Surface anomaly-detection and analysis method

• US 6,028,948 A
• Filed: 12/29/1997
• Issued: 02/22/2000
• Est. Priority Date: 12/29/1997
• Status: Expired due to Term

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Kil, David H., Rose, David W., Shin, Frances B.
• Primary Examiner(s): Au, Amelia
• Assistant Examiner(s): Dastouri, Mehrdad
• Application Number:
  US08/999,216
• Time in Patent Office: 785 Days
• Field of Search: 382/104, 382/108, 382/149, 382/259, 382/260, 382/263, 382/282, 382/283, 382/100, 404/17, 404/72, 404/107, 356/392, 356/394, 702/40, 73/146
• US Class Current: 382/108
• CPC Class Codes: E01C 23/01 Devices or auxiliary means for setting-out or checking the configuration of new surfacing, e.g. templates, screed or reference line supports form rails E01C19/50; Applications of apparatus for measuring, indicating, or recording the surface configuration of existing surfacing, e.g. profilographs E01C23/07 takes precedence; measuring roughness or irregularity in general G01B
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 7/0008 checking presence/absence

Abstract

A method (30) is provided for the detection and analysis of anomalies (32) in a road surface (36). An image (34) of the road surface (36) is obtained (82) wherein traffic control markings (76) are masked (88). The image (34) is filtered (90) and a pixel map (92) is produced (98). The pixel map (92) is partitioned (112) into a multiplicity of subimages (108). For each subimage (108), anomaly parameters are identified (120) and a status characteristic is determined (124) and assigned (136). A subimage map (138) is produced (142) depicting the subimages (108) and their status characteristics. A determination (156) is made as to which subimages (108) contain anomalies (32). Anomaly-containing subimages (108) are grouped (158) into anomalous objects (152). For each anomalous object (152), an object type (162) is determined (160) and assigned (164). An object map (154) is then produced (166) depicting the anomalous objects (152).

Claims

1. A method for the detection and analysis of anomalies in a surface, said method comprising the steps of:

obtaining an image of said surface including at least one of said anomalies, wherein said one anomaly is a line anomaly;

filtering said image to enhance a depiction of said line anomaly by removing non-line-anomalous features of said image, said filtering step producing a line pixel map;

filtering said image to enhance a depiction of an area anomaly by removing non-area-anomalous features of said image, said filtering step producing an area pixel map;

combining said line pixel map and said area pixel map to produce a combined pixel map;

partitioning said combined pixel map into a multiplicity of subimages, wherein each of said subimages includes a plurality of pixels;

assigning a status characteristic to each of said subimages in response to any anomaly depicted therein;

forming an anomalous object in response to said status characteristics of said subimages, said anomalous object corresponding to said one anomaly in said surface; and

producing an object map of said surface, said object map depicting a location of said anomalous object.

2. A method as claimed in claim 1 additionally comprising, prior to said line-anomaly filtering step, the step of masking, within said image, a predetermined non-anomalous feature of said surface so as to prevent one of said line-anomaly and area-anomaly filtering steps from enhancing a depiction of said predetermined non-anomalous feature.

3. A method as claimed in claim 2 wherein said surface is a road surface and said predetermined non-anomalous feature is a traffic control marking.

4. A method as claimed in claim 1 wherein, for each of said subimages, said assigning step comprises the steps of:

identifying parameters of said one anomaly;

comparing said identified parameters against predetermined parameters of a hypothetical anomaly; and

determining said status characteristic of said subimage in response to said comparing step.

5. A method as claimed in claim 4 additionally comprising the step of defining said hypothetical anomaly by teaching an expert system.

6. A method as claimed in claim 4 wherein:

said determining step assigns a no-anomaly status characteristic to each of said subimages for which said comparing step indicates an absence of said one anomaly therein; and

said determining step assigns an anomaly status characteristic to each of said subimages for which said comparing step indicates a presence of said one anomaly therein.

7. A method as claimed in claim 4 wherein said determining step comprises the step of examining, when results of said comparing step are ambiguous, adjacent subimages to refine said results.

8. A method as claimed in claim 1 wherein said forming step comprises the steps of:

inspecting said subimages to determine which of said subimages have said one anomaly contained therein;

grouping anomaly-containing subimages into said anomalous object when said anomaly-containing subimages are proximate each other;

determining an object type for said anomalous object from spatial relationships between said anomaly-containing subimages and from said status characteristics of each of said anomaly-containing subimages; and

assigning said object type to said anomalous object.

