This is a follow-up of my previous post: Setting up OpenCV 2.3 and Netbeans 7.0 on Ubuntu Oneiric 11.10. A couple of days ago Gijzelaar’s PPA for OpenCV has updated from 2.3 to version 2.3.1. So I thought I’d share this with you guys. I also think this is a slightly better approach

OpenCV (Open Source Computer Vision Library) is a library of programming functions mainly aimed at real time computer vision, developed by Intel and now supported by Willow Garage. It is free for use under the open source BSD license. The library is cross-platform. It focuses mainly on real-time image processing. (Wikipedia)

The upgrade did leave my machine with some broken packages, so be cautious! I managed to fix the broken packages, but that’s another story and it probably was a rare case.

Okay, so here is a quick guide:

1. Installation of OpenCV 2.3.1

Launch a terminal and execute these commands. This adds the PPA of our friend Gijs Molenaar, which now contains OpenCV 2.3.1.

$ sudo add-apt-repository ppa:gijzelaar/cuda
$ sudo add-apt-repository ppa:gijzelaar/opencv2.3
$ sudo apt-get update
$ sudo apt-get install libopencv-dev

2. Installation of Netbeans 7.1

Ubuntu 11.10 also comes with Netbeans 6.9, just download the Netbeans 7.1 from the Netbeans Homepage and you’re all set. Be sure you download the version with C/C++ support.

3. Set the project properties

Create a new C/C++ project in Netbeans 7.1. First, set the project properties correctly. These settings worked for me.

In the C++ Compiler dialog add “/usr/locale/opencv2″ to the “Include Directories” field.

In the Linker Dialog, Click on the “…” button for Libraries, a new Dialog appears, choose “Add PkgConfig Library…”. Search for “opencv” and add it.

This should add the following library expression

`pkg-config --libs opencv`

(Note that this is a better solution than in my previous post)

4. Set the includes

Note: this has also changed since my previous post!

#include "opencv2/highgui/highgui.hpp"
#include "opencv2/core/core.hpp" 
#include "opencv2/imgproc/imgproc.hpp"

I find it easy to work with the cv and std namespace by default, so I don’t have to type “cv::” before every function.

using namespace cv;
using namespace std;

And if we re-use our previous example (yea, I’m lazy), a little program that shows an image, you have something like this:

#include "opencv2/highgui/highgui.hpp"
#include "opencv2/core/core.hpp" 
#include "opencv2/imgproc/imgproc.hpp"
#include <iostream>
using namespace cv;
using namespace std;
 
int main(int argc, char *argv[]) {
 
    Mat image;
    image = imread(argv[1], CV_LOAD_IMAGE_COLOR);
 
    if (!image.data) // Check for invalid input
    {
        cout << "Could not open or find the image" << endl;
        return -1;
    }
 
    namedWindow("Display window", CV_WINDOW_AUTOSIZE); // Create a window for display.
    imshow("Display window", image);
 
    return 0;
}

OpenCV has really poor/decentralized documentation, which makes these simple things over-complicated. I hope this helps some people struggling with getting OpenCV to work. Let me know if it works!

Cheers!
Ruben Verhack

OpenCV (Open Source Computer Vision Library) is a library of programming functions mainly aimed at real time computer vision, developed by Intel and now supported by Willow Garage. It is free for use under the open source BSD license. The library is cross-platform. It focuses mainly on real-time image processing.

1. Installation of OpenCV 2.3

First of all, Ubuntu 11.10 comes with OpenCV 2.1, and I wanted 2.3. Luckely our friend Gijs Molenaar provides us with a PPA for OpenCV 2.3.

$ sudo add-apt-repository ppa:gijzelaar/cuda
$ sudo add-apt-repository ppa:gijzelaar/opencv2.3
$ sudo apt-get update
$ sudo apt-get install libopencv-dev

(Little notice: the package that comes with Ubuntu 11.10 is called libcv-dev, however in the PPA of Gijs the package is called libopencv-dev)

You might also want to install other packages related to OpenCV, just search for them in the Ubuntu Software Center. However, libopencv-dev will do.

