Method and apparatus for generating a database of road sign images and positions
First Claim
Patent Images
1. A method for automatically generating a database of road sign images and positions comprising the steps of:
- providing video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location metrics for those video images;
analyzing said video images to identify a plurality of road sign images, wherein analyzing said video images initially analyzes each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis;
storing in a database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images, including the further processing for each image frame identified as having a road sign image present of;
segmenting said image frame into at least one region of interest;
applying at least one filter to each of said regions of interest which differentiates between various types of road sign images so as to classify each of said regions of interest by type; and
storing in said database as a unique record in said database at least a portion of a bitmap corresponding to each region of interest in said image frame, said record including information identifying the type of road sign image classified from said at least one filter; and
storing in said database a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded.
2 Assignments
0 Petitions
Accused Products
Abstract
A method and apparatus for automatically generating a database of road sign images and positions where the road signs are identified from video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location information.
191 Citations
71 Claims
-
1. A method for automatically generating a database of road sign images and positions comprising the steps of:
-
providing video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location metrics for those video images;
analyzing said video images to identify a plurality of road sign images, wherein analyzing said video images initially analyzes each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis;
storing in a database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images, including the further processing for each image frame identified as having a road sign image present of;
segmenting said image frame into at least one region of interest;
applying at least one filter to each of said regions of interest which differentiates between various types of road sign images so as to classify each of said regions of interest by type; and
storing in said database as a unique record in said database at least a portion of a bitmap corresponding to each region of interest in said image frame, said record including information identifying the type of road sign image classified from said at least one filter; and
storing in said database a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded.
-
-
2. A method for automatically generating a database of road sign images and positions comprising the steps of:
-
providing video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location metrics for those video images;
analyzing said video images to identify a plurality of road sign images, wherein analyzing said video images initially analyzes each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis;
storing in a database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images, including the further processing for each image frame identified as having a road sign image of;
segmenting said image frame into a plurality of non-overlapping image segments;
identifying a segment that exhibits a differentiable characteristic as a search space; and
expanding said search space to include segments adjacent to said segment that exhibits the differentiable characteristic. - View Dependent Claims (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
applying at least two separate filters to said image frame to generate a discrete output value from each filter, wherein each of at least two of said filters screens for a differentiable characteristic associated with a road sign image that is unique to that filter;
comparing the discrete output values for each of said at least two filters to at least one reference; and
identifying said image frame as potentially having a road sign image if at least one of the discrete output values indicates a differentiable characteristic is present for said image frame when the discrete output values are compared to the at least one reference.
-
-
5. The method of claim 4 wherein each of said at least two filters is selected from the set of the following filters:
- an edge filter, a color-pair filter, a color filter operating in the L*u*v color space, an edge filter combined with a line extender, or a color filter operating in the LCH color space.
-
6. The method of claim 4 wherein the differentiable characteristics of each of said at least two filters is selected from the set comprising:
- color characteristics, edge characteristics, texture characteristics, symmetry, convexity, lack of three dimensional volume, number and orientation of side edges, characteristic corner angles, frequency and luminescence.
-
7. The method of claim 6 wherein the step of expanding further comprises the step of:
utilizing morphology techniques to grow and erode said search space by adding or subtracting segments adjacent said search space until either said search space meets or fails to meet uniform criteria for a differentiable characteristic.
-
8. The method of claim 3 further comprising for at least two image frames that each depict at least a single common road sign the steps of:
-
applying a fuzzy logic color filter to said at least two image frames to segment as separate image portions from said at least two image frames each region that contains a group of pixels all having a color-set from a set of at least one pre-selected road sign color-sets; and
saving to as separate record in said database each of said separate image portions.
-
-
9. The method of claim 8 further comprising prior to completing the step of applying the fuzzy logic color filter, the step of converting said at least two image frames from a native color space to a single color space portion of a L*u*v color space and wherein the fuzzy logic color filter provides a value output signal that is at a maximum for only said set of at least one pre-selected road sign color-sets.
-
10. The method of claim 9, wherein the value output signals of the fuzzy logic color filter for each pixel are determined by location in said L*u*v color space and wherein the value output signals are assigned to a minimal set of mathematically described colors representing all of the legal color names and combinations of said set of at least one pre-selected road sign color sets.
