Image Processing & Correction
39
Terms
Term Reference
39 terms| Term | Description | |
|---|---|---|
| Anisotropic filter Edge-preserving smoothing via Perona-Malik nonlinear diffusion | An iterative edge-preserving smoothing filter formulated as a partial differential equation. The image evolves under non... | |
| Arithmetic image operations Pixel-wise math — add, subtract, multiply, divide, and the variants | Pixel-wise math on images: addition, subtraction, multiplication, division, exponential, logarithm, polynomial, square, ... | |
| Average filter Box-mean smoothing — every output pixel is the unweighted mean of its neighborhood | A convolutional smoothing filter where every neighborhood pixel is weighted equally. The output pixel is the arithmetic ... | |
| Background Removal | A pre-processing engine in StrataQuest that models and subtracts the background intensity pattern from imaging channels,... | |
| Beer-Lambert law The exponential relationship between light absorption, path length, and absorber concentration | The physical law relating transmitted light intensity to absorber concentration: I = I₀ · exp(−ε·c·l). Taking th... | |
| Bitwise logical operations AND, OR, NOT, XOR — combining masks by truth table | Pixel-wise logical operations on binary masks. AND keeps pixels in both inputs. OR keeps pixels in either. XOR keeps pix... | |
| Channel One dimension of a multi-dimensional image | One dimension of a multi-channel image. RGB has three channels (red, green, blue). A fluorescence acquisition has one ch... | |
| Classifier | A machine-learning engine in StrataQuest that learns to automatically identify and classify tissue structures based on u... | |
| Color Separation | A pre-processing engine that separates the individual stain contributions from a composite brightfield RGB image. By def... | |
| Color space conversions Re-coordinating color information for the analysis at hand | Engines that convert color images between coordinate systems: BGR/RGB, HLS, HSV, Lab, Optical Density, Grayscale. Differ... | |
| Convolution The slide-and-sum operation underlying most image filters | Slide a small grid of weights (a kernel) across the image. At each position, multiply each pixel by its weight and sum. ... | |
| Custom filter User-defined convolution kernel | A convolutional filter where the user supplies the kernel weights directly. Every other convolutional filter (Gauss, Sob... | |
| Field of View (FOV) One acquisition tile — the unit the camera captures, which may not be the unit the analysis cares about | A single acquisition frame from the microscope camera — the field the optics image at one position. A whole-slide acqu... | |
| Fill holes The topological operation that closes any background region surrounded by foreground | An operation on binary masks that converts every fully-enclosed background region into foreground. Unlike Close (a size-... | |
| Foreground / Background The two regions a mask partitions an image into | The two regions a binary mask partitions an image into. Foreground is what you want; background is everything else. The ... | |
| Grayscale image Single-channel intensity image | An image with one intensity value per pixel — no color, just brightness. The native format of fluorescence channels an... | |
| Image bit depth 8-bit, 16-bit, 32-bit — and what gets lost between them | The number of bits used per pixel to store intensity. 8-bit holds 256 values; 16-bit holds 65,536; 32-bit holds billions... | |
| Image channel operations Pulling channels apart, recombining them, swapping their order | Engines that manipulate the channel structure of multi-channel images: extract a single channel as grayscale, extract th... | |
| Image filters The family of operations that recompute each pixel from its neighborhood | Operations that produce a new image where each output pixel is a function of the input pixel and its neighbors. Differen... | |
| Laplace filter Discrete second derivative — edge response via the Laplacian operator | A convolutional filter approximating the discrete Laplacian (∇²I). Output is bright where intensity changes rapidly a... | |
| Lookup table (LUT) A function from input intensity to output intensity, applied per pixel | A function that maps each input intensity value to an output value, applied per pixel. LUTs encode display colors (false... | |
| Mask Binary mask | An image where each pixel is either on or off — typically 0/255 or 0/1. The same data structure plays two roles: marki... | |
| Otsu Threshold Otsu Threshold Engine (Global) | A global pre-processing engine that computes a single optimal intensity threshold for the entire sample using Otsu's met... | |
| Pixel-wise Min and Max Per-pixel reduction across an image stack | Two engines that produce an output image where each pixel takes the maximum (or minimum) value across a stack of input i... | |
| Projection Z-Stack Projection Engine | A pre-processing engine that fuses multiple z-stack slices into a single image with greater depth of field, using pixel-... | |
| Re-Label operations Renumbering label IDs after a coded image has been edited | Engines that renumber labels in a coded image to produce a clean sequence — closing gaps left by previous removal step... | |
| Region of Interest (ROI) A user-defined sub-region that scopes which pixels an analysis acts on | A user-defined sub-region of an image — a rectangle, polygon, or arbitrary mask — that restricts subsequent analysis... | |
| Removal operations Filtering a labeled image by criterion — size, intensity, contact, seeding | Engines that remove labels or objects from a detection result based on criteria such as size, intensity, edge contact, o... | |
| Scaling factor The output multiplier that recurs across arithmetic and filter engines | A multiplier applied to an engine's output, allowing the result to be scaled up or down before it's written. Appears in ... | |
| Sharp filter Detail amplification via box-average high-pass | A sharpening filter that amplifies fine detail by adding back the high-frequency component of an image. The high-frequen... | |
| Skeleton & spur operations The medial-axis representation of a mask, and the eight engines that manipulate it | Operations that compute and refine the skeleton of a binary mask — the 1-pixel-wide medial axis of each foreground reg... | |
| Spectral Unmixing | A mathematical technique for separating overlapping fluorescence signals into individual channel contributions. In multi... | |
| Stitching Combining adjacent acquisition tiles into a single composite image | The process of combining multiple adjacent FOVs into a single stitched image. Approaches differ by how positions are det... | |
| Structuring element The shape morphology uses to probe a binary mask | A small shape (3×3, disk, line, etc.) that morphology operations slide across a mask. It defines the neighborhood used ... | |
| The StrataQuest Workspace Layer · Section · Engine — how a StrataQuest analysis is composed | Every StrataQuest analysis is built inside a layer. A layer has four engine slots — Pre-Processing, Detection, Measure... | |
| Thresholding & Comparison Turning grayscale intensities into binary decisions | The family of operations that produces a binary mask by testing each pixel against a threshold value, an interval, or an... | |
| Top-hat transform Original minus opening — the bright structures the opening removed | An operation that subtracts a morphologically-opened image from the original. The result isolates bright features smalle... | |
| Unsharp filter Unsharp mask — sharpening via Gaussian high-pass | The classic photographic unsharp mask. Two independent parameters: σ (Gaussian radius — sets the scale of detail bein... | |
| Virtual Channel Virtual Channel Engine | A pre-processing utility engine that generates a virtual channel by combining information from selected grayscale images... |