2. Installation of Netbeans 7.0

Ubuntu 11.10 also comes with Netbeans 6.9, just download the new version from http://netbeans.org and you’re all set. Be sure you download the version with C/C++ support.

3. Set the project properties

Then create a new C/C++ project in Netbeans 7.0. First, set the project properties correctly. These settings worked for me.

In the C++ Compiler dialog add “/usr/locale/opencv2″ to the “Include Directories” field, also do this in the Linker dialog. In the Linker Dialog, also add the libraries opencv_core and opencv_highgui. You can find these libraries directly under “/usr/lib/”.

4. Set the includes

#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>

I find it easy to work with the cv and std namespace by default, so I don’t have to type “cv::” before every function.

using namespace cv;
using namespace std;

And if we put this all together in to a little program that shows an image, you have something like this:

#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <iostream>
using namespace cv;
using namespace std;
 
int main(int argc, char *argv[]) {
 
    Mat image;
    image = imread(argv[1], CV_LOAD_IMAGE_COLOR);
 
    if (!image.data) // Check for invalid input
    {
        cout &lt;&lt; "Could not open or find the image" &lt;&lt; endl;
        return -1;
    }
 
    namedWindow("Display window", CV_WINDOW_AUTOSIZE); // Create a window for display.
    imshow("Display window", image);
 
    return 0;
}

There we go: Our first OpenCV program in Netbeans! I hope this will work for you too

grep -ir 'invalid user' /var/log/*

Example output

/var/log/auth.log.0:Aug 15 18:14:40 sshd[20296]: Invalid user production from 85.214.40.85
/var/log/auth.log.0:Aug 15 18:14:42 sshd[20296]: Failed password for invalid user production from 85.214.40.85 port 54327 ssh2

That’s when I noticed usernames like

minecraft, eggbreaker2 ,batman ,sir , queen, elmo, frenzy, christmas, idiot, birdseed, einstein123, breast, knight, cookie, eminem, asshole123, googol, denied

and decided to make some graphs with the data.
Using perl I extracted the information from the log files.
This is what I got from ~20.000 failed attempts (spread over couple days)

Popular usernames

Existing usernames

Invalid usernames

Bots tried ~5500 invalid usernames in ~15500 attempts

Invalid vs. existing usernames

Overall

It’s not surprising ‘root’ came out on top, it was used in 18.8% off all attempts.
That doesn’t seem much but it is, since the bots tried that many different usernames.

You can disable root login by opening sshd_config and changing the ‘PermitRootLogin’ setting to ‘no’.

# Authentication:
PermitRootLogin no

Don’t forget to restart the SSH daemon.
P.S. don’t lock yourself out by forgetting you other administrator users passwords

SSH Recommendations

Found out Symantec did the same thing a year ago, you can check their findings and recommendations if you want to protect against these attacks.

Setup a chroot jail

We have no FTP running on our server, users need to connect through SFTP.
Everyone gets a unix user and home folder on the server, to hide the server structure and
limit the users to their home folder we had to setup chroot. It also blocks those user to login
with SSH, SFTP is the only allowed service.

Creating a chroot jail is fairly easy providing you use openssh 5.x.
This explanations is only valid for those versions, and I assume openssh is correctly installed.
If you need to install openssh I suggest you visit http://adamsworld.name/chrootjailv5.php

Modify sshd_config

Mine is located at /etc/ssh/sshd_config (depending on your installation)
Comment all lines starting with ‘Subsystem’ and add ‘Subsystem sftp internal-sftp’.
Add the 4 lines containing the match block at the end of the file.

# Subsystem sftp /usr/lib/openssh/sftp-server
# Setting up chroot jail
Subsystem sftp internal-sftp    
 
Match group users
       ChrootDirectory /home/%u
       ForceCommand internal-sftp
       AllowTcpForwarding no

This jails all users in the group ‘users’ to their home folder. You can also jail induviduals with another match rule : ‘Match user username’.