-
11. The method of claim 8 wherein the fuzzy logic color filter segments the separate image portions such that each image portion contains a range of colors related to said color set whereby variations between colors of individual pixels are allowed to exist in said group of pixels that comprise each image portion.
-
12. The method of claim 2 wherein the step of storing in said database said location data metric comprises storing at least an orientation of a camera while recording said image fame, a focal length of the camera and the location of the camera.
-
13. The method of claim 2 wherein said location data metric is determined for said road sign image by using triangulation techniques that assume a known location of at least one camera and a set of known camera parameters.
-
14. The method of claim 2 wherein the step of providing video images of roadside scenes utilizes multiple cameras in said vehicle to record said video images.
-
15. The method of claim 2 wherein the step of storing in said database said at least a pointer to a portion of said image frame includes the steps selected from the set consisting of:
- storing a pointer to a portion of said image frame in said frame buffer, setting a flag that a road sign image is present, storing a description of a portion of said image frame;
storing a portion of said image frame, storing said image frame, storing multiple image frames of a common road sign image, or any combination thereof.
- storing a pointer to a portion of said image frame in said frame buffer, setting a flag that a road sign image is present, storing a description of a portion of said image frame;
-
16. A method for automatically generating a database of road sign images and positions comprises the steps of:
-
providing video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location metrics for those video images;
analyzing said video images to identify a plurality of road sign images, wherein analyzing said video images initially analyzes each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis;
storing in a database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images;
iteratively comparing at least two frames of said video images having a road sign image present to determine whether the road sign image in each of said at least two frames matches;
if there is a match, storing in said database at least an indication for at least one of said video images having a road sign image present that a common road sign image is depicted in multiple frames; and
storing in said database a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded. - View Dependent Claims (17, 18, 19)
placing all members of the first pixel set and the second pixel set that match in an identified field of said database;
synchronizing a location data metric to the identified field; and
storing in said database at least a pointer to a representative bitmap image of either the first pixel set or the second pixel set in conjunction with said location data metric.
-
-
19. The method of claim 18 wherein the said location data metric is determined for said common road sign image by using triangulation techniques that assume a known location of at least one camera and a set of known camera parameters.
-
20. A computer system for automatically generating a database of road sign images and positions comprising:
-
a frame buffer that stores at least one frame of a series of digital images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores information corresponding to a location of said vehicle when those digital images were recorded;
a processor that analyzes the each frame in the frame buffer to identify a plurality of road sign images; and
a database into which the processor stores at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images identified by the computer system and a location data metric corresponding to a location for each of said plurality of road sign images. - View Dependent Claims (21, 22, 23, 24, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38)
at least two separate filters operably connected to said frame buffer to generate a discrete output from each filter in response to each of said frames in said frame buffer, wherein each of at least two of said filters screens for a differentiable characteristic associated with a road sign that is unique to that filter; and
a neural network operably connected to said discrete outputs of said at least two separate filters to identify a frame of said digital images as potentially having a road sign present in response to said discrete outputs of said at least two separate filters.
-
-
23. The system of claim 22 wherein each of said at least two filters is selected from the set of the following filters:
- an edge filter, a color-pair filter, a color filter operating in the L*u*v color space, an edge filter combined with a line extender, or a color filter operating in the LCH color space.
-
24. The system of claim 22 wherein the differentiable characteristics of each of said at least two filters is selected from the set comprising:
- color characteristics, edge characteristics, texture characteristics, symmetry, convexity, lack of three dimensional volume, number and orientation of side edges, characteristic corner angles, frequency and luminescence.
-
28. The system of claim 21 wherein for at least two image frames that each depict at least a single common road sign the processor further comprises:
-
a fuzzy logic color filter that is applied to said at least two image frames to segment as separate image portions from said at least two image frames each region that contains a group of pixels all having a color-set from a set of at least one pre-selected road sign color-sets; and
means for saving as a separate record in said database each of said separate image portions.
