|
Thank you for really useful tips. If I try to decode everything between % and _ I get meaningless text. I know ceasar's cipher because I've read Digital Fortress  .
This algorithm also changes numbers. I couldn't include a number example but it does. I think we've made progress, we will solve this. 
|
|
|
|
|
Have you by any chance tried the common Algoprithms? Blowfish, Catfish etc?
BHM
|
|
|
|
|
|
Originally, I want to use the Markov model to realize object tracking and I have found Markov Random Field can be used for the optimizing Multiple target tracking.
However, the background of my tracking project is changing because of the movement of the camera.
So, I want to know whether mean-shift(or HMM, Markov model)can be used for target tracking at this kind of moving background.
And, if it is recommendable for me to combine Markov Random Field with mean-shift for producing a result of robust muti-target tracking.
Thank you.
modified on Friday, September 11, 2009 11:12 AM
|
|
|
|
|
Greetings,
I'm trying to determine the best way to get a polygon from a set of points. The points are not in any particularly useful order (they are the endpoints of individual lines that fill the polygon in question). How can I easily take such a set of points and figure out the vertices of a polygon from them? The polygons may have gaps inside of them. Any ideas?
Thanks,
Will
PS. Sorry about the repost from the Graphics forum. I wasn't sure whether to put it here or there, but was informed that here was a better choice.
|
|
|
|
|
A quick search of the codeproject site turned up this:
Convex Hull[^]
Best Wishes,
-David Delaune
|
|
|
|
|
This is a really dificult... perspective...
if the shape is solid(with enclosed points) then you can use a contracting concentric circle approach.
if you have a form of wrapping region... good luck(unless there is other structure to your data that you didn't include)
I recently did something similar (the convex polygon approach to making a vooroni tesselation(delanuy triagulation)).
you can alternativly use a line crossing approach, but this is also tricky.
it is best if the points are "ordered" either inpriciple, or by allowence, and then building up the polygon vertex(pair) by vertex(pair).
|
|
|
|
|
I have search almost every ware and to be honest I can't find on site that could answer my question.
Here is the situation, I have a .EXE, lets call it BOB.exe. BOB.exe has a portion of it's code when opened in a hex editor that is readable which leaves it open to be hacked. One think that it has is an IP that sometimes need to be changed by the administer. this app connects online but all this is relevant right now.
Now what I want to do is create a program with my own algorithms that can encrypt BOB.exe yet make BOB.exe executable by any user and at the same time when opened with a hex editor you can't understand anything in it because it will be encrypted.
So far from what I read i need to encrypt bit by bit of the program.
Please help, need this in ether C++ or C# or VB.Net
|
|
|
|
|
By what you describe it seems your problem is having strings in your exe that can be read by opening it with an hex editor.
If that's the case, then you can encrypt your strings using any encryption algorithm you like, and implement a decoding routine in your software. Each time you use an encrypted string, you must decode it. For example (C#):
string MyEncodedString = "#@SIRJI3893N";
MyTextBox.Text = DecodeString(MyEncodedString);
There are many encryption algorithms you can use. The simplest are based on XORing and byte swapping. They are very fast and easy to implement but must contain a fair number of operations to make it less easy to understand the encryption method.
If you use .NET, you can also have a look at the System.Security.Cryptographic namespace, which provides standard encryption algorithms. These are stronger but depend on an encryption/decryption key, which must be embedded in the exe and must be obfuscated in some way (XORing and swapping again or similar).
About .NET, you can also make your life easier by simply using an obfuscator which provides string encryption, like eazfuscator[^]. It will do all the work for you.
You should also be aware that there is no way to stop a determined attacker with enough patience and time: he will reverse engineer your decoding routine and apply it to your strings. That said, if you implement it properly, it will take time and sweat to do that.
2+2=5 for very large amounts of 2
(always loved that one hehe!)
|
|
|
|
|
Hello all,
first of all im sorry if its the wrong place for this question.
