← Back
Home › Systems › Energy, industry and public infrastructure

machine-vision-inspection

factory optical inspection and metrology

Cognex, Keyence, Zeiss

SystemTechniques · 14Primitives · 40 machine-vision-inspecti… Clock synchronizationdistributed · medium: Estimate a clock's offset from four timestamps, filter out queueing noise, and steer the clock smoothly; NTP gets about a millisecond, PTP with hardware timestamps under a microsecond.›Fpga dataflow pipelineshardware-os · medium: Fully pipelined fixed-latency FPGA logic: initiation interval, HLS, BRAM and DSP budgetingGeometric pattern matchingmachine-learning · medium: Find a model contour in an image under rotation and scale using gradient-orientation features and a pyramid search with a match scoreImage segmentationmachine-learning · medium: Pixel/instance segmentation networks and promptable segmentation (Mask R-CNN, SAM-style)Object detectionmachine-learning · medium: A network turns an image into scored, classified boxes in one pass; anchor-based, anchor-free and set-prediction heads differ in how they assign labels and remove duplicates.›Optical character recognition (text…machine-learning · medium: Find text lines in an image, read each crop with a CTC or attention model, then correct with a lexicon and score confidence for review routing.›Levenberg marquardtnumerics · medium: Damped Gauss-Newton nonlinear least squares for model calibrationConvolutional networksml-training · low: Layers that slide a small shared kernel over a grid, giving translation equivariance and few parameters; still the most compute-efficient choice for many vision tasks.›Data augmentationmachine-learning · low: Generate label-preserving variants of each training example on the fly to enlarge the effective dataset and bake in invariances.›Convolution image filtersgraphics · low: Separable 2D convolution kernels for blur, sharpen (unsharp mask) and edges with border handlingBarcode symbologiesdata-structures · low: Encode/decode 1D and 2D barcodes (Code 128, DataMatrix, QR) with their error correctionReal time fieldbusembedded · low: Deterministic cyclic fieldbuses (EtherCAT, PROFINET IRT): on-the-fly frames, distributed clocksImage pyramidsgraphics · low: Gaussian/Laplacian pyramids and tiled multi-resolution images for zoom and blendingImage warping remapgraphics · low: Resampling with lookup maps: resize, lens undistortion, homographies, bilinear/bicubicSubpixel edge measurementmachine-learning · low: Locate edges to a fraction of a pixel by interpolating the gradient peak along caliper projections, then fit lines or circlesLine buffer streamingmedia · low: Stream image processing with a few line buffers instead of full frames (hardware ISPs)Weight quantizationml-training · low: Store weights as 8/4-bit integers with per-group scales; decode is memory-bound, so 4-bit weights give about 3-4x tokens/s and fit 4x larger models.›Ransacrobotics · low: Random sample consensus for robust model fitting with outliers (homographies, planes, poses)Structured light depthrobotics · low: Depth from projected dot patterns or time-of-flight for 3D face capturePID controllerrobotics · low: Drive a measured value to a setpoint with three terms on the error: proportional, integral, derivative. The default feedback loop for motors, heaters, drones and gimbals.›Phase-locked loop (digital)signal-processing · low: Feedback loop that locks a local oscillator to the phase and frequency of an input tone; the core of carrier recovery, clock recovery and GNSS tracking.›Robust statisticsstatistics · low: Location and scale estimates (median, MAD, trimmed mean, Huber M-estimator) that stay correct when a minority of the data is garbage.›Loss functionsmachine-learning · low: The scalar a model minimizes: cross-entropy for classes and tokens, MSE or Huber for values, each with a stable form and simple gradient.›+ 7 more techniques+ 24 more primitives

Similar systems

Share the most building blocks with this one.

camera-isp 19 %wafer-inspection-yield 15 %