IP Library › Granted Patent US 12,711,734
Granted Patent B2
US 12,711,734 · App. 18/231,712 · Granted Aug 18, 2026

Difference recognition method and difference recognition system

Inventors: Takuma Nakabayashi (Foster City, CA); Chih-Hung E. Yeh (Alameda, CA); Anirban Roy (San Francisco, CA)
Assignee: OBAYASHI CORPORATION
G06V10/761G06V20/70
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,711,734
App. No.
18/231,712
Filed
Aug 8, 2023
Granted
Aug 18, 2026
Kind
B2
Examiner
CRUZ, IRIANA
Art Unit
2681
USPC
382/218
Abstract

A difference recognition method and a difference recognition system that correctly recognizes the latest progress status. A difference recognition system recognizes the latest progress status of a construction site from a difference, which indicates progress of a construction site. The difference recognition system includes a data management unit that acquires first captured-image data captured in a first state at a first time point and second captured-image data captured in a second state at a second time point that is subsequent to the first time point, and a work determination unit serving as a difference recognition unit that recognizes a newly installed building element from the first captured-image data and second captured-image data as a difference between the first state and the second state. The task determination unit determines the latest progress status from the recognized difference.

Claims (42)

1 . A difference recognition method for recognizing a difference, which indicates progress of a construction site, the difference recognition method comprising:

a first state acquiring step of acquiring first captured-image data captured in a first state at a first time point;

a second state acquiring step of acquiring second captured-image data captured in a second state at a second time point subsequent to the first time point;

a first difference recognition step of recognizing a first difference corresponding to a change from the first state to the second state based on the first captured-image data and the second captured-image data using a trained machine-learning model that outputs labels indicating newly installed building elements or removed building elements;

a middle state acquiring step of acquiring middle captured-image data captured in a middle state at a middle time point between the first time point and the second time point;

a second difference recognition step of recognizing a second difference corresponding to a change from the first state to the middle state based on the first captured-image data and the middle captured-image data; and

a work determination step of determining a first work corresponding to a process from the first time point to the middle time point based on the first difference and the second difference.

2 . The difference recognition method according to claim 1 , wherein the difference recognition step includes acquiring a change in recognition possibility of the difference when exchanging a chronological relationship of the first captured-image data and the second captured-image data, and recognizing a rework as the difference in accordance with the change in the recognition possibility.

3 . The difference recognition method according to claim 1 , wherein the difference recognition step includes recognizing a rework as the difference when an element included in the first captured-image data is determined as having been removed in the second captured-image data.

4 . The difference recognition method according to claim 1 , wherein the difference recognition step includes recognizing the rework by comparing a change in work from the first captured-image data to the second captured image data with a chronological order of works prepared in advance.

5 . The difference recognition method according to claim 1 , wherein the difference recognition step includes using a model that learns the difference so as to recognize the difference.

6 . The difference recognition method according to claim 5 , wherein the difference recognition step includes extracting a portion changed between the first-captured image data and the second captured-image data and then using the model to recognize the difference.

7 . The difference recognition method according to claim 5 , wherein the learning of the model includes assigning a label indicating the difference.

8 . The difference recognition method according to claim 5 , wherein the learning of the model includes generating captured-image data distinguishing a region indicating the difference.

9 . The difference recognition method according to claim 1 , wherein the difference recognition step includes categorizing the difference.

10 . The difference recognition method according to claim 1 , wherein the work determination step comprises:

specifying a second work corresponding to the second difference; and

determining the first work based on the second work and the first difference.

11 . A difference recognition system for recognizing a difference, which indicates progress of a construction site, the difference recognition system comprising:

a data management unit that acquires first captured-image data captured in a first state at a first time point, second captured-image data captured in a second state at a second time point that is subsequent to the first time point, and middle captured-image data captured in a middle state at a middle time point between the first time point and the second time point;

a difference recognition unit that recognizes a first difference and a second difference, the first difference corresponding to a change from the first state to the second state based on the first captured-image data and the second captured-image data using a trained machine-learning model that outputs labels indicating newly installed building elements or removed building elements, and the second difference corresponding to a change from the first state to the middle state based on the first captured-image data and the middle captured-image data; and

a work determination unit that determines a first work corresponding to a process from the first time point to the middle time point based on the first difference and the second difference.