-
-
29. The system of claim 28 wherein the processor further comprises means for converting said at least two image frames from a native color space to a single color space portion of a L*u*v color space and wherein the fuzzy logic color filter provides a value output signal that is at a maximum for only said set of at least one pre-selected road sign color-sets.
-
30. The system of claim 29, wherein the value output signals of the fuzzy logic color filter for each pixel are determined by location in said L*u*v color space and wherein the value output signals are assigned to a minimal set of mathematically described colors representing all of the legal color names and combinations of said set of at least one pre-selected road sign color sets.
-
31. The system of claim 28 wherein the fuzzy logic color filter segments the separate image portions such that each image portion contains a range of colors related to said color set whereby variations between colors of individual pixels are allowed to exist in said group of pixels that comprise each image portion.
-
32. The system of claim 20 wherein said location data metric stored in said database comprises storing at least an orientation of a camera while recording said image frame, a focal length of the camera and the location of the camera.
-
33. The system of claim 20 wherein the processor determines said location data metric for said road sign image by using triangulation techniques that assume a known location of at least one camera and a set of known camera parameters.
-
34. The system of claim 20 wherein said vehicle includes multiple cameras to generate said series of digital images.
-
35. The system of claim 20 wherein the processor stores into said database information selected from the set consisting of:
- a pointer to a portion of said frame in said frame buffer;
a flag that a road sign image is present, a description of a portion of said frame;
a portion of said frame, said frame, multiple frames of a common road sign image, or any combination thereof.
- a pointer to a portion of said frame in said frame buffer;
-
36. The system of claim 21 wherein the processor analyzes at least two frames of said series of digital images to determine whether the same road sign image is present in each of said at least two frames.
-
38. The system of claim 21 wherein the processor synchronizes a location data metric for each road sign image based on at least two frames of said series of digital images determined to have the same road sign image present.
-
25. A computer system for automatically generating a database of road sign images and positions comprising:
-
a frame buffer that stores at least one frame of a series of digital images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores information corresponding to a location of said vehicle when those digital images were recorded;
a processor that analyzes each frame in the frame buffer to identify a plurality of road sign images, with the processor initially analyzing each frame in said frame buffer to determine whether a road sign image is present and storing at least a pointer to a portion of a road sign image in a database if a road sign image is present in that image in said frame buffer, wherein for each image frame identified as having a road sign image present the processor comprises;
means for segmenting said image frame into at least one region of interest;
means for applying at least one filter to each of said regions of interest which differentiates between various types of road sign images so as to classify each of said regions of interest by type; and
means for storing in said database as a unique record in said database at least a portion of a bitmap corresponding to each region of interest in said image frame, said record including information identifying the type of road sign image classified from said at least one filter and a location data metric corresponding to a location for each of said plurality of road sign images.
-
-
26. A computer system for automatically generating a database of road sign images and positions comprising:
-
a frame buffer that stores at least one frame of a series of digital images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores information corresponding to a location of said vehicle when those digital images were recorded;
a processor that analyzes each frame in the frame buffer to identify a plurality of road sign images, with the processor initially analyzing each frame in said frame buffer to determine whether a road sign image is present and storing at least a pointer to a portion of a road sign image in a database if a road sign image is present in that image in said frame buffer, wherein for each image frame identified as having a road sign image present the processor comprises;
means for segmenting said image frame into a plurality of non-overlapping image segments;
means for identifying a segment that exhibits a differentiable characteristic as a search space; and
means for expanding said search space to include segments adjacent to said segment that exhibits the differentiable characteristic. - View Dependent Claims (27)
means for utilizing morphology techniques to grow and erode said search space by adding or subtracting segments adjacent said search space until either said search space meets or fails to meet uniform criteria for a differentiable characteristic.
-
-
37. A computer system for automatically generating a database of road sign images and positions comprising:
-
a frame buffer that stores at least one frame of a series of digital images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores information corresponding to a location of said vehicle when those digital images were recorded;
a processor that initially analyzes each frame in said frame buffer to determine whether a road sign image is present then analyzes at least two frames of said series of digital images to determine whether the same road sign image is present in each of said at least two frames, wherein the processor iteratively compares bitmap frames of said at least two frames of said series of digital images to determine if a first bitmap pixel set matches a second bitmap pixel set in terms of reflectance, color, shape or any combination thereof for a road sign image; and
a database into which the processor stores at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images only if a road sign image is present in that image in said frame buffer; and
a location data metric corresponding to a location for each of said plurality of road sign images.
