Processing

Please wait...

Settings

Settings

Goto Application

1. WO2019049060 - FROTH SEGMENTATION IN FLOTATION CELLS

Publication Number WO/2019/049060
Publication Date 14.03.2019
International Application No. PCT/IB2018/056802
International Filing Date 06.09.2018
IPC
B03D 1/02 2006.01
BPERFORMING OPERATIONS; TRANSPORTING
03SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
DFLOTATION; DIFFERENTIAL SEDIMENTATION
1Flotation
02Froth-flotation processes
G06K 9/34 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
20Image acquisition
34Segmentation of touching or overlapping patterns in the image field
CPC
B03D 1/028
BPERFORMING OPERATIONS; TRANSPORTING
03SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
DFLOTATION; DIFFERENTIAL SEDIMENTATION
1Flotation
02Froth-flotation processes
028Control and monitoring of flotation processes; computer models therefor
G06K 9/34
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
20Image acquisition
34Segmentation of touching or overlapping patterns in the image field
G06K 9/6274
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
627based on distances between the pattern to be recognised and training or reference patterns
6271based on distances to prototypes
6274based on distances to neighbourhood prototypes, e.g. Restricted Coulomb Energy Networks
G06T 7/12
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
12Edge-based segmentation
Applicants
  • STONE THREE DIGITAL (PTY) LTD [ZA]/[ZA]
Inventors
  • IRWIN, Shaun George
  • BOTHA, Kristo
  • VAN DER BIJL, Leendert
Agents
  • VON SEIDELS INTELLECTUAL PROPERTY ATTORNEYS
Priority Data
2017/0611408.09.2017ZA
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) FROTH SEGMENTATION IN FLOTATION CELLS
(FR) SEGMENTATION DE MOUSSE DANS DES CELLULES DE FLOTTATION
Abstract
(EN)
System (100) and method for generating a bubble segmentation image from a digital froth image of a froth phase of a flotation cell (107). The method comprises receiving the froth image, applying one or more deep learning networks to the froth image, the deep learning networks having been trained with one or more datasets of training images of labelled, predetermined bubble segmentation images to learn to identify features useful for identifying bubble boundaries automatically. The method further comprises generating the bubble segmentation image utilizing the deep learning networks so that the bubble segmentation image includes identified boundary data representing boundaries between bubbles present in the froth image, and outputting the bubble segmentation image.
(FR)
La présente invention concerne un système (100) et un procédé pour générer une image de segmentation de bulles à partir d'une image de mousse numérique d'une phase de mousse d'une cellule de flottation (107). Le procédé consiste à recevoir l'image de mousse, et à appliquer un ou plusieurs réseaux d'apprentissage profond à l'image de mousse, les réseaux d'apprentissage profond ayant été entraînés avec un ou plusieurs ensembles de données d'images d'apprentissage d'images de segmentation de bulle prédéterminées marquées, pour apprendre à identifier des caractéristiques utiles pour identifier automatiquement des limites de bulle. Le procédé consiste en outre à générer l'image de segmentation de bulles à l'aide des réseaux d'apprentissage profond de telle sorte que l'image de segmentation de bulles comprend des données de limite identifiées représentant des limites entre des bulles présentes dans l'image de mousse, et à diffuser l'image de segmentation de bulles.
Also published as
PE000321-2020
Latest bibliographic data on file with the International Bureau