Posts Tagged ‘audio stream’

How to digital watermark to a picture – Protect pic with copyright image or text with composite, convert and Phatch on GNU / Linux

Friday, March 23rd, 2012

Watermarking is a technique to identify a physical or non-physical object with its owner (creator). First watermarks in history originates from very ancient times.

There are two basic types of Watermark types:

I. Physical Watermarks (classical)

II. Digital Watermarks

Historically Classical Watermarks, were mostly important. As we tend to use more and more visible and we switch to use of more invisible stuff, nowdays the importance and use of digital watermarks is steadily raising.

You have most likely already seen pictures from websites which contain a copyright holder message stamp or website logo on it.
As you could imagine the picture watermark is placed in order to prevent pictures from being re-used in another internet space ,without the picture copyright holder explicit permission…

Watermarks have entered most if not all areas of our life, but we often don't recognize they're there / rarely think about them.
Few of the many "physical watermarks" we use daily are:
 

  • paper money watermark (to protect against anti money forgery)
  • bank debit / credit cards stamps near the card chip
  • postcards paper stamps

There are too many different kind of "physical watermarks" and since this is not the accent of this article, I will continue straight to explaiin a bit on Digital (picture) watermarks and how to watermark images with ImageMagick image editting command line suit.

Just like with physical watermarks, there are different kinds of digital watermarks. There are:
 

  • Picture (Images) digital watermarks
  • – Steganography

  • Video watermarks
  • Audio stream digital watermarks
  • Visual digital watermarks
  • – Visible area of text or picture over another text picture or video

  • Invisible digital watermarks
  • – digital information (files) metadata with watermark content etc.

The topic of watermarking is quite wide, so I will stop here and focus on the main idea of this article – to show how to place digital watermark on graphic image or collection of pictures.

The most straightway non-interactive way to do picture watermarking is with ImageMagick's composite command line tool. This little handy tool is capable of creating watermarks in single and multiple pictures.

If you prefer to have a simple text as a watermark, then you should use imagick's convert cmd.

1. Putting a watermark of picture in the right bottom corner

$ composite -gravity southeast -dissolve 100 \
watermark_picture.png image-to-watermark.png \
output-watermarked-image.png

Snoopy Writting pc freak watermark picture text watermark on the right bottom corner with composite

2. Placing watermark to picture in the bottom right corner

$ composite -gravity northeast -dissolve 80 \
watermark_picture.png image-to-watermark.png \
output-watermarked-image.png

Snoopy Writting pc freak picture text watermark on the right bottom corner with composite

3. Watermarking picture in the bottom left corner

$ composite -gravity southwest -dissolve 90 \
watermark_picture.png image-to-watermark.png \
output-watermarked-image.png

Snoopy writting watermarking picture on the bottom left corner imagemagick (composite)

4. Watermarking picture in the top left corner

$ composite -gravity northwest -dissolve 100 \
watermark_picture.png image-to-watermark.jpg \
output-watermarked-image.jpg

As you see from above example, composite even accept mixing up input / output between PNG and JPEG pictures 🙂

Output Watermarked Image picture on top left corner with pc-freak logo image Imagick composite

5. Put a watermark in the image center

$ composite -gravity center -dissolve 100 \
watermark_picture.png image-to-watermark.png \
output-watermarked-image.png

position watermark on the picture middle (center) composite output picture

6. Sealing image with custom text / Text Watermarking a picture

a) Writting text watermark to an image centered in "footer"

$ convert image-to-watermark.png -pointsize 20 \
-draw "gravity south fill black text 0,12 \
'hip0s Watermark'" output-watermarked-image.jpg

This will place a watermark in position 0,12, meaning the text will be added in the bottom center of the watermarked image.

Watermarking a picture sealing with custom text image imagick composite pic

-pointsize 20 defines the text font size. hip0s Watermark is the actual text that will be stamped.

b) Writting image watermark with font type customization (Arial Tahoma etc.):

To list all available fonts ready to be used by convert, type:

$ convert -list font
$ convert -list font |grep -i arial
Font: Arial-Black-Regular
family: Arial Black
glyphs: /usr/share/fonts/truetype/msttcorefonts/Arial_Black.ttf
Font: Arial-Bold
family: Arial
glyphs: /usr/share/fonts/truetype/msttcorefonts/Arial_Bold.ttf
Font: Arial-Bold-Italic
family: Arial
glyphs: /usr/share/fonts/truetype/msttcorefonts/arialbi.ttf
Font: Arial-Italic
family: Arial
glyphs: /usr/share/fonts/truetype/msttcorefonts/ariali.ttf
Font: Arial-Regular
family: Arial
glyphs: /usr/share/fonts/truetype/msttcorefonts/Arial.ttf

