Patent Application
20220358683
This application is based on and claims priority under 35 USC 119 from Japanese Patent Application No. 2021-078993 filed May 7, 2021.
The present disclosure relates to an information processing apparatus and a non-transitory computer readable medium.
Japanese Unexamined Patent Application Publication No. 2013-69148 describes an information processing apparatus including the following: a belonging probability calculating unit that obtains, for each scene category, a belonging probability which is a probability that an input image belongs to the scene category; an obtaining unit that obtains environment information that is information indicating an environment at the time of image capturing; a correction value calculating unit that calculates a composite correction value by reflecting the environment information in the belonging probability of each scene category; and an image correction unit that corrects the image using the composite correction value.
Japanese Unexamined Patent Application Publication No. 2004-234069 describes an image processing method including the following: identifying a subject present in an image and separating the image into a plurality of separated images; and for each separated image, obtaining a subject pattern that is image-processable by a method determined from a relationship with other images.
A user wants to correct an image to achieve a sense of unity in image quality with a plurality of images, but that image may happen to include two or more objects. In such a case, the user may correct the image according to any of the two or more objects. However, if the user adopts a configuration that corrects the image without outputting according to which of the objects the image is corrected, the user will not be informed of according to which of the objects the image is corrected.
Aspects of non-limiting embodiments of the present disclosure relate to informing a user of according to which of two or more objects included in an image the image is corrected in order to achieve a sense of unity in image quality with a plurality of images.
Aspects of certain non-limiting embodiments of the present disclosure address the above advantages and/or other advantages not described above. However, aspects of the non-limiting embodiments are not required to address the advantages described above, and aspects of the non-limiting embodiments of the present disclosure may not address advantages described above.
According to an aspect of the present disclosure, there is provided an information processing apparatus including a processor configured to: obtain a plurality of images each including any of a plurality of objects; and outputting report information that is information generated based on an analysis result regarding the plurality of objects in the plurality of images, and that is information for reporting according to which of two or more objects, among the plurality of objects, included in an image the image is to be corrected.
An exemplary embodiment of the present disclosure will be described in detail based on the following figures, wherein:
Hereinafter, an exemplary embodiment of the present disclosure will be described in detail with reference to the accompanying drawings.
The present exemplary embodiment provides an information processing apparatus that obtains a plurality of images each including any of a plurality of objects, and that outputs report information that is information generated based on an analysis result regarding the plurality of objects in the plurality of images, and that is information for reporting according to which of two or more objects, among the plurality of objects, included in an image the image is corrected.
Here, although the information processing apparatus may be one that obtains a plurality of images from data in any unit, such as the original or pages of printed matter that includes a plurality of images, the following description assumes that the information processing apparatus is one that obtains a plurality of images from an original of printed matter that includes a plurality of images.
In that case, it is considered that an image that is a target for determining according to which of the two or more objects the image is corrected may be included or not included in the plurality of images obtained from the original. The former is the case where images in the original are corrected to achieve a sense of unity in the original, and the latter is the case where images outside the original are corrected according to the sense of unity in the original. Although the information processing apparatus may perform either of the two types of correction, the following description assumes that the information processing apparatus performs the former type of correction.
In addition, the report information may be one that includes the order of two or more objects included in an image for determining, before correcting the image, according to which of the two or more objects the image is corrected. Alternatively, the report information may be one that includes the order of a plurality of objects for determining, before correcting a plurality of images each including any of the plurality of objects, according to which of two or more objects included in each image the image is corrected. Hereinafter, the case where the report information is the latter will be described by way of example.
In that case, at first, the order of the plurality of objects may be the order of all the objects or the order of some of the objects. The former is the case where the report information includes the order of the plurality of objects without omitting some of the objects, and the latter is the case where the report information includes the order of the plurality of objects with some omissions.
On this occasion, some of the objects may be determined by any method; however, the following description describes an exemplary case where some of the objects are determined by the following two methods.
A first method is a method in the case where the analysis result indicates the importance of each of the plurality of objects in the plurality of images. This method is such that, in the case where the analysis result indicates that the importance of a specific object among the plurality of objects is less than or equal to a criterion, a portion excluding the specific object serves as some of the plurality of objects.
