image subtraction on intensity graph

[hllio]

the acquisition is realized in this mode:

first we acquisit the background (see figure for understanding what i mean with background)

after we acquisit the object + background...

In my Vi if the sensor work with axis ortoganl to the object, i don't doing a real background subtraction but i use the z axis for take only the data that represent the object.

I don't know if is correct talk about background subtraction but i want to delete the background and take only the object...

THE RESULT IS ON MY ATTACHMENT... IS A TEST 

[hllio]

with my solution work correctly..i have done test,for assure what i mean...

but my vi function good only when the sensor is ortogonal to the plane of the object....

and not work correctly when the sensor is rotated...ok?

[altenbach]

Your two images have different sizes. How should they align?

[hllio]

because the image is taken by the sensor using a ROI selector (using two control on the graph with reference):

if you have seen the image you can see that there's an area that is more near coloured by blue...this is the selected area..ok? 

However the size is not a problem because i can solve this problem..with roi,look only in the interest area

[altenbach]

Then please give the offsets for alignment. Everything is blue one way or another.

[hllio]

NO SORRY..is not correct the data that i have sent...

[hllio]

ok...the new file

[altenbach]

Maybe something like this? You can adjust the threshold accordingly.

[hllio]

..OK ...is so noised image but with roi on the object can resolve this type of problem..thankss

[hllio]

sorry,but the solution is not correct and i have enough problem...and i help all the community of labview if there are someone that is  interested

The problem is this:

i  have a sensor that giving a measure of distance from his axis (that for me is z-axis using carthesian coordinates)...

Z axis is the axis coaxial with the sensor (for me). 

http://www.ifm.com/obj/gb_o3d200.swf

The result of the output is a matrix of  z-data (in meters),between the object and the sensor. 

I want to use this sensor in an enviroment that is dirty,and my problem is about the background..(that for me is a wall)...

i want to exclude the wall from my measure and i want to consider only the object (target of the measure) that is above the wall...

For realized this type of application,i want to save an image of the background (only once or when the wall is dirty) and compare the background image,with the image of the object+background: the result that i want is seeing only the object.

I want to exclude the value of the matrix that having a values that are more bigger than the quote between sensor and object...

I have post the pseudo result and 2 data acquired from the sensor:

The sensor is not with the plane parallel to the surface of the object but is inclinated of about 40 degrees..This type of orientation is for avoid that object with high reflective behaviour can noised the measure...