WO2006036026A1 - 画像補正装置および方法、ならびに画像補正プログラム - Google Patents
画像補正装置および方法、ならびに画像補正プログラム Download PDFInfo
- Publication number
- WO2006036026A1 WO2006036026A1 PCT/JP2005/018576 JP2005018576W WO2006036026A1 WO 2006036026 A1 WO2006036026 A1 WO 2006036026A1 JP 2005018576 W JP2005018576 W JP 2005018576W WO 2006036026 A1 WO2006036026 A1 WO 2006036026A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- subject
- subject type
- input
- calculated
- correction value
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000003702 image correction Methods 0.000 title claims abstract description 40
- 238000004364 calculation method Methods 0.000 claims abstract description 131
- 238000012935 Averaging Methods 0.000 claims abstract 4
- 238000012937 correction Methods 0.000 claims description 103
- 238000012545 processing Methods 0.000 claims description 67
- 238000001514 detection method Methods 0.000 claims description 58
- 230000006870 function Effects 0.000 description 46
- 238000006243 chemical reaction Methods 0.000 description 39
- 238000010586 diagram Methods 0.000 description 13
- 238000000605 extraction Methods 0.000 description 13
- 230000015572 biosynthetic process Effects 0.000 description 7
- 238000003786 synthesis reaction Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 239000003086 colorant Substances 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000012447 hatching Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/46—Colour picture communication systems
- H04N1/56—Processing of colour picture signals
- H04N1/60—Colour correction or control
- H04N1/6027—Correction or control of colour gradation or colour contrast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/46—Colour picture communication systems
- H04N1/56—Processing of colour picture signals
- H04N1/60—Colour correction or control
- H04N1/62—Retouching, i.e. modification of isolated colours only or in isolated picture areas only
- H04N1/628—Memory colours, e.g. skin or sky
Definitions
- the present invention relates to an image correction apparatus, a method, and an image correction program.
- Japanese Laid-Open Patent Publication No. 2000-0 1960 6900 discloses that a face image area included in an input image is detected, and the average density of the detected face image falls within a predetermined target density range. A method of image conversion (image correction) so as to enter is disclosed.
- An object of the present invention is to enable appropriate image correction for an image having a plurality of subject types.
- Another object of the present invention is to correct the color (brightness) of a desired image as an entire image for an image having a plurality of subject types.
- the image correction apparatus is an importance input means for accepting an input of importance for each of a plurality of object types, and based on the given image data, Predetermined Based on the image data, each of the subject type detection means for detecting which one of the plurality of subject types is included, and the subject type detected by the subject type detection means.
- Input feature value calculating means for calculating the input feature value by the calculation process determined for each subject type, the input feature value calculated by the input feature value calculating means, and the target feature value determined for each subject type Based on the above, the correction value calculation means for calculating the correction value for each of the subject types, the correction value calculated by the correction value calculation means, the importance received for each of the subject types by the importance input means Applied correction value calculation means for calculating the weighted average applied correction value using, and the above applied correction value calculation An image correction means for correcting the image data based on the applied correction value calculated by the means is provided.
- a plurality of subject types to be detected are determined in advance.
- This subject type is the type of subject that can be identified based on the image data (which may be referred to as the shooting scene type) and should be corrected differently in image correction (one that has performed different correction processing). (It is empirically known that the image after correction is appropriate.)
- Multiple types of subjects are set. For example, “Blue Sky”, “Human Face”, “Underwater”, “High Saturation”, “Evening Scenery”, “High Key”, “Night Scenery”, etc. can be set as multiple subject types.
- the process of detecting which of the predetermined object types is included in the image represented by the given image data is performed in advance for each of the object types. This can be done using the set identification data. For example, multiple image data (blue sky sample images) representing the subject type “blue sky” are prepared, and appropriate identification conditions (feature type and range (hue range, luminance range, (Saturation range, etc.)) H) If one or more feature values are extracted from the given image data, and the extracted one or more feature values meet the identification conditions obtained from the learning process, the image represented by the given image data Can be determined to include the subject type “blue sky”.
- an input feature value determined for each subject type is calculated based on the image data.
- the input feature value is a characteristic value (quantity) specific to the subject type for each subject type, and the calculation process is different for each subject type. For example, if the subject type is “blue sky”, the average value of the RGB values of the blue sky image pixels (blue sky pixels) contained in the image represented by the given image data is the subject type “blue sky”. The input feature value for.
- the input feature value calculation method may be determined for each subject type in accordance with the characteristics of the image represented by a plurality of subject types determined in advance.
- the correction value is a value for correcting the input feature value to the target feature value.
- the calculated correction values include various correction values that can be used for image correction, for example, density (luminance) gradation correction values, saturation correction values, and other correction values.
- the ratio (gain) between the input feature value and the target feature value is also included in the correction value obtained by this invention.
