Title:
METHOD AND APPARATUS FOR DETERMINING DEGREE OF PROGRESS IN CONSTRUCTION THROUGH DEEP LEARNING-BASED CLASSIFICATION
Document Type and Number:
WIPO Patent Application WO/2023/204528
Kind Code:
A1
Abstract:
The present invention relates to a method for determining the degree of progress in construction. A deep learning-based method for determining the degree of progress in construction implemented by a server, according to the present invention, comprises the steps of: receiving an image of a construction site as an input; performing a pixel-by-pixel classification of the construction site image by using a trained construction site classification model; and calculating the degree of progress in construction by means of the ratio of pixels per class. According to the present invention, a trained neural network enables the calculation of an object indicator and accordingly more convenient determination of the degree of progress in construction.
Inventors:
BANG SEUNG ON (KR)
SUNG BEOM KYU (KR)
SUNG BEOM KYU (KR)
Application Number:
PCT/KR2023/005063
Publication Date:
October 26, 2023
Filing Date:
April 14, 2023
Export Citation:
Assignee:
OGQ CORP (KR)
LOTTE ENG & CONSTRUCTION CO LTD (KR)
LOTTE ENG & CONSTRUCTION CO LTD (KR)
International Classes:
G06Q50/08; G06N3/08; G06T7/11; G06T7/187
Foreign References:
| KR102366840B1 | 2022-02-24 | |||
| KR102354684B1 | 2022-01-21 | |||
| KR102338765B1 | 2021-12-15 | |||
| KR102244978B1 | 2021-04-28 | |||
| KR20210081618A | 2021-07-02 |
Attorney, Agent or Firm:
LEE, Jang Hoon (KR)
Download PDF: