Computer readable medium, image processing apparatus, and image processing method for learning images based on classification information

• US 8,923,628 B2
• Filed: 08/29/2011
• Issued: 12/30/2014
• Est. Priority Date: 02/08/2011
• Status: Expired due to Fees

• Current Assignee: Fujifilm Business Innovation Corp. (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Inventors: Fukui, Motofumi, Kato, Noriji
• Primary Examiner(s): DRENNAN, BARRY T
• Application Number:
  US13/220,201
• Publication Number:
  US 20120201464A1
• Time in Patent Office: 1,219 Days
• Field of Search: None
• US Class Current: 382/224
• CPC Class Codes: G06F 18/2321 using statistics or function optimisation, e.g. modelling of probability density functions
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529

Abstract

A computer readable medium stores a program causing a computer to execute a process for image processing. The process includes: calculating, on the basis of image feature information of a plurality of image areas each set with a classification information item, a probability distribution of the image feature information for each classification information item; acquiring a target image; calculating an evaluation value of each of pixels included in the target image relating to a specified classification information item, on the basis of the image feature information of an image area including the pixel and the probability distribution of the image feature information calculated for the specified classification information item; and extracting, from the target image, an image area relating to the specified classification information item, on the basis of the evaluation value calculated for each of the pixels included in the target image.

Claims

1. A non-transitory computer readable medium storing a program causing a computer to execute a process for image processing, the process comprising:

calculating, on the basis of image feature information of a plurality of learning image areas each set with a classification information item, a probability distribution of the image feature information for each classification information item;

acquiring a target image;

calculating an evaluation value of each of pixels included in the target image relating to a specified classification information item of the set classification information items corresponding to the plurality of learning image areas, on the basis of the image feature information of a target image area corresponding to the specified classification information item including the pixel and the probability distribution of the image feature information calculated for the specified classification information item; and

extracting, from the target image, the target image area relating to the specified classification information item, on the basis of the evaluation value calculated for each of the pixels included in the target image,

wherein the image feature information comprises a feature vector formed by an arrangement of mean image feature values extracted from a plurality of unit areas in the target image area corresponding to the specified classification information item,

wherein the mean image features values are color information and texture information extracted from the unit areas.

2. The non-transitory computer readable medium according to claim 1, wherein the extracting extracts, from the pixels included in the target image, the target image area including a connected pixel group connecting pixels, the calculated evaluation values of which are larger than a threshold value.

3. The computer readable medium according to claim 2, wherein the calculating the evaluation value of the pixel calculates, for each of a plurality of specified classification information items, the evaluation value of each of the pixels included in the target image relating to the corresponding classification information item, and

wherein the extracting extracts, for each of the plurality of specified classification information items, the target image area relating to the classification information item from the target image.

4. The non-transitory computer readable medium according to claim 1, the process further comprising:

setting another plurality of image areas in the target image; and

calculating, for each of the another plurality of set image areas, an evaluation value of an image area in the another plurality of set image areas on the basis of an evaluation value calculated for each of pixels included in the target image area,

wherein the extracting extracts, from the another plurality of image areas, the target image area relating to the specified classification information item, on the basis of the evaluation value calculated for each of the another plurality of image areas.

5. The computer readable medium according to claim 4, wherein the calculating the evaluation value of the pixel calculates, for each of a plurality of specified classification information items, the evaluation value of each of the pixels included in the target image relating to the corresponding classification information item, and

wherein the extracting extracts, for each of the plurality of specified classification information items, the target image area relating to the classification information item from the target image.

6. The non-transitory computer readable medium according to claim 1, wherein the calculating the evaluation value of the pixel calculates, for each of a plurality of specified classification information items, the evaluation value of each of the pixels included in the target image relating to the corresponding classification information item, and

wherein the extracting extracts, for each of the plurality of specified classification information items, the target image area relating to the classification information item from the target image.