Change ownership & rights

Create the group ‘users’, and add the persons you want to jail to the group.
The jail needs to be owned by the root user, so lets say you want to jail ‘john’ (homefolder : /home/john.com)

chmod 755 /home/john.com
chown root:root /home/john.com

Restart SSH daemon

# Debian, Ubuntu
sudo /etc/init.d/ssh restart
 
#Fedora, CentOS
sudo /etc/init.d/sshd restart

And you’re done!
The files and directories inside the home-folder can have whatever rights and owner you please.

It’s been a while since I wrote something, but I liked to share this with you all. I bought a new Dell Studio 1558 and all the fn-keys were working except the one to switch the displays…

I enjoy switching between dual-screen, using only my external 22″ or using only my laptop, when I’m on the road. So I was looking for a workaround.

This laptop, is my first machine with Ubuntu that uses an ATI graphics chip, so this was all quite new to me. But I have to say; I’m very very pleased with it.

Ok, so what do you have to do to get things working?

1. Create the script
Create the script, I called it toggleDisplay.sh and this is how it looks in my case:

#!/bin/bash
# Script to toggle display configuration on Dell Studio 1558
# author: Ruben Verhack
 
config="/tmp/display.conf"
current=`cat $config`
 
# Check if config exists
if [ ! -a $config ]
then
   # Empty file executes default
   touch $config;
fi
 
# Check if CRT2 is connected
if xrandr -q | grep "CRT2 connected"
then
   # Toggle between states
   case "$current" in
   '')
      # default
      xrandr --output LVDS --auto --output CRT2 --left-of LVDS --output CRT2 --auto
      echo "dual" > $config
      ;;
   'dual')
      # was dual, now external only
      xrandr --output LVDS --off --output CRT2 --auto
      echo "external" > $config
      ;;
   'external')
      # was external, now laptop only
      xrandr --output LVDS --auto --output CRT2 --off
      echo "laptop" > $config
      ;;
   'laptop')
      # was laptop, now both
      xrandr --output LVDS --auto --output CRT2 --left-of LVDS --output CRT2 --auto
      echo "dual" > $config
      ;;
   esac
else
   xrandr --output LVDS --auto
fi

This toggles your displays from dual screen to external only to laptop only, and from there to dual screen again. Note that my monitors were called LVDS and CRT2, you can easily check how your monitors are called by executing:

xrandr -q

It’s pretty easy, huh?

2. Bind the key to the script
Make sure you given the script rights to be executed, let’s assume you’ve put the shell script in your home-folder.
chmod +x ~/toggleDisplay.sh

Next launch Gnome Keybinding by launching System > Preferences > Keybindings .

Add a custom action:
- Name: Toggle Displays
- Action: sh ~/toggleDisplay.sh

Select it, click on the shortcut and press the key you want to bind to this action and presto!

I hope this helped someone!

Here are some of the most important new features :

• Minimize/Close to tray (WINDOWS)
• Local Books
• Online Books Download
• Inline Search
• Performance Upgrade (SQLITE)
• Custom tree icons
• Complete design makeover

Take a look on our website to learn more …

Somethings been bugging me while skinning a flex application :

The label from the selected tab from the TabNavigator component is 1px lower than the other tabs. As you can see in the example.
(It has something to to with the bottom border I’m guessing).

This can help as a visual support you are on a different tab, but I found it rather annoying, so here is my fix for it.
A quick fix by changing the “paddingTop” property.

View source is enabled so can download the example there.
http://ninetynine.be/blog/wp-content/uploads/2009/12/srcview/index.html

Code after the jump.