-
-
39. In a computer system, a computer-readable storage media storing:
-
at least one computer program that operates to automatically generate a database of road sign images and positions from video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location information at least one camera recording those video images comprising the steps of;
analyzing said video images to identify a plurality of road sign images;
storing in a database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images; and
storing in said database a location data metric corresponding to a location for each of said plurality of road sign images. - View Dependent Claims (40, 41, 44, 45, 46, 47, 48, 50, 52, 53)
applying at least two separate filters to said image frame to generate a discrete output value from each filter, wherein each of at least two of said filters screens for a differentiable characteristic associated with a road sign image that is unique to that filter;
comparing the discrete output values for each of said at least two filters to at least one reference; and
identifying said image frame as potentially having a road sign image if at least one of the discrete output values indicates a differentiable characteristic is present for said image frame when the discrete output values are compared to the at least one reference.
-
-
44. The storage media of claim 40 wherein the at least one computer program further comprises for at least two image frames that each depict at least a single common road sign the steps of:
-
applying a fuzzy logic color filter to said at least two image frames to segment as separate image portions from said at least two image frames each region that contains a group of pixels all having a color-set from a set of at least one pre-selected road sign color-sets; and
saving as a separate record in said database each of said separate image portions.
-
-
45. The storage media of claim 39 wherein said location data metric stored in said database comprises storing at least an orientation of a camera while recording said image frame, a focal length of the camera and the location of the camera.
-
46. The storage media of claim 39 wherein the step of storing said location data metric determines said location data metric for said road sign image by using triangulation techniques that assume a known location of at least one camera and a set of known camera parameters.
-
47. The storage media of claim 39 wherein the step of storing in said database said at least a pointer to a portion of said image frame includes the steps selected from the set consisting of:
- storing a pointer to a portion of said image frame in said frame buffer;
setting a flag that a road sign image is present, storing a description of a portion of said image frame;
storing a portion of said image frame, storing said image frame, storing multiple image frames of a common road sign image, or any combination thereof.
- storing a pointer to a portion of said image frame in said frame buffer;
-
48. The storage media of claim 40 wherein the step of analyzing said video images further comprises the steps of:
-
iteratively comparing at least two frames of said video images having a road sign image present to determine whether the road sign image in each of said at least two frames matches; and
if there is a match, storing in said database at least an indication for at least one of the video images having a road sign image present that a common road sign image is depicted in multiple frames.
-
-
50. The storage media of claim 48 further comprising the step of synchronizing a location data metric for each road sign image based on at least two frames of said series of digital images determined to have the same road sign image present.
-
52. The storage media of claim 41 wherein each image portion contains a range of colors related to said color set whereby variations between colors of individual pixels are allowed to exist in said group of pixels that comprise each image portion.
-
53. The storage media of claim 41 wherein said location data metric includes information identifying at least an orientation of a camera while recording said image frame, a focal length of the camera and the location of the camera.
-
42. In a computer system, a computer-readable storage media storing:
-
at least one computer program that operates to automatically generate a database of road sign images and positions from video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location information at least one camera recording those video images comprising the steps of;
analyzing said video images to identify a plurality of road sign images, wherein the at least one computer program initially analyzes each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis;
storing in said database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images, wherein for each image frame identified as having a road sign image present;
segmenting said image frame into at least one region of interest;
applying at least one filter to each of said regions of interest which differentiates between various types of road sign images so as to classify each of said regions of interest by type; and
storing in said database as a unique record in said database at least a portion of a bitmap corresponding to each region of interest in said image frame, said record including information identifying the type of road sign image classified from said at least one filter; and
storing in said database a location data metric corresponding to a location for each of said plurality of road sign images.
