Map information processing method and apparatus for correlating road location on a road network map

• US 5,848,374 A
• Filed: 03/01/1996
• Issued: 12/08/1998
• Est. Priority Date: 03/06/1995
• Status: Expired due to Term

• Current Assignee: Nippon Telegraph and Telephone Corporation
• Original Assignee: Nippon Telegraph and Telephone Corporation
• Inventors: Iwata, Masahiko, Yasuda, Tsuneo, Nunobiki, Tadashi, Wakabayashi, Kaoru
• Primary Examiner(s): Jacques, Jacques H. Louis
• Assistant Examiner(s): ARTHUR JEANGLAUDE, GERTRUDE
• Application Number:
  US08/609,451
• Time in Patent Office: 1,012 Days
• Field of Search: 701/200, 701/208, 701/212, 701/202, 701/210, 701/211, 701/207, 701/209, 701/55-56, 340/990, 340/988, 340/995
• US Class Current: 701/212
• CPC Class Codes: G01C 15/00 Surveying instruments or accessories not provided for in groups G01C1/00 - G01C13/00
• CPC Class Codes: G01C 21/3819 Road shape data, e.g. outline of a route
• CPC Class Codes: G06V 10/757 Matching configurations of points or features
• CPC Class Codes: G06V 30/422 Technical drawings; Geographical maps

Abstract

A map processing apparatus may receive information on a road network map and a housing map. A coordinate transforming section may absorb difference in reduction scale and coordinate system as much as possible between the road network map and the housing map by means of geometric transformation. A corresponding candidate detecting section may detect a plurality of corresponding candidate points as candidates of corresponding points on the other arbitrary map to the road constituent points by collating road images near respective road constituent points on the road network map with road images of corresponding location on the other arbitrary map. A correspondence determining section may determines one corresponding candidate point based on derived similarities corresponding to respective corresponding candidate points by selecting one of the plurality of corresponding candidate points one by one with respect to respective road constituent points, and verifying similarity between shifted entire profile of road network and original entire profile of road network based on feature amounts representing intersecting angles of roads to be connected mutually on respective road constituent points while shifting respective road constituent points to locations of selected corresponding candidate points.

Claims

1. An automatic road recognition method employing a computer, comprising the steps of:

inputting road margin line information representing road profile on a map, to thus detect, with regard to portions being put between road margin lines, road plane candidate indices including indices as to parallelism between road margin lines and indices as to distance between road margin lines;

estimating that either side of the road margin lines is the road plane based on the road plane candidate indices on both sides of the road margin lines; and

determining as the road plane a region which resides between two road margin lines opposing to each other on the side estimated as the road plane side.

2. An automatic road recognition method employing a computer, comprising the steps of:

inputting road margin line information representing road profile on a map and non-road figure information which cannot reside in road planes, to thus detect, with regard to portions being put between road margin lines, road plane candidate indices including parallelism indices as to parallelism between road margin lines, distance indices as to distance between road margin lines and non-road indices as to whether non-road figure resides in;

estimating that either side of the road margin lines is the road plane based on the road plane candidate indices on both sides of the road margin lines; and

determining as the road plane a region which resides between two road margin lines opposing to other on the side estimated as the road plane side and in which each the non-road figure cannot reside.

3. An automatic road recognition apparatus comprising:

road candidate detecting means for detecting, with regard to portions being put between road margin lines, road plane candidate indices including parallelism indices as to parallelism between road margin lines and distance indices as to distance between road margin lines by inputting road margin line information representing road profile on a map;

road plane side estimating means for estimating that either side of the road margin lines is the road plane based on the road plane candidate indices on both sides of the road margin lines; and

road plane determining means for determining as the road plane a region which resides between two road margin lines opposing to each other on the side estimated as the road plane side.

4. A map information processing method employing a computer, comprising the steps of:

inputting a road network map which includes road constituent points formed of starting points, end points and interpolation points, and information indicating locations of the road constituent points and represents connection relations between respective roads as a graphic structure, and another arbitrary map which includes road profile information;

detecting a plurality of corresponding candidate points as candidates of corresponding points on the other arbitrary map to the road constituent points by collating road images near respective road constituent points on the road network map with road images of corresponding location on the other arbitrary map; and

correlating road location on the road network map with road location on the other arbitrary map, by selecting one of the plurality of corresponding candidate points one by one with respect to respective road constituent points, summing up differences between feature amounts representing intersecting angles of roads to be connected mutually on respective road constituent points of shifted road network and feature amounts on respective road constituent points of original road network while shifting respective road constituent points to locations of selected corresponding candidate points, and determining one corresponding candidate point based on amount of the summed differences corresponding to respective corresponding candidate points.

5. A map information processing method according to claim 4, further comprising, after determining one corresponding candidate point with respect to respective road constituent points, the steps of:

selecting two from respective corresponding candidate points at a time;

determining that either of both corresponding candidate points has high correspondence rate to a corresponding road on the other arbitrary map; and

rotating and shifting other corresponding candidate point while a corresponding candidate point which is determined to have the high correspondence rate being fixed until the correspondence rate of other corresponding candidate point to road on the other arbitrary map takes a maximum value.

6. A map information processing method according to claim 4, further comprising, after determining one corresponding candidate point with respect to respective road constituent points, the steps of:

selecting two from respective corresponding candidate points at a time;

determining that either of both corresponding candidate points has high correspondence rate to a corresponding road on the other arbitrary map; and

rotating and shifting other corresponding candidate point while a corresponding candidate point which is determined to have the high correspondence rate being fixed until an angle between an angle of road network obtained based on both corresponding candidate points and an angle of a road on the other arbitrary map takes a minimum value.

