WO2010070804A1 - 画像検索装置及び画像検索方法 - Google Patents
画像検索装置及び画像検索方法 Download PDFInfo
- Publication number
- WO2010070804A1 WO2010070804A1 PCT/JP2009/005866 JP2009005866W WO2010070804A1 WO 2010070804 A1 WO2010070804 A1 WO 2010070804A1 JP 2009005866 W JP2009005866 W JP 2009005866W WO 2010070804 A1 WO2010070804 A1 WO 2010070804A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- interpolation
- interpolated
- images
- range
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims description 26
- 238000004364 calculation method Methods 0.000 claims abstract description 60
- 238000003384 imaging method Methods 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims description 46
- 239000000284 extract Substances 0.000 claims description 14
- 238000010586 diagram Methods 0.000 description 26
- 238000012545 processing Methods 0.000 description 21
- 230000015654 memory Effects 0.000 description 16
- 238000004891 communication Methods 0.000 description 6
- 238000007796 conventional method Methods 0.000 description 3
- 238000010187 selection method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Definitions
- the present invention relates to an image search apparatus and an image search method for searching for image content for interpolating between shooting positions of image content shot by a user from among a large amount of stored image content.
- Patent Document 1 a method has been proposed in which a user can easily create a memorable album using photos of sightseeing spots stored in advance on a server (see, for example, Patent Document 1).
- a user reads a high-quality photograph of the point stored in advance in a server by using a tag reader installed at each photography spot in a sightseeing spot.
- a tag reader installed at each photography spot in a sightseeing spot.
- the user cannot acquire a landscape photograph viewed from the window when the user is moving by train. Further, when the user moves by car or bicycle, the user needs to get off the car or bicycle and then stop at the photo spot. Therefore, in the conventional method, the load given to the user to acquire a photograph becomes high.
- the present invention solves the above-described problem, and it is desired to interpolate between the shooting positions of the image content selected by the user without stopping at the shooting spot from among a large amount of stored image content.
- An image search apparatus and an image search method capable of searching for images are provided.
- an image search apparatus provides an interpolation image for interpolating between positions where each of at least two interpolation target images is captured from a plurality of images.
- An image search device to search for a geographical location including a first shooting position that is a position where a first interpolation target image is shot and a second shooting position that is a position where a second interpolation target image is shot
- An interpolation range calculation unit that calculates a range as an interpolation range, and an image including a shooting position within the interpolation range calculated by the interpolation range calculation unit as a candidate image, from among the plurality of images stored in the image database
- the still picture and an interpolated image selecting section for selecting from among the candidate images.
- the user can acquire the interpolated image without constructing a special infrastructure such as a tag reader.
- FIG. 1 is a block diagram showing a characteristic functional configuration of an image search apparatus according to Embodiment 1 of the present invention.
- FIG. 2 is a diagram illustrating an example of image content and content-attached information.
- FIG. 3 is a diagram illustrating an example of an image database.
- FIG. 4 is a diagram illustrating a screen example of the input unit.
- FIG. 5 is a diagram illustrating an example of the interpolation range.
- FIG. 6 is a flowchart showing the flow of overall processing executed by the image search device according to Embodiment 1 of the present invention.
- FIG. 7 is a flowchart showing a detailed process flow of step S105 of FIG.
- FIG. 8 is a diagram illustrating an example of a processing result of the interpolation image selection unit.
- FIG. 1 is a block diagram showing a characteristic functional configuration of an image search apparatus according to Embodiment 1 of the present invention.
- FIG. 2 is a diagram illustrating an example of image content and content-attached
- FIG. 9 is a diagram illustrating an example of a screen output to the output medium by the output unit.
- FIG. 10 is a diagram illustrating an example of a screen output to the output medium by the output unit.
- FIG. 11 is a block diagram showing a characteristic functional configuration of the image search apparatus according to Embodiment 2 of the present invention.
- FIG. 12 is a diagram illustrating an example of a relevance database.
- FIG. 13 is a conceptual diagram showing hierarchized areas.
- FIG. 14 is a flowchart showing an overall flow of processing executed by the image search device according to Embodiment 2 of the present invention.
- FIG. 15 is a flowchart showing a detailed process flow of step S301 of FIG.
- FIG. 16 is a diagram illustrating an example of a relevance database.
- FIG. 17 is a flowchart showing another detailed process flow of step S301 of FIG.
- FIG. 18 is a diagram illustrating an example of the subject distance database.
- FIG. 19 is a diagram illustrating an example of
- FIG. 1 is a block diagram showing a characteristic functional configuration of an image search apparatus 100 according to Embodiment 1 of the present invention.
- the image search apparatus 100 is an apparatus that searches for an interpolated image from a plurality of images.
- the interpolated image is an image for interpolating between the shooting positions of the two images.
- the interpolated image is an image photographed at a position between the photographing positions of the two images.
- the image search apparatus 100 includes an image database 101, an input unit 102, an interpolation range calculation unit 103, an interpolation image candidate acquisition unit 104, a moving speed calculation unit 105, an interpolation image selection unit 106, and an output unit 107.
- the image database 101 is a storage device including, for example, a hard disk, a memory, and the like.
- the image database 101 is a database that stores the image content 20 to be searched and the content attached information 21 corresponding to the image content 20.
- the image content 20 indicates an image stored in an electronic medium.
- the content attached information 21 is information indicating the content of the image content 20.
- the content-attached information 21 includes Exif (Exchangeable image file format) information (shooting date and time, latitude / longitude information, etc.) information added to the image content 20 at the time of image shooting, and image content after the user has shot the image content.
- 20 is image information such as a tag added to 20.
- the user ID 22 indicates information for identifying the user who has taken the image content 20.
- the image ID 23 indicates information for identifying the image content 20.
- the image address 24 indicates information on a location where the image content 20 is stored.
- the shooting date and time 25 indicates the date and time when the image content 20 was shot.
- the shooting position 26 indicates a position where the image content 20 is shot. Note that the position where the image content 20 is captured is specified by latitude and longitude.
- the subject distance 28 indicates the distance to the subject when the image content 20 is photographed.
- the subject distance 28 is automatically given to the image as Exif information at the time of shooting.
- the subject distance a calculated using the following equation may be included in the content attached information 21.
- a is the subject distance
- f is the focal length
- M is the photographing magnification. That is, among image contents stored in the image database 101, for image contents in which the subject distance is not given as Exif information, the shooting magnification and the focal length included in the Exif information given to each image content are expressed by equations.
- the subject distance a obtained by substituting for (1) may be stored in the image database 101.
- the tag 27 indicates a word (keyword) indicating the content of the image content 20.
- the image search apparatus 100 may automatically add a region name or a building name acquired based on latitude and longitude information included in the shooting position 26 as a tag 27.
- the image search apparatus 100 may automatically add weather and temperature information as a tag 27 using the shooting date and time 25, the shooting position 26, and a weather database.
- the image search device 100 may automatically add the image representative color obtained from the pixel value of the image content 20 as the tag 27.
- the image search apparatus 100 may automatically add information such as the number of people detected by performing person detection in the image content 20 as the tag 27.
- FIG. 3 is a diagram illustrating an example of the image database 101. As shown in FIG. 3, in the image database 101, each content attached information 21 shown in FIG. 2 is listed so as to be easily searched.
- the image database 101 is provided in the image search apparatus 100 .
