Mixin interface for performing blocked image automation
BlockedImageAutomation mixin class provides an interface for
attaching blocked image properties to an automation algorithm. You can add this class only to
automation algorithms used by the Image Labeler
To use the
vision.labeler.mixin.BlockedImageAutomation object, open
the Image Labeler app,
load a blocked image, and create at least one label. Then, click Select
Algorithm in the Automate Labeling section of the app
toolstrip. Select Add Blocked Image Algorithm and then choose
Create New Blocked Image Algorithm. A class template opens in the
editor that inherits from both the
class and the
vision.labeler.mixin.BlockedImageAutomation mixin class.
Edit this template to create your blocked image automation algorithm.
BlockedImageAutomation mixin class predefines this
set of properties.
BatchSize — Number of blocks supplied to the
apply function at a time
Number of blocks supplied to the
apply function at a time,
specified as an integer valued vector of length equal to the
NumDimensions property of the blocked image.
BorderSize — Size of additional border elements
1-by-D numeric vector
Size of additional border elements to add around a block in each dimension,
specified as a 1-by-D numeric vector, where D is
equal to the
NumDimensions property of the first
InclusionThreshold — Ratio that determines whether a block should be considered for processing
Ratio that determines whether a block should be considered for processing, specified as a value in the range [0 1]. The value specifies the minimum percentage of nonzero pixels in the block. A value of 0 means that only one pixel in the block must be true/nonzero for it to be included in processing. A value of 1 requires that all pixels in the block be true/nonzero for it to be included in processing.
PadMethod — Method used to obtain padding pixels
Method used to obtain padding pixels to honor the
PadPartialBlocks parameters, specified as scalar
PadPartialBlocks — Pad partial blocks that might exist on the edges
Pad partial blocks that might exist on the edges, specified as a logical scalar,
Resume — Continue processing from where a previous run stopped
Continue processing from where a previous run stopped, specified as a logical
|Implements the blocked image automation algorithm
Introduced in R2021a