(3553.04032020.158000154) L.M.4.0 (Q4081958): Difference between revisions
Jump to navigation
Jump to search
(Changed label, description and/or aliases in en: Setting new description) |
(Changed label, description and/or aliases in en, and other parts: Adding English translations) |
||||||||||||||
| label / en | label / en | ||||||||||||||
(3553.04032020.158000154) L.M.4.0 | |||||||||||||||
| Property / summary | |||||||||||||||
THE PROBLEM OF CLASSIFYING THE DIFFERENT TYPES OF DEFECTS THAT CAN OCCUR ON RAW MATERIALS OF DIFFERENT NATURE IS TYPICALLY ADDRESSED WITH CLASSIC COMPUTER VISION TECHNIQUES THAT EXPLOIT VISUAL FEATURES? FIRST IDENTIFIED BY HUMAN EXPERTS BASED ON THEIR EXPERIENCE AND INTUITION (A PHASE THAT IS NORMALLY CALLED?FEATURE ENGINEERING?). THESE TECHNIQUES THAT EXPLOIT CHARACTERISTICS ENGINEERED BY HUMAN EXPERTS (?HAND-ENGINEERED?), HAVE LED TO THE CREATION OF A VAST SET OF DESCRIPTORS THAT ARE INVARIANT BOTH TO THE TYPICAL GEOMETRIC AND RADIOMETRIC DEFORMATIONS INTRODUCED BY THE ACQUISITION SYSTEMS, AND TO THE VARYING ENVIRONMENTAL AND LIGHTING CONDITIONS. ALGORITHMS SUCH AS, FOR EXAMPLE: INVARIANT SCALES FEATURE TRANSFORM? (SIFT),?SPEEDED UP ROBUST FEATURES? (SURF) OR?HISTOGRAM OF ORIENTED GRADIENT? (HOG) HAVE BECOME WIDESPREAD PRECISELY BECAUSE OF THEIR CHARACTERISTICS OF INVARIANCE TO THE MAIN TRANSFORMATIONS (English) | |||||||||||||||
| Property / summary: THE PROBLEM OF CLASSIFYING THE DIFFERENT TYPES OF DEFECTS THAT CAN OCCUR ON RAW MATERIALS OF DIFFERENT NATURE IS TYPICALLY ADDRESSED WITH CLASSIC COMPUTER VISION TECHNIQUES THAT EXPLOIT VISUAL FEATURES? FIRST IDENTIFIED BY HUMAN EXPERTS BASED ON THEIR EXPERIENCE AND INTUITION (A PHASE THAT IS NORMALLY CALLED?FEATURE ENGINEERING?). THESE TECHNIQUES THAT EXPLOIT CHARACTERISTICS ENGINEERED BY HUMAN EXPERTS (?HAND-ENGINEERED?), HAVE LED TO THE CREATION OF A VAST SET OF DESCRIPTORS THAT ARE INVARIANT BOTH TO THE TYPICAL GEOMETRIC AND RADIOMETRIC DEFORMATIONS INTRODUCED BY THE ACQUISITION SYSTEMS, AND TO THE VARYING ENVIRONMENTAL AND LIGHTING CONDITIONS. ALGORITHMS SUCH AS, FOR EXAMPLE: INVARIANT SCALES FEATURE TRANSFORM? (SIFT),?SPEEDED UP ROBUST FEATURES? (SURF) OR?HISTOGRAM OF ORIENTED GRADIENT? (HOG) HAVE BECOME WIDESPREAD PRECISELY BECAUSE OF THEIR CHARACTERISTICS OF INVARIANCE TO THE MAIN TRANSFORMATIONS (English) / rank | |||||||||||||||
Normal rank | |||||||||||||||
| Property / summary: THE PROBLEM OF CLASSIFYING THE DIFFERENT TYPES OF DEFECTS THAT CAN OCCUR ON RAW MATERIALS OF DIFFERENT NATURE IS TYPICALLY ADDRESSED WITH CLASSIC COMPUTER VISION TECHNIQUES THAT EXPLOIT VISUAL FEATURES? FIRST IDENTIFIED BY HUMAN EXPERTS BASED ON THEIR EXPERIENCE AND INTUITION (A PHASE THAT IS NORMALLY CALLED?FEATURE ENGINEERING?). THESE TECHNIQUES THAT EXPLOIT CHARACTERISTICS ENGINEERED BY HUMAN EXPERTS (?HAND-ENGINEERED?), HAVE LED TO THE CREATION OF A VAST SET OF DESCRIPTORS THAT ARE INVARIANT BOTH TO THE TYPICAL GEOMETRIC AND RADIOMETRIC DEFORMATIONS INTRODUCED BY THE ACQUISITION SYSTEMS, AND TO THE VARYING ENVIRONMENTAL AND LIGHTING CONDITIONS. ALGORITHMS SUCH AS, FOR EXAMPLE: INVARIANT SCALES FEATURE TRANSFORM? (SIFT),?SPEEDED UP ROBUST FEATURES? (SURF) OR?HISTOGRAM OF ORIENTED GRADIENT? (HOG) HAVE BECOME WIDESPREAD PRECISELY BECAUSE OF THEIR CHARACTERISTICS OF INVARIANCE TO THE MAIN TRANSFORMATIONS (English) / qualifier | |||||||||||||||
point in time: 27 January 2022
| |||||||||||||||
Revision as of 10:19, 27 January 2022
Project Q4081958 in Italy
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | (3553.04032020.158000154) L.M.4.0 |
Project Q4081958 in Italy |
Statements
71,876.34 Euro
0 references
143,752.68 Euro
0 references
50.0 percent
0 references
SEEONE VISION TECHNOLOGY S.R.L.
