- Document Number:
20120274782
- Appl. No:
13/454689
- Application Filed:
April 24, 2012
- نبذة مختصرة :
A target fish-eye image for object recognition is split into regions in accordance with the distortion direction. An object recognition portion performs object recognition by using databases prepared in accordance with the regions respectively. The same process is applied to a plurality of target fish-eye images created by rotation of the target fish-eye image. A detected coordinate transformation portion restores the obtained position of an object into the original position by inverse rotation, and outputs a thus obtained result as a detection result. Detection accuracy in object recognition is improved, and the quality of data in databases is reduced. Rotated object images required for database creation are used for creation of databases for object recognition for the fish-eye image. Thus, a distorted fish-eye image for object recognition is directly used without projective conversion or image conversion based on corrective operation to the fish-eye image.
- Inventors:
KITAGUCHI, Takahiro (Ashigarakami-gun, JP); Suzuki, Kunihiko (Ashigarakami-gun, JP); Uejima, Toshiaki (Ashigarakami-gun, JP)
- Assignees:
Hitachi Information & Communication Engineering, Ltd. (Yokohama-shi, JP)
- Claim:
1. An object recognition method in an object recognition apparatus directly using a fish-eye image taken with a fish-eye camera for recognizing an object contained in the fish-eye image, wherein: the object recognition apparatus includes an object recognition portion which recognizes a detection-target object contained in the fish-eye image from the fish-eye image inputted from the fish-eye camera, and region-specific databases which are used for recognition of detection-target objects correspondingly to split regions which are determined in accordance with a direction of distortion of the input fish-eye image so that the input fish-eye image is split into the split regions; and the object recognition portion detects and recognizes the detection-target object contained in the fish-eye image from the input fish-eye image by using the region-specific databases corresponding to the split regions of the input fish-eye image.
- Claim:
2. An object recognition method according to claim 1, wherein: the object recognition apparatus further includes an image rotation portion which rotates the input fish-eye image; the image rotation portion rotates the input fish-eye image at intervals of a predetermined angle to thereby create a plurality of fish-eye images; and the object recognition portion performs detection and recognition of the detection-target object for each of the input not-rotated fish-eye image and the plurality of rotated fish-eye images in accordance with each split region.
- Claim:
3. An object recognition method according to claim 2, wherein: the object recognition apparatus further includes a detected coordinate transformation portion; and the detected coordinate transformation portion transforms detected coordinates of the object as a result of detection into coordinates in the fish-eye image before rotation in accordance with which of the not-rotated fish-eye image and the plurality of rotated fish-eye images the target object detected and recognized by the object recognition portion is detected from.
- Claim:
4. An object recognition method according to claim 1, wherein: the region-specific databases are created in such a manner that the fish-eye image containing the detection-target object and used for database creation is rotated and data of the detection-target object are extracted from the fish-eye image.
- Claim:
5. An object recognition apparatus directly using a fish-eye image taken with a fish-eye camera for recognizing an object contained in the fish-eye image, the object recognition apparatus comprising: an object recognition portion which recognizes a detection-target object contained in the fish-eye image from the fish-eye image inputted from the fish-eye camera; and region-specific databases which are used for recognition of detection-target objects correspondingly to split regions which are determined in accordance with a direction of distortion of the input fish-eye image so that the input fish-eye image is split into the split regions; wherein: the object recognition portion detects and recognizes the detection-target object contained in the fish-eye image from the input fish-eye image by using the region-specific databases corresponding to the split regions of the input fish-eye image.
- Claim:
6. An object recognition apparatus according to claim 5, further comprising: an image rotation portion which rotates the input fish-eye image; wherein: the image rotation portion rotates the input fish-eye image at intervals of a predetermined angle to thereby create a plurality of fish-eye images; and the object recognition portion performs detection and recognition of the detection-target object for each of the input not-rotated fish-eye image and the plurality of rotated fish-eye images in accordance with each split region.
- Claim:
7. An object recognition apparatus according to claim 6, further comprising: a detected coordinate transformation portion; wherein: the detected coordinate transformation portion transforms detected coordinates of the object as a result of detection into coordinates in the fish-eye image before rotation in accordance with which of the not-rotated fish-eye image and the plurality of rotated fish-eye images the target object detected and recognized by the object recognition portion is detected from.
- Claim:
8. An object recognition apparatus according to claim 5, wherein: the region-specific databases are created in such a manner that the fish-eye image containing the detection-target object and used for database creation is rotated and data of the detection-target object are extracted from the fish-eye image.
- Claim:
9. An object recognition apparatus according to claim 6, wherein: the region-specific databases are created in such a manner that the fish-eye image containing the detection-target object and used for database creation is rotated and data of the detection-target object are extracted from the fish-eye image.
- Claim:
10. An object recognition apparatus according to claim 7, wherein: the region-specific databases are created in such a manner that the fish-eye image containing the detection-target object and used for database creation is rotated and data of the detection-target object are extracted from the fish-eye image.
- Claim:
11. An object recognition method according to claim 2, wherein: the region-specific databases are created in such a manner that the fish-eye image containing the detection-target object and used for database creation is rotated and data of the detection-target object are extracted from the fish-eye image.
- Claim:
12. An object recognition method according to claim 3, wherein: the region-specific databases are created in such a manner that the fish-eye image containing the detection-target object and used for database creation is rotated and data of the detection-target object are extracted from the fish-eye image.
- Current U.S. Class:
348/169
- Current International Class:
04
- الرقم المعرف:
edspap.20120274782
Processing Request
No Comments.