- the applied correction value applied to the image data is detected.
- the correction values obtained for each subject type (that is, included in the image represented by the image data to be processed) are weighted averages based on the importance of each subject type.
- the correction value obtained for each subject type corrects the input feature value based on the given image data to the target feature value. Since the multiple correction values calculated for each of the detected multiple types of subjects are weighted and averaged based on the received importance, the applied correction value applied to the image data is
- the input feature values of multiple subject types are generally brought closer to the target feature value corresponding to the input feature value.
- the color (or brightness) of the given image can be corrected to the color (or brightness) of the desired image.
- the input feature cannot be made to completely match the target feature when attention is paid to each of the subject types.
- any of the plurality of subject types can be used.
- the importance given to each type of subject is used to calculate the weighted average. It is possible to obtain an applied correction value that largely reflects the correction value for the type of subject that the user regards as important. This makes it possible to obtain an applied correction value that realizes image correction according to user preferences.
- the correction value and the applied correction value may be unique for each color image data of multiple colors (for example, RGB), or may be common to each of the color image data of multiple colors. .
- how accurately each of the images representing the plurality of subject types is included in the image represented by the image data.
- subject-specific existence probability calculating means for calculating the existence probability representing the above for each of the plurality of subject types.
- the applied correction value calculation means uses the first importance and the second importance obtained in accordance with the existence probability for each of the subject types calculated by the existence probability calculation means.
- the second importance according to the existence probability is used for the weighted average in addition to the first importance input, the content of the given image data (existence of subject type) Image correction according to the accuracy of the image.
- subject area area calculating means for calculating the area of the image representing the subject type in the image represented by the image data for each detected subject type.
- the correction value calculation means uses the first importance degree and the third importance degree obtained according to the area for each of the subject types calculated by the area calculation means, and uses the correction value calculation means.
- the correction values for each calculated subject type are weighted averaged. Image correction according to the area of the subject type in the image represented by the given image data is realized according to the user's preference.
- the image correction apparatus includes a parameter input unit that accepts an input of a parameter for changing the target feature amount, and a target that corrects the target feature amount based on the parameter input by the parameter input unit.
- a feature amount correcting means is further provided. Since the target feature value is modified according to the user's request, the image represented by the image data after correction can be made according to the user's preference.
- Input parameters may be accepted by numerical values, or abstract words (quality-oriented, productivity-oriented) Etc.) may be accepted.
- the image correction apparatus is represented by importance input means for receiving input of importance for each of a plurality of predetermined object types, based on the given image data. For each of the subject types detected by the subject type detecting means and the subject type detecting means for detecting which subject type is included among a plurality of subject types in the image, Based on the input feature value calculation means for calculating the input feature value by the calculation process determined for each subject type, and the input feature value for each subject type calculated by the input feature value calculation means is determined for each subject type.
- Condition judging means for judging whether or not a predetermined predetermined condition is satisfied, and the above condition judging means It is determined that the input feature quantity does not satisfy the predetermined condition.
- a correction value calculation means for calculating a correction value for the subject type so that the input feature quantity for the subject type satisfies the predetermined condition;
- the correction value for each of the subject types calculated by the correction value calculation means is calculated by applying a weighted average to the correction values for each of the subject types by the importance input means.
- an image correction means for correcting the image data based on the applied correction value calculated by the applied correction value calculation means.
- the correction value is calculated based on whether or not the input feature value for each detected subject type satisfies a predetermined condition. If the input feature value for one detected subject type satisfies the specified condition, it is not necessary to correct the input feature value for that one subject type. If the input feature value for one detected subject type does not satisfy the above specified condition, A correction value that satisfies the predetermined condition is calculated.
- the above predetermined conditions can also be determined by a learning process that is performed in advance.
- the applied correction value applied to the image data is an input feature that does not satisfy the predetermined condition.
- the image data is corrected so that the input feature amount approaches the predetermined condition, and image correction according to the user's preference is realized.
- the color (or brightness) of the given image can be corrected to the desired color (or brightness) of the image.
- parameter input means for receiving an input of a parameter for changing the predetermined condition
- condition correction means for correcting the predetermined condition based on the parameter input by the parameter input means are further provided.
- the present invention also provides an image correction method and an image correction program corresponding to the first and second image correction apparatuses described above.
- Figure 1 is a block diagram showing the hardware configuration of the image correction system.
- FIG. 2 is a block diagram showing the electrical configuration of the image correction apparatus in the first embodiment.
- Fig. 3 is a functional block diagram of the subject type detection circuit, the input feature quantity calculation circuit for each subject type, and the target feature color conversion condition calculation circuit for each subject type.
- Figure 4 shows the subject type detection circuit for the subject type “blue sky”, the feature amount calculation circuit for each subject type, and the target feature no color conversion condition calculation for each subject type. It is a flowchart which shows the flow of each process of an output circuit.
