Skip to main content

[JAI] Warp Operation

5 replies [Last post]
gwampole
Offline
Joined: 2005-02-15
Points: 0

Good day, I have a question regarding the "warp" JAI operation. I've noticed that the super-class: Warp defines a method warpSparseRect that allows a period in the x and y directions to be specified.

Further, I notice that I must create an instance of one of Warp's subclasses to be used in the ParameterBlock that is passed to the create method of JAI, in order to create the operation.

My question is this: how can I specify the period values? Let's say I want to create an instance of WarpAffine, it's not immediately clear to me how I might specify the periodX and periodY values that will be used in the invocation of warpSparseRect during the operation. Of course, I might be missing the point altogether, in which case, clarification would be appreciated!

Thanks!

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Brian Burkhalter

You can't specify the values of the period. The intent of the methods
containing period{X,Y} parameters is to allow the computation to be effected
when only a subset of the boundary points guarantee sufficient accuracy. This
might be used by some type of warp which was computationally expensive for
example. In point of fact I don't think that any of our implementations
actually call these methods anywhere. They are there for completeness.

Brian

On Fri, 27 May 2005 jai-interest@javadesktop.org wrote:

> Good day, I have a question regarding the "warp" JAI operation. I've noticed that the super-class: Warp defines a method warpSparseRect that allows a period in the x and y directions to be specified.
>
> Further, I notice that I must create an instance of one of Warp's subclasses to be used in the ParameterBlock that is passed to the create method of JAI, in order to create the operation.
>
> My question is this: how can I specify the period values? Let's say I want to create an instance of WarpAffine, it's not immediately clear to me how I might specify the periodX and periodY values that will be used in the invocation of warpSparseRect during the operation. Of course, I might be missing the point altogether, in which case, clarification would be appreciated!

----------------
Brian Burkhalter
Advanced Development, Graphics and Media
Software Chief Technology Office
Sun Microsystems, Inc.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This email message is for the sole use of the intended recipient(s)
and may contain confidential and privileged information. Any
unauthorized review, use, disclosure or distribution is prohibited.
If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@jai.dev.java.net
For additional commands, e-mail: interest-help@jai.dev.java.net

gwampole
Offline
Joined: 2005-02-15
Points: 0

Hey Brian, thanks for the reply. It has helped me to formulate a better phrased question.

In my case, it appears that the warp operation is computationally intensive, I believe it is because of the size of the image. So: using your implementations; what would be the best way to scale down the scope of the operation?

Brian Burkhalter

Are you using an instance of one of the warp classes already provided by JAI
or are you using an instance of a Warp subclass that you implemented? If the
former, what are the specific parameters provided?

Brian

On Mon, 6 Jun 2005 jai-interest@javadesktop.org wrote:

> In my case, it appears that the warp operation is computationally intensive, I believe it is because of the size of the image. So: using your implementations; what would be the best way to scale down the scope of the operation?

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@jai.dev.java.net
For additional commands, e-mail: interest-help@jai.dev.java.net

gwampole
Offline
Joined: 2005-02-15
Points: 0

I am using a Warp implementation provided by JAI, WarpPolynomial to be exact. Although implementing a new subclass of Warp is not out of the question if no suitable method of scaling down the operation exists.

Message was edited by: gwampole

Brian Burkhalter

Depending on the order of your polynomial you might be able to get better
performance by using a WarpGrid object wherein the grid nodes are calculated
exactly using your polynomial. This would mean in effect that your mapping
would in effect be piecewise linear.

Brian

On Thu, 9 Jun 2005 jai-interest@javadesktop.org wrote:

> I am using a Warp implementation provided by JAI, WarpPolynomial to be exact. Although implementing a new subclass of Warp is not out of the question if no suitable method of scaling down the operation.
> ---
> [Message sent by forum member 'gwampole' (Garrett Wampole)]

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@jai.dev.java.net
For additional commands, e-mail: interest-help@jai.dev.java.net