- the image search apparatus to which the present invention is applied is not limited to such an image search apparatus.
- the image database 101 may be provided on a server connected to the image search apparatus via a network.
- the content attached information 21 may be included in the image content 20.
- the image database 101 may store the moving speed of the user between images as the content attached information 21.
- the moving speed is a value calculated by dividing the distance between the shooting positions of two interpolation target images by the difference time between the shooting times of the two interpolation target images.
- the moving speed of the image content 20 is calculated by dividing the distance between the shooting positions with the image content shot next to the image content 20 by the difference time of the shooting time. Then, the calculated moving speed is stored as the content attached information 21.
- the moving speed between images may be stored in the image database 101.
- the moving speed calculation unit 105 may calculate the moving speed every time an input from the user is received.
- the input unit 102 acquires, for example, information input by a user from a keyboard, a mouse, or the like in a memory or the like. Specifically, the input unit 102 includes a plurality of interpolation targets that the user wants to interpolate between the positions at which the images were taken from the image contents 20 taken by the user, the image IDs 23 thereof, and the image contents 20. Information about the image and the interpolation target image (image ID 23, etc.) is acquired. Then, the input unit 102 notifies the interpolation range calculation unit 103 and the movement speed calculation unit 105 of the acquired image ID 23 of the interpolation target image.
- FIG. 4 is a diagram illustrating a screen example of the input unit 102.
- the user selects a plurality of image contents 20 (interpolation target images) to be interpolated between the positions where the images are photographed, from the image contents 20 photographed by the user.
- the user selects the first interpolation target image (Kiyomizu-dera) 40 and the second interpolation target image (Ginkaku-ji) 41 out of the six image contents 20 photographed in Kyoto. That is, the input unit 102 has received a request to interpolate an image of a place between Kiyomizu Temple and Ginkakuji from the user.
- the interpolation target image is not necessarily selected directly by the user.
- the image search device may estimate the interpolation target image based on the user's preference or the like.
- the image search apparatus may select an interpolation target image using the input place information.
- the interpolation range calculation unit 103 includes, for example, a CPU (Central Processing Unit), a memory storing a program, and the like.
- the interpolation range calculation unit 103 refers to the image database 101 to acquire the shooting position 26 and the shooting date / time 25 corresponding to the image ID 23 acquired by the input unit 102. Then, the interpolation range calculation unit 103 calculates the interpolation range 52 indicating the range of the location to be interpolated using the acquired shooting position 26 and shooting date / time 25. That is, the interpolation range calculation unit 103 calculates a range of a location including the shooting position 26 of the interpolation target image as the interpolation range.
- a CPU Central Processing Unit
- the interpolation range calculation unit 103 calculates the interpolation range so that the location range becomes wider as the elapsed time of the photographing date and time 25 of the interpolation target image increases. Further, the interpolation range calculation unit 103 notifies the interpolation image selection unit 106 of the calculated interpolation range 52. A specific interpolation range calculation method will be described later.
- FIG. 5 is a diagram illustrating an example of the interpolation range 52 calculated based on the first interpolation target image (Kiyomizu-dera) 40 and the second interpolation target image (Ginkaku-ji) 41.
- the interpolation range calculation unit 103 calculates the latitude and longitude of the diagonal points of the rectangle as the range information 50 and the range information 51.
- the interpolation image candidate acquisition unit 104 includes, for example, a CPU, a memory storing a program, and the like.
- the interpolated image candidate acquisition unit 104 detects an image group photographed within the interpolation range calculated by the interpolation range calculation unit 103.
- the interpolation image candidate acquisition unit 104 refers to the image database 101 to acquire the image ID of the image group included in the latitude and longitude information calculated by the interpolation range calculation unit 103. Then, the interpolation image candidate acquisition unit 104 notifies the interpolation image selection unit 106 of the acquired image ID.
- the moving speed calculation unit 105 includes, for example, a CPU and a memory storing a program.
- the movement speed calculation unit 105 calculates the movement speed of the user between the interpolation target images acquired from the input unit 102. Specifically, the moving speed calculation unit 105 calculates a linear distance between the shooting positions 26 of the interpolation target image. Further, the moving speed calculation unit 105 calculates time from the difference of the shooting date and time 25 of the interpolation target image. Then, the moving speed calculation unit 105 calculates the moving speed by dividing the calculated distance by time. In addition, the moving speed calculation unit 105 notifies the interpolation image selection unit 106 of the shooting position information between the interpolation target images and the calculated moving speed information.
- the moving speed is a speed at which the user moves from the shooting position of one interpolation target image to the shooting position of the other interpolation target image when the same user has shot the interpolation target image.
- the moving speed is a value calculated by dividing the distance between the shooting positions of the two interpolation target images by the difference time between the shooting times of the two interpolation target images.
- the interpolation image selection unit 106 includes, for example, a CPU, a memory storing a program, and the like.
- the interpolation image selection unit 106 uses the image ID acquired from the interpolation image candidate acquisition unit 104, the shooting position information and movement speed information acquired from the movement speed calculation unit 105, and the image database 101 to acquire the interpolation image candidate acquisition unit 104. Only necessary images are selected from the selected image group according to the moving speed of the user. That is, the interpolation image selection unit 106 selects among the candidate images acquired by the interpolation image candidate acquisition unit 104 so that only images with a larger subject distance are selected as the movement speed calculated by the movement speed calculation unit 105 increases. To select an interpolated image.
- the interpolation image selection unit 106 selects all the image groups acquired from the interpolation image candidate acquisition unit 104 when the moving speed is equal to or less than a predetermined threshold. On the other hand, when the moving speed is greater than a predetermined threshold, the interpolation image selection unit 106 selects a photograph (distant view) whose subject distance is equal to or greater than a predetermined threshold from the image group acquired from the interpolation image candidate acquisition unit 104. Just get a photo). Then, the interpolation image selection unit 106 notifies the output unit 107 of the image ID of the selected image. Note that the interpolation image selection unit 106 may acquire the movement speed of the user between the interpolation target images acquired from the input unit 102 from the image database 101. A specific interpolation image selection method will be described later.
- the interpolated image selection unit 106 selects an image having a larger subject distance as the moving speed increases.
- the moving speed of the photographer is relatively slow (walking or the like)
- the photographer can appreciate both a short-distance subject and a long-distance subject.
- the distant image which is an image with a large subject distance
- the moving speed is relatively fast (car, train, etc.)
- the photographer cannot appreciate the subject at a short distance. Therefore, when the moving speed is fast, the foreground image does not match the image that the user wants to interpolate.
- the interpolation image selection unit 106 selects both the distant view image and the foreground image when the moving speed is slow, and selects only the distant view image when the moving speed is fast.
- the image search apparatus 100 can search only images that match the image that the user wants to interpolate.
- the output unit 107 outputs the image content 20 acquired by the interpolation image selection unit 106 to an output medium.
- the output medium is a monitor such as a liquid crystal display or a television.
- FIG. 6 is a flowchart showing the overall flow of processing executed by the image search apparatus 100.
- the input unit 102 receives the image ID 23 of the image content 20 photographed by the user and the image IDs 23 of a plurality of interpolation target images that the user wants to perform interpolation from among the image IDs 23. Then, the input unit 102 notifies the image ID 23 of the received interpolation target image to the interpolation range calculation unit 103 and the interpolation image candidate acquisition unit 104 (step S101).
