Example Based Single-frame Image Super-resolution by Support Vector Regression

 
Thread Tools Search this Thread
Special Forums News, Links, Events and Announcements UNIX and Linux RSS News Example Based Single-frame Image Super-resolution by Support Vector Regression
# 1  
Old 10-23-2010
Example Based Single-frame Image Super-resolution by Support Vector Regression

HPL-2010-157 Example Based Single-frame Image Super-resolution by Support Vector Regression - Li, Dalong; Simske, Steven
Keyword(s): Support Vector Regression, single-frame image super-resolution, ill-posed problem, example-based, machine learning
Abstract: As many other inverse problems, single-frame image super-resolution is an ill-posed problem. The problem has been approached in the context of machine learning. However, the proposed method in this paper is different from other learning based methods regarding how the input/output are formulated as ...
Full Report

More...
Login or Register to Ask a Question

Previous Thread | Next Thread

6 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Need support to frame the output in line by line

Hi I run the command "x" and the output looks like this a;b;c;d;e;f;g;h;i;j;k;l; but how can i get the output as below a b c d e f g h pleae help me in this... (4 Replies)
Discussion started by: kumar85shiv
4 Replies

2. UNIX for Dummies Questions & Answers

Order based on timestamp in a single file

Hi All, I have a large text file which is a combination of multiple files. This is what I used and it worked. for i in /home/docs/text/* do cat $i >> Single_File done Now wondering, if there is a way to sort that single large file based on timestamps in ascending order. Text file... (11 Replies)
Discussion started by: prrampalli
11 Replies

3. UNIX for Dummies Questions & Answers

Remove duplicate rows when >10 based on single column value

Hello, I'm trying to delete duplicates when there are more than 10 duplicates, based on the value of the first column. e.g. a 1 a 2 a 3 b 1 c 1 gives b 1 c 1 but requires 11 duplicates before it deletes. Thanks for the help Video tutorial on how to use code tags in The UNIX... (11 Replies)
Discussion started by: informaticist
11 Replies

4. Shell Programming and Scripting

remove duplicates based on single column

Hello, I am new to shell scripting. I have a huge file with multiple columns for example: I have 5 columns below. HWUSI-EAS000_29:1:105 + chr5 76654650 AATTGGAA HHHHG HWUSI-EAS000_29:1:106 + chr5 76654650 AATTGGAA B@HYL HWUSI-EAS000_29:1:108 + ... (4 Replies)
Discussion started by: Diya123
4 Replies

5. Shell Programming and Scripting

duplicate row based on single column

I am a newbie to shell scripting .. I have a .csv file. It has 1000 some rows and about 7 columns... but before I insert this data to a table I have to parse it and clean it ..basing on the value of the first column..which a string of phone number type... example below.. column 1 ... (2 Replies)
Discussion started by: mitr
2 Replies

6. HP-UX

BAD SUPER BLOCK - Run fsck with alternate super block number

Error received when I tried to restore a blank disk with an 'auto recovery' DDS tape via HP-UX recovery system 2.0 onto a 1Gb SCSI. I assumed it would do the setup, wrong. Could someone tell me the procedure to initial disk for recovering files using cpio. The system is a HP-UX 9.04 version on a... (1 Reply)
Discussion started by: admin wanabee
1 Replies
Login or Register to Ask a Question
fb.modes(8)						     Linux frame buffer utils						       fb.modes(8)

NAME
fb.modes - frame buffer modes file DESCRIPTION
/etc/fb.modes contains an unlimited number of video mode descriptions. The general format of a video mode is: mode geometry <xres> <yres> <vxres> <vyres> <depth> timings <pixclock> <left> <right> <upper> <lower> <hslen> <vslen> options <value> endmode OPTIONS
geometry options: xres visible horizontal resolution (in pixels) yres visible vertical resolution (in pixels) vxres virtual horizontal resolution (in pixels) vyres virtual vertical resolution (in pixels) depth display depth (in bits per pixel) timing options: pixclock length of one pixel (in picoseconds) left left margin (in pixels) right right margin (in pixels) upper upper margin (in pixel lines) lower lower margin (in pixel lines) hslen horizontal sync length (in pixels) vslen vertical sync length (in pixel lines) other options: the first value of this options is the default hsync {low|high} the horizontal sync polarity vsync {low|high} the vertical sync polarity csync {low|high} the composite sync polarity extsync {false|true} enable or disable external resync. If enabled the sync timings are not generated by the frame buffer device and must be pro- vided externally instead. Note that this option may not be supported by every frame buffer device laced {false|true} enable or disable interlace. If enabled the display will be split in two frames, each frame contains only even and odd lines respectively. These two frames will be displayed alternating, this way twice the lines can be displayed and the vertical fre- quency for monitor stays the same, but the visible vertical frequency gets halved double {false|true} enable or disable doublescan. If enabled every line will be displayed twice and this way the horizontal frequency can easily be doubled, so that the same resolution can be displayed on different monitors, even if the horizontal frequency specifica- tion differs. Note that this option may not be supported by every frame buffer device INTERNALS
Generally a frame buffer display is organized as follows: +---+---------------+---+---+ | | ^ | | | | | | 5 | | | | | v | | | +---#################---+---+ | # ^ # | | | # | # | | | # | # | | | 1 # | 2 # 3 | 4 | |<->#<------+------>#<->|<->| | # | # | | | # | 6 # | | | # | # | | | # v # | | +---#################---+---+ | | ^ | | | | | | 7 | | | | | v | | | +---+---------------+---+---+ | | ^ | | | | | | 8 | | | | | v | | | +---+---------------+---+---+ 1 left margin 2 xres 3 right margin 4 horizontal sync len 5 upper margin 6 yres 7 lower margin 8 vertical sync len The area bordered with `#' is the visible display area. Horizontal and vertical frequencies can now easily be calculated, for this the sum of horizontal or vertical values are important htotal = left + xres + right + hslen vtotal = upper + yres + lower + vslen The length of one line can now be calculated with pixclock line = pixclock * htotal and we have the horizontal frequency hfreq = 1E12 / line = 1E12 / (pixclock * htotal) To get the vertical frequency vtotal must eventually adjusted. If the display is laced, vtotal must be halved or if the display is a dou- blescan one, vtotal must be doubled. Now we can calculate the length of one frame if (lace) vtotal /= 2 if (double) vtotal *= 2 frame = vtotal * line and we get also the vertical frequency vfreq = 1E12 / frame = hfreq / vtotal SEE ALSO
fbset(8), fbdev(4) local Aug 1996 fb.modes(8)