I am trying to decaptcha the following captcha:
[IMG]http://i26.tinypic.com/wrazt.png[/IMG]
what i am trying to do is removing the letters from the rest of the image. I am using Matlab atm but will do it in C# i think. its where i stucked, since all texture on background and letters are same, i can't find a proper way of doing it.
Im looking for your suggestions, thanks in advance
ps: its not a hacking project or so, never will be used for spamming or other purposes.
|
|
|
|
|
After blurring it and turning up the contrast a bit it suddenly looked much easier to do OCR on, so that may help
|
|
|
|
|
Write an algorithm that finds the largest blocks of white. Once you got that list, start coloring those blocks using the floodfill-algo. Lastly, drop all other colors
Some things Man was never meant to know. For everything else, there's Google.
|
|
|
|
|
that is probably the best captcha i've ever seen!
|
|
|
|
|
I know right..?
the only thing that could be better would be a well know object (instead of the grid) .. but yes brilliant.
|
|
|
|
|
do you found the solution ???
i have same issue
http://www.codeproject.com/Questions/127525/image-procesing-get-text-from-image.aspx
please help if you can
Palestine
|
|
|
|
|
hi everyone 
i am doing work on image processing.So i am doing a project on color segmentation.
i have identified human skin color.but now i am trying to do labeling of
identified region.
can anyone explain me ,how can i do the labeling?
|
|
|
|
|
Use the skin color to find the face in the image, and get a bounding rectangle on the face pixels.
Get a large sample of images that exhibit more variation than you're likely to encounter in practice, and compile statistics on where in the bounding rectangle the features you're looking for generally appear.
To zero in on the features, you can use thresholding and mathematical morphology to simplify the complexity of the image.
|
|
|
|
|
As i have used skin color to find the face,so it's identified the hand too.
I have find the skin color and change it's color to black.
Now i want to make a rectangle on the detected area.ex-make a rectangle on hand,face.
As i want to make the rectangle,so,i think i need to find the x,y axis of detected area.
How can i detect the x,y axis of detected area
|
|
|
|
|
Your question isn't clear.
|
|
|
|
|
kabirbdboy wrote: As i want to make the rectangle,so,i think i need to find the x,y axis of detected area.
How can i detect the x,y axis of detected area
Areas don't have axes, coordinate systems do. What you primarily need are the x and y centroids, extents, possibly area and moments. OpenCV provides a function for calculating the moments of an area, check into it.
You measure democracy by the freedom it gives its dissidents, not the freedom it gives its assimilated conformists.
|
|
|
|
|
my friend sent me a message and I have no idea how to decode it...he claims it's an extension of binary, but this only confuses me more...can anyone help? 
363336313645323036383631373636332303734 3638363532303633364637353732363136373635 3230363136453634323037343638363532303646
i apologize if this is in the wrong forum, I tried my best to get it in the right place!
thanks in advance 
|
|
|
|
|
I don't see any 9's in the string. Base 9?
You measure democracy by the freedom it gives its dissidents, not the freedom it gives its assimilated conformists.
|
|
|
|
|
Haven't been able to crack it, but some observations:
1) The three parts are almost the same size - Are you sure you aren't missing a digit in the first part? The next observation also seems to suggest that the "3323" near the end should really be "33323"
2) With the exception of a few "45" and "46" occurrences, with with the change mentioned above, the entire message could be read as a series of 3x pairs (30-38). The fact that there are 45s and 46s, but no pairs 40-44, suggest that the 4x values might be separators of some kind. With those taken out, and the leading "3" in each pair removed, the remaining message might be further decoded into something readable.
|
|
|
|
|
I've seen stuff like that when hex files get loaded as binary. I think you're missing a "5" between the two consecutive "3"'s near the end of the first group, though.
Take pairs of characters as hex, translate to ASCII, take pairs of those characters as hex, and translate to ASCII. "can have the courage and the o".
|
|
|
|
|
Wow... What can I say, but...
3734363836313734323737333230363132
3037323635363136433643373932303633
36463646364332303633364636343635
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
