# US 2013/0156313 A1 · System and method for identifying physical markings on objects > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2013/0156313 A1 - Document ID: US20130156313A1 - Status: Published application - Kind: A1 - Application number: 13/328,604 - Filed: 2011-12-16 - Published: 2013-06-20 - Earliest claimed: 2011-12-16 - Assignees: General Electric Company - Inventors: Joseph Salvo; John Carbone; Lynn Ann Derose; Daniel Messier; Bouchra Bouqata; Adam McCann; William Leonard - Web page: https://patentgrep.com/patent/US20130156313A1 ## CPC - G06V 10/96: Management of image or video recognition tasks - G06V 20/63: {Scene text, e.g. street names} - G06V 30/10: Character recognition - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A system and method for identifying characters using a processor and a sparse distributed memory (SDM) module. The system and method are configured to receive image data relating to an object having a surface with physical markings thereon. The physical markings include characters-of-interest. The system and method are also configured to analyze the image data to convert at least one of the characters-of-interest in the image data into a corresponding feature vector. The system and method are also configured to identify the characters-of-interest using the feature vector and the SDM module. A suggested identity for the characters-of-interest is provided. ## Claims ### Claim 1 A non-transitory computer readable medium for identifying characters using at least one processor and a sparse distributed memory (SDM) module, the computer readable medium including instructions to command the processor to: receive image data relating to an object having a surface with physical markings thereon, the physical markings including characters-of-interest; analyze the image data to convert at least one of the characters-of-interest in the image data into a corresponding feature vector, the feature vector including a vector address and a data pattern; and identify at least one of the characters-of-interest using the SDM module, the SDM module having hard locations that include a stored vector location address and a stored content counter, wherein said identifying operation includes: determining a relative distance between the vector address of the feature vector and the stored location addresses of the hard locations, wherein the hard locations that are within a predetermined distance from the vector address are activated locations; and providing a suggested identity of the corresponding characters-of-interest that is based upon the stored content counters of the activated locations. ### Claim 2 (depends on 1) The computer readable medium of claim 1, further comprising instructions to command the processor to store the data pattern of the feature vector in the stored content counters of the activated locations. ### Claim 3 (depends on 1) The computer readable medium of claim 1, further comprising instructions to command the processor to process the character(s)-of-interest to generate a binary image, the binary image having cells that are defined by a cell location and a value, the data pattern of the feature vector being based upon (a) the cell locations and (b) the values of the cells. ### Claim 4 (depends on 3) The computer readable medium of claim 3, wherein each value is either a first value or a different second value. ### Claim 5 (depends on 3) The computer readable medium of claim 3, wherein the binary image has less than 100 cells. ### Claim 6 (depends on 1) The computer readable medium of claim 1, wherein the image data includes lines of the physical markings in which at least two of the lines include a plurality of the characters-of-interest, wherein the computer readable medium includes instructions to command the processor to separate the lines horizontally and separate the lines vertically to isolate the characters-of-interest. ### Claim 7 (depends on 1) The computer readable medium of claim 1, wherein the vector address of the feature vector is also the data pattern of the feature vector. ### Claim 8 (depends on 1) The computer readable medium of claim 1, wherein the computer readable medium includes instructions to command the processor to provide the suggested identity only after a plurality of objects have been imaged. ### Claim 9 (depends on 1) The computer readable medium of claim 1, wherein the suggested identity includes an alphanumeric character. ### Claim 10 (depends on 1) The computer readable medium of claim 1, wherein the relative distance is a Hamming distance between the vector address and the corresponding location address. ### Claim 11 A system for identifying characters-of-interest from physical markings on a surface of an object, the system comprising: a segmentation module configured to receive and analyze an image of the object having physical markings that include characters-of-interest, the segmentation module also converting at least one of the characters-of-interest into a corresponding feature vector, the feature vector including a vector address and a data pattern; an identification module configured to identify the character(s)-of-interest using a sparse distributed memory (SDM) module, the SDM module having hard locations that include a stored vector location address and a stored content counter, wherein the identification module is configured to: determine a relative distance between the vector address of the feature vector and the stored vector location addresses of the hard locations, wherein the hard locations that are within a predetermined distance from the vector address are activated locations; and providing a suggested identity of the character-of-interest that is based upon the stored content counters of the activated locations. ### Claim 12 (depends on 11) The system of claim 11, further comprising an imager configured to acquire an image of the physical markings on the object. ### Claim 13 (depends on 12) The system of claim 12, further comprising a device housing that is sized and shaped to be handheld and carried by an individual, the device housing holding the imager and a non-transitory computer readable medium that includes the segmentation and identification modules. ### Claim 14 (depends on 11) The system of claim 11, wherein the segmentation module is configured to process the character(s)-of-interest to generate a binary image, the binary image having cells that are defined by a cell location and a value, the data pattern of the feature vector being based upon (a) the cell locations and (b) the values of the cells. ### Claim 15 (depends on 11) The system of claim 11, wherein the acquired image includes lines of the physical markings in which at least two of the lines include a plurality of the characters-of-interest, wherein the segmentation module is configured to separate the image horizontally to separate the lines and separate the image vertically to isolate the characters-of-interest. ### Claim 16 (depends on 11) The system of claim 11, wherein the vector address of the feature vector is also the data pattern of the feature vector. ### Claim 17 (depends on 11) The system of claim 11, wherein the suggested identity includes an alphanumeric character. ### Claim 18 (depends on 11) The system of claim 11, wherein the relative distance is a Hamming distance between the vector address and the corresponding stored vector location address. ### Claim 19 A method of identifying characters from an object having physical markings, the method comprising: receiving image data relating to an object having a surface with physical markings thereon, the physical markings including characters-of-interest; analyzing the image data to convert at least one of the characters-of-interest into a corresponding feature vector, the feature vector including a vector address and a data pattern; providing hard locations that include including a stored vector location address and a stored content counter; identifying activated locations by determining a relative distance between the feature vectors and the stored vector location addresses of the hard locations, wherein the relative distances between the feature vector and the activated locations are less than a predetermined distance; and providing a suggested identity of the character-of-interest that is based upon the stored content counters of the activated locations. ### Claim 20 (depends on 19) The method of claim 19, further comprising storing the data pattern of the feature vector in the stored content of the activated locations. ### Claim 21 (depends on 19) The method of claim 19, further comprising generating a binary image, the binary image having cells that are defined by a cell location and a value, the data pattern of the feature vector being based upon (a) the cell locations and (b) the values of the cells. ### Claim 22 (depends on 19) The method of claim 19, wherein the image includes a plurality of lines of the physical markings in which at least two of the lines include a plurality of the characters-of-interest, wherein the analyzing operation includes separating the image horizontally to separate the lines and separating the image vertically to isolate the characters-of-interest. ### Claim 23 (depends on 19) The method of claim 19, wherein the vector address of the feature vector is also the data pattern of the feature vector. ### Claim 24 (depends on 19) The method of claim 19, wherein the determining the relative distance operation includes determining a Hamming distance between the vector address and the corresponding stored vector location address.