Main Content


Put data in specific block of big image



setBlock(bigimg,level,locationWorld,data) sets the pixel data in the block of big image bigimg that contains coordinate locationWorld at the specified resolution level.


collapse all

Create a bigimage using a modified version of image "tumor_091.tif" from the CAMELYON16 data set. The original image is a training image of a lymph node containing tumor tissue. The original image has eight resolution levels, and the finest level has resolution 53760-by-61440. The modified image has only three coarse resolution levels. The spatial referencing of the modified image has been adjusted to enforce a consistent aspect ratio and to register features at each level.

bim = bigimage('tumor_091R.tif');

Display the bigimage, then create a circle ROI over the displayed image.

h = bigimageshow(bim);
hROI = drawcircle(gca,'Radius',470,'Position',[1477 2284]);

Choose the level at which to create a writeable bigimage. Level 3 is the coarsest resolution level.

maskLevel = 3;

Get the spatial referencing and pixel extents from the specified level.

ref = bim.SpatialReferencing(maskLevel);
pixelExtent = [ref.PixelExtentInWorldX,ref.PixelExtentInWorldY];

Create a writeable bigimage by specifying the spatial referencing instead of image data. This big image has one channel and is of data type logical.

bmask = bigimage(ref,1,'logical');

Loop through all blocks in the writeable big image to create a mask image. For each block, set the pixel values as 1 (true) for pixels inside the ROI and 0 (false) for pixels outside the ROI.

for cStart = 1:bmask.BlockSize(2):ref.ImageSize(2)
    for rStart = 1:bmask.BlockSize(1):ref.ImageSize(1)

        % Get the center of top left pixel of this block in world units.
        xyStart = [cStart,rStart].*pixelExtent;

        % Get the block size. The |'BlockSize'| property represents the
        % size as a 2-element vector of the form [row,column]. Switch the
        % order of the elements so that the block size is represented as
        % [x,y].
        bsize = bmask.BlockSize;
        numRows = bsize(1);
        numCols = bsize(2);

        % Determine which pixels have coordinates inside the ROI.
        roiPositions = hROI.Vertices;

        % Transform |roiPositions| from world coordinates to the intrinsic
        % image indices at the given resolution level.
        roiPositions = (roiPositions - xyStart) ./ pixelExtent + 1;

        blockMask = poly2mask(roiPositions(:,1),roiPositions(:,2), ...
                              numRows, numCols);

        % Set the pixel values of the block.

Display the mask.


Input Arguments

collapse all

Big image, specified as a bigimage object.

Resolution level, specified as a positive integer that is less than or equal to the number of resolution levels of bigimg.

Coordinate of a point, specified as a 1-by-2 numeric vector of the form [x y]. The location is specified in world coordinates, which are the pixel locations relative to the highest resolution level. The position must be a valid position within bigimg.

Pixel data, specified as a numeric array of the same data type as the big image, bigimg.ClassUnderlying. The first two dimensions of the data must match the block size at the specified level.


  • Create a writeable bigimage by using a syntax that does not initialize image data. If you create a bigimage by specifying the file name, directory name, or variable name of image data, or by using the apply function, then the bigimage is not writeable and you cannot use the setBlock function.

  • If the size of data is less than the block size bigimg.BlockSize, then setBlock pads the data with the default value, bigimg.UnloadedValue.

  • setBlock trims data for partial edge blocks.

Introduced in R2019b