- the interpolation range calculation unit 103 acquires the shooting position 26 and the shooting date / time 25 corresponding to the image ID 23 of the interpolation target image notified from the input unit 102 from the image database 101. Then, the interpolation range calculation unit 103 uses the acquired shooting position 26 and shooting date / time 25 to calculate an interpolation range indicating the range of the location to be interpolated. Further, the interpolation range calculation unit 103 notifies the interpolation image candidate acquisition unit 104 of the calculated interpolation range (step S102).
- the interpolation range calculation unit 103 sets a distance that is centered on the center of gravity of the shooting position 26 (latitude, longitude) of the interpolation target image and that is proportional to the elapsed time obtained from the shooting date 25 of the interpolation target image.
- the radius range is calculated as the interpolation range.
- the elapsed time is the difference between the shooting date and time 25 of the interpolation target image.
- the elapsed time is the difference between the oldest shooting date and time 25 and the latest shooting date and time 25 in the shooting date and time 25 of the interpolation target image.
- the interpolation range calculation unit 103 may select a range including the shooting position 26 (latitude, longitude) of the interpolation target image from among a plurality of preset ranges as the interpolation range. In this case, the interpolation range calculation unit 103 may add a range adjacent to the selected range to the interpolation range according to the elapsed time of the interpolation target image.
- the interpolation range calculation unit 103 uses an area shown in FIG. 5 in order to allow the user to specify the interpolation range. May be presented to the user. Thereby, the interpolation range calculation unit 103 can prevent an interpolation range from becoming too large and an interpolation range calculation error.
- the interpolation image candidate acquisition unit 104 detects an image group included in the interpolation range calculated by the interpolation range calculation unit 103, and notifies the interpolation image selection unit 106 (step S103). Specifically, the image ID of the image group included in the latitude and longitude information calculated by the interpolation range calculation unit 103 is acquired by referring to the image database 101.
- the moving speed calculation unit 105 calculates the moving speed of the user between the interpolation target images using the shooting date and time 25 and the shooting position 26 of each interpolation target image acquired from the input unit 102, and the interpolation target image.
- the interpolated image selection unit 106 is notified of the shooting position information and the calculated speed information (step S104).
- the moving speed calculation unit 105 calculates the distance x between the shooting positions from the difference between the shooting positions of the first interpolation target image (Kiyomizu Temple) 40 and the second interpolation target image (Ginkakuji) 41 in FIG. To do. On the other hand, the moving speed calculation unit 105 calculates the time t from the difference in shooting time. Then, the moving speed calculation unit 105 calculates the moving speed by dividing the distance x by the time t.
- the moving speed calculation unit 105 may perform a route search connecting the shooting positions using a separately prepared map database, and calculate the distance x along the searched route. Thereby, the moving speed calculation unit 105 can calculate the distance between the shooting positions more accurately. Therefore, the moving speed calculation unit 105 can improve the calculation accuracy of the moving speed, and can prevent unnecessary interpolation images from being presented to the user due to erroneous calculation of the moving speed.
- the interpolated image selection unit 106 selects a candidate image suitable for the moving speed from the shooting position information and speed information of the interpolation target image notified from the moving speed calculating unit 105, and notifies the output unit 107 of the candidate image. (Step S105).
- the output unit 107 outputs the interpolation image acquired by the interpolation image selection unit 106 to the output medium (step S106).
- FIG. 7 is a flowchart showing a detailed process flow of the interpolation image selection process (step S105 in FIG. 6) by the interpolation image selection unit 106.
- step S105 in FIG. 6 an image selection method adapted to the moving speed will be described with reference to FIG.
- the interpolation image selection unit 106 acquires the image ID of the candidate image from the interpolation image candidate acquisition unit 104 (step S201).
- the interpolated image selection unit 106 acquires the moving speed of the user from the moving speed calculation unit 105 (step S202).
- the interpolation image selection unit 106 determines whether or not the moving speed is greater than a predetermined threshold (step S203). If the moving speed is greater than the threshold (Yes in step S203), the process proceeds to step S204. On the other hand, when the moving speed is equal to or less than the threshold (No in step S203), the process proceeds to step S205.
- a standard person's limit walking speed (about 12 km / hour) may be used as the threshold of the moving speed. Thereby, the interpolation image selection unit 106 can determine whether or not the user is walking and moving between the shooting positions of the interpolation target image.
- the interpolated image selection unit 106 selects all candidate images acquired in step S201 as interpolated images (step S205).
- the moving speed of the photographer is slow (walking or the like)
- the photographer can appreciate both a short-distance subject and a long-distance subject. Therefore, both the foreground image and the foreground image match the experience of the photographer and can be an interpolation image. Therefore, when the moving speed is slow, the interpolation image selection unit 106 can select an image that matches the image that the user wants to interpolate by selecting both the distant view image and the foreground image.
- FIG. 8 is a diagram illustrating an example of a processing result of the interpolation image selection unit 106 in which different interpolation images are selected according to the moving speed by the processing in steps S201 to S205.
- the user selects the first interpolation target image (Kiyomizu Temple) 40 and the second interpolation target image (Ginkakuji) 41 shown in FIG. 4, and the interpolation range is the interpolation range 52 shown in FIG.
- an image whose shooting position is included in the interpolation range 52 among the images existing in the image database 101 is the candidate image 80.
- the interpolation image ( In the case of low speed movement) 81 all candidate images are selected. That is, when the moving speed is “3 km / h”, the interpolated image selection unit 106 selects the interpolated image 81 (during low-speed movement).
- the interpolated image selection unit 106 selects the interpolated image (at high speed movement) 82.
- the interpolation image selection unit 106 can select only an image that matches the moving speed of the user as an interpolation image.
- FIG. 9 and 10 are diagrams illustrating examples of screens output to the output medium by the output unit 107.
- the output unit 107 may display a list of interpolation images on a map or the like. Also, as shown in FIG. 10, the output unit 107 may display the interpolated images in a slide show format in the order from the closest shooting position.
- the image search apparatus 100 can estimate the captured content of the interpolated image. It is possible to acquire an image of a user who performs shooting similar to that as an interpolation image.
- the image search apparatus 100 can select an interpolated image from images stored in the image database 101, so that the user can select an interpolated image without stopping at the shooting spot. It can be acquired.
- the image search apparatus 100 can select an image having a larger subject distance as the moving speed between the image capturing positions increases, it is possible to efficiently search for an image that the user wants to interpolate.
- the image search apparatus 200 is characterized in that an image highly relevant to the user's shooting preference is extracted from the candidate images and presented.
- the photographing preference indicates the preference of the subject to be photographed and the preference of the image quality such as the image color and the angle of view.
- the image search apparatus 100 acquires an image suitable for the moving speed of the user from images whose shooting positions are included in the interpolation range, and selects all the acquired images as interpolation images.
- the image search apparatus 100 cannot present an interpolation image that is more suitable for the user's shooting preference from among the interpolation images to be presented. The possibility arises. For example, consider a case where the user is a user who takes a lot of photographs with the theme of “garden”.
- the image search device presents the “garden” image if the user's preference is not known It is not always possible. Therefore, if the user's shooting preference is known from the image captured by the user, the image search device can preferentially present an image highly relevant to the user's shooting preference from the interpolation images to be presented. Convenience can be improved.
- the image search apparatus 200 extracts an image highly relevant to the user's shooting preference using the tag, shooting time, shooting position, and the like of the image shot by the user.