- Figure 5 shows the contents of the detection condition table.
- Figure 6 shows the blue sky extraction conditions in a graph.
- Figure 7 shows the contents of the feature quantity calculation condition table.
- Figure 8 shows the boundary function for the subject type “blue sky”.
- Figure 9 shows the boundary function for the subject type “blue sky”.
- Figure 10 shows the correction function
- Fig. 1 1a shows the correction function for the subject type "Night scene” (shadow), the subject type “Night scene” (highlight) and the subject type “Face”, and Fig. 1 1 b shows the correction determined by the applied gain. Each function is shown.
- Figure 12 shows an example of a screen for inputting user requests.
- FIG. 13 is a block diagram showing the electrical configuration of the image correction apparatus according to a modification of the first embodiment.
- FIG. 14 is a block diagram showing the electrical configuration of the image correction apparatus in the second embodiment.
- Fig. 15 is a functional block diagram of the existence probability calculation circuit for each subject type, input feature value calculation circuit for each subject type, and target feature value Z color conversion condition calculation circuit for each object type.
- Figure 16 shows the contents of the existence probability calculation table.
- Figure 17 shows the flow of creating the existence probability calculation table.
- Fig. 18 is a graph showing the relationship between the existence probability and the importance corresponding to the existence probability.
- Figure 1 is a block diagram showing the hardware configuration of the image processing system. is there.
- Fig. 2 is a block diagram showing the electrical configuration of the image processing device 1 included in the image processing system together with the storage device 4.
- the image processing system includes an image processing device 1, an input device 2 (keyboard, mouse, etc.) connected to the image processing device 1, a display device 3 (CRT display, liquid crystal display, etc.), a storage device 4 (hard disk, etc.) And printer 5.
- an input device 2 keyboard, mouse, etc.
- a display device 3 CRT display, liquid crystal display, etc.
- a storage device 4 hard disk, etc.
- Image data is stored in the storage device 4 connected to the image processing device 1.
- the specified image data is read from the storage device 4 using the input device 2 and is input to the image processing device 1 through the input interface (not shown) (hereinafter, input to the image processing device 1).
- the image data is called “input image data”, and the image represented by the input image data is called “input image”.
- image data recorded on a CD-ROM, DVD-ROM, memory card or other recording medium may be input to the image processing device 1.
- a CD-ROM drive, a DVD-ROM drive, etc. are connected to the image processing device 1 of the image processing system.
- the image data transmitted through the network may be input to the image processing device 1.
- the image processing device 1 of the image processing system is connected to a transmission / reception device (such as a modem) that transmits and receives image data over the network.
- the image processing apparatus 1 includes a subject type detection circuit ⁇ , an input feature amount calculation circuit 12 for each subject type, a target feature amount for each subject type, a color conversion condition calculation circuit 13, a color conversion condition synthesis circuit 14, and an image correction circuit 15. .
- the subject type detection circuit 11 is a circuit that detects (identifies) the subject type included in the input image represented by the input image data.
- the input image data input to the image processing device 1 is first given to the subject type detection circuit 11.
- the subject type detection circuit 11 includes the subject type detection circuit in the input image. Detects whether the subject types “blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key”, and “night scene” are included.
- a detection condition table 21 is connected to the subject type detection circuit 11, and based on the data stored in the detection condition table 21, any type of subject is included in the input image in the subject type detection circuit 11. Is detected. Details of the processing of the subject type detection circuit 11 and the contents of the detection condition table 21 will be described later.
- the subject type input feature quantity calculation circuit 12 calculates a predetermined feature quantity (eg, average RGB value, RGB variance value, etc.) specific to each subject type for the input image represented by the input image data. Circuit.
- the feature value calculated by the input feature value calculation circuit 12 for each subject type using the input image data is called the “input feature value”.
- a feature quantity calculation table 22 is connected to the input feature quantity calculation circuit 12 for each subject type.
- the input feature quantity calculation circuit 12 for each subject type refers to this feature quantity calculation table 22 and calculates a unique input feature quantity for each subject type. Details of the processing of the input feature quantity calculation circuit 12 for each subject type and the contents of the feature quantity calculation table 22 will be described later.
- Target feature value by subject type The color conversion condition calculation circuit 13 is a circuit that calculates a target feature value (eg, R GB value) for each subject type.
- the target feature value is a parameter (hereinafter referred to as input parameter 2) input using the input device 2 by the operator (user) of the input feature value image processing system 12 calculated by the input feature value calculation circuit 12 for each subject type. , And input parameters).
- the target feature / color conversion condition calculation circuit 13 for each subject type calculates (determines) the target feature amount corresponding to the input feature amount, thereby determining each of the subject types detected by the subject type detection circuit 11. It can be said that the circuit calculates the gain of.