-
-
43. In a computer system, a computer-readable storage media storing:
-
at least one computer program that operates to automatically generate a database of road sign images and positions from video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location information at least one camera recording those video images comprising the steps of;
analyzing said video images to identify a plurality of road sign images, wherein the at least one computer program initially analyzes each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis;
storing in said database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images, wherein for each image frame identified as having a road sign image;
segmenting said image frame into a plurality of non-overlapping image segments;
identifying a segment that exhibits a differentiable characteristic as a search space; and
expanding said search space to include segments adjacent to said segment that exhibits the differentiable characteristic utilizing morphology techniques to grow and erode said search space by adding or subtracting segments adjacent said search space until either said search space meets or fails to meet uniform criteria for a differentiable characteristic; and
storing in said database a location data metric corresponding to a location for each of said plurality of road sign images.
-
-
49. In a computer system, a computer-readable storage media storing:
-
at least one computer program that operates to automatically generating a database of road sign images and positions from video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location information at least one camera recording those video images comprising the steps of;
analyzing said video images to identify a plurality of road sign images by initially analyzing each image frame of said video images to determine whether a road sign image is present and if a road sign image is not present that image frame is discarded from further analysis, then iteratively comparing at least two frames of said video images having a road sign image present to determine whether the road sign image in each of said at least two frames matches by iteratively comparing bitmap frames of said at least two frames of said video image to determine if a first bitmap pixel set matches a second bitmap pixel set in terms of reflectance, color, shape or any combination thereof for the road sign image; and
if there is a match, storing in said database at least an indication for at least one of the video images having a road sign image present that a common road sign image is depicted in multiple frames;
storing in said database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images; and
storing in said database a location data metric corresponding to a location for each of said plurality of road sign images.
-
-
51. In a computer system, a computer-readable storage media storing:
-
a database comprising;
a portion of an image frame corresponding to each of said plurality of road sign images identified from video images depicting roadside scenes that are recorded from a vehicle navigating a road and having a system that stores location metrics for those video images; and
a location data metric corresponding to a location for each of said plurality of road sign images.
-
-
54. A method for generating a database of road sign images and positions comprising the steps of:
-
recording video images depicting roadside scenes that are recorded from a vehicle while navigating a road;
determining a location data metric of said vehicle corresponding to a location of said vehicle along said road when each of at least a portion of the image frames of said video images are recorded;
using a computer processor system to analyze said video images to identify a plurality of road sign images and store in a database;
at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images; and
a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded. - View Dependent Claims (55, 56, 57, 58, 61, 62, 63, 64, 65, 66, 67)
applying at least two separate filters to said image frame to generate a discrete output value from each filter, wherein each of at least two of said filters screens for a differentiable characteristic associated with a road sign image that is unique to that filter;
comparing the discrete output values for each of said at least two filters to at least one reference; and
identifying said image frame as potentially having a road sign image if at least one of the discrete output values indicates a differentiable characteristic is present for said image frame when the discrete output values are compared to the at least one reference.
-
-
57. The method of claim 56 wherein each of said at least two filters is selected from the set of the following filters:
- an edge filter, a color-pair filter, a color filter operating in the L*u*v color space, an edge filter combined with a line extender, or a color filter operating in the LCH color space.
-
58. The method of claim 56 wherein the differentiable characteristics of each of said at least two filters is selected from the set comprising:
- color characteristics, edge characteristics, texture characteristics, symmetry, convexity, lack of three dimensional volume, number and orientation of side edges, characteristic comer angles, frequency and luminescence.
-
61. The method of claim 54 wherein the computer processor analyzes at least two image frames that each depict at least a single common road sign by:
-
applying a fuzzy logic color filter to said at least two image frames to segment as separate image portions from said at least two image frames each region that contains a group of pixels all having a color-set from a set of at least one pre-selected road sign color-sets; and
saving as a separate record in said database each of said separate image portions.
-
-
62. The method of claim 61 wherein prior to applying the fuzzy logic color filter, the computer processor converts said at least two image frames from a native color space to a single color space portion of a L*u*v color space and wherein the fuzzy logic color filter provides a value output signal that is at a maximum for only said set of at least one pre-selected road sign color-sets.
-
63. The method of claim 62, wherein the value output signals of the fuzzy logic color filter for each pixel are determined by location in said L*u*v color space and wherein the value output signals are assigned to a minimal set of mathematically described colors representing all of the legal color names and combinations of said set of at least one pre-selected road sign color sets.