- an example is described in which an image stored on the server is interpolated with an image taken by a photographer during a Kyoto trip to create a memorable album. To do.
- the relevance database 201 is a storage device composed of, for example, a hard disk, a memory, or the like.
- the relevance level database 201 stores a relevance level indicating the level of relevance between the image contents 20. Specifically, the degree of association between the subject information of the image content 20 is stored.
- the subject information is information on a word (keyword) representing the subject in the image content 20 taken by the user.
- the subject information includes, for example, the name of the subject set in the tag 27, the name of the area where the subject exists, the number of subjects, the color of the subject, and the like.
- the relevance level database 201 stores the relevance level of subject information in units of areas. Specifically, the degree of association database 201 stores the degree of association between the tags 27 of the image content 20 belonging to the area for each area.
- the area indicates a place divided based on a prefecture, a municipality, a 50 km square, or the like.
- the relevance is, for example, “metadata space generation method for associative search based on relevance between words appearing in a document” (Non-patent Document 1: Hidenori Honma et al., 16th Data Engineering Workshop (DEWS 2005). ), 6A-o2, The Institute of Electronics, Information and Communication Engineers, 2005).
- FIG. 12 is a diagram showing an example of the relevance database 201 of the area around Higashiyama Ward and Sakyo Ward in Kyoto City.
- the relevance level between the row direction tag 27 and the column direction tag 27 is stored at a position where the row direction and the column direction intersect. This degree of association is calculated based on the co-occurrence between the tags 27 of the image content 20 belonging to each area.
- whether or not the image content 20 belongs to the area is determined based on the shooting position 26 of the image content 20.
- region in the relevance degree database 201 may be hierarchized into two or more.
- the relevance level database 201 may store the relevance level of the tag 27 using three stages of a small area, a medium area, and a large area for the same point.
- FIG. 13 is a conceptual diagram showing hierarchized areas.
- the relevance level database 201 may store the relevance level between the tags 27 of the image content 20 belonging to Kyoto Prefecture as the relevance level corresponding to the large area 130 indicating the area in units of prefectures. . Further, the relevance level database 201 may store the relevance level between the tags 27 of the image content 20 belonging to Kyoto city as the relevance level corresponding to the middle area 131 indicating the area of the municipality unit. Further, the relevance level database 201 may store the relevance level between the tags 27 of the image content 20 belonging to Shimogyo Ward or the like as the relevance level corresponding to the small area 132 indicating the area in units of wards.
- the association image extraction unit 202 described later determines the degree of association in the degree of association database 201 according to the area where the user wants to interpolate the image.
- the area to be acquired can be selected.
- the related image extraction unit 202 can accurately estimate the captured content of the image that the user wants to interpolate.
- the related image extraction unit 202 includes, for example, a CPU, a memory storing a program, and the like.
- the related image extraction unit 202 uses the image database 101 and the relevance degree database 201 to extract a related image highly relevant to the user's captured image from the interpolation images acquired from the interpolation image selection unit 106 and extract the extracted images.
- the image ID is notified to the output unit 107. A specific related image extraction method will be described later.
- the related image is defined as an image highly relevant to the image information given to the image taken by the user.
- an image having a tag highly relevant to a tag attached to an image taken by the user is an example.
- the tag attached to the image is information indicating an object and an impression about the object. It can be said that an image in which information on these subjects matches is an image highly relevant to the user's preference.
- An image that is highly related to the shooting conditions image information such as shooting time, color information (white balance, RGB ratio), aperture, subject distance, resolution, F value, shutter speed, etc.
- image information such as shooting time, color information (white balance, RGB ratio), aperture, subject distance, resolution, F value, shutter speed, etc.
- image information such as shooting time, color information (white balance, RGB ratio), aperture, subject distance, resolution, F value, shutter speed, etc.
- image information such as shooting time, color information (white balance, RGB ratio), aperture, subject distance, resolution, F value, shutter speed, etc.
- an image highly relevant to an image captured by the user is extracted using the relevance level database 201 in which the relevance level between tags of images registered in the image database 101 is stored.
- the processing to be performed is different from the processing in the image search device 100 of the first embodiment.
- other processes are the same as those of the image search apparatus 100 of the first embodiment.
- the processing of steps S101 to S106 executed by the image search device 200 is the same as the processing of the image search device 100 of Embodiment 1 shown in FIG.
- the related image extraction unit 202 uses the image database 101 and the relevance level database 201 to extract an image highly relevant to the user's shooting preference from the interpolation images notified from the interpolation image selection unit 106 (step S301). ).
- FIG. 15 is a flowchart showing a detailed processing flow of the image extraction processing (step S301 in FIG. 14) by the related image extraction unit 202.
- an image extraction method for extracting an image highly relevant to the user's shooting preference will be described with reference to FIG.
- the related image extraction unit 202 acquires the image ID of the interpolation image from the interpolation image selection unit 106 (step S402).
- the related image extraction unit 202 selects one image ID from the image IDs of the interpolated images acquired in step S402 (step S403). That is, the related image extraction unit 202 selects one interpolation image.
- the related image extraction unit 202 acquires a tag corresponding to the image ID acquired in step S403 from the image database 101 (step S404).
- the tag acquired in step S404 is referred to as a tag set B.
- the related image extraction unit 202 calculates the degree of association between the tag set A acquired in step S401 and the tag set B acquired in step S404 (step S405).
- the related image extraction unit 202 refers to the relevance level database 201 of the region to which the shooting position 26 of the tag set A belongs, thereby acquiring the relevance level of the tag set B with respect to the tag 27 included in the tag set A. .
- the related image extraction unit 202 calculates the degree of association between the tag set A and the tag set B by total addition of the obtained degree of association between the tags.
- the tag set A is a tag attached to an image taken by the user himself, and the tag reflects the user's preference for the image and the viewpoint of the photograph.
- the tag set B is a tag given by a person other than the user to an image taken by a person other than the user, and these tags reflect the preferences and viewpoints of the person other than the user. That is, it can be said that the image including the tag set B having a high degree of association with the tag set A is an image having high relevance to the user's preference.
- FIG. 16 shows the interpolated image tag 160 (“Heian Shrine”, “Torii”, and “Red” (tag set B)) selected in step S403, and the captured image tag 161 (“Ginkakuji”, It is a figure which shows an example of the relevance degree database 201 in which the relevance degree with "garden” and "Kiyomizu-dera” (tag set A)) was memorize
- the related image extraction unit 202 calculates the degree of association between the interpolation image and the user-captured image in step S405 will be described with reference to FIG.
- Heian Jingu” in tag set B and tag set A (“Ginkakuji”, “Garden”, “Kiyomizu Temple”) are related to each tag in “Heian Jingu” and tag set A ( It can be calculated by adding the hatched area 133).
- the degree of association between “torii” and tag set A is 0.5
- the degree of association between “red” and tag set A is 0.5.
- the degree of association between the user-captured image and the interpolated image is obtained by adding all the degrees of association between the tags assigned to both images, which is obtained when the related image extraction unit 202 refers to the association degree database 201. Is calculated by
- the related image extraction unit 202 stores the relevance calculated in step S405 together with the image ID of the interpolated image selected in step S403 in the image ID-relevance level list (step S406).
- the related image extraction unit 202 determines whether all the interpolated images acquired in step S402 have been selected in step S403 (step S407).
