Region according to value

I have a pixellated 8*8 data set in which the pixels have value either 1 or 2.

1 1 1 1 1 1 1 1,

1 1 c2 2 2 1 1 1,

1 1 2 2 2 1 1 1,

1 1 2 1 2 2 1 1,

1 1 1 2 2 2 1 1,

1 1 1 1 2 1 1 1,

1 1 1 2 2 1 c1 1,

1 1 1 1 1 1 1 1,

There are 2 control points C1 and C2 which have their value as 0. I need to create a continuous region around the control points such that the region around C1 has a summation of the pixel value as 32 and the region around c1 has the value of 44. The summation of values in the dataset is 32+44=76. As I need to create a region, I thought of using the region grow algorithm using my control points as the seed point but it grows the region according to the similar intensity of pixels. Is there any way to do this? Thanks in advance.

[Image Analyst]

Explain in detail exactly how you get the 32 and 44 from a surrounding group of 1's and 2's.

[Rutuja Kate]

If I start from c1 and start growing around it, acquiring neighbouring pixels, the pixels shown in bold satisfy the criteria. Once I finish this region, then I start from c2 and the rest of the pixels give me a value of 44.

1 1 1 1 1 1 1 1

1 1 c2 2 2 1 1 1

1 1 2 2 2 1 1 1

1 1 2 1 2 2 1 1

1 1 1 2 2 2 1 1

1 1 1 1 2 1 1 1

1 1 1 2 2 1 c1 1

1 1 1 1 1 1 1 1

Answer 1

[Image Analyst]

You can set c1 and c2 as "true" in a binary image and then repeatedly call imdilate() and then call reqionprops() to find out the resulting area after a layer of pixels has been grown around it. Or just compute the area you need and make a circle with that area centered at that point. See the FAQ

http://matlab.wikia.com/wiki/FAQ#How_do_I_create_a_circle.3F

What do you really want to do? What's the use case?

[Rutuja Kate]

c1 and c2 are two facilities in two towns and all the other pixels are small villages. The pixel positions are manipulated such that the position of the pixel(village) from the facility is to the scale to the actual distance on land. The pixel values represent the population(1 and 2).c1 and c2 can cater to the needs of a specific amount of population (in this case 32 and 44). I need to assign the villages to the facilities satisfying the above criteria and to optimise the travelling distance.

[Image Analyst]

How are you going to manipulate the pixels? The pixels must be in a fixed rectangular grid if you're using a matrix. And what do the 1 and 2 represent? Your growing around c1 and c2 doesn't seem to care what the values are.

[Rutuja Kate]

1 and 2 represent the population of the particular village.

when growing around c1, it will follow the 8-neighbour approach and calculate the total value of those 9 pixels. if the summation of values of the 9 pixels is less than the criteria of 32, it will again grow and occupy next neighbouring pixels. again it will calculate the summation of the values, and check against the criteria, it will continue to grow till the summation of the values is 32, and then it will terminate.

[Image Analyst]

So did you try what I said - to use imdilate? If not, why not?