Image analysis – Image transformation or preprocessing – Combining image portions
Reexamination Certificate
1999-03-18
2001-09-18
Boudreau, Leo (Department: 2621)
Image analysis
Image transformation or preprocessing
Combining image portions
Reexamination Certificate
active
06292593
ABSTRACT:
TECHNICAL FIELD
The present invention relates to an image synthesizing apparatus and method for obtaining an image which has the number of pixels larger than an image obtained by only one imaging operation using an image pickup device and which has high resolution, or has a wide angle of view and wide range.
BACKGROUND ART
In recent years, in an image synthesizing apparatus used as a so-called scanner, a one-dimensional image pickup device in which light receiving areas are linearly arranged, such as a CCD line sensor, is used. When an output image is obtained by this image synthesizing apparatus, first, while the image pickup device is moved relative to a subject on a two-dimensional plane in the direction perpendicular to the arrangement direction of the light receiving areas, an image of light from the subject is taken at a plurality of different positions. Next, the images obtained in the respective imaging operations are combined on the two-dimensional plane so that they are arranged with the same positional relation as the position of the image pickup device at the respective imaging operations, and the output image is obtained.
In Japanese Unexamined Patent Publication JP-A 5-260264, the present applicant has proposed a technique which uses a two-dimensional image pickup device with light receiving areas arranged in matrix and obtains an output image composed of pixels the number of which is larger than the number of the light receiving areas of the image pickup device. In an image processing apparatus of the publication, an operator of the apparatus takes an image of light while horizontally moving an image input means of a so-called CCD camera so that a plurality of images are obtained. At every time when a new image is obtained, a processing circuit first makes a check on matching of the new image and a combined image, and obtains a difference between these images. Next, on the basis of the obtained difference, a new image is deformed, and finally, the image after deformation and the combined image are combined, so that a new combined image is obtained.
With respect to the image obtained by the two apparatuses described above, as compared with an image directly obtained from an image pickup device, although the number of pixels arranged in the direction along the moving direction of the image pickup device is increased, the number of pixels in the direction orthogonal to the moving direction is not changed. As a technique for increasing the number of pixels in the orthogonal direction of the image, there is a technique in which the plurality of output images are arranged along the orthogonal direction, and are further combined.
Japanese Unexamined Patent Publication JP-A 63-64180 discloses an image synthesizing method of the technique mentioned above. The apparatus of the picture image synthesizing method uses a hand scanner which is provided with a one-dimensional image pickup device and can be manually moved. In the method, first, an original image of an input subject is divided into regions having an appropriate size to read by one scanning using the hand scanner, and image pickup is carried out while the hand scanner is moved along the center axis of the regions in the state that the hand scanner is made to be in contact with the original image, so that images of the respective regions are obtained. The image of each of the regions is equivalent to the output image of the apparatus using the one-dimensional image pickup device, and includes an overlapped region in which the same portion of the original image is read. Next, a check on matching of the overlapped regions of the images in the respective regions is carried out, and regions in the images where images of the subject are coincident with each other are obtained. Finally, superimposition is made after translation of the respective images is carried out to superimpose these images, so that a combined image of a picture image is obtained.
In the picture image synthesizing method described above, since the hand scanner including the image pickup device is manually moved by the operator of the apparatus, at the movement, there is a case where an unintentional movement of the operator's hands is applied to the hand scanner. Moreover, from the same reason, there is a case where the relative position between the image pickup device in the hand scanner and the center axis of the regions of the original image of the input subject is tilted. Moreover, although the hand scanner is provided with a pair of rollers at a portion which is brought into contact with an original image so that the hand scanner smoothly moves on the surface of the original image, when the smoothness of one of the rollers is different from that of the other, there is a case where the moving speed of the hand scanner is different between a portion in the vicinity of one roller and a portion in the vicinity of the other roller.
When these happen, in the image obtained by the apparatus, distortion of an image, such as deviation of an image of the subject and partial compression of the image, occurs. In the image in which such distortion occurred, images of the subject of a plurality of images are not smoothly connected by only the horizontal movement of the images, so that distortion occurs in the image of the subject in the combined image.
In the image processing apparatus of JP-A 5-260264, the image input means takes image light of an subject from a position apart from the subject by a predetermined distant. This image input means is supported by only a hand of an operator of the apparatus and is held in the air, so that the relative position to the subject is apt to be shifted. When the relative position is shifted like this, distortion of an image, such as deviation, tilt, and change in magnification of an image, occurs in the obtained images of the subject in the images-to-be-combined. Although the apparatus combines the images while correcting the distortions of the images of the subject in combining the images, there is a case where the distortion remains since all distortions of the image can not be corrected by, for example, an error in matching and a correction error.
An object of the invention is to provide an image synthesizing apparatus and method capable of obtaining an image in which an image of a subject has a small distortion, and which has high resolution or a wide angle of view and wide range, by using an image pickup device with a small number of light receiving areas.
DISCLOSURE OF THE INVENTION
The present invention provides an image synthesizing apparatus comprising:
image-to-be-combined producing means for producing a plurality of signals of images-to-be-combined, the signals representing the images-to-be-combined including different portions of a subject, the means producing signals so that an image-to-be-combined including a portion of the subject and an adjacent image-to-be-combined including a portion adjacent to said portion of the subject have an overlapped region where a same portion of the subject is imaged in the images-to-be-combined;
first matching means for making a check on matching of the images of the subject in the overlapped region for each set of adjacent images-to-be-combined among the image-to-be-combined signals produced by the image-to-be-combined producing means, and detecting a relative positional deviation between the overlapped regions on the basis of a result of the check on matching;
second matching means for setting a plurality of reference regions of a predetermined size in the overlapped region of one of the images-to-be-combined whose relative positional deviation is detected by the first matching means, for setting a search region larger than the reference region for each of the reference regions in the overlapped region of the other of the images-to-be-combined on the basis of the relative positional deviation, for making a check on matching of the images of the subject between the reference region and the search region corresponding to the reference region, and for detecting a relati
Akagi Hiroshi
Hirosawa Masashi
Kitamura Yoshihiro
Nakamura Mitsuaki
Nako Kazuyuki
Boudreau Leo
Nixon & Vanderhye P.C.
Patel Kanji
Sharp Kabushiki Kaisha
LandOfFree
Image synthesizing device and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image synthesizing device and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image synthesizing device and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2548098