- the related image extraction unit 202 extracts images with higher relevance levels from the interpolated image-relevance level list until the number of images that can be output is reached. (Step S408).
- the related image extraction unit 202 when the interpolated images are arranged in descending order in the interpolated image-relevance list, the interpolated images that are equal to or higher than the relevance of the interpolated images appearing in the same order as the number of outputtable images. Extract.
- the outputtable number is the number of images that can be output by the output unit 107, which is determined by the storage capacity of the output medium, the display area, and the like.
- the related image extraction unit 202 repeats the processes in steps S403 to S406.
- the related image extraction unit 202 extracts an interpolation image highly relevant to the user's shooting preference from the interpolation images acquired by the interpolation image selection unit 106. can do.
- the image search apparatus 200 can perform even when the number of interpolation images selected by the interpolation image selection unit 106 exceeds the number that can be presented. It is possible to preferentially present an image highly relevant to the user's shooting preference.
- step S ⁇ b> 301 the related image extraction unit 202 uses the image database 101 to generate an image (an image acquired by the input unit 102) captured by the user from among the interpolation images notified from the interpolation image selection unit 106.
- You may extract an image with high relevance (henceforth a similarity degree) with imaging conditions.
- shooting conditions are also reflected in the user's shooting preferences.
- the shooting conditions indicate shooting parameters that are automatically given to an image at the time of shooting such as color information (white balance, RGB ratio), aperture, subject distance, resolution, F value, shutter speed, and the like.
- FIG. 17 is a flowchart showing a flow of processing performed by the related image extraction unit 202 to extract an image having a high degree of similarity with the shooting condition of the image shot by the user.
- an image extraction method for extracting an image highly relevant to the user's shooting conditions will be described with reference to FIG.
- the related image extraction unit 202 refers to the image database 101 to acquire the shooting parameters given to the image shot by the user (the image acquired in step S101) (step S501).
- the shooting parameters are information automatically added to each image at the time of shooting such as color information (white balance, RGB ratio), aperture, subject distance, resolution, F value, shutter speed, and the like.
- the related image extraction unit 202 acquires the interpolation image selected by the interpolation image selection unit 106 (step S503).
- the related image extraction unit 202 selects one image from the interpolated images acquired in step S503 (step S504).
- the related image extraction unit 202 refers to the image database 101 to acquire the imaging parameter (P2) of the interpolated image acquired in step S504 (step S505).
- the related image extraction unit 202 uses the shooting parameter representative value (P1) calculated in step S502 and the shooting parameter (P2) acquired from step S505, and the shooting parameter of the user shooting image and the selected interpolation image.
- the similarity e is calculated (step S506).
- the similarity e can be calculated using, for example, the following equation (2).
- n is the number of types of shooting parameters to be compared
- P1i is the i-th shooting parameter of P1
- P2i is the i-th shooting parameter of P2.
- the sum of values obtained by normalizing the difference values of the parameters P1 and P2 using the parameter values of P1 is calculated as the similarity e. Therefore, the closer the value of similarity e is to 0, the more similar the shooting parameters of the user-captured image and the interpolation image selected in step S504. That is, the closer the value of similarity e is to 0, the higher the relevance of the shooting conditions between the image acquired in step S101 and the interpolated image selected in step S504.
- the related image extraction unit 202 lists and stores combinations of the interpolation image selected in step S504 and the similarity calculated in step S506 (step S507).
- step S508 determines whether or not all the interpolated images acquired in step S503 have been selected in step S504 (step S508). If all the images are selected (Yes in step S508), the image search device 200 proceeds to step S509. If all are not selected (No in step S508), the image search device 200 proceeds to step S504.
- the related image extraction unit 202 refers to the list of interpolated images and similarities stored in step S507, and extracts as many images as possible that can be output (step S509).
- the related image extraction unit 202 extracts an interpolation image highly relevant to the user's shooting conditions from the interpolation images acquired by the interpolation image selection unit 106. can do.
- the image search apparatus 200 can extract an interpolated image having a high relevance between the interpolation target image and the shooting conditions, the user can more efficiently search for an image to be interpolated.
- the image search device has been described based on the embodiments.
- the present invention is not limited to these embodiments. Unless it deviates from the gist of the present invention, various modifications conceived by those skilled in the art have been made in the present embodiment, and forms constructed by combining components in different embodiments are also included in the scope of the present invention. .
- the image search apparatuses 100 and 200 include the image database 101, the input unit 102, the moving speed calculation unit 105, the output unit 107, and the like. .
- the interpolation range calculation unit 103 receives a plurality of interpolation target images (including the first image and the second image) and information (first information) about each interpolation target image via a network or the like from an external device or the like. And the second information are included).
- the interpolation image candidate acquisition unit 104 and the interpolation image selection unit 106 may refer to an image database provided in a server or the like via a network or the like.
- the interpolation image selection unit 106 may acquire the moving speed between the shooting positions of each interpolation target image from the image database, information on each interpolation target image, or the like.
- the image search apparatus can compare the images that the user wants to interpolate between the shooting positions of the images selected by the user from among a plurality of images without having to stop at the shooting spot without including the moving speed calculation unit 105 or the like. It is possible to search easily.
- the image search device determines the subject distance threshold value by comparing the moving speed with a predetermined threshold value.
- the image search device uses such an image search method. It is not limited to a device.
- the image search device may determine a subject distance threshold by referring to a subject distance database that stores a table storing a relationship between a moving speed and a subject distance. good.
- FIG. 18 is a diagram showing an example of the subject distance database.
- the subject distance database stores a threshold of the subject distance with respect to the moving speed. For example, when the moving speed is “20 km / h”, the image search device selects an image having a subject distance of “10 m” or more as an interpolated image.
- the image search apparatus may use the subject distance threshold value for the closest moving speed stored in the subject distance database. Note that the image search apparatus according to an aspect of the present invention may determine the subject distance threshold value using a mathematical expression showing the relationship between the moving speed and the subject distance instead of the subject distance database.
- the related image extraction unit extracts an image from the interpolation images selected by the interpolation image selection unit. It is not limited to an image search device.
- the related image extraction unit may extract an image from candidate images.
- the interpolation image selection unit may select an interpolation image from the images extracted by the related image extraction unit.
- the subject distance is included in the content ancillary information 21.
- the interpolated image selection unit 106 may calculate the subject distance using Expression (1).
- the present invention can be realized as an image search method for executing processing performed by the characteristic components of such an image search apparatus, or as a program for causing a computer to execute the method as shown in FIG. You can also Such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.
- FIG. 19 is a diagram illustrating an example of a hardware configuration of a computer.
- a program for searching for an interpolated image from a plurality of images is stored in, for example, a CD-ROM 515 that is a computer-readable medium, and is read out via the CD-ROM device 514. Further, for example, a program for searching for an interpolated image from a plurality of images is transmitted via a wired or wireless network, broadcast, or the like.
- the CPU 501 executes a program read via the CD-ROM device 514 or a program received via the communication interface 505. Specifically, the CPU 501 expands a program read via the CD-ROM device 514 or a program received via the communication interface 505 in the RAM 503. Then, the CPU 501 executes each coded instruction in the program expanded in the RAM 503.
- the ROM 502 is a read-only memory that stores programs and data necessary for the operation of the computer 500.
- the RAM 503 is used as a work area when the CPU 501 executes a program. Specifically, the RAM 503 temporarily stores data such as parameters at the time of program execution, for example.