-
64. The method of claim 61 wherein the fuzzy logic color filter segments the separate image portions such that each image portion contains a range of colors related to said color set whereby variations between colors of individual pixels are allowed to exist in said group of pixels that comprise each image portion.
-
65. The method of claim 54 wherein the location data metric stored in said database comprises at least an orientation of a camera while recording said image frame, a focal length of the camera and the location of the camera.
-
66. The method of claim 54 wherein the step of storing said location data metric determines a location data metric for said road sign image by using triangulation techniques that assume a known location of at least one camera and a set of known camera parameters.
-
67. The method of claim 54 wherein the step of storing in said database said at least a pointer to a portion of said image frame includes the steps selected from the set consisting of:
- storing a pointer to a portion of said image frame in said frame buffer;
setting a flag that a road sign image is present, storing a description of a portion of said image frame;
storing a portion of said image frame, storing said image frame, storing multiple image frames of a common road sign image, or any combination thereof.
- storing a pointer to a portion of said image frame in said frame buffer;
-
59. A method for generating a database of road sign images and positions comprising the steps of:
-
recording video images depicting roadside scenes that are recorded from a vehicle while navigating a road;
determining a location data metric of said vehicle corresponding to a location of said vehicle along said road when at least a portion of the image frames of said video images are recorded;
using a computer processor system to analyze said video images to identify a plurality of road sign images and store to a database, wherein the computer processor system analyzes each image frame by;
determining whether said image frame has at least one region of interest and, if so, segmenting said image frame into at least one region of interest;
applying at least one filter to each of said regions of interest which differentiates between various types of road sign images so as to classify each of said regions of interest by type; and
storing in said database as a unique record in said database at least a portion of a bitmap corresponding to each region of interest in said image frame, said record including information identifying the type of road sign image classified from said at least one filter; and
storing in said database a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded.
-
-
60. A method for generating a database of road sign images and positions comprising the steps of:
-
recording video images depicting roadside scenes that are recorded from a vehicle while navigating a road;
determining a location data metric of said vehicle corresponding to a location of said vehicle along said road when at least a portion of the image frames of said video images are recorded;
using a computer processor system to analyze said video images to identify a plurality of road sign images and store to a database, wherein the computer processor system analyzes each image frame by;
discarding image frames that do not have at least one region of interest;
segmenting said remaining image frames into a plurality of non-overlapping image segments;
identifying a segment that exhibits a differentiable characteristic as a search space; and
expanding said search space to include segments adjacent to said segment that exhibits the differentiable characteristic by adding or subtracting segments adjacent said search space until either said search space meets or fails to meet uniform criteria for a differentiable characteristic;
storing in said database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images; and
storing in said database a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded.
-
-
68. A method for generating a database of road sign images and positions comprising the steps of:
-
recording video images depicting roadside scenes that are recorded from a vehicle while navigating a road;
determining a location data metric of said vehicle corresponding to a location of said vehicle along said road when at least a portion of the image frames of said video images are recorded;
using a computer processor system to analyze said video images to identify a plurality of road sign images and store to a database, wherein analyzing said video images comprises the steps of;
iteratively comparing at least two frames of said video images having a road sign image present to correlate common road sign images in each of said at least two frames; and
if there is a correlation, storing in said database at least an indication for at least one of the video images having a road sign image present that a common road sign image is depicted in multiple frames;
storing in said database at least a pointer to a portion of an image frame of said video images corresponding to each of said plurality of road sign images; and
storing in said database a location data metric corresponding to a location of said vehicle for each of said plurality of road sign images when said image frame corresponding to that road sign image was originally recorded. - View Dependent Claims (69, 70, 71)
placing all members of the first pixel set and the second pixel set that match in an identified field of said database;
synchronizing a location data metric to the identified field; and
storing in said database at least a pointer to a representative bitmap image of either the first pixel set or the second pixel set in conjunction with said location data metric.
-
-
71. The method of claim 70 wherein the said location data metric is determined for said common road sign image by using triangulation techniques that assume a known location of at least one camera and a set of known camera parameters.
Specification