A second method is a method that, in the case where there is no image that includes a specific object and another object among the plurality of objects, a portion excluding the specific object serves as some of the plurality of objects.
In that case, next, the information processing apparatus may change the order of the plurality of objects in accordance with a user operation. Although it is not always necessary for the information processing apparatus to change the order of the plurality of objects, the following description assumes that the information processing apparatus changes the order of the plurality of objects.
In addition, the information processing apparatus may be one that outputs impact information indicating the impact of a change in the order of a first object and a second object among the plurality of objects on correction of an image including the first object and the second object. Although it is not always necessary for the information processing apparatus to output the impact information, the following description assumes that the information processing apparatus outputs the impact information.
The impact information may be any information that indicates such an impact. Hereinafter, as the impact information, one that includes information for comparing a corrected image obtained by correcting the image including the first object and the second object according to the first object, and a corrected image obtained by correcting the image including the first object and the second object according to the second object will be described by way of example.
Hereinafter, the information processing apparatus will be described as an image processing apparatus by way of example. Here, the image processing apparatus may be a personal computer (PC) where image processing software is installed, and may be connected to a printer that prints an image-processed image.
The original obtaining unit 21 obtains an original of printed matter that includes a plurality of images. Here, the printed matter is one that is printed by a printer on a recording medium such as paper and that is used for a specific application. Examples of the printed matter include photobooks and food commodities. The original is data output to a printer for generating the printed matter. Examples of the original include data, generated by software, of photobooks and food commodities.
The image obtaining unit 22 obtains a plurality of images from the original obtained by the original obtaining unit 21. Here, it is preferable that the plurality of images be all the images included in the original; however, if all the images are too numerous, some of the images included in the original may be obtained. In addition, each of the plurality of images includes any of a plurality of scenes prepared in advance. Here, the scenes are objects included in images. For example, if an image is a photograph, a scene is a subject, i.e., an image-capturing target, included in the image. The scenes include, for example, landscapes, people, animals, and dishes. As the scenes, it is preferable that scenes that may be discriminated by image discrimination technology based on artificial intelligence (AI) be prepared in advance. In the present exemplary embodiment, processing of the image obtaining unit 22 is performed as an example of obtaining a plurality of images each including any of a plurality of objects.
The scene determination unit 23 determines one or more scenes included in each of the plurality of images obtained by the image obtaining unit 22, thereby generating scene information regarding scenes in the original. The scene determination unit 23 may determine one or more scenes using, for example, image discrimination technology based on AI. For each scene, the scene determination unit 23 counts the number of images including that scene, and, for each scene, generates information on the number of images including that scene as scene information.
Here, it is not always necessary that a scene determined by image discrimination technology based on AI and a scene managed by scene information have a one-to-one correspondence. To improve the accuracy of learning in AI, many subdivided labels may be prepared as labels representing each scene. However, if labels are displayed as they are, there will be numerous items and the user has difficulty in selecting scenes. Therefore, scenes discriminated by AI are organized into groups of scenes to be displayed to the user. For example, it is conceivable to organize mammals and reptiles into animals, dishes and ingredients into food, trees and flowers into plants, glasses and metals into industrial products, and so forth.
In addition, the scene determination unit 23 may separately count the number of images including only a certain scene and the number of images including that scene and another scene. Alternatively, the scene determination unit 23 may separately count the number of images at least including only a certain scene and the number of images including that scene and another scene. Hereinafter, the former counting method will be described by way of example. However, regardless of which counting method is adopted, the number of images per scene refers to the number of images including at least that scene. In addition, the scene determination unit 23 associates, for a scene combining a certain scene and another scene, not only the number of images but also the images themselves.
Every time the scene determination unit 23 determines one or more scenes included in an image obtained by the image obtaining unit 22, the scene determination unit 23 passes that image to the correction content determination unit 28.
The scene information memory 24 stores the scene information generated by the scene determination unit 23. In the present exemplary embodiment, the scene information is used as an example of an analysis result regarding the plurality of objects in the plurality of images. In the present exemplary embodiment, the scene information is also used as an example of an analysis result indicating the importance of each of the plurality of objects in the plurality of images.