- the hard disk 504 stores programs, data, and the like.
- the communication interface 505 communicates with other computers via a network.
- the bus 506 connects the CPU 501, ROM 502, RAM 503, hard disk 504, communication interface 505, display 511, keyboard 512, mouse 513 and CD-ROM device 514 to one another.
- the image search apparatus can be used as an image search apparatus that allows a user to search for an image that the user wants to interpolate from a large amount of stored image content.
- the image search apparatus can be used as an apparatus for automatically creating a memory album by searching for an image that the user wants to interpolate from a database in which a large amount of image data is stored on the web or the like. .
- the present invention can be used as an image search device for searching image content for interpolating between shooting positions of image content shot by a user from among a large amount of stored image content.
- Image Content 21 Content Attached Information 22 User ID 23 Image ID 24 Image address 25 Shooting date and time 26 Shooting position 27 Tag 28 Subject distance 40 First interpolation target image (Kiyomizu Temple) 41 Second interpolation target image (Ginkakuji) 50 Range information 51 Range information 52 Interpolation range 80 Candidate image 81 Interpolated image (during low-speed movement) 82 Interpolated image (at high speed) DESCRIPTION OF SYMBOLS 100,200 Image search apparatus 101 Image database 102 Input part 103 Interpolation range calculation part 104 Interpolation image candidate acquisition part 105 Movement speed calculation part 106 Interpolation image selection part 107 Output part 130 Large area 131 Middle area 132 Small area 133 Diagonal area 160 Interpolation Image tag 161 Photographed image tag 201 Relevance database 202 Related image extraction unit
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Processing Or Creating Images (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
図1は、本発明の実施の形態1に係る画像検索装置100の特徴的な機能構成を示すブロック図である。図1に示すように、画像検索装置100は、複数の画像の中から補間画像を検索する装置である。ここで、補間画像とは、二つの画像のそれぞれの撮影位置の間を補間するための画像である。具体的には、補間画像とは、二つの画像のそれぞれの撮影位置の間の位置において撮影された画像である。画像検索装置100は、画像データベース101、入力部102、補間範囲算出部103、補間画像候補取得部104、移動速度算出部105、補間画像選択部106、及び出力部107を備える。
次に、本発明の実施の形態2に係る画像検索装置200について説明する。
図11は、本発明の実施の形態2に係る画像検索装置200の特徴的な機能構成を示すブロック図である。図11において、図1と同じ構成要素については同じ符号を付し、説明を省略する。図11に示す画像検索装置200が図1に示す画像検索装置100と相違する点は、画像データベース101に登録されている画像コンテンツのコンテンツ付属情報間の関連性の高さを示す関連度を記憶している関連度データベース201と、補間画像選択部106により選択された補間画像の中からユーザの撮影嗜好と関連性の高い画像を抽出する関連画像抽出部202を有する点である。
21 コンテンツ付属情報
22 ユーザID
23 画像ID
24 画像アドレス
25 撮影日時
26 撮影位置
27 タグ
28 被写体距離
40 第一補間対象画像(清水寺)
41 第二補間対象画像(銀閣寺)
50 範囲情報
51 範囲情報
52 補間範囲
80 候補画像
81 補間画像(低速移動時)
82 補間画像(高速移動時)
100、200 画像検索装置
101 画像データベース
102 入力部
103 補間範囲算出部
104 補間画像候補取得部
105 移動速度算出部
106 補間画像選択部
107 出力部
130 大領域
131 中領域
132 小領域
133 斜線領域
160 補間画像タグ
161 撮影画像タグ
201 関連度データベース
202 関連画像抽出部
Claims (10)
- 少なくとも2つの補間対象画像のそれぞれが撮影された位置の間を補間するための補間画像を複数の画像の中から検索する画像検索装置であって、
第一補間対象画像が撮影された位置である第一撮影位置と第二補間対象画像が撮影された位置である第二撮影位置とが含まれる地理的場所の範囲を補間範囲として算出する補間範囲算出部と、
前記補間範囲算出部により算出された補間範囲内に撮影位置が含まれる画像を候補画像として、画像データベースに格納された前記複数の画像の中から取得する補間画像候補取得部と、
前記第一撮影位置及び前記第二撮影位置の間の移動速度が大きいほど、画像が撮影されたときの撮像装置と被写体との距離である被写体距離が大きい画像を前記候補画像の中から選択する補間画像選択部と、
を備える画像検索装置。 - さらに、前記第一撮影位置と前記第二撮影位置との距離と、前記第一補間対象画像が撮影された日時と前記第二補間対象画像が撮影された日時との差分時間と、を用いて、前記第一撮影位置及び前記第二撮影位置の間の移動速度を算出する移動速度算出部を備え、
前記補間画像選択部は、前記移動速度算出部により算出された移動速度が大きいほど、被写体距離が大きい画像を選択する、
請求項1に記載の画像検索装置。 - さらに、前記補間画像選択部により選択された補間画像と前記第一補間対象画像との関連度である第一関連度と、前記補間画像選択部により選択された補間画像と前記第二補間対象画像との関連度である第二関連度と、を取得し、取得した第一関連度及び第二関連度が示す関連性の高さが所定閾値以上となる前記補間画像を抽出する関連画像抽出部を備え、
前記関連度は、画像に付属する情報である付属情報間の関連性の高さを示す、
請求項1に記載の画像検索装置。 - さらに、被写体に関するキーワード間の関連度を記憶している関連度データベースを備え、
前記関連画像抽出部は、前記関連度データベースを参照することにより、前記補間画像から得られる被写体に関するキーワードと前記第一補間対象画像から得られる被写体に関するキーワードとの関連度を前記第一関連度として取得し、前記補間画像から得られる被写体に関するキーワードと前記第二補間対象画像から得られる被写体に関するキーワードとの関連度を前記第二関連度として取得する、
請求項3に記載の画像検索装置。 - 前記関連度データベースは、所定範囲の場所を示す領域毎に前記関連度を記憶しており、
前記関連画像抽出部は、前記第一補間対象画像の撮影位置が含まれる領域に対応する関連度を前記第一関連度として取得し、前記第二補間対象画像の撮影位置が含まれる領域に対応する関連度を前記第二関連度として取得する、
請求項4に記載の画像検索装置。 - 前記関連画像抽出部は、前記補間画像が撮影されたときの撮影条件と前記第一補間対象画像が撮影されたときの撮影条件との関連度を前記第一関連度として取得し、前記補間画像が撮影されたときの撮影条件と前記第二補間対象画像が撮影されたときの撮影条件との関連度を前記第二関連度として取得する、
請求項3に記載の画像検索装置。 - 前記補間画像選択部は、前記第一撮影位置及び前記第二撮影位置の間の移動速度が予め定められた第一閾値より大きい場合、被写体距離が予め定められた第二閾値以上の前記候補画像を補間画像として選択する、
請求項1に記載の画像検索装置。 - 前記補間範囲算出部は、前記第一補間対象画像が撮影された日時と前記第二補間対象画像が撮影された日時との差分時間が大きいほど場所の範囲が広くなるように、前記補間範囲を算出する、
請求項1に記載の画像検索装置。 - 少なくとも2つの補間対象画像のそれぞれが撮影された位置の間を補間するための補間画像を複数の画像の中からコンピュータが検索する画像検索方法であって、
第一補間対象画像が撮影された位置である第一撮影位置と第二補間対象画像が撮影された位置である第二撮影位置とが含まれる地理的場所の範囲を補間範囲として算出する補間範囲算出ステップと、
前記補間範囲算出ステップにおいて算出された補間範囲内に撮影位置が含まれる画像を候補画像として、画像データベースに格納された前記複数の画像の中から取得する補間画像候補取得ステップと、
前記第一撮影位置及び前記第二撮影位置の間の移動速度が大きいほど、画像が撮影されたときの撮像装置と被写体との距離である被写体距離が大きい画像を前記候補画像の中から選択する補間画像選択ステップと、