7. A non-transitory computer readable medium storing a program causing a computer to execute a process for image processing, the process comprising:

calculating, on the basis of image feature information of a plurality of learning image areas each set with a classification information item, a probability distribution of the image feature information for each classification information item;

setting a sub-area in a target image;

calculating, for each of a plurality of specified classification information items, an evaluation value of each of pixels included in the sub-area relating to the classification information item of the set classification information items corresponding to the plurality of learning image areas, on the basis of the image feature information of a target image area corresponding to the specified classification information item including the pixel and the probability distribution of the image feature information calculated for the classification information item;

calculating, for each of the plurality of specified classification information items, an evaluation value of the sub-area on the basis of the evaluation value calculated for each of the pixels included in the sub-area; and

determining, from the plurality of specified classification information items, a classification information item relating to the sub-area, on the basis of the evaluation value of the sub-area calculated for each of the plurality of specified classification information items,

wherein the image feature information comprises a feature vector formed by an arrangement of mean image feature values extracted from a plurality of unit areas in the target image area corresponding to the specified classification information item,

wherein the mean image features values are color information and texture information extracted from the unit areas.

8. The non-transitory computer readable medium according to claim 7, wherein the setting sets another plurality of sub-areas each including a point specified in the target image,

wherein the calculating the evaluation value of a sub-area in the another plurality of sub-areas calculates, for each of the plurality of specified classification information items, an evaluation value of each of the another plurality of sub-areas, and

wherein the determining determines the sub-area and a classification information item corresponding to the largest calculated evaluation value.

9. An image processing apparatus comprising:

at least one processor which executes;

a probability distribution calculating unit that calculates, on the basis of image feature information of a plurality of learning image areas each set with a classification information item, a probability distribution of the image feature information for each classification information item;

an acquiring unit that acquires a target image;

a pixel evaluation value calculating unit that calculates an evaluation value of each of pixels included in the target image relating to a specified classification information item of the set classification items corresponding to the plurality of learning image areas, on the basis of the image feature information of a target image area corresponding to the specified classification information item including the pixel and the probability distribution of the image feature information calculated by the probability distribution calculating unit for the specified classification information item; and

an image area extracting unit that extracts, from the target image, the target image area relating to the specified classification information item, on the basis of the evaluation value calculated by the pixel evaluation value calculating unit for each of the pixels included in the target image,

wherein the image feature information comprises a feature vector formed by an arrangement of mean image feature values extracted from a plurality of unit areas in the target image area corresponding to the specified classification information item,

wherein the mean image features values are color information and texture information extracted from the unit areas.

10. An image processing apparatus comprising:

at least one processor which executes;

a probability distribution calculating unit that calculates, on the basis of image feature information of a plurality of learning image areas each set with a classification information item, a probability distribution of the image feature information for each classification information item;

a setting unit that sets a sub-area in a target image;

a pixel evaluation value calculating unit that calculates, for each of a plurality of specified classification information items, an evaluation value of each of pixels included in the sub-area relating to the classification information item of the set classification items corresponding to the plurality of learning image areas, on the basis of the image feature information of a target image area corresponding to the specified classification information item including the pixel and the probability distribution of the image feature information calculated by the probability distribution calculating unit for the classification information item;

an area evaluation value calculating unit that calculates, for each of the plurality of specified classification information items, an evaluation value of the sub-area on the basis of the evaluation value calculated by the pixel evaluation value calculating unit for each of the pixels included in the sub-area; and

a determining unit that determines, from the plurality of specified classification information items, a classification information item relating to the sub-area, on the basis of the evaluation value of the sub-area calculated by the area evaluation value calculating unit for each of the plurality of specified classification information items,

wherein the image feature information comprises a feature vector formed by an arrangement of mean image feature values extracted from a plurality of unit areas in the target image area corresponding to the specified classification information item,

wherein the mean image features values are color information and texture information extracted from the unit areas.

11. An image processing method comprising:

calculating, on the basis of image feature information of a plurality of learning image areas each set with a classification information item, a probability distribution of the image feature information for each classification information item;

acquiring a target image;

calculating an evaluation value of each of pixels included in the target image relating to a specified classification information item of the set classification information items corresponding to the plurality of learning image areas, on the basis of the image feature information of a target image area corresponding to the specified classification information item including the pixel and the probability distribution of the image feature information calculated for the specified classification information item; and

extracting, from the target image, the target image area relating to the specified classification information item, on the basis of the evaluation value calculated for each of the pixels included in the target image,

wherein the image feature information comprises a feature vector formed by an arrangement of mean image feature values extracted from a plurality of unit areas in the target image area corresponding to the specified classification information item,

wherein the mean image features values are color information and texture information extracted from the unit areas.