The priority determination unit 25 determines, on the basis of the scene information stored in the scene information memory 24, the priority of scenes for determining according to which of two or more scenes included in an image the image is corrected. For example, the priority determination unit 25 determines the priority so that the priority of a scene included in many images in the original will be high. In addition, the priority determination unit 25 determines the priority so that, if there are two or more scenes included in the same number of images in the original, the priority of a scene whose priority in predetermined priority of recommendation is higher will be higher. Here, as the predetermined priority of recommendation, the order of people, landscapes, animals, food, plants, and industrial products is given as an example. This is the ascending order of the degree of impact, which is obtained by taking into consideration the degree of impact of correction.
The priority memory 26 stores the priority determined by the priority determination unit 25.
The priority changing unit 27 generates a priority checking screen including the priority stored in the priority memory 26 as it is, and outputs the priority checking screen to the display device 15. In the present exemplary embodiment, the priority checking screen is used as an example of report information that is information generated based on an analysis result and that is information for reporting according to which of two or more objects, among the plurality of objects, included in an image the image is corrected. In addition, in the present exemplary embodiment, the priority checking screen is also used as an example of report information that includes an order of the plurality of objects for determining according to which of the two or more objects the image is corrected. Furthermore, in the present exemplary embodiment, this processing of the priority changing unit 27 is performed as an example of outputting the report information.
In addition, the priority changing unit 27 may generate a priority checking screen that includes the priority stored in the priority memory 26 that is in a narrowed-down state, and output the priority checking screen to the display device 15. As this priority checking screen, there is one that includes the priority of scenes after excluding scenes that are included in only a few images, and one that includes the priority of scenes after excluding scenes that are not included with another scene in an image. In the former case, if the number of images for a certain scene is less than or equal to a criterion on the basis of the scene information stored in the scene information memory 24, the priority changing unit 27 simply excludes that scene. In the latter case, if the number of images for a scene combining a certain scene and another scene is 0 on the basis of the scene information stored in the scene information memory 24, the priority changing unit 27 simply excludes that scene. In the present exemplary embodiment, this priority checking screen is used as an example of report information that includes the order of the plurality of objects which is the order of some of the plurality of objects. In addition, in the present exemplary embodiment, this priority checking screen is also used as an example of report information that includes the order of the plurality of objects which is the order of a portion excluding, in the case where the analysis result indicates that the importance of a specific object among the plurality of objects is less than or equal to a criterion, the specific object. Furthermore, in the present exemplary embodiment, this priority checking screen is also used as an example of report information that includes the order of the plurality of objects which is the order of a portion excluding, in the case where there is no image that includes a specific object and another object among the plurality of objects, the specific object.
Furthermore, the priority changing unit 27 changes the priority stored in the priority memory 26 in response to a user operation on the priority checking screen displayed on the display device 15. In addition, the priority changing unit 27 may generate, before changing the priority, a change result checking screen that indicates the impact of changing the priority on correction of the image, and output the change result checking screen to the display device 15. On that occasion, the priority changing unit 27 may extract, on the basis of the scene information stored in the scene information memory 24, an image associated with a scene whose priority has been changed and a scene combined with another scene as an image affected by changing the priority. The priority changing unit 27 may generate information for comparing a corrected image before changing the priority of the image and a corrected image after changing the priority of the image, and output this information as part of the change result checking screen to the display device 15. In the present exemplary embodiment, this processing of the priority changing unit 27 is performed as an example of changing the order of the plurality of objects in accordance with a user operation. In addition, in the present exemplary embodiment, the change result checking screen is used as an example of impact information indicating the impact of a change in the order of a first object and a second object among the plurality of objects on correction of an image including the first object and the second object. Furthermore, in the present exemplary embodiment, the change result checking screen is also used as an example of impact information including information for comparing a corrected image obtained by correcting an image including the first object and the second object according to the first object, and a corrected image obtained by correcting an image including the first object and the second object according to the second object. Furthermore, in the present exemplary embodiment, this processing of the priority changing unit 27 is performed as an example of outputting the impact information.
The correction content determination unit 28 determines, on the basis of the priority stored in the priority memory 26, according to which of two or more scenes included in an image, passed from the scene determination unit 23, the image is corrected. The correction content determination unit 28 determines the correction content of correction according to the scene on the basis of the association between predetermined scenes and correction content.
When the image and the correction content are passed from the correction content determination unit 28, the correction processor 29 performs correction processing of the passed image with the passed correction content.
