image_enhance.analysisRun.applyPipeline

Description

Converts every input image to RGB double data, then applies each history step in order. The same step sequence is applied independently to every image. The input arrays and step records are not modified, so the function can be used outside the GUI for reproducible preview or export processing.

Inputs

images

One numeric image or a cell array of numeric images. Grayscale and multichannel inputs are converted to M-by-N-by-3 RGB double images and limited to [0,1]. The returned cell order matches the input order.

steps

Structure array created by image_enhance.analysisRun.makeStep. Steps are reshaped to a column and executed in array order. An empty array performs normalization only.

contexts

Optional cell array with one entry per image. Context is used by "White ROI calibration" to find whiteRoi, either directly as context.whiteRoi or as context.item.whiteRoi. Other step kinds ignore it. Default: one empty context per image.

Step Fields

kind

One label returned by image_enhance.imagePreview.presentationData.toolKinds: "Brightness/contrast", "Local contrast", "Sharpen", "Hue/saturation", "White balance", "White ROI calibration", or "Subject-preserving enhance".

amount

Primary numeric control. It is brightness percentage, local- contrast strength, sharpening strength, hue rotation in degrees, white-balance strength, or protected-enhancement blend strength, depending on kind.

secondary

Secondary numeric control. It is contrast percentage, radius in pixels, saturation percentage, warm/cool percentage, or target background white level in percent, depending on kind.

referenceIndex

Preserved history metadata; applyPipeline does not use it.

label

Human-readable history text; applyPipeline does not branch on it.

Outputs

processed

Column cell array of M-by-N-by-3 double images in [0,1]. Each output preserves its corresponding source height and width.

Errors

labkit_ImageEnhance_app:UnknownEnhancementStep

A step kind is not one of the supported labels.

labkit_ImageEnhance_app:MissingWhiteRoi

"White ROI calibration" is used without a valid [x y width height] ROI in that image's context. MATLAB indexing errors - A nonempty contexts array has fewer entries than images.

Example

imageData = repmat(linspace(0.1, 0.8, 20), 12, 1);
steps = [ ...
image_enhance.analysisRun.makeStep("Brightness/contrast", 10, 20, 0); ...
image_enhance.analysisRun.makeStep("Sharpen", 30, 1, 0)];
processed = image_enhance.analysisRun.applyPipeline(imageData, steps);
assert(iscell(processed) && isequal(size(processed{1}), [12 20 3]))