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

import mx.events.IndexChangedEvent;
import mx.events.FlexEvent;
 
/*
*	@author: Michiel Vancoillie
*	http://www.ninetynine.be
*/
 
private function init():void{
 
	//fix the first selected tab
	fixedTabNav.addEventListener(Event.ADDED_TO_STAGE,onTabNavCreated);
	//fix the newly selected tab
	fixedTabNav.addEventListener(IndexChangedEvent.CHANGE,onTabNavChanged);
 
	//JUST ADD THE SAME EVENTLISTENERS TO ALL TABNAVIGATORS YOU WANT FIXED
	//notfixedTabNav.addEventListener(Event.ADDED_TO_STAGE,onTabNavCreated);
	//notfixedTabNav.addEventListener(IndexChangedEvent.CHANGE,onTabNavChanged);
}
 
private function onTabNavCreated(e:Event):void{
	var tabNav:TabNavigator = e.target as TabNavigator;
	tabNav.getTabAt(tabNav.selectedIndex).setStyle('paddingTop',-1)
}
 
private function onTabNavChanged(e:IndexChangedEvent):void{
	var tabNav:TabNavigator = e.target as TabNavigator;
	//reset the old tab
	tabNav.getTabAt(e.oldIndex).setStyle('paddingTop',0);
	//fix the new tab
	tabNav.getTabAt(e.newIndex).setStyle('paddingTop',-1);
}

MamppControl

Hey,

Yesterday I had some spare time, so I decided to take a look at the Cocoa collection of frameworks, APIs, and accompanying runtimes that make up the development layer of Mac OS X.
I’ve created this first application with XCode in Objective-C and named it MamppControl.

It allows you to start, stop, reload Mampp from the tray, with a handy tray icon & menu.
Mampp is the other name for Xampp on Mac.

XAMPP is an easy to install Apache distribution containing MySQL, PHP and Perl. XAMPP is really very easy to install and to use – just download, extract and start.

More information about Xampp here.

I know that there is a launcher includes in the xampp package, but it has no tray icon and stay’s visible in the dock.

Default Xampp Control Panel

At the moment Mampp supports only the basic commands (start, stop, reload). If there is need for the other start/stop settings, contact me through the contact page on this blog or shout it here in the comments.

Mampp needs root access to start the xampp, that’s why you get the security login at start.

1

sudo /Applications/xampp/xamppfiles/mampp start

You only have to do this once for the lifetime of the process. And when you change the path setting.

Also keep in mind that there could be some delay when starting or stopping Xampp, since ProFTPD usually takes a few seconds to launch (at least on my laptop it does ).

MamppControl checks if Xampp is already running on startup, and behaves appropriately with that knowledge …

Screenshots

 

 Quicksilver

 

 Security login

 

 Tray Icon

 

 Tray Menu

Download

MamppControl is licensed under the MIT license.

Copyright (c) 2009 Ninetynine.be

Permission is hereby granted, free of charge, to any person
obtaining a copy of this software and associated documentation
files (the “Software”), to deal in the Software without
restriction, including without limitation the rights to use,
copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following
conditions:

The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.

Download : MamppControl - 116.76 kB


Current version: 2.0
Requirements:

   • Mac OS X 10.5 or higher.
   • Xampp installation @( /Applications/xampp/xamppfiles/mampp )

Version History

v1.0

• First public release

v2.0

• Added support for latest Xampp for Mac OSX (1.0.1)
• Added option to choose path to “xampp” executable
• Added complete events for starting & stopping & reloading
• Some minor changes internally

As a PHP developer it comes in handy to run your websites without having to upload them to your server. You can run them on your own computer if you want to, there isn’t much to it. I also like using the CodeIgniter framework for my projects, this is a really speeds up the whole process!

I will take you through the basic steps to set up your localhost to get CodeIgniter running.

1) Use Synaptic package manager to install “apache2″, “php5″ and “mysql-server”.

During the installation you will be asked to enter the password for the root user for your mysql-server.

2) Create a symbolic link, so that when you go to localhost in your browser, you get the website you want.

sudo cd /var/
sudo rm -r www/
sudo ln -s /path/to/website www

Always look before you delete something with “rm -r”, normally there should only be a index.html file in your www/ folder.

I prefer creating a link to my website that I put somewhere in my home directory, as this is being back-upped frequently, and is on another partition.



3) Install the package “phpmyadmin” through synaptic.

It is important that you wait to install phpmyadmin until you have installed the other packages and created the symbolic link. Otherwise it can be a mess to get things running.

