Method and apparatus for processing image

1. A method of processing an image by using an image processing apparatus, the method comprising:

acquiring, by the image processing apparatus, a target image;

extracting a shape of a target object included in the target image;

determining a category including the target object based on the extracted shape; and

storing the target image by mapping the target image with additional information including at least one keyword related to the category,

wherein the method further comprises:

acquiring first position information of the image processing apparatus,

acquiring spatial information related to the category, and

generating second position information of the image processing apparatus based on the spatial information and the first position information.

2. The method of claim 1 , further comprising:

receiving a keyword from a user;

searching for the target image mapped with the additional information related to the received keyword; and

displaying the target image.

3. The method of claim 1 ,

wherein the acquiring of the target image comprises acquiring information indicating a time when the target image is acquired, and

wherein the additional information further comprises at least one keyword related to the information indicating the time.

4. The method of claim 1 ,

wherein the extracting of the shape of the target object included in the target image comprises extracting a feature map indicating an intensity gradient of pixels of the target image, and

wherein the determining of the category including the target object comprises comparing the extracted feature map with at least one of feature map models that are previously learned with respect to a shape of an object included in a first category.

5. The method of claim 4 , wherein the comparing of the extracted feature map with the at least one of the feature map models comprises:

calculating a reaction value of a filter designed based on the at least one of the feature map models with respect to the extracted feature map; and

if the reaction value is less than a critical value allotted to the at least one of the feature map models, determining that the target object is not included in the first category.

6. The method of claim 4 , wherein the comparing of the extracted feature map with the at least one of the feature map models comprises:

performing a first determination process of determining whether the target object is included in the first category by comparing the extracted feature map with a root model that is a previously learned feature map model with respect to an entire shape of an object included in the first category; and

if the target object is determined to be included in the first category according to a result of the first determination process, comparing the extracted feature map with at least one part model of part models that are previously learned feature map models with respect to shapes of parts of an object included in the first category.

7. The method of claim 6 , wherein the root model is previously learned by:

obtaining positive samples including a plurality of images related to the first category;

obtaining negative samples including a plurality of images related to the first category;

generating a plurality of feature maps from the positive samples and the negative samples; and

learning the root model of the first category by using the plurality of feature maps generated from the positive samples and the negative samples.

8. The method of claim 6 , wherein the extracting of the feature map comprises:

extracting a first feature map from the target image at a first resolution to compare with the root model; and

extracting a second feature map from the target image at a second resolution to compare with the at least one part model of the part models, the first resolution being lower than the second resolution.

9. The method of claim 6 , wherein the comparing of the extracted feature map with the at least one part model of the part models comprises:

selecting the at least one part model of the part models based on a priority order of the part models; and

comparing the extracted feature map with the selected part model.

10. The method of claim 6 , wherein the comparing of the extracted feature map with the at least one part model of the part models comprises comparing the extracted feature map with the at least part model one of the part models considering position information of the part models with respect to the root model.

11. An image processing apparatus comprising:

a storage device configured to store instructions, and a target image mapped with additional information, therein;

a processor, executing the stored instructions configured to:

extract a shape of a target object included in the target image,

determine a category including the target object based on the extracted shape,

map the target image with additional information including at least one keyword related to the category, and

control the storage device to store the target image by mapping the target image with additional information including at least one keyword related to the category; and

a position detection unit configured to acquire first position information of the image processing apparatus,

wherein the processor is further configured to:

acquire spatial information related to the category and

generate second position information of the image processing apparatus based on the spatial information and the first position information.

12. The image processing apparatus of claim 11 , further comprising:

a user input device configured to receive a keyword from a user; and

a display configured to display the target image mapped with the additional information related to the received keyword,

wherein the processor is further configured to search the storage unit for the target image mapped with the additional information related to the received keyword.

13. The image processing apparatus of claim 11 , further comprising:

an image sensor configured to acquire the target image and information indicating a time when the target image is acquired,

wherein the additional information further comprises at least one keyword related to the information indicating the time.

14. The image processing apparatus of claim 11 , wherein the processor is further configured to:

extract a feature map indicating an intensity gradient of pixels of the target image, and

compare the extracted feature map with at least one of feature map models that are previously learned with respect to a shape of an object included in a first category.

15. The image processing apparatus of claim 14 ,

wherein the processor is further configured to calculate a reaction value of a filter designed based on the at least one of the feature map models with respect to the extracted feature map, and

wherein, if the reaction value is less than a critical value allotted to the at least one of the feature map models, the target object is determined not to be included in the first category.

16. The image processing apparatus of claim 14 , wherein the processor is further configured to:

perform a first determination process of determining whether the target object is included in the first category by comparing the extracted feature map with a root model that is a previously learned feature map model with respect to an entire shape of an object included in the first category, and

if the target object is determined to be included in the first category according to a result of the first determination process, compare the extracted feature map with at least one part model of part models that are previously learned feature map models with respect to shapes of parts of an object included in the first category.

17. The image processing apparatus of claim 16 , wherein the processor is further configured to:

generate the part models with respect to parts of the object by extracting a first feature map from the target image at a first resolution to compare with the root model, and

extract a second feature map from the target image at a second resolution to compare with the at least one part model of the part models, the first resolution being lower than the second resolution.

18. The image processing apparatus of claim 16 , wherein the processor is further configured to:

select at least one part model of the part models based on a priority order of the part models, and

compare the extracted feature map with the selected part model.

19. A non-transitory computer readable storage medium having stored thereon a program, which when executed by a computer, performs a method comprising:

acquiring, by the image processing apparatus, a target image;

extracting a shape of a target object included in the target image;

determining a category including the target object based on the extracted shape; and

storing the target image by mapping the target image with additional information including at least one keyword related to the category,

wherein the method further comprises:

acquiring first position information of the image processing apparatus,

acquiring spatial information related to the category, and

generating second position information of the image processing apparatus based on the spatial information and the first position information.