Systems and methods for obtaining and using information from map images
First Claim
1. A non-transitory computer-readable storage medium carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to at least perform the following:
- causing, at least in part, a rendering of an image of a region of a metric vector space in a map image;
determining, within the map image of the region of the metric vector space, a pixel extent containing one or more sets of contiguous pixels having a text color, wherein the pixel extent comprises a raster label extract (RLE);
determining a confidence score for each RLE representing a likelihood that the pixel extent actually contains a textual annotation;
causing, at least in part, a generation of a text string for each RLE based on the pixel extent, a detection of a user input entering the text string based on the pixel extent, or a combination thereof;
causing to display, on a user interface, a list of RLEs with the corresponding confidence score and an associated image of the RLE, a set of extraction state controls for each RLE, wherein a extraction state control for each RLE displays information about the RLE; and
causing, at least in part, a change to a state of the RLE through detection of a user input modifying a state associated with the pixel extent via the user interface, wherein the modifying the state includes at least one of entering a new text string, correcting the incorrect text string, validating the correct text string, or a combination thereof.
4 Assignments
0 Petitions
Accused Products
Abstract
Under one aspect, an interface program stored on a computer-readable medium causes a computer system with a display device to perform the functions of: displaying an image of a region of a metric vector space; obtaining a set of pixel extents in the image that contain apparent textual annotations; for each pixel extent of the set, obtaining a confidence score that the pixel extent actually contains a textual annotation; and displaying the set of pixel extents in a ranked list based on the corresponding confidence scores. In some embodiments, the program further causes the computer system to perform the functions of, for at least one pixel extent of the set, at least one of automatically generating a text string based on the apparent textual annotation, and accepting user input entering a text string based on the apparent textual annotation.
-
Citations
16 Claims
-
1. A non-transitory computer-readable storage medium carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to at least perform the following:
-
causing, at least in part, a rendering of an image of a region of a metric vector space in a map image; determining, within the map image of the region of the metric vector space, a pixel extent containing one or more sets of contiguous pixels having a text color, wherein the pixel extent comprises a raster label extract (RLE); determining a confidence score for each RLE representing a likelihood that the pixel extent actually contains a textual annotation; causing, at least in part, a generation of a text string for each RLE based on the pixel extent, a detection of a user input entering the text string based on the pixel extent, or a combination thereof; causing to display, on a user interface, a list of RLEs with the corresponding confidence score and an associated image of the RLE, a set of extraction state controls for each RLE, wherein a extraction state control for each RLE displays information about the RLE; and causing, at least in part, a change to a state of the RLE through detection of a user input modifying a state associated with the pixel extent via the user interface, wherein the modifying the state includes at least one of entering a new text string, correcting the incorrect text string, validating the correct text string, or a combination thereof. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A method comprising:
-
causing, at least in part, a rendering of an image of a region of a metric vector space in a map image; determining, within the map image of the region of the metric vector space, a pixel extent containing one or more sets of contiguous pixels having a text color, wherein the pixel extent comprises a raster label extract (RLE); determining a confidence score for each RLE representing a likelihood that the pixel extent actually contains a textual annotation; causing, at least in part, a generation of a text string for each RLE based on the pixel extent, a detection of a user input entering the text string based on the pixel extent, or a combination thereof; causing to display, on a user interface, a list of RLEs with the corresponding confidence score and an associated image of the RLE, a set of extraction state controls for each RLE, wherein a extraction state control for each RLE displays information about the RLE; and causing, at least in part, a change to a state of the RLE through detection of a user input modifying a state associated with the pixel extent via the user interface, wherein the modifying the state includes at least one of entering a new text string, correcting the incorrect text string, validating the correct text string, or a combination thereof. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
Specification