を含む画像検索方法。 - 少なくとも2つの補間対象画像のそれぞれが撮影された位置の間を補間するための補間画像を複数の画像の中から検索するプログラムであって、
第一補間対象画像が撮影された位置である第一撮影位置と第二補間対象画像が撮影された位置である第二撮影位置とが含まれる地理的場所の範囲を補間範囲として算出する補間範囲算出ステップと、
前記補間範囲算出ステップにおいて算出された補間範囲内に撮影位置が含まれる画像を候補画像として、画像データベースに格納された前記複数の画像の中から取得する補間画像候補取得ステップと、
前記第一撮影位置及び前記第二撮影位置の間の移動速度が大きいほど、画像が撮影されたときの撮像装置と被写体との距離である被写体距離が大きい画像を前記候補画像の中から選択する補間画像選択ステップと、
をコンピュータに実行させるプログラム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/935,296 US8694515B2 (en) | 2008-12-19 | 2009-11-05 | Image search device and image search method |
CN200980114967XA CN102016909A (zh) | 2008-12-19 | 2009-11-05 | 图像检索装置以及图像检索方法 |
JP2010503293A JP4545828B2 (ja) | 2008-12-19 | 2009-11-05 | 画像検索装置及び画像検索方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008-324700 | 2008-12-19 | ||
JP2008324700 | 2008-12-19 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010070804A1 true WO2010070804A1 (ja) | 2010-06-24 |
Family
ID=42268488
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/005866 WO2010070804A1 (ja) | 2008-12-19 | 2009-11-05 | 画像検索装置及び画像検索方法 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8694515B2 (ja) |
JP (1) | JP4545828B2 (ja) |
CN (1) | CN102016909A (ja) |
WO (1) | WO2010070804A1 (ja) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012103963A (ja) * | 2010-11-11 | 2012-05-31 | Buffalo Inc | コンテンツデータ管理装置、コンテンツデータ管理システム、コンテンツデータ管理プログラム、コンテンツデータ管理方法 |
JP2012244634A (ja) * | 2011-05-24 | 2012-12-10 | Canon Inc | 画像クラスタリング方法 |
JP2013182332A (ja) * | 2012-02-29 | 2013-09-12 | Nikon Corp | 画像表示プログラムおよび画像表示装置 |
JP5934653B2 (ja) * | 2010-11-29 | 2016-06-15 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカPanasonic Intellectual Property Corporation of America | 画像分類装置、画像分類方法、プログラム、記録媒体、集積回路、モデル作成装置 |
JP2019067076A (ja) * | 2017-09-29 | 2019-04-25 | 劉謹銘 | 多重画像源の処理方法 |
CN110377791A (zh) * | 2018-04-11 | 2019-10-25 | 刘谨铭 | 快速影像排序方法 |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080320384A1 (en) * | 2007-06-25 | 2008-12-25 | Ramesh Nagarajan | Automated addition of images to text |
JP5277230B2 (ja) * | 2010-11-11 | 2013-08-28 | 株式会社バッファロー | タグ情報管理装置、タグ情報管理システム、タグ情報管理プログラム、タグ情報管理方法 |
US9424765B2 (en) * | 2011-09-20 | 2016-08-23 | Sony Corporation | Image processing apparatus, image processing method, and program |
KR20140093957A (ko) * | 2011-11-24 | 2014-07-29 | 마이크로소프트 코포레이션 | 상호작용 멀티-모달 이미지 검색 기법 |
JP2015500345A (ja) * | 2011-12-16 | 2015-01-05 | トランスポサジェン・バイオファーマシューティカルズ・インコーポレイテッドTransposagen Biopharmaceuticals, Inc. | 疼痛の治療または予防に使用されるtrpc4調節因子 |
JP5806623B2 (ja) * | 2012-02-07 | 2015-11-10 | オリンパス株式会社 | 撮像装置、撮像方法およびプログラム |
JP6070307B2 (ja) * | 2012-05-21 | 2017-02-01 | 株式会社リコー | パターン抽出装置、画像投影装置、パターン抽出方法およびプログラム |
JP6031924B2 (ja) * | 2012-09-28 | 2016-11-24 | オムロン株式会社 | 画像検索装置、画像検索方法、制御プログラムおよび記録媒体 |
KR101561628B1 (ko) * | 2013-12-30 | 2015-10-20 | 주식회사 케이티 | 스마트 글래스의 영상 정보를 제공하는 검색 장치 및 검색 방법 |
US9972121B2 (en) * | 2014-04-22 | 2018-05-15 | Google Llc | Selecting time-distributed panoramic images for display |
US10171770B2 (en) * | 2014-05-09 | 2019-01-01 | Maxell, Ltd. | Image playback device, display device, and transmission device |
WO2016004330A1 (en) | 2014-07-03 | 2016-01-07 | Oim Squared Inc. | Interactive content generation |
CN104536990B (zh) * | 2014-12-10 | 2018-03-27 | 广东欧珀移动通信有限公司 | 一种图片显示方法及终端 |
CN104866194B (zh) * | 2015-05-21 | 2018-07-13 | 百度在线网络技术(北京)有限公司 | 图片搜索方法及装置 |
CN108431800B (zh) * | 2015-12-17 | 2021-12-14 | 株式会社日立制作所 | 图像处理装置以及图像检索界面的显示方法 |
DE102019200180A1 (de) | 2018-07-25 | 2020-01-30 | Audi Ag | Filterelement für einen Partikelfilter, Abgaspartikelfilter, Verfahren zum Herstellen eines Filterelements sowie Verwendung eines Beschichtungsmaterials |
JP7087844B2 (ja) * | 2018-08-31 | 2022-06-21 | トヨタ自動車株式会社 | 画像処理装置、画像処理システム及び車両 |
CN110659376A (zh) * | 2019-08-14 | 2020-01-07 | 浙江大华技术股份有限公司 | 图片查找方法、装置、计算机设备和存储介质 |
WO2021155000A1 (en) | 2020-01-29 | 2021-08-05 | ReportsNow, Inc. | Systems, methods, and devices for image processing |
US11158031B1 (en) | 2021-05-24 | 2021-10-26 | ReportsNow, Inc. | Systems, methods, and devices for image processing |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001282813A (ja) * | 2000-03-29 | 2001-10-12 | Toshiba Corp | マルチメディアデータ検索方法、インデックス情報提供方法、マルチメディアデータ検索装置、インデックスサーバ及びマルチメディアデータ検索サーバ |
JP2002278993A (ja) * | 2001-03-16 | 2002-09-27 | Nippon Telegr & Teleph Corp <Ntt> | 画像データ登録・再生方法、システム、プログラムおよびその記録媒体 |
JP2004220420A (ja) * | 2003-01-16 | 2004-08-05 | Fuji Photo Film Co Ltd | 画像検索方法および装置並びにプログラム |
JP2005128785A (ja) * | 2003-10-23 | 2005-05-19 | Sony Corp | 記録媒体及び電子アルバム作成装置 |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020047798A1 (en) * | 1999-06-25 | 2002-04-25 | Timothy James Platt | Image acquisition and retrieval system employing position data |
EP1125608A3 (en) * | 2000-01-21 | 2005-03-30 | Sony Computer Entertainment Inc. | Entertainment apparatus, storage medium and object display method |