9. A method as claimed in claim 1 additionally comprising the step of identifying, prior to said producing step, a type for said anomalous object.

10. A method as claimed in claim 9 wherein said surface is a road surface, and said anomalous-object type is one of a crack, a seam, an edge, a sealed crack, a patch, and a pothole of said road surface.

11. A method as claimed in claim 10 wherein said crack is one of a dark crack and a light crack.

12. A method as claimed in claim 10 wherein said crack is one of a longitudinal crack, a transverse crack, an oblique crack, a fatigue crack, and a block crack.

13. A method as claimed in claim 9 wherein said object map additionally depicts said anomalous-object type.

14. A method for the detection and analysis of anomalies in a surface, said method comprising the steps of:

obtaining an image of said surface including at least one of said anomalies, wherein said one anomaly is a line anomaly;

filtering said image to enhance a depiction of said line anomaly by removing non-line-anomalous features of said image, said line-anomaly filtering step producing a line pixel map;

partitioning said line pixel map into a multiplicity of line subimages, wherein each of said line subimages includes a plurality of pixels;

assigning a status characteristic to each of said line subimages in response to any anomaly depicted therein;

forming an anomalous object in response to said status characteristics of said line subimages, said anomalous object corresponding to said line anomaly in said surface;

producing an object map of said surface, said object map depicting a location of said anomalous object;

producing a line-subimage map depicting said line subimages and said status characteristics thereof;

filtering said image to enhance a depiction of an area anomaly of said surface by removing non-area-anomalous features of said image, said area-anomaly filtering step producing an area pixel map;

partitioning said area pixel map into a multiplicity of area subimages, wherein each of said area subimages includes a plurality of pixels;

assigning a status characteristic to each of said area subimages in response to any area anomaly depicted therein;

producing an area-subimage map depicting said area subimages and said status characteristics thereof; and

fusing said line-subimage map and said area-subimage map to produce a fused subimage map so that each of said corresponding line and area subimages forms a fused subimage, and so that said forming step forms said anomalous object in response to said status characteristics of said fused subimages.

15. A method as claimed in claim 14 wherein:

said surface is a road surface;

said line anomaly is one of a crack, a seam, and an edge of said road surface; and

said area anomaly is one of a sealed crack, a patch, and a pothole of said road surface.

16. A method as claimed in claim 14 wherein said area anomaly is a linear-area anomaly, said non-area-anomalous features are non-linear-area-anomalous features, and said area pixel map is a linear-area pixel map, and wherein said method additionally comprises the steps of:

filtering said image to enhance a depiction of a block-area anomaly of said surface by removing non-block-area-anomalous features of said image, said block-area filtering step producing a block-area pixel map; and

combining said linear-area pixel map and said block-area pixel map to produce a combined area pixel map so that said area-pixel-map partitioning step partitions said combined area pixel map.

17. A method as claimed in claim 16 wherein:

said surface is a road surface;

said linear-area anomaly is a sealed crack of said road surface; and

said block-area anomaly is one of a patch and a pothole of said road surface.

18. A method for the detection and analysis of anomalies in a road surface, said method comprising the steps of:

a) obtaining an image of said road surface including any number of said anomalies;

b) filtering said image to enhance a depiction of a line anomaly of said surface by removing non-line-anomalous features of said image, said filtering step b) producing a line pixel map;

c) partitioning said line pixel map into a multiplicity of line subimages, wherein each of said line subimages includes a plurality of pixels;

d) assigning a status characteristic to each of said line subimages in response to any line anomaly depicted therein;

e) producing a line-subimage map depicting said line subimages and said characteristics thereof;

f) filtering said image to enhance a depiction of an area anomaly of said surface by removing non-area-anomalous features of said image, said filtering step f) producing an area pixel map;

g) partitioning said area pixel map into a multiplicity of area subimages, wherein each of said area subimages includes a plurality of pixels;

h) assigning a status characteristic to each of said area subimages in response to any area anomaly depicted therein;

i) producing an area-subimage map depicting said area subimages and said characteristics thereof;

j) fusing said line-subimage map and said area-subimage map to produce a fused subimage map so that each of said corresponding line and area subimages forms a fused subimage;

k) forming an anomalous object in response to said status characteristics of said fused subimages; and

l) producing an object map of said road surface, said object map depicting said object.

