Method and system for image processing for automatic road sign recognition
AssigneeUniversity of Connecticut
InventorsBahram Javidi, Elisabet Perez
Life of this patent
Estimating the term…
Abstract
The invention may comprise is a method and system of processing an image such as a road sign wherein the image is viewed from a host such as an automobile. The method begins with the initiating of a scan of an object to obtain an input. The input is converted to a signal before comparing the input with a set of stored inputs to determine a match. The converting step occurs via a fourier or similar transform to produce a transformed input; and, then filters the transformed input using nonlinear filtering.
Claims23
A method of processing an image wherein said image is viewed from a host, comprising:
- (a) initiating a scan of an object within a field so as to obtain an input,
- (b) converting said input to a signal representative of said input;
- (c) comparing said input with each one of a set of stored inputs to determine a match;
- and
- (i) if a match is determined, then causing an action to be performed in accordance with a set of instructions associated with said stored input;
- and
- (ii) if a match is not determined then continuing to scan said field for a second or subsequent image to be subjected to said comparing;
- and
- wherein said converting further comprises:
- (a) converting said input via a fourier transform of said input to produce a transformed input;
- and
- (b) filtering said transformed input using nonlinear filtering.
A method of automatically identifying a road sign from a moving vehicle, said method comprising:
- (a) initiating a scan of a road sign so as to obtain an identifying input wherein said identifying input is capable of being converted to a signal representative of said road sign;
- (b) comparing said identifying input with each one of a set of stored road sign inputs to determine a match;
- and
- (i) if a match is determined, then causing an action to be performed in accordance with a set of instructions associated with said stored input;
- and
- (ii) if a match is not determined then continuing to scan for a second or subsequent road sign to be subjected to said comparison step wherein said converting further comprises:
- (a) converting said input via a fourier transform of said input to produce a transformed input;
- and
- (b) filtering said transformed input using nonlinear filtering.
A system mounted on a host for processing an image comprising:
- (a) means for scanning an image;
- (b) conversion means for converting said scanned image to a set of data for comparison with one or more stored sets of data wherein said one or more stored sets of data are representative of one or more expected images
- wherein said conversion means further comprises:
- means for converting said scanned image via a fourier transform of said scanned image to produce a transformed input;
- and
- means for filtering said transformed input using nonlinear filtering;
- (c) a set of instructions associated with each one of stored sets of data wherein said set of instructions is indicative of an action to be performed by said host if a match is determined between said set of data and a one of said stored sets of data;
- (d) comparison means for comparing said set of data to each one of said stored sets of one or more images;
- (e) determination means for determining whether or not said set of data matches one of said stored sets of data;
- and
- (f) transmission means for transmitting a signal from said system to said host to react in accordance with said set of instructions.
Dependencies are inferred from the claim text.