Note that it is not always necessary for the image processing apparatus 10 to include all of the original obtaining unit 21, the image obtaining unit 22, the scene determination unit 23, the scene information memory 24, the correction content determination unit 28, and the correction processor 29. For example, the image processing apparatus 10 need not include the original obtaining unit 21. In that case, in the image processing apparatus 10, the image obtaining unit 22 may obtain a plurality of images before these images are included in an original. Alternatively, the image processing apparatus 10 need not include the correction processor 29. In that case, the image processing apparatus 10 may convey the correction content determined by the correction content determination unit 28 to another apparatus, and this other apparatus may correct a to-be-corrected image with the conveyed correction content.
As illustrated in
Next, the image obtaining unit 22 obtains a plurality of images from the original obtained in step S201 (step S202).
Next, the scene determination unit 23 pays attention to one of the plurality of images obtained in step S202 (step S203). The scene determination unit 23 determines one or more scenes in the image to which attention has been paid in step S203 (step S204). Accordingly, the scene determination unit 23 counts up the number of images per scene on the basis of the scene information stored in the scene information memory 24 (step S205). After that, the scene determination unit 23 determines whether there is any unprocessed image in the images obtained in step S202 (step S206). If it is determined that there is an unprocessed image, the scene determination unit 23 returns the process back to step S203; if it is determined that there is no unprocessed image, the scene determination unit 23 advances the process to step S207.
Next, the priority determination unit 25 determines the priority of the scenes by referring to the scene information stored in the scene information memory 24. Specifically, the priority determination unit 25 determines the priority so that the priority becomes higher in descending order of the number of images per scene (step S207). The priority determination unit 25 determines whether there are plural scenes with the same number of images per scene on the basis of the scene information stored in the scene information memory 24 (step S208). If it is determined that there are plural scenes with the same number of images per scene, the priority determination unit 25 determines the priority so that the priority of a scene whose priority in predetermined priority of recommendation is higher will be higher (step S209). If it is determined that there are no plural scenes with the same number of images per scene, the priority determination unit 25 advances the process as it is to step S210. Then, the priority determination unit 25 stores the priority determined as above in the priority memory 26 (step S210).
Next, as illustrated in
Accordingly, the user may perform an operation to change the priority on the priority checking screen. Then, the priority changing unit 27 determines if there has been any operation performed by the user to change the priority (step S222). If it is determined that there has been no operation performed by the user to change the priority, the priority changing unit 27 advances the process to step S226. If it is determined that there has been an operation performed by the user to change the priority, the priority changing unit 27 outputs the changed priority to the display device 15 so that the changed priority will be displayed on the priority checking screen (step S223).
Accordingly, the user may check the result of changing the priority, that is, how correction of the image will be affected by changing the priority. Then, the priority changing unit 27 determines if there has been any operation performed by the user to request checking of the change result (step S224). If it is determined that there has been no operation performed by the user to request checking of the change result, the priority changing unit 27 advances the process to step S226. If it is determined that there has been an operation performed by the user to request checking of the change result, the priority changing unit 27 outputs a change result checking screen for allowing the user to check the result of changing the priority to the display device 15 (step S225). Then, the priority changing unit 27 returns the process back to step S224. For example, the priority changing unit 27 may return the process back to step S224 in the case where an approval entry is performed on the change result checking screen or a predetermined time has elapsed since the change result checking screen was displayed.
After that, the priority changing unit 27 determines if an operation has been performed by the user to confirm the priority (step S226). If it is determined that no operation has been performed by the user to confirm the priority, the priority changing unit 27 returns the process back to step S221. If it is determined that an operation has been performed by the user to confirm the priority, the priority changing unit 27 confirms the priority (step S227). Specifically, if step S227 is executed since there has been no operation performed by the user to change the priority in step S222, the priority changing unit 27 allows the priority stored in the priority memory 26 to be confirmed as it is. Alternatively, if step S227 is executed since there has been no operation performed by the user to request checking of the change result in step S224, the priority changing unit 27 overwrites the priority stored in the priority memory 26 with the priority changed by the user and allows the changed priority to be confirmed.
