NAME 2D Ikusmen Laborategia
Ceitek 2D ikusmen laborategia dauka, non IA aplikatzen den ikusmen artifizialaren arazoei eta irudiak prozesatzeko teknika aurreratuei: -objektuen detekzioa -sailkapena -segmentazioa -ikuskapena -metrologia Teknika hauek irtenbide berritzaileak garatzera bideratuta daude, hala nola prozesuen automatizazioaren barruan kalitatearen kontrola eta ikuskapena, doitasun metrologia, desmuntatze automatizatua, etab. Hala ere, arlo honetan dugun ezagutzak hainbat eremu desberdinetan lan egitea ahalbidetzen digu, nahiz eta Ceitek esperientzia zabala duen Industria 4.0 aplikazioetan, garraioa eta mugikortasun adimentsua lantzen dituen proiektuetan ere ari da.
FIELDS OF APPLICATION
Deployment and Application
Developments
MOST OUTSTANDING EQUIPMENT AND COMPONENTS
-
2D vision metrology
To achieve high-precision 2D dimensional control (hundredth or thousandth of a mm), telecentric optics of different sizes and precision benches are available.
-
Defectology with 2D vision
Lighting and cameras that operate in the visible spectrum, as well as a variety of optics to cover different object sizes, working distances, etc. Once the product images are obtained, artificial intelligence algorithms are applied for classification and defect detection
-
High-temperature process monitoring
High-level thermographic camera that allows monitoring temperatures up to 2,000 °C.
-
Material classification
Hyperspectral camera operating in the visible and infrared spectrum, allowing materials to be classified based on their spectral signature (the material's response to these wavelengths).
SERVICES OFFERED BY THE ASSET
2D parts inspection for precision metrology
Computer vision techniques are used to measure different aspects of the object.
Classification of materials using a hyperspectral camera.
Computer vision techniques are used to differentiate between various materials in the image based on their spectrum.
Detection of visual defects using 2D vision, thermographic cameras, or deflectometry.
Computer vision techniques are used to detect visual defects such as marks, dirt, etc.
Video or image analysis for applications related to automatic dismantling tasks
Deep learning-based algorithms for the detection and precise localization of different screws or other class elements in dismantling.
Video or image analysis for CCAM-related applications
Different deep learning-based algorithms for the detection of (1) infrastructure assets (signs, traffic lights, construction cones, etc.), (2) defects (potholes, cracks, faded road markings, etc.); (3) classification of visibility level due to fog scenarios; (4) vegetation segmentation (trees and grass).
ENTITY MANAGING THE ASSET
Contact person:
Diego Borro Yágüez
dborro@ceit.es