$ convert -list type
Bilevel
ColorSeparation
ColorSeparationMatte
Grayscale
GrayscaleMatte
Optimize
Palette
PaletteBilevelMatte
PaletteMatte
TrueColorMatte
TrueColor

On my system, I have 392 of fonts installed, to check the number of installed fonts ready for use by convert I used:

$ convert -list font|grep -i 'font:'|wc -l
392

To only check exact fonts names usable in convert:

$ convert -list font|grep -i 'font:'

To use the red marked Arial-Regular for font of the text picture timestamp issue;

$ convert watermark_picture.jpg -font Arial-Regular \
-pointsize 20 -draw "gravity south fill black text 0,12 'hip0s Watermark'" \
output-watermarked-image.jpg

Watermark and with Arial-Regular font image magick convert screenshot dog type writter

c) Using external font with convert to place image text watermark

Lets say you would like to use an external font (arhangai.ttf) not listed in convert -font list usable fonts:

$ convert image-to-watermark.png -pointsize 20 \
-font /usr/share/fonts/truetype/arhangai/arhangai.ttf \
-draw "gravity south fill black text 0,12 \
'hip0s Watermark'" output-watermarked-image_7.png

Talking about fonts, if you would like to add some external, nice free-fonts (ttf) files to your current logged in user, exec:

hipo@noah:~$ cd ~/fonts
hipo@noah:/fonts$ for i in \
$(lynx -dump http://www.webpagepublicity.com/free-fonts.html|grep -i .ttf|grep -i http|awk '{ print $2 }'); \
do wget -r -l2 -nd -Nc $i;
done

This will add 85 new nice looking fonts. Putting fonts in .fonts/ directory, are red while fonts are looked up by applications installed on respective the server or desktop GNU / Linux systems. Any font put there is ready to be used across all ImageMagick command line tools, as well as will be added across the list of possible fonts to use in GIMP and the rest of gui editors installed on the system.

According to the (watermark) texts font size passed to convert on some pictures the text written will exceed the picture dimensions and only partially some of the text intended as watermark will be visible.
If you encounter the exceed picture text problem, take few minutes and play with fonts sizes until you have a good font size to fit the approximate dimensions of the (expected minimum / maximum – horizontal and vertical) stamped picture dimensions.

For the sake of clarity, here is a list with arguments used in above, composite and convert examples.
 

  • composite — The ImageMagick command that combines two images.
  • -dissolve 80 — The number after the option determines the brightness of the watermark.  100 is full strength.
  • -gravity southeast — Determines the placement of watermark.
    Possible options are; north, west, south, east, northwest, northeast, southeast, southwest, center
  • watermark_picture.png — The watermark image is the first argument.
  • image-to-watermark.jpg — The second argument is the original image to be watermarked.
  • output-watermarked-image.jpg — The third argument is the new composite image to be created.
    N. B. !  If you don't specify a new file, be careful, the original file will be overwritten.

As ImageMagick is cross platform graphic editting suit – it runs on both *nix (Linux,BSD) and Windows. I have tested it on Linux, only but on FreeBSD and other BSDs it should work without any problem.
The composite and convert above examples should be easily rewritten to run on achieve watarmarking on MS Windows too.

7. Watermarking multiple pictures in a directory

To watermark multiple pictures within a directory use, a short bash loop in combination with either convert or composite could be used:

$ cd your-directory/
$ for i in *; do
convert $i -pointsize 20 -draw "gravity south fill black text 0,12 'hip0s Watermark'" output-watermarked-image.jpg
done

convert and composite also support wildcards like '*.JPG, *.PNG', but I'm not sure if this syntax can be used for mass picture marking?

8. Adding watermark and doing other various advanced Image Edit, Convert and Compose stuff with Phatch GUI program

Another program that is capable to put watermarks on pictures and besides that doing a number of routine graphic manipulation operations achievable with expert Image manipulation programs like GIMP / Inkscape is PHATCH = PHOTO & BATCH

Phatch is swiss army knife for doing web design or or graphics design on Linux.

Phatch is really great and easy to use program. Tt makes putting basic designer effects on pictures with no requirement for any design skills.
With Phatch you can become a designer for a day literally 😉

If you haven't used it yet, make sure you try it!
Below, are two screenshots of Phatch running on my Debian G* / Linux

Phatch Linux Debian Squeeze Screenshot

Phatch Linux Debian Squeeze Screenshot Watermark effect

Phatch is installable via apt on Debian and Ubuntu Linux. It has also a phatch-cli tools, which are a possible substitute to ImageMagick's composite / convert tools.