4) To allow mod-rewrite, to get the clean CodeIgniter URL’s, you need to do the following.

sudo a2enmod rewrite
sudo nano /etc/apache2/sites-available/default

You will see something like this:

        DocumentRoot /var/www
        <Directory />
                Options FollowSymLinks
                AllowOverride None
        </Directory>
        <Directory /var/www/>
                Options Indexes FollowSymLinks MultiViews
                AllowOverride None
                Order allow,deny
                allow from all
        </Directory>

Change the AllowOverride’s to All.

5) Restart Apache

sudo /etc/init.d/apache2 restart

This wasn’t so hard, and it really makes life easier!

Note: You can’t use the function mail() or CodeIgniter’s mail-class without further configuration. I won’t go into details about this.

Also note that there are many ways to tackle this, this is only one easy suggestion. If you need to switch a lot between different websites, then changing the symbolic link each time is too much work. You can check out this article here.

Happy coding!

I did it on a fresh install of Ubuntu 9.04 RC. It has known to work on Ubuntu 8.10, but it gave me a segmentation fault, probably because I had been messing around . I tested it with the trial version downloaded from adobe.com.

To do so, you need Wine. I will take you through the steps that I’ve taken:

1. If you have used wine before, please start of with a clean .wine folder. Or bottle you Photoshop by doing every action with a specific wineprefix.
2. Install Wine 1.1.16 from here.
3. Download winetricks and install necessary libraries. Also make sure you have a decent driver for your graphics card installed.

wget http://www.kegel.com/wine/winetricks
sh winetricks msxml6 gdiplus gecko vcrun2005
sudo apt-get install msttcorefonts

3. If you have a non-english Ubuntu installed please put LANG=C before the following command like this:

LANG=C wine Setup.exe

4. Install Wine 1.1.19 from here, to fix the menu bar bug.
5. To fix the text-layer bug, put atmlib.dll in “~/.wine/drive_c/windows/system 32″. You can download it from several places if you search it on google.
6. Run it!

It was tested on an amd64 and i386.

The reason why you have to set the “LANG” variable, is that Photoshop searches for files in “Program files”, if you don’t set it, wine installs it into a local translation of Program files.

Very happy that Wine made such a big progress over the last years. But there are a lot of regressions in the last versions. That is the reason you sometimes have to try other versions of wine for the installer. But 1.1.16 and 1.1.19 are not stable versions, so I forgive them. Last version that was declared stable till today was 1.0.1.

I also got Ableton 7 and MS Office Excel, Word and Powerpoint 2007 working with wine!

Big thanks to Quitte for helping me out.

‘Why is this useful?’, you could ask. Well one situation is you’re creating an online rich text editor, and you want to add a font
to the font list. You would only have to edit the css and upload the new swf instead of the whole application. Or you want a client
to be able to add fonts without him having the source code from the application.

Luckily it’s possible in Flex to load runtime CSS compiled as an SWF with the StyleManager class.
StyleManager on LiveDocs

Fonts.css Example :

1
2
3
4

@font-face {
    src:url("fonts/Arial.ttf");
    fontFamily:"Arial";
}

How to compile the CSS to SWF?



In Flex Builder you can right click any CSS inside your source and check the “compile CSS to SWF” option.


An alternative way to compile the CSS to SWF is using the command line :

1

mxmlc Fonts.css

Loading the compiled stylesheet

For the example where we just compiled the Fonts.css to Fonts.swf

1
2
3

import mx.styles.StyleManager;
 
StyleManager.loadStyleDeclarations("Fonts.swf");

It’s as simple as that… you can now start using the font.
The nice thing about this is you can now load your fonts visually using an intermediate loader or animation.


Loading the stylesheet visually with intermediate loader

First thing we do is create a custom titlewindow with a preloader in it to show the loading.
We can catch the StyleEvent.COMPLETE with an IEventDispatcher.
Right click the example for the source code.

You can use this to ‘preload’ any stylesheet.

Compiling CSS to SWF and loading it runtime is also used for application with theme support.
You can read more about that and view an example here : Understanding runtime CSS

Happy coding!