IP Library › Granted Patent US 12,737,508
Granted Patent B2
US 12,737,508 · App. 17/775,940 · Granted Sep 15, 2026

Centerline correction apparatus, centerline correction method, and spatial network data generation system and program

Inventors: Mitsuru Mochizuki (Musashino, JP); Hitoshi Seshimo (Musashino, JP); Osamu Matsuda (Musashino, JP)
Assignee: NTT, Inc.
G06F30/13
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 12,737,508
App. No.
17/775,940
Granted
Sep 15, 2026
Kind
B2
Abstract

A centerline correction apparatus according to one embodiment corrects a centerline of a path that is acquired from shape information about an indoor space or road shape information about an outdoor space and that is a travelable region in the indoor space or the outdoor space. This centerline correction apparatus includes a centerline correction unit that performs: a center point extraction process that extracts a center point of an individual side of the centerline; a centerline correction process that moves the centerline to be at a center with respect to a distance to a boundary of an original shape based on the extracted center point; and an unnecessary centerline deletion process that deletes an unnecessary one of the moved centerlines.

Claims (26)

1 . A centerline correction apparatus configured to correct a centerline of a path that is acquired from shape information about an indoor space or road shape information about an outdoor space and that is a travelable region in the indoor space or the outdoor space, the centerline correction apparatus comprising:

at least one memory and

one or more processors configured to perform operations including:

extracting a center point of an individual side of the centerline;

moving the centerline to be at a center with respect to a distance to a boundary of an original shape based on the extracted center point by rotating the individual side of the centerline around the center point, and translating the rotated centerline; and

deleting an unnecessary one of the moved centerlines.

2 . The centerline correction apparatus according to claim 1 , wherein the one or more processors are configured to perform operations further comprising:

dividing an outer frame shape of the path;

extracting straight line portions of the outer frame shape in order of length;

grouping the straight line portions based on respective angles formed by the extracted straight line portions and a reference line and that generates a region per group with a plurality of the straight line portions included in each group; and

performing the correction of the centerline per region.

3 . The centerline correction apparatus according to claim 2 , wherein the one or more processors are configured to perform the centerline correction process by performing operations including:

excluding the region from a correction target of the centerline correction process when an area of the region is equal to or less than a threshold; and

removing, when different ones of the regions overlap with each other, a smaller region from a figure included in a larger region and further performs division of the region.

4 . The centerline correction apparatus according to claim 3 , wherein the one or more processors are configured to perform a division process that translates, along a circumscribed rectangle of the larger region, a straight line parallel to the reference line from one end to the other end of the circumscribed rectangle and divides the larger region at a position where a significant change occurs in the region to be divided.

5 . A centerline correction method performed by a centerline correction apparatus which corrects a centerline of a path that is acquired from shape information about an indoor space or road shape information about an outdoor space and that is a travelable region in the indoor space or the outdoor space, the centerline correction method, comprising:

extracting a center point of an individual side of the centerline;

moving the centerline to be at a center with respect to a distance to a boundary of an original shape based on the extracted center point by rotating the individual side of the centerline around the center point, and translating the rotated centerline; and

deleting an unnecessary one of the moved centerlines.

6 . A network data generation system comprising:

at least one memory; and

one or mor processors configured to perform operations comprising:

generating spatial network data which is information about a path based on a centerline corrected by at least one of the one or more processors;

extracting a center point of an individual side of the centerline;

moving the centerline to be at a center with respect to a distance to a boundary of an original shape based on the extracted center point by rotating the individual side of the centerline around the center point, and translating the rotated centerline; and

deleting an unnecessary one of the moved centerlines.

Assignments (2)
CHANGE OF NAME Recorded Aug 14, 2025
From: NIPPON TELEGRAPH AND TELEPHONE CORPORATION
To: NTT, INC.
Reel/Frame 072412/0498 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 11, 2023
From: MOCHIZUKI, MITSURU; SESHIMO, HITOSHI; MATSUDA, OSAMU
To: NIPPON TELEGRAPH AND TELEPHONE CORPORATION
Reel/Frame 062346/0338 →
Continuity (1)
Related Publication 20220382928A1 · Dec 1, 2022
References Cited (8)
US 6622085B1 · Amita et al. · 2003 [cited by applicant]
JP 2009230518 · 2009 [cited by applicant]
WO WO2000043953 · 2000 [cited by applicant]
Haunert, Jan-Henrik et al., “Area Collapse and Road Centerlines based on Straight Skeletons”, Aug. 17, 2007, Geoinformatica, Sprnger. (Year: 2007). [cited by examiner]
Poullis, C. et al. “Delineation and geometric modeling of road networks”, ISPRS Journal of Photogrammetry and Remote Sensing, Elsevier B.V. (Year: 2009). [cited by examiner]
Yang, Xiaofei et al., “Road Detection and Centerline Extraction Via Deep Recurrent Convolutional Neural Network U-Net”, Sep. 2019, IEEE Transactions on Geoscience and Remote Sensing, vol. 57, No. 9, IEEE. (Year: 2019). [cited by examiner]
Mortari, “Automatic Extraction of Improved Geometrical Network Model from CityGML for Indoor navigation,” University of L'Aquila department of engineering and information sciences and mathematics degree course in comput… [cited by applicant]
Okuaki, “Method to Generate the Line data of the Road Centerline Automatically along with the Attribute Data of Width of Road,” Proceedings of the Geographic Information Systems Society, 2012, retrieved from URL <https:… [cited by applicant]