On deb based distros install Phatch with:

noah:~# apt-get --yes install phatch phatch-cli

In Phatch it is also possible, to create a combination of filters to be later applied to an image file or a group of image files all in a directory. The program capabilities are really outstanding, it is pure joy to work with it.

Using Phatch GUI interface is hard to comprehend in the beginning, I needed few minutes until I can get the idea how to use it. Anyhow once you know the basics, its very easy to use onwards.

Phatch currently can perform the following actions:
 

  • Auto Contrast – Maximize image contrast
  • Border – Crop or add border to all sides
  • Brightness – Adjust brightness from black to white
  • Canvas – Crop the image or enlarge canvas without resizing the image
  • Colorize – Colorize grayscale image
  • Common – Copies the most common pixel value
  • Contrast – Adjust from grey to black & white
  • Convert Mode – Convert the color mode of an image (grayscale, RGB, RGBA or CMYK)
  • Effect – Blur, Sharpen, Emboss, Smooth, ..
  • Equalize – Equalize the image histogram
  • Fit – Downsize and crop image with fixed ratio
  • Grayscale – Fade all colours to gray
  • Invert – Invert the colors of the image (negative)
  • Maximum – Copies the maximum pixel value
  • Median – Copies the median pixel value
  • Minimum – Copies the minimum pixel value
  • Offset – Offset by distance and wrap around
  • Posterize – Reduce the number of bits of colour channel
  • Rank – Copies the rank'th pixel value
  • Rotate – Rotate with random angle
  • Round – Round or crossed corners with variable radius and corners
  • Saturation – Adjust saturation from grayscale to high
  • Save – Save an image with variable compression in different types
  • Scale – Scale an image with different resample filters.
  • Shadow – Drop a blurred shadow under a photo with variable position, blur and color
  • Solarize – Invert all pixel values above threshold
  • Text – Write text at a given position
  • Transpose – Flip or rotate an image by 90 degrees
  • Watermark – Apply a watermark image with variable placement (offset, scaling, tiling) and opacity

Most of the function / effects Phatch in the up list works fine as I tested them to get to know the program.
The only effect that didn't worked for me is Blender effect.
Trying to apply the Blending effect I got error:

Can not apply action Blender:
'dict' object has no attribute 'rfind'

dict object has not attribiture rfind error screenshot my linux

Its really a pity blender filter don't work. I've seen on Phatch's website some pictures showing the blender effect in action and it looks really awesome.

In attempt to work around the err, I tried downloading Phatch's latest release and running it with python interpreter but it didn't work out …
I tried also to install some packages to the system that somehow seemed to be related to blenderversatile 3D modeller/renderer program but this worked neither.
I suspect Phatch blender effect is not working on Ubuntu too as I've red complains in some Ubuntu forums.
If someone succeeding making blend effect work please let me know how?

Interesting feature of Phatch is the program support for applying its predefined filters using a cli interface.
The syntax for phatch cli, should be something like:

phatch -console action_list.phatch

Where action_list.phatch is a Phatch predefined filter. Anyways I didn't manage to figure out how to use the program CLI. I'll be glad to hear if someone succeeded in using the program console, if so please share with me how?

9. Adding Watermark to pictures with GIMP

To add a watermark text or picture in GIMP, there are plenty of ways but is more time consuming by both Phatch or convert, composite..
There is a script in gimp plugin registry site – watermark.scm which adds watermarking capability to GIMP

On my system this script was installed with the deb package gimp-data-extras. To apply the plugin on a pic, I used GIMP menus:

Filters -> Eg -> Copyright Placer

GIMP Copyright Holder plugin Watermark Screenshot

If someone knows about better or quicker ways to do watermarking, please share 🙂

Best software available today for Linux video Desktop capturing on Debian

Tuesday, April 19th, 2011

I’ve been experimenting since some time in order to understand better what is the current situation with Video Desktop Capturing Software available today for Linux and FreeBSD.

My previous investigations has led me to write an article about xvidcap called How to make Video from your Linux Desktop with xvidcap

Though xvidcap works pretty well, it is currently uncapable of capturing the audio stream of a Linux Desktop env and hence part of the interactivity of the videos is missing when used.

A bit of further investigation on the topic has pointed me to 3 free software programs which are capable to record Desktop environment on Linux with sound embedded

The interesting screen video capturing Desktop tools I’ve found are:

1. recordMyDesktop
2. Istanbul
and
3. vnc2swf

Installing them on a Debian based distribution is pleasable, as there are installable debian packages of each one which I installed easily with apt:

debian:~# apt-get install istanbul recordmydesktop gtk-recordmydesktop vnc2swf

RecordMyDesktop‘s package contains a command line little tool which when started directly starts capturing video and audio of the Linux Desktop. After a Ctrl+C is pressed the program quits, saves and encodes the video in ogg-encapsulated theora-vorbis file format.