0 references
IL PROBLEMA DI CLASSIFICARE LE DIVERSE TIPOLOGIE DI DIFETTI CHE SI POSSONO PRESENTARE SU MATERIALI RAW DI DIVERSA NATURA VIENE TIPICAMENTE AFFRONTATO CON TECNICHE CLASSICHE DI COMPUTER VISION CHE SFRUTTANO ?FEATURE VISUALI? PRELIMINARMENTE IDENTIFICATE DA ESPERTI UMANI SULLA BASE DELLA LORO ESPERIENZA ED INTUIZIONE (UNA FASE CHE VIENE NORMALMENTE CHIAMATA FASE DI ?FEATURE ENGINEERING?). QUESTE TECNICHE CHE SFRUTTANO CARATTERISTICHE INGEGNERIZZATE DA ESPERTI UMANI (?HAND-ENGINEERED?), HANNO PORTATO ALLA NASCITA DI UN VASTISSIMO INSIEME DI DESCRITTORI CHE RISULTANO INVARIANTI SIA ALLE TIPICHE DEFORMAZIONI GEOMETRICHE E RADIOMETRICHE INTRODOTTE DAI SISTEMI DI ACQUISIZIONE, SIA AL VARIARE DELLE CONDIZIONI AMBIENTALI E DI ILLUMINAZIONE. ALGORITMI COME, PER ESEMPIO: ?SCALE INVARIANT FEATURE TRANSFORM? (SIFT), ?SPEEDED UP ROBUST FEATURES? (SURF) OPPURE ?HISTOGRAM OF ORIENTED GRADIENT? (HOG) SI SONO MOLTO DIFFUSI PROPRIO PER LE LORO CARATTERISTICHE DI INVARIANZA ALLE PRINCIPALI TRASFORMAZIONI (Italian)
0 references
THE PROBLEM OF CLASSIFYING THE DIFFERENT TYPES OF DEFECTS THAT CAN OCCUR ON RAW MATERIALS OF DIFFERENT NATURE IS TYPICALLY ADDRESSED WITH CLASSIC COMPUTER VISION TECHNIQUES THAT EXPLOIT VISUAL FEATURES? FIRST IDENTIFIED BY HUMAN EXPERTS BASED ON THEIR EXPERIENCE AND INTUITION (A PHASE THAT IS NORMALLY CALLED?FEATURE ENGINEERING?). THESE TECHNIQUES THAT EXPLOIT CHARACTERISTICS ENGINEERED BY HUMAN EXPERTS (?HAND-ENGINEERED?), HAVE LED TO THE CREATION OF A VAST SET OF DESCRIPTORS THAT ARE INVARIANT BOTH TO THE TYPICAL GEOMETRIC AND RADIOMETRIC DEFORMATIONS INTRODUCED BY THE ACQUISITION SYSTEMS, AND TO THE VARYING ENVIRONMENTAL AND LIGHTING CONDITIONS. ALGORITHMS SUCH AS, FOR EXAMPLE: INVARIANT SCALES FEATURE TRANSFORM? (SIFT),?SPEEDED UP ROBUST FEATURES? (SURF) OR?HISTOGRAM OF ORIENTED GRADIENT? (HOG) HAVE BECOME WIDESPREAD PRECISELY BECAUSE OF THEIR CHARACTERISTICS OF INVARIANCE TO THE MAIN TRANSFORMATIONS (English)
27 January 2022
0 references
SCANDICCI
0 references