7. A map information processing method according to any of claims 4 to 6, further comprising the steps of:

integrating the road network map with the other arbitrary map based on correlation of road location between the road network map and the other arbitrary map by effecting correlation with respect to map information other than road.

8. A map information processing method according to any of claims 4 to 6, further comprising the steps of:

making new maps based on correlation of road location between the road network map and the other arbitrary map by effecting correlation with respect to map information other than road.

9. A map information processing method according to any of claims 4 to 6, wherein said other arbitrary map is a high precision map or a large-scale map including road profile information which has higher location precision than that of the road network map.

10. A map information processing method according to claim 7, wherein said other arbitrary map is a high precision map or a large-scale map including road profile information which has higher location precision than that of the road network map.

11. A map information processing method according to claim 8, wherein said other arbitrary map is a high precision map or a large-scale map including road profile information which has higher location precision than that of the road network map.

12. A map information processing method according to claim 4, further comprising, if the road profile information on the other arbitrary map comprise disconnected incomplete road margin lines, the steps of:

inputting road margin line information representing road profile on a map, to thus detect, with regard to portions being put between road margin lines, road plane candidate indices including parallelism indices as to parallelism between road margin lines and distance indices as to distance between road margin lines;

estimating that either side of the road margin lines is the road plane based on the road plane candidate indices on both sides of the road margin lines;

determining as the road plane a region which resides between two road margin lines opposing to each other on the side estimated as the road plane side; and

performing the input of the other arbitrary map which being simultaneously performed with the input of the road network map.

13. A map information processing method according to claim 4, further comprising, if the road profile information on the other arbitrary map comprise disconnected incomplete road margin lines, the steps of:

inputting road margin line information representing road profile on a map and non-road figure information which cannot reside in road planes, to thus detect, with respect to portions being put between road margin lines, road plane candidate indices including parallelism indices as to parallelism between road margin lines, distance indices as to distance between road margin lines and non-road indices as to whether non-road figure resides in;

estimating that either side of the road margin lines is the road plane based on the road plane candidate indices on both sides of the road margin lines;

determining as the road plane a region which resides between two road margin lines opposing to each other on the side estimated as the road plane side and in which the non-road figure cannot reside; and

performing the input of the other arbitrary map which being simultaneously performed with the input of the road network map.

14. A map information processing apparatus for correlating road location on the road network map with road location on another arbitrary map by inputting a road network map which includes road constituent points formed of starting points, end points and interpolation points, and information indicating locations of the road constituent points and represents connection relations between respective roads as a graphic structure, and another arbitrary map which includes road profile information, the apparatus comprising:

corresponding candidate detecting means for detecting a plurality of corresponding candidate points as candidates of corresponding points on the other arbitrary map to the road constituent points by collating road images near respective road constituent points on the road network map with road images of corresponding location on the other arbitrary map; and

correspondence determining means for selecting one of the plurality of corresponding candidate points one by one with respect to respective road constituent points, summing up differences between feature amounts representing intersecting angles of roads to be connected mutually on respective road constituent points of shifted road network and feature amounts on respective road constituent points of original road network while shifting respective road constituent points to locations of selected corresponding candidate points, and determining one corresponding candidate point based on amount of the summed differences corresponding to respective corresponding candidate points.

15. A map information processing apparatus according to claim 14, wherein, after determining one corresponding candidate point with respect to respective road constituent points, the correspondence determining means

selecting two from respective corresponding candidate points at a time,

determining that either of both corresponding candidate points has high correspondence rate to a corresponding road on the other arbitrary map, and

rotating and shifting other corresponding candidate point while a corresponding candidate point which is determined to have the high correspondence rate being fixed until the correspondence rate of other corresponding candidate point to road on the other arbitrary map takes a maximum value.

16. A map information processing apparatus according to claim 14, wherein, after determining one corresponding candidate point with respect to respective road constituent points, the correspondence determining means

selecting two from respective corresponding candidate points at a time,

determining that either of both corresponding candidate points has high correspondence rate to a corresponding road on the other arbitrary map, and

rotating and shifting other corresponding candidate point while a corresponding candidate point which is determined to have the high correspondence rate being fixed until an angle between an angle of road network obtained based on both corresponding candidate points and an angle of a road on the other arbitrary map.

17. A map information processing apparatus according to any of claims 14 to 16, wherein, if the road profile information on the other arbitrary map comprise disconnected incomplete road margin lines, prior to process in the corresponding candidate detecting section,

road margin line information representing road profile on a map are input to thus detect, with respect to portions being put between road margin lines, road plane candidate indices including parallelism indices as to parallelism between road margin lines and distance indices as to distance between road margin lines,

either side of the road margin lines is estimated as the road plane based on the road plane candidate indices on both sides of the road margin lines, and

a region which resides between two road margin lines opposing to each other on the side estimated as the road plane side is determined as the road plane.

18. A map information processing apparatus according to any of claims 14 to 16, wherein, if the road profile information on the other arbitrary map comprise disconnected incomplete road margin lines, prior to process in the corresponding candidate detecting section,

road margin line information representing road profile on a map and non-road figure information which cannot reside in road planes are input to thus detect, with respect to portions being put between road margin lines, road plane candidate indices including parallelism indices as to parallelism between road margin lines, distance indices as to distance between road margin lines and non-road indices as to whether non-road figure resides in,

either side of the road margin lines is estimated as the road plane based on the road plane candidate indices on both sides of the road margin lines, and

a region which resides between two road margin lines opposing to each other on the side estimated as the road plane side and in which the non-road figure cannot reside is determined as the road plane.