|
pavanip wrote: develop add-on to internet explorer and mozilla firebox.
Two different interfaces. Not exactly a neat solution, but doable...
pavanip wrote: If we develop in dotnet it will not work with out having Dotnet frame work
True.
pavanip wrote: all users no need to install framework
They have to install the .NET Framework in order to run managed code (C#, VB.NET, C++/CLI, ...). There are utilities out there that will compile the app down to native code and include the .NET Framework, but there are no free ones and they cost at least $1,000 to get.
pavanip wrote: Is there any way to develop a toolbar in dotnet,It will work with out having dotnet framework.
Without said utility, no.
|
|
|
|
|
I want to grab a portion of the screen with just text (already accomplished) and save it as a bitonal tiff images and then image enhance that file so that OCR programs can effectively convert it to text. Is there a surefire method using GDI+ that will give me a great looking image that a OCR program can sucessfully read? By great looking I mean where the text is readable and not blurry.
Right now, I'm creating a 600x400 image canvas and pasting the text I grabed from the desktop onto that image. The reason why the image canvas (white space) is so big is that I'm concerned that some OCR engines won't read really small images. However, most OCR programs return garbled text even if the text on the image is pixel perfect. Why is this? Is there something wrong with the resolution, DPI, quality of the image? For example, I have an image with the text "Color.White" centered in the middle in 10 font size. When I OCR it, it reads "C0l0t.Uhic&"
I also ned to know how to save a GDI+ image as a bitonal tiff.
Also, does anyone reccommend a good OCR engine I can use with the .NET framework?
|
|
|
|
|
The OCR library you're using is making the mistakes. It's probably got nothing to do with the image itself, but the font used in the image and the quality of the OCR software.
As for TIFF support in GDI+, I'd recommend using a third party library to do the image manipulation. TIFF support is a bit spotty in GDI+ and it doesn't have a suite of functions for cleaning up images.
|
|
|
|
|
Hi,
The limited experience I have with OCR told me you need 16 pt or larger to get reasonable results.
At 10pt each characters is only a few pixels wide and high, insufficient for OCR to work properly.
Luc Pattyn [Forum Guidelines] [My Articles]
- before you ask a question here, search CodeProject, then Google
- the quality and detail of your question reflects on the effectiveness of the help you are likely to get
- use the code block button (PRE tags) to preserve formatting when showing multi-line code snippets
modified on Wednesday, February 11, 2009 6:14 PM
|
|
|
|
|
How are you converting it to a bitonal representation? This is an information loss and thus probably an error source.
|
|
|
|
|
Trying to read freshly-generated textual information off a screen tends to be a different problem than reading a scanned document; programs that are designed for one problem may not be optimal for the other. Programs that are seeking to read information off the screen generally have to cope with a lower-resolution version of characters than programs which are dealing with a scanned page, but in many cases have the huge advantage that the letter forms on screen will often be much more consistent than those in a scan. This allowed for much more efficient operation than would otherwise be possible. Even the original IBM PC could read text off the low-resolution graphics screen; indeed, that's how the BASIC interpreter worked in graphics mode. The "Insert" key was a bit sluggish in graphics mode, but that's because each character had to be read off the screen and rewritten one space over. Of course, most applications don't use monospaced text aligned on byte boundaries, but modern PC's don't have to run at 4.77MHz (4*chroma/3).
|
|
|
|
|
hlo everyone..
how can a tune sung by a user (humming) be compared with the songs in the database & produce similar results??
since the pitch of man & woman varies widely, how can the results can be made accurate?
|
|
|
|
|
I would have used Artficial Neural Nettworks for a task like that. If you don't wanna code it from scratch you can use a library like AForge.
|
|
|
|
|
I give up - How do I get the caller host from a WCF context when the service is hosted as a ServiceHost running as an NT Service.
The best I can come up with is:
string host;
if (Demon.IsDemon) {
//Thought this might do the trick - but doesn't
OperationContext oc = OperationContext.Current;
host = oc.RequestContext.RequestMessage.Headers.From.Uri.Host;
} else {
//This works in IIS
host = HttpContext.Current.Request.UserHostName;
}
I have searched for hours and just can't find the solution - Is it really crazy to want to know the host that is calling my service?
There are 10 kinds of people that understand binary, those that do and those that don't...
|
|
|
|
|
Okay, finally found the answer here... thanks Phil
http://blogs.msdn.com/phenning/archive/2007/08/08/remoteendpointmessageproperty-in-wcf-net-3-5.aspx[^]
It seems this was an after thought in WCF, but it is imho essential to have the ability to know who is calling my public methods, and how - even if it is just for simple logging or debugging.