Here is the output I got in saving a sample file by launching recordmydesktop without any arguments:

hipo@debian:~/Desktop$ recordmydesktop
Initial recording window is set to:
X:0 Y:0 Width:1024 Height:768
Adjusted recording window is set to:
X:0 Y:0 Width:1024 Height:768
Your window manager appears to be Metacity

Initializing…
Buffer size adjusted to 4096 from 4096 frames.
Opened PCM device hw:0,0
Recording on device hw:0,0 is set to:
2 channels at 22050Hz
Capturing!
Broken pipe: Overrun occurred.
Broken pipe: Overrun occurred.
Broken pipe: Overrun occurred.
^C
*********************************************

Cached 5 MB, from 207 MB that were received.
Average cache compression ratio: 97.3 %

*********************************************
Saved 69 frames in a total of 69 requests
Shutting down..Broken pipe: Overrun occurred.

STATE:ENCODING
Encoding started!
This may take several minutes.
Pressing Ctrl-C will cancel the procedure (resuming will not be possible, but
any portion of the video, which is already encoded won’t be deleted).
Please wait…
Output file: out.ogv
[100%]
Encoding finished!
Wait a moment please…

Done.
Written 692529 bytes
(635547 of which were video data and 56982 audio data)

Cleanning up cache…
Done!!!
Goodbye!

The captured file as I you see in the above output is saved in file out.ogv

RecordMyDesktop has also a GUI interface (written in Python) called gtk-recordmydesktop

Below you see a screenshot of the GUI gtk-recordmydesktop:

RecordMyDesktop GTK interface entry screen

gtk-recordmydesktop is a super-easy to use as you already see in the picture, you can either configure it with Advanced button or use Save As button to select where you want the Desktop captured video and audio to be stored.

In Debian Squeeze 6.0, the Advanced GUI button interface button is not working but that’s not such an issue, as the rest of the buttons works fine.
After the recordmydesktop‘s Record button is pressed it will start capturing from your Desktop and the window seen in the above screenshot will disappear/hide in the system tray:

recordmydesktop recording minimized in system tray
When you press over the white little square in the system tray the screen capturing will be interrupted and a window will pop-up informing you that the captured video and audio is being encoded, here is another screenshot of recordmydesktop encoding a saved Desktop video stream:

recordmydesktop saving captured desktop video

After the final .ogv file is encoded and saved to further transfer it into (.flv) I used ffmpeg;

debian:~# ffmpeg -i test.ogv test.flv
...

Now let’s evaluate a bit on the the final results, the produced test.ogv‘s synchronization between sound and video was not good as the sound was starting earlier than the video and therefore even though recordmydesktop used to be highly praised on the net, the proggie developers still needs to do some bug fixing
Further on, I continued and (gave the other Desktop screen capturer) Istanbul a try hoping that at least with it the video and audio of my Linux desktop will be properly captured. But guess what, the results with Istanbul was even more unsatisfactory as the produced videos and sounds, were slow and a lot of frames from the screens were missing completely.
Moreover the sound which was supposed to accompany the video was completely 🙁

Thus I will skip on talking about Istanbul as in my view, this piece of software is far away from being production ready.

I also tested vnc2swf , launched it by: pressing alt+f2 and typing in vnc2swf in GNOME’s run application prompt, just to be surprised by an error …:

vnc2swf error no 111 Connection refused

The reason for this error is caused by the xserver (Xorg) port 5900 is being closed by default on Debian

However this error is easily solvable, by making the Xserver to listen to a the port 5900, to make the Xorg server on Debian to listen on this port you need to edit the file:

/etc/X11/xinit/xserverrc

and change inside it:

exec /usr/bin/X -nolisten tcp "$@"

with:

exec /usr/bin/X "$@"

and either reboot your Linux or restart only the Xorg server by pressing Ctrl+Alt+BackSpace

Now let me conclude, the results from my desktop video screen capturing experiments prooved that vnc2swf is superior (as it is capable of properly saving a movie with sound and video from a Linux Desktop). It appears this soft is actually the best one you can use to make a video of your Linux desktop.

Sadly my testing has proven that Linux is still lacking behind Windows and Mac in even doing the most simple tasks …
Let’s hope that situation will get better soon and Gnome or KDE developers will soon provide us with better software capable to save properly a video and audio captured from the Linux Desktop.