Basic Operations Module
39
Terms
Term Reference
39 terms| Term | Description | |
|---|---|---|
| Anisotropic filter Edge-preserving smoothing via Perona-Malik nonlinear diffusion | Edge-preserving smoothing by nonlinear diffusion: fast inside uniform regions, slow across edges. | |
| Arithmetic image operations Pixel-wise math: add, subtract, multiply, divide, and the variants | Pixel-wise math on one or two images, or an image and a constant: add, subtract, multiply, log, and more. | |
| Arithmetic Operations | The Basic Operations Module group for pixel-wise math: add, subtract, multiply, divide, log, exp, and more. | |
| Average filter Box-mean smoothing: every output pixel is the unweighted mean of its neighborhood | Smoothing in which every pixel of an N×N neighborhood counts equally: the output is their mean. | |
| Bilateral Filter | Edge-preserving smoothing: neighbors count by both distance and intensity similarity, so boundaries stay sharp. | |
| Bitwise logical operations AND, OR, NOT, XOR: combining masks by truth table | Pixel-wise AND, OR, XOR, and NOT on binary masks: intersections, unions, and inversions. | |
| Canny Edge Detector | Multi-stage edge detector (smooth, gradient, thin, hysteresis) that yields thin, connected edge contours. | |
| Color space conversions Re-coordinating color information for the analysis at hand | Converts images between RGB, HLS, HSV, Lab, optical density, and grayscale to isolate the property you need. | |
| Color Space Conversions | The Basic Operations Module group that converts between color spaces (BGR, HLS, HSV, Lab, grayscale). | |
| Connected Component Labeling | Connected-component labeling: gives each separate foreground region its own ID, producing a coded image. | |
| Convolution The slide-and-sum operation underlying most image filters | Slide a small grid of weights over the image; each output pixel is the weighted sum of its neighborhood. | |
| Custom filter User-defined convolution kernel | A convolution filter whose kernel weights you enter yourself; named filters are fixed kernels of this kind. | |
| Fill holes The topological operation that closes any background region surrounded by foreground | Turns every fully enclosed background region of a mask into foreground, however large the hole. | |
| Filters Overview | The Basic Operations Module filters: a kernel slides over the image, computing each pixel from its neighborhood. | |
| Fuse Events | Merges events from several coded images into one, optionally restricted by a mask. | |
| Gaussian Filter | Smoothing with a Gaussian-weighted kernel; a single parameter, sigma, sets how far the blur reaches. | |
| Image channel operations Pulling channels apart, recombining them, swapping their order | Engines that extract, swap, or recombine the channels of a multi-channel image. | |
| Image Data Operations | Utilities that convert bit depth, extract or swap channels, apply masks, and build RGB composites. | |
| Image filters The family of operations that recompute each pixel from its neighborhood | Operations in which each output pixel depends on the input pixel and its neighbors. | |
| Label Management | Operations that refine coded images after labeling: cut bottlenecks, fill holes, grow or shrink, re-label. | |
| Laplace filter Discrete second derivative: edge response via the Laplacian operator | A second-derivative filter: bright where intensity changes rapidly, zero where it is locally uniform. | |
| Logical Operations | Pixel-by-pixel AND, OR, NOT, and XOR for combining masks into intersections, unions, and exclusions. | |
| Median Filter | Replaces each pixel with its neighborhood's median, setting aside isolated extreme pixels while keeping edges. | |
| Membrane detector | A Basic Operations filter that finds membrane-like lines along chosen directions, dark-on-bright or bright-on-dark. | |
| Morphological Operations | Dilate, erode, open, and close: the four basic ways to grow, shrink, and reshape a binary mask. | |
| Optical Density Conversion | Converts brightfield RGB values to optical density, in which the value is proportional to the amount of stain. | |
| Pixel-wise Min and Max Per-pixel reduction across an image stack | Each output pixel takes the larger, or the smaller, of two input images’ values at that position. | |
| Post Processing | Refines masks between thresholding and measurement: removes fragments, fills gaps, smooths boundaries. | |
| Re-Label operations Renumbering label IDs after a coded image has been edited | Renumbers the labels of a coded image into a clean sequence, closing gaps left by removals. | |
| Removal operations Filtering a labeled image by criterion: size, intensity, contact, seeding | Removes labels or objects from a detection by size, intensity, or seed overlap. | |
| Sharp filter Detail amplification via box-average high-pass | Sharpens by adding back fine detail: the difference between the image and its local average. | |
| Skeleton & spur operations The medial-axis representation of a mask, and the eight engines that manipulate it | Reduce mask regions to 1-pixel-wide center lines, then trim, extend, or split their branches. | |
| Sobel Filter | A gradient filter that measures edge strength and direction with 3×3 kernels. | |
| Statistical Operations | The Basic Operations Module statistics group: the pixel-wise maximum or minimum across images. | |
| Structuring element The shape morphology uses to probe a binary mask | The small shape morphology slides across a mask; it plays the role a kernel plays for filters. | |
| Threshold & Compare | Turns grayscale images into binary masks by thresholding, and compares images pixel by pixel. | |
| Thresholding & Comparison Turning grayscale intensities into binary decisions | Makes a binary mask by testing each pixel against a value, an interval, or an Otsu threshold. | |
| Top-hat transform Original minus opening: the bright structures the opening removed | Subtracts the opened image from the original, isolating small bright features on an uneven background. | |
| Unsharp filter Unsharp mask: background removal by subtracting a blurred copy | Subtracts a Gaussian or Average blur of the image to lower a high background; Kernel radius sets the blur. |