Next, as illustrated in
If it is determined in step S242 that the image includes two scenes, the correction content determination unit 28 obtains the priority stored in the priority memory 26 (step S244). By referring to the priority obtained in step S244, the correction content determination unit 28 determines correction content according to a scene of higher priority out of the two scenes (step S245).
Next, the correction processor 29 corrects the image to which attention has been paid in step S241 with the correction content determined in step S243 or step S245 (step S246). After that, the correction processor 29 determines whether there is any unprocessed image in the images used in scene determination in step S204 (step S247). If it is determined that there is an unprocessed image, the correction processor 29 returns the process back to step S241; if it is determined that there is no unprocessed image, the correction processor 29 ends the process. Specific Example of Operation of Image Processing Apparatus
The scene name is the name of a scene. The scene name incudes the scene name of only one scene, and the scene name of a combination of two scenes.
The number of images is the number of images including a corresponding scene.
The image ID is the identification information of an image including a corresponding scene.
Although it is assumed that the scene name includes the scene name of only one scene and the scene name of a combination of two scenes, the scene name may include the scene name at least including one scene, and the scene name of a combination of two scenes. In this case, if one image includes a plurality of scenes, each of the scenes is counted as one image. For example, if a certain image includes people and dishes, the image is counted as one image including people and also as one image including dishes.
It is displayed in the priority display field 401 that people, food, landscapes, animals, and plants are prioritized in this order. In addition, as information for allowing the user to recognize the degree of impact of each scene, the number of images including that scene is also displayed in the priority display field 401. Since the number of images that only include people is forty, the number of images that include people and animals is two, and the number of images that include people and food is eight in
The cancellation button 411 is a button that is pressed when aborting this process, and the confirmation button 412 is a button that is pressed when confirming the priority displayed at that time in the priority display field 401. In addition, the priority checking screen 400 is not provided with a button for checking the result of changing the priority. Therefore, the priority checking screen 400 is a screen displayed in the case where, after step S223 in
Only people, food, and landscapes are displayed in the priority display field 421. In contrast, neither animals nor plants are displayed since images including them are both less than or equal to 5% of the total. The cancellation button 431 and the confirmation button 432 are the same as the cancellation button 411 and the confirmation button 412 in
In the priority display field 441, only people, food, and animals that overlap any other scene in one image are displayed. In contrast, because landscapes and plants do not overlap any other scene in one image and, even if their priority is changed, this does not affect correction; thus, neither landscapes nor plants are displayed. In
As illustrated in
For example, the case of correcting an image including people and food will be considered. To correct the image according to people, brightness correction, contrast correction, noise correction, and skin-color correction are performed. In contrast, to correct the image according to food, brightness correction, contrast correction, and saturation correction are performed, and sharpness correction is weakly performed. Note that the combination of types of correction for each scene illustrated in
In the embodiments above, the term “processor” refers to hardware in a broad sense. Examples of the processor include general processors (e.g., CPU: Central Processing Unit) and dedicated processors (e.g., GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array, and programmable logic device).
In the embodiments above, the term “processor” is broad enough to encompass one processor or plural processors in collaboration which are located physically apart from each other but may work cooperatively. The order of operations of the processor is not limited to one described in the embodiments above, and may be changed.
Processing performed by the image processing apparatus 10 in the present exemplary embodiment is prepared as a program such as application software.
That is, a program realizing the present exemplary embodiment is regarded as a program causing a computer to realize the following functions: obtaining a plurality of images each including any of a plurality of objects, and outputting report information that is information generated based on an analysis result regarding the plurality of objects in the plurality of images, and that is information for reporting according to which of two or more objects, among the plurality of objects, included in an image the image is corrected.
Needless to say, the program realizing the present exemplary embodiment may be provided by a communication unit or by being stored in a recording medium such as compact-disc read-only memory (CD-ROM) or the like.
The foregoing description of the exemplary embodiments of the present disclosure has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosure to the precise forms disclosed. Obviously, many modifications and variations will be apparent to practitioners skilled in the art. The embodiments were chosen and described in order to best explain the principles of the disclosure and its practical applications, thereby enabling others skilled in the art to understand the disclosure for various embodiments and with the various modifications as are suited to the particular use contemplated. It is intended that the scope of the disclosure be defined by the following claims and their equivalents.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2021-078993 | May 2021 | JP | national |