I know the request may have been proxied accross the world, and that the physical 'sender' may not be reachable, and that the underlaying protocol may not be 'whatever', but I will 'always' want to access this kind of information.
It is simple really, someone calls me and I give them an answer - but only if I can see their number in the display. Maybe I just want to call them back later.
So to simply sum up what I wanted this for:
public void Subscribe(string MyCallbackProtocol, int MyCallbackPort, string MyCallbackInterfaceName){
string YourAddress = (I want to determine this! I want to call you back later.);
}
There are 10 kinds of people that understand binary, those that do and those that don't...
|
|
|
|
|
Hi,
I have a WebService-WS, WebApp-WA and DTOproject-DTO. I made a DTO, SZ, serializable and added it as reference to WS and WA. When I call webmethod WS.oneWS.method() it returns an array of WS.SZ where I need DTO.SZ. Am I missing something here apart from making SZ serializable? Please help me.
Thanks in advance.
|
|
|
|
|
Thanks in Adv.
How to forcefully Kill a Thread in VB dot net 2005
Best Of Regards,
Mirza Rahman
|
|
|
|
|
It's not really the proper way to stop a thread, but you just call it's Abort method.
someThreadIStarted.Abort
|
|
|
|
|
Thanks for Replying I tried this
like
Thread.abort
thread.resetabort
thread.suspend
But nothing works
It give error that Call is from some outside referenced Thread
|
|
|
|
|
What's the exact error message?? Are you trying to kill a thread your code didn't create?? Or are you trying to have a thread kill itself??
|
|
|
|
|
U need to create an event handler with a delegate and call it when needed.
Just google it
Best Regards
3ala2
|
|
|
|
|
I have a setup.ini file has the information below, created in .NET setup project
[MSILoader]
MSIFileName=Allianz-PHI-Setup-SW2.msi
[FXSection]
SupportedRuntimes=1.1.4322
InstallUrl=http://go.microsoft.com/fwlink/?LinkId=9832
Message=1.1.4322
UseDefaultMessage=1
The out put of the above setup projects are setup.exe, setup.ini & Allianz-PHI-Setup-SW2.msi
I need to install one more msi package during the setup.exe installation(that is internally calling Allianz-PHI-Setup-SW2.msi) like the same I need to add one more msi called sqlxml.msi.
I need to add the sqlxml.msi details into the setup.ini file so that both the packages will get installed.
Please advise
-
|
|
|
|
|
NICE TO MEET wrote: I need to add the sqlxml.msi details into the setup.ini file
Who told you that and where did they get their information?
|
|
|
|
|
Hi,
I have a VS2005 Setup & Deployment project and I want to use a Windows installer condition to check to see if a folder already exists (sub-folder in the Application Folder hierarchy). I used the condition 'NOT Installed' in the properties for my sub-folder in the File System tab, but it looks like it's installing the sub-folder every time. Is 'NOT Installed' the right property to check to see if a folder already exists, or is that only used to check if the application is installed or not?
Thanks
|
|
|
|
|
JBAK_CP wrote: I want to use a Windows installer condition to check to see if a folder already exists
See if this helps[^]
|
|
|
|
|
Where I work we have a support tool that does many custom fixes.
Our desktops in the business are all locked down and have restricted/minimal access to the Control Panel.
We currently log on as Admins to machines to get full functionality or do a RunAs on the require .cpl applet.
Is there a way we can remotley contol another machines control panel or individual control panel applets; so the other machines control panel appears on our machines, so we dont have to remote to other machines or log users off
Thanks in advance
|
|
|
|
|
What does this have to do with the .NET Framework.
And, no, you can't do this. Remote code execution is a HUGE security risk, and is therefore HIGHLY restricted. I don't know of any control panels that support remote management.
|
|
|
|
|
I have a .NET 1.1 project I am trying to create msi package in .NET 2005 version setup project. After the build of the setup project msi gets created but the exe is giving error.
Also one more thing, if I run an msi package that should call one more msi (that is that need s to install ane more package). How to add the second msi to this one.
-
|
|
|
|
|
NICE TO MEET wrote: but the exe is giving error
What is the error?
Vasudevan Deepak Kumar
Personal Homepage Tech Gossips
The woods are lovely, dark and deep,
But I have promises to keep,
And miles to go before I sleep,
And miles to go before I sleep!
|
|
|
|
|
if I open the exe it is saying, It has encountered a problem it needs to close
-
|
|
|
|