US6915273B1 (en) * | 2000-05-23 | 2005-07-05 | Eastman Kodak Company | Method for providing customized photo products over a network using images captured from a digital camera |
JP2003099434A (ja) | 2001-09-25 | 2003-04-04 | Sanyo Electric Co Ltd | 電子アルバム装置 |
JP2004139576A (ja) | 2002-09-27 | 2004-05-13 | Fuji Photo Film Co Ltd | アルバム作成装置および方法並びにプログラム |
CN100391248C (zh) * | 2002-09-27 | 2008-05-28 | 富士胶片株式会社 | 相册制作方法及其装置 |
US20060080286A1 (en) | 2004-08-31 | 2006-04-13 | Flashpoint Technology, Inc. | System and method for storing and accessing images based on position data associated therewith |
JP2006185242A (ja) | 2004-12-28 | 2006-07-13 | Fuji Photo Film Co Ltd | 画像合成装置、画像合成プログラム、画像合成プログラム記憶媒体、および画像合成方法 |
EP1899920B1 (en) | 2005-05-12 | 2020-04-01 | FUJIFILM Corporation | Album creating apparatus, album creating method and program |
JP4379409B2 (ja) * | 2005-11-08 | 2009-12-09 | ソニー株式会社 | 撮像装置、および情報処理方法、並びにコンピュータ・プログラム |
JP4706844B2 (ja) | 2006-02-08 | 2011-06-22 | 日本電気株式会社 | アルバム作成システム、サーバー、アルバム作成方法およびプログラム |
JP2007323315A (ja) | 2006-05-31 | 2007-12-13 | Nippon Telegr & Teleph Corp <Ntt> | 協調フィルタリング方法、協調フィルタリング装置、および協調フィルタリングプログラムならびにそのプログラムを記録した記録媒体 |
-
2009
- 2009-11-05 US US12/935,296 patent/US8694515B2/en active Active
- 2009-11-05 JP JP2010503293A patent/JP4545828B2/ja active Active
- 2009-11-05 WO PCT/JP2009/005866 patent/WO2010070804A1/ja active Application Filing
- 2009-11-05 CN CN200980114967XA patent/CN102016909A/zh active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001282813A (ja) * | 2000-03-29 | 2001-10-12 | Toshiba Corp | マルチメディアデータ検索方法、インデックス情報提供方法、マルチメディアデータ検索装置、インデックスサーバ及びマルチメディアデータ検索サーバ |
JP2002278993A (ja) * | 2001-03-16 | 2002-09-27 | Nippon Telegr & Teleph Corp <Ntt> | 画像データ登録・再生方法、システム、プログラムおよびその記録媒体 |
JP2004220420A (ja) * | 2003-01-16 | 2004-08-05 | Fuji Photo Film Co Ltd | 画像検索方法および装置並びにプログラム |
JP2005128785A (ja) * | 2003-10-23 | 2005-05-19 | Sony Corp | 記録媒体及び電子アルバム作成装置 |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012103963A (ja) * | 2010-11-11 | 2012-05-31 | Buffalo Inc | コンテンツデータ管理装置、コンテンツデータ管理システム、コンテンツデータ管理プログラム、コンテンツデータ管理方法 |
JP5934653B2 (ja) * | 2010-11-29 | 2016-06-15 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカPanasonic Intellectual Property Corporation of America | 画像分類装置、画像分類方法、プログラム、記録媒体、集積回路、モデル作成装置 |
JP2012244634A (ja) * | 2011-05-24 | 2012-12-10 | Canon Inc | 画像クラスタリング方法 |
JP2013182332A (ja) * | 2012-02-29 | 2013-09-12 | Nikon Corp | 画像表示プログラムおよび画像表示装置 |
JP2019067076A (ja) * | 2017-09-29 | 2019-04-25 | 劉謹銘 | 多重画像源の処理方法 |
CN110377791A (zh) * | 2018-04-11 | 2019-10-25 | 刘谨铭 | 快速影像排序方法 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2010070804A1 (ja) | 2012-05-24 |
US20110022634A1 (en) | 2011-01-27 |
US8694515B2 (en) | 2014-04-08 |
JP4545828B2 (ja) | 2010-09-15 |
CN102016909A (zh) | 2011-04-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4545828B2 (ja) | 画像検索装置及び画像検索方法 | |
JP4475547B2 (ja) | 画像検索装置、画像検索方法及び画像推定装置 | |
US8254727B2 (en) | Method and apparatus for providing picture file | |
JP4796435B2 (ja) | 画像ビューア | |
US8094974B2 (en) | Picture data management apparatus and picture data management method | |
US9286624B2 (en) | System and method of displaying annotations on geographic object surfaces | |
JP5134664B2 (ja) | アノテーション装置 | |
KR100882025B1 (ko) | 지피에스 이미지 검색 방법, 지리적 위치 서비스 기반의지피에스 이미지 검색 방법, 블로그 서비스 기반의지피에스 이미지 검색 방법 및 지역별 블로그 서비스 제공방법 | |
JP4457988B2 (ja) | 画像管理装置、画像管理方法およびコンピュータプログラム | |
KR20090047487A (ko) | 그래픽 사용자 인터페이스, 컴퓨터 판독가능 매체, 및 시선방향 제시 방법 | |
JP2005174060A (ja) | 画像分類装置ならびに画像分類方法およびプログラム | |
JP2011078008A (ja) | コンテンツ共有装置、コンテンツ編集装置、コンテンツ共有プログラム、およびコンテンツ編集プログラム | |
JP6210807B2 (ja) | 表示制御装置及び表示制御装置の制御方法 | |
US20110202582A1 (en) | Method and apparatus for managing tag of multimedia content | |
JP2004221647A (ja) | 旅行アルバム作成方法および装置並びにプログラム | |
JP6179315B2 (ja) | 情報処理装置、画像処理システム、情報処理装置における画像処理方法及びプログラム | |
JP2013126107A (ja) | デジタル放送受信装置 | |
KR100563085B1 (ko) | 수치지도와 사진영상의 복합 디스플레이 방법 | |
US20150379040A1 (en) | Generating automated tours of geographic-location related features | |
US8571357B2 (en) | Image data management apparatus, method and program | |
JP5708868B1 (ja) | プログラム、情報処理装置及び方法 | |
JP2008242682A (ja) | メタ情報自動付与システム、メタ情報自動付与方法、及びメタ情報自動付与プログラム | |
JP5440197B2 (ja) | 撮影対象スポットデータベースの構築装置、構築システム、及び構築方法 | |
JP5058201B2 (ja) | 情報管理システム及び情報管理方法 | |
JP6436191B2 (ja) | サーバ装置、サーバ装置を用いた情報処理方法及びプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200980114967.X Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010503293 Country of ref document: JP |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09833114 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12935296 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 09833114 Country of ref document: EP Kind code of ref document: A1 |