much! Dimitris