Code recognition device
According to one embodiment, a code recognition device includes a reader, a processor for region detection, for first extraction, and for second extraction. The reader photographs a code image. The processor detects a code region and a letter region included in the code image from the code image photographed by the reader. The processor extracts first code information indicating the code image from the code region. The processor also extracts second code information indicating the code image from the letter region.
1. A code recognition device comprising: a reader for photographing a code image; a processor for detecting a code region and a letter region included in the code image from the code image photographed by the reader, for extracting a first code information represented by the code image from the code region, and for extracting a second code information represented by the code image from the letter region, wherein the processor is configured to perform an OCR process to recognize a letter string in the letter region after performing a distortion correction process to correct geometric distortion of the letter region; wherein when the reader photographs a code image with a camera that has pan and tilt functions, and wherein the device further comprises a processor for detecting a position of the code image based on a position of the code region in the code image and pan and tilt information indicating a direction of the camera adjusted using the pan and tilt functions.
2. The device according to claim 1 ,
wherein the processor extracts the second code information from the letter region when the first extraction may not extract the first code information.
3. The device according to claim 2 , further comprising:
a processor for integrating the first code information with the second code information.
4. The device according to claim 2 ,
wherein the reader photographs a code image with a camera that has pan and tilt functions, and
wherein the device further comprises:
a processor for detecting a position of the code image based on a position of the code region in the code image and pan and tilt information indicating a direction of the camera adjusted using the pan and tilt functions.
5. The device according to claim 1 , further comprising:
a processor for integrating the first code information with the second code information.
6. The device according to claim 5 ,
wherein the reader photographs a code image with a camera that has pan and tilt functions, and
wherein the device further comprises:
a processor for detecting a position of the code image based on a position of the code region in the code image and pan and tilt information indicating a direction of the camera adjusted using the pan and tilt functions.
7. The device according to claim 1 , wherein the reader photographs a code image with a camera that has pan and tilt functions; wherein the processor detects the code region from a plurality of code images photographed by the reader at a plurality of different positions, and wherein the processor detects a position of a code equivalent to the code image in 3-dimensional space based on a position of the code region detected from the plurality of images and pan and tilt information indicating a plurality of directions of the camera.
8. A method for performing code recognition, comprising: photographing a code image by a reader detecting, by a processor, a code region and a letter region included in the code image from the code image photographed by the reader; extracting, by the processor, a first code information represented by the code image from the code region; extracting, by the processor, a second code information represented by the code image from the letter region, wherein extracting the second code information comprises: performing an OCR process, by the processor, to recognize a letter string in the letter region after performing a distortion correction process to correct geometric distortion of the letter region; detecting, by the processor, the code region from a plurality of code images photographed by the reader at a plurality of different positions; and detecting, by the processor, a position of a code equivalent to the code image in a 3-dimensional space based on a position of the code region detected from the plurality of images and pan and tilt information indicating a plurality of directions of the camera.
9. The method according to claim 8 , further comprising:
extracting, by the processor, the second code information from the letter region when the first extraction does not extract the first code information.
10. The method according to claim 9 , further comprising:
integrating, by the processor, the first code information with the second code information.
11. The method according to claim 9 , further comprising:
photographing, by the reader, a code image with a camera that has pan and tilt functions; and
detecting, by the processor, a position of the code image based on a position of the code region in the code image and pan and tilt information indicating a direction of the camera adjusted using the pan and tilt functions.
12. The method according to claim 8 , further comprising:
integrating, by the processor, the first code information with the second code information.
13. The method according to claim 12 , further comprising:
photographing, by the reader, a code image with a camera that has pan and tilt functions; and
detecting, by the processor, a position of the code image based on a position of the code region in the code image and pan and tilt information indicating a direction of the camera adjusted using the pan and tilt functions.
14. The method according to claim 8 , further comprising:
photographing, by the reader, a code image with a camera that has pan and tilt functions; and
detecting, by the processor, a position of the code image based on a position of the code region in the code image and pan and tilt information indicating a direction of the camera adjusted using the pan and tilt functions.