Image processing method and apparatus using trained dictionary
US-9443287-B2 · Sep 13, 2016 · US
US9875532B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9875532-B2 |
| Application number | US-201414911556-A |
| Country | US |
| Kind code | B2 |
| Filing date | Jul 18, 2014 |
| Priority date | Aug 15, 2013 |
| Publication date | Jan 23, 2018 |
| Grant date | Jan 23, 2018 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
The present invention provides an information processing device in which a degradation process of an input image is accurately estimated and a dictionary necessary for generating a desired restored image from the input image can be obtained. The information processing device is provided with: an image acquisition means that acquires a plurality of study images and an input image; and an estimation means that, on the basis of similarity between an arbitrary region of the input image and each of a plurality of degradation images in a case where regions of the study images corresponding to the arbitrary region are degraded on the basis of each of the plurality of degradation processes, outputs an estimated degradation process corresponding to the degradation process corresponding to the region of the input image.
Opening claim text (preview).
The invention claimed is: 1. An information processing device comprising circuitry configured to: acquire a plurality of first study images and an input image; output an estimated degradation process on a basis of first similarities between an arbitrary region in the input image and respective ones of a plurality of first degraded images, wherein corresponding regions in the first study images are degraded on a basis of respective ones of a plurality of degradation processes; and create a dictionary that stores a plurality of patch pairs each of which includes a degraded patch and a restored patch, the degraded patch being a patch in a second degraded image into which a second study image is degraded on a basis of the estimated degradation process, the restored patch being a patch in the second study image, wherein the estimated degradation process corresponds to a degradation process in the degradation processes, the degradation process being related to the region in the input image. 2. The information processing device according to claim 1 , wherein the circuitry is further configured to: on a basis of correspondence relations between respective ones of the first degraded images and the degradation processes from the first study images to respective ones of the first degraded images, output information discriminating the degradation process related to the region in the input image as the estimated degradation process. 3. The information processing device according to claim 1 , wherein the first similarities correspond to relations between feature vectors corresponding to the first degraded images and a feature vector corresponding to the region in the input image. 4. The information processing device according to claim 3 , wherein the circuitry is further configured to: output the estimated degradation process that corresponds to one of k (k is a natural number equal to or greater than 1) degradation processes of k-nearest neighbors, the k degradation processes being related to the region in the input image. 5. The information processing device according to claim 1 , wherein the circuitry is further configured to: generate a restored image from the input image by using the dictionary, and output the generated restored image; select the first study images on a basis of second similarities between the first study images and the restored image; and on a basis of the first similarities between respective ones of a plurality of first degraded images into which the regions in the selected first study images are degraded and the region in the input image, output the estimated degradation process. 6. The information processing device according to claim 5 , wherein the circuitry is further configured to: on a basis of correspondence relations between respective ones of the first degraded images and degradation processes from the selected first study images to respective ones of the first degraded images, output information discriminating the degradation process related to the region in the input image as the estimated degradation process. 7. The information processing device according to claim 5 , wherein the second similarities correspond to relations between feature vectors corresponding to the second study images and a feature vector corresponding to the restored image. 8. The information processing device according to claim 1 , wherein the circuitry is further configured to: generate the first degraded images from the first study images on a basis of the degradation processes; generate the feature vectors corresponding to the first degraded images; generate correspondence information that indicates relations between the feature vectors corresponding to the first degraded images and the degradation processes corresponding to the feature vectors corresponding to the first degraded images; calculate the feature vector corresponding to the region in the input image; classify the feature vector corresponding to the region in the input image into one of classes of the degradation processes on a basis of relations between the feature vector corresponding to the region in the input image and the feature vectors included in the correspondence information; and output the estimated degradation process corresponding to the class into which the feature vector corresponding to the region in the input image is classified to the outside. 9. The information processing device according to claim 8 , wherein the circuitry is further configured to: compress the feature vectors corresponding to the first degraded images; and generate the correspondence information that indicates relations between the compressed feature vectors and the degradation processes. 10. The information processing device according to claim 1 , wherein the circuitry is further configured to: output a first degraded image in the first degraded images as the estimated degradation process, the first degraded image being related to the degradation process corresponding to the region in the input image. 11. The information processing device according to claim 1 , wherein the circuitry is further configured to: allow a user to input degradation information of the input image, and further on a basis of the degradation information, output the estimated degradation process. 12. An image processing method using a computer implementing the image processing method, the image processing method comprising: acquiring a plurality of first study images and an input image; outputting an estimated degradation process on a basis of first similarities between an arbitrary region in the input image and respective ones of a plurality of first degraded images, wherein corresponding regions in the first study images are degraded on a basis of respective ones of a plurality of degradation processes; and creating a dictionary that stores a plurality of patch pairs each of which includes a degraded patch and a restored patch, the degraded patch being a patch in a second degraded image into which a second study image is degraded on a basis of the estimated degradation process, the restored patch being a patch in the second study image. 13. A non-transitory computer-readable recording medium storing a set of instructions which, when executed, causes a process of a computer to perform a method of information processing, the method comprising: acquiring a plurality of first study image and an input image; and outputting an estimated degradation process on a basis of first similarities between an arbitrary region in the input image and respective ones of a plurality of first degraded images, wherein corresponding regions in the first study images are degraded on a basis of respective ones of a plurality of degradation processes; and creating a dictionary that stores a plurality of patch pairs each of which includes a degraded patch and a restored patch, the degraded patch being a patch in a second degraded image into which a second study image is degraded on a basis of the estimated degradation process, the restored patch being a patch in the second study image. 14. An information processing device comprising: a memory storing instructions; and at least one processor configured to: acquire a plurality of first study images and an input image; output an estimated degradation process on a basis of first similarities between an arbitrary region in the input image and respective ones of a plurality of first degraded images, wherein corresponding regions in the first study images are degraded on a basis of respective ones of a plurality of degradation processes; and create a dictio
Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries, e.g. user dictionaries · CPC title
Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches · CPC title
Inspection of images, e.g. flaw detection · CPC title
Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title
Matching criteria, e.g. proximity measures · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.