I got the image on the vi,now I want to take rectangle potion from that image and compare with other image

[Image comparison]

Hi,

Now i have a image on VI ,i want to take rectangale postion of the image and compare with the other image(In this image also i need to take a small postionof it) and  show the result.

Both image are equal or not....if it is equal show pass if the image is not ok fail.

Regards,

Sri.

[MikeS81]

Hi Sri,

what do you mean with you have an image on vi? Do you read it from a file?
See this link, http://forums.ni.com/ni/board/message?board.id=170&message.id=320659#M320659

Hope it helps.

Mike

[jpaul]

Hi,

I also wanted to follow up to see if I could get a little more clarification on your issue.

1. Since you refer to picture width and height (I assume in pixels) are you using picture controls in LabVIEW to load the two pictures (see attached screenshot)?
2. Did you want to compare the pixels by RGB convention?
3. Do you know the location of the pixels?

Anyother information would help to narrow down the advice that I can provide to you on this issue. Thanks!

Cheers,

Jonah
Applications Engineer
National Instruments

[Image comparison]

First i would like to thanks for the inputs,

I have made one Vi for comparing the two images and display  equal or not.

Problem:if there is a small change in the picture it is showing false ,but the image is ok only the contract ,clarity ,inclination is the problem... i need your help in controlling that image......like controlling the colors and telling if the color , shade, projection of the image of the image is in between this range show it as ok....

i started with cluster taking the out put of the BMP file, but i am not able to do....can u help me on this.

i am attaching the file what i have made...

Regards,

Sri

Color

[jpaul]

Hi Sri,

I expanded my original VI that I attached to include view the color pixel by pixel of an image (see attached VI). In the VI there are three arrays one for Red, Green and Blue. The elements of the arrays start at the upper left hand corner of the picture going left to right and then down the picture. You could use this breakdown to define an acceptable variation between the pixel colors of the two images (i.e. if the acceptable range was within 10% red; therefore, since red spans from 0 to 255, the two pictures could only differentiate by ~25 to be equal). I hope this helps to point you in the right direction.

Cheers,

Jonah
Applications Engineer
National Instruments