19. A method as claimed in claim 18 additionally comprising, prior to said filtering steps b) and g), the step of masking, within said image, traffic control markings upon said road surface so as to prevent said filtering steps b) and g) from enhancing a depiction of said traffic control markings.

20. A method as claimed in claim 18 wherein said area anomaly is a linear-area anomaly, said non-anomalous features are non-linear-area-anomalous features, and said area pixel map is a linear-area pixel map, and wherein said method additionally comprises the steps of:

m) filtering said image to enhance a depiction of a block-area anomaly of said surface by removing non-block-area-anomalous features of said image, said filtering step m) producing a block-area pixel map; and

n) combining said linear-area pixel map and said block-area pixel map to produce a combined area pixel map so that said partitioning step h) partitions said combined area pixel map.

21. A method as claimed in claim 18 additionally comprising the step of identifying, prior to said producing step 1), a type for said anomalous object, wherein said type is one of a crack, a seam, an edge, a sealed crack, a patch, and a pothole of said surface.

22. A method as claimed in claim 18 wherein, for each of said line subimages, said assigning step d) comprises the steps of:

identifying line parameters of said line anomaly;

comparing said line parameters against predetermined parameters of a hypothetical line anomaly; and

determining said status characteristic of each of said line subimages in response to said comparing step for each of said line subimages.

23. A method as claimed in claim 18 wherein, for each of said area subimages, said assigning step i) comprises the steps of:

identifying area parameters of said area anomaly;

comparing said area parameters against predetermined parameters of a hypothetical area anomaly; and

determining said status characteristic of each of said area subimages in response to said comparing step for each of said area subimages.

24. A method for the detection and analysis of anomalies in a road surface, said method comprising the steps of:

a) obtaining an image of said road surface including any number of said anomalies;

b) masking, within said image, traffic control markings upon said road surface;

c) filtering said image to enhance a depiction of a line anomaly, which includes a crack, a seam, and an edge of said road surface, by removing non-line-anomalous features of said image, said filtering step c) producing a line pixel map;

d) partitioning said line pixel map into a multiplicity of line subimages, wherein each of said line subimages includes a plurality of pixels;

e) identifying, for each of said line subimages, parameters of said line anomaly;

f) comparing, for each of said line subimages, said identified parameters against predetermined parameters of a hypothetical line anomaly;

g) assigning a status characteristic to each of said line subimages in response to said comparing step f);

h) producing a line-subimage map depicting said line subimages and said status characteristics thereof;

i) filtering said image to enhance a depiction of a linear-area anomaly, which includes a sealed crack of said road surface, by removing non-linear-area-anomalous features of said image, said linear-area-anomaly filtering step i) producing a linear-area pixel map;

j) filtering said image to enhance a depiction of a block-area anomaly, which includes one of a patch and a pothole of said road surface, by removing non-block-area-anomalous features of said image, said block-area-anomaly filtering step j) producing a block-area pixel map;

k) combining said linear-area pixel map and said block-area pixel map to produce a combined area pixel map;

l) partitioning said combined area pixel map into a multiplicity of area subimages, wherein each of said area subimages includes a plurality of pixels;

m) identifying, for each of said area subimages, parameters of an area anomaly contained therein, said area anomaly being one of said linear-area anomaly and said block-area anomaly;

n) comparing, for each of said area subimages, said identified parameters against predetermined parameters of a hypothetical area anomaly;

o) assigning a status characteristic to each of said area subimages in response to said comparing step n);

p) producing an area-subimage map depicting said area subimages and said status characteristics thereof;

q) fusing said line-subimage map and said area-subimage map to produce a fused subimage map so that each of said corresponding line and area subimages forms a fused subimage;

r) inspecting each of said fused subimages to determine which of said fused subimages is an anomaly-containing subimage having one of said line anomalies and said area anomalies contained therein;

s) grouping said anomaly-containing subimages into an object when said anomaly-containing subimages are proximate each other;

t) determining an object type, which indicates one of said crack, said seam, said edge, said sealed crack, said patch, and said pothole of said surface, for said object from spatial relationships between said anomaly-containing subimages and from said status characteristics thereof;

u) assigning said object type to said anomalous object; and

v) producing an object map of said road surface, said object map depicting said anomalous object and said object type.