US 12,254,614 B2
Method for detecting glass-ceramic materials
Hugues Chenneviere, Colombes (FR); and Ezzeddine Ouerghemmi, Antony (FR)
Assigned to SAINT-GOBAIN ISOVER, Courbevoie (FR)
Appl. No. 17/604,647
Filed by SAINT-GOBAIN ISOVER, Courbevoie (FR)
PCT Filed Apr. 9, 2020, PCT No. PCT/EP2020/060247
§ 371(c)(1), (2) Date Oct. 18, 2021,
PCT Pub. No. WO2020/212266, PCT Pub. Date Oct. 22, 2020.
Claims priority of application No. 1904247 (FR), filed on Apr. 19, 2019.
Prior Publication US 2022/0196544 A1, Jun. 23, 2022
Int. Cl. G06T 7/00 (2017.01); G06T 7/90 (2017.01)
CPC G06T 7/0008 (2013.01) [G06T 7/90 (2017.01); G06T 2207/10024 (2013.01)] 7 Claims
OG exemplary drawing
 
1. A detection method for automated detection of materials of a glass-ceramic type among a cullet, comprising:
a step of detecting a glass-ceramic material in cullet, during which fragments thought to be glass-ceramic material are identified in the cullet,
a step of obtaining a digital image resulting from the preceding step of detecting the glass-ceramic material, said image comprising at least one group of pixels corresponding to a fragment thought to be glass-ceramic material,
a step of colorimetrically processing the digital image, during which at least the group of pixels of said digital image corresponding to fragments thought to be glass-ceramic material is processed by a colorimetric image processing module according to an RGB model,
wherein the colorimetric image processing module according to an RGB model calculates a ratio B/R between the blue datum B and the red datum R for each pixel or groups of pixels corresponding to fragments of the cullet having been thought to be fragments of glass-ceramic material, the colorimetric image processing module according to an RGB model comparing the ratios B/R between the blue datum B and the red datum R as the ratios were calculated beforehand at a threshold value, the threshold value being 0.5.