12 . The difference recognition system according to claim 11 , wherein the work determination unit is further configured to:

specify a second work corresponding to the second difference; and

determine the first work based on the second work and the first difference.

13 . A difference recognition method for recognizing a difference, which indicates progress of a construction site, the difference recognition method comprising:

a first state acquiring step of acquiring first captured-image data captured in a first state at a first time point;

a second state acquiring step of acquiring second captured-image data captured in a second state at a second time point subsequent to the first time point;

a first difference recognition step of recognizing a first difference corresponding to a change from the first state to the second state based on the first captured-image data and the second captured-image data using a trained machine-learning model that outputs labels indicating newly installed building elements or removed building elements;

a third state acquiring step of acquiring third captured-image data captured in a third state at a third time point subsequent to the first time point and the second time point;

a second difference recognition step of recognizing a second difference corresponding to a change from the second state to the third state based on the second captured-image data and the third captured-image data; and

a work determination step of determining a first work corresponding to a process from the first time point to the second time point based on the first difference and the second difference.

14 . The difference recognition method according to claim 13 , wherein the work determination step comprises:

specifying a second work corresponding to the first difference; and

determining the first work based on the second work and the second difference.

15 . A difference recognition system for recognizing a difference, which indicates progress of a construction site, the difference recognition system comprising:

a data management unit that acquires first captured-image data captured in a first state at a first time point, second captured-image data captured in a second state at a second time point that is subsequent to the first time point, and third captured-image data captured in a third state at a third time point subsequent to the first time point and the second time point;

a difference recognition unit that recognizes a first difference and a second difference, the first difference corresponding to a change from the first state to the second state based on the first captured-image data and the second captured-image data using a trained machine-learning model that outputs labels indicating newly installed building elements or removed building elements, and the second difference corresponding to a change from the second state to the third state based on the second captured-image data and the third captured-image data; and

a work determination unit that determines a first work corresponding to a process from the first time point to the second time point based on the first difference and the second difference.

16 . The difference recognition system according to claim 15 , wherein the work determination unit is further configured to:

specify a second work corresponding to the first difference; and

determine the first work based on the second work and the second difference.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 27, 2023
From: SRI INTERNATIONAL
To: OBAYASHI CORPORATION
Reel/Frame 065378/0090 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 14, 2023
From: YEH, CHIH-HUNG E.; ROY, ANIRBAN
To: SRI INTERNATIONAL
Reel/Frame 064577/0284 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 11, 2023
From: NAKABAYASHI, TAKUMA
To: OBAYASHI CORPORATION
Reel/Frame 064569/0765 →
Continuity (1)
Related Publication 20250054275A1 · Feb 13, 2025
References Cited (23)
US 12576541B2 · Telleria · 2026 [cited by examiner]
US 20140307974A1 · Tanaka · 2014 [cited by examiner]
US 20190028676A1 · Koga et al. · 2019 [cited by applicant]
US 20200034592A1 · Kawasaki et al. · 2020 [cited by applicant]
US 20200410424A1 · Bellaish · 2020 [cited by examiner]
US 20200413011A1 · Zass · 2020 [cited by examiner]
US 20210027110A1 · Takahashi · 2021 [cited by examiner]
US 20210240957A1 · Kawasaki et al. · 2021 [cited by applicant]
US 20230394405A1 · Zass · 2023 [cited by examiner]
JP 2001349059A · 2001 [cited by applicant]
JP 2004005237A · 2004 [cited by applicant]
JP 2007108880A · 2007 [cited by applicant]
JP 2008225596A · 2008 [cited by applicant]
JP 2010157206A · 2010 [cited by applicant]
JP 2015212916A · 2015 [cited by applicant]
JP 2015212917A · 2015 [cited by applicant]
JP 2017107443A · 2017 [cited by applicant]
JP 2020024658A · 2020 [cited by applicant]
JP 2021149650A · 2021 [cited by applicant]
JP 2022070512A · 2022 [cited by applicant]
JP 2022089260A · 2022 [cited by applicant]
JP 2022135733A · 2022 [cited by applicant]
WO 2017170651A1 · 2017 [cited by applicant]