- Target feature value by subject type Z color conversion condition calculation circuit 13 is connected to a boundary function table 23 in which data representing the boundary function defined for each subject type is stored.
- the target feature value (gain) is , Calculated (determined) based on the boundary function table 23. Details of processing function boundary function table 23 of target feature amount Z color conversion condition calculation circuit 13 by subject type will be described later.
- the color conversion condition composition circuit 14 obtains the gain (R component, B component, and G component gain for each of the detected object types obtained in the subject type target feature value / color conversion condition calculation circuit 13. Coefficient) and the importance W i for each type of subject input by the operator (user) of the image processing system using the input device 2, and a gain (R, G, B) Each of these is a circuit for calculating Details of the processing of the color conversion condition synthesis circuit 14 will be described later.
- the image correction circuit 15 is a circuit that corrects (converts) each R GB value of the pixels constituting the input image based on the gain (applied gain coefficient) calculated by the color conversion condition synthesis circuit 14.
- FIG. 3 is a functional block diagram showing the functions of the subject type detection circuit 11, the input feature value calculation circuit 12 for each subject type, and the target feature color conversion condition calculation circuit 13 for each subject type.
- subject type means seven types of subjects: “blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key J and ⁇ night scene” as described above. (Shooting scene) is a general term that refers to.
- subject type detection circuit 11 the image data to be processed read from the storage device 4 Based on the above, it is detected whether these seven types of subject types are included in the image represented by the image data (function blocks 11a to 11g).
- the input feature value calculation circuit 12 for each subject type inputs for each of the seven subject types (“blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key”, and “night scene”). It is configured so that feature quantities can be calculated (functional blocks 12a to 12 g).
- the value used as the input feature varies depending on the subject type.
- the definition of what value to use as the input feature is stored in the feature calculation table 22 (described later).
- Target feature value (gain) corresponding to the input feature value for each subject type calculated by the subject type input feature value calculation circuit 12 Calculated (determined) by the target feature type target feature value Z color conversion condition calculation circuit 13 (Functional blocks 13a to 13g).
- the gain (Gri, Ggi, Gbi) calculated for each detected subject type is given from the target feature type color conversion condition calculation circuit 13 to the color conversion condition synthesis circuit 14 (see Fig. 2). Here, one gain (applicable gain) is calculated.
- subject type detection circuit 11 Focusing on the subject type “blue sky”, which is one of the subject types that can be detected by the image processing apparatus 1 in the first embodiment, subject type detection circuit 11, subject type input feature quantity calculation circuit 12 and target type specific target
- subject type detection circuit 11 Processing of the feature quantity / color conversion condition calculation circuit 13 (processing of the function blocks lla, 12a, and 13a) is explained.
- Figure 4 shows the subject type detection circuit 11 for the subject type “blue sky” 11, the input feature value calculation circuit 12 for each subject type, and the target feature value color conversion condition calculation circuit 13 for each subject type (function block) 11 a, 12a and 13 It is a flowchart which shows the flow of (processing of a).
- the image data to be processed (input image data) is read from the storage device 4.
- the read input image data is given to the subject type detection circuit 11 (step 31).
- the subject type detection circuit 11 refers to the detection condition table 21.
- FIG. 5 shows the contents of the detection condition table 21.
- the detection condition table 21 stores data representing pixel extraction conditions and data representing detection conditions for each subject type.
- the blue sky detection function 11a uses a set of pixel extraction conditions and detection conditions for the subject type “blue sky”.
- the pixel extraction conditions stored in the detection condition table 21 define the conditions (definitions) for extracting the pixels representing each of the above-mentioned subject types from among the pixels composing the input image.
- the data (blue sky pixel extraction condition) representing the pixel extraction condition defined for the subject type “blue sky” will be described.
- the blue sky pixel is a pixel that satisfies the following equation among the pixels that make up the input image.
- Luminance Y is calculated based on the R GB value of each pixel according to the following equation.
- the saturation S is calculated based on the RGB value of each pixel by the following equation.
- S SQRT (C r 2 + C b 2 )
- Equation 3 C r and C b in Equation 3 are calculated by the following equations
- Equation 1 Y raax is the maximum luminance of the entire input image, and S raax is the maximum saturation of the entire input image.
- Figure 6 shows the blue sky pixel extraction condition (blue sky pixel extraction condition. D at) (Equation 1), with the vertical axis representing luminance Y and the horizontal axis representing saturation S. Pixels that belong to the range defined by Equation 1, that is, pixels that belong to the range indicated by hatching in Fig. 6, are extracted pixels for the subject type “blue sky” among the pixels that constitute the input image (hereinafter referred to as blue sky). This is called a pixel). Based on the blue sky pixel extraction conditions stored in the detection condition table 21, blue sky pixels are extracted from the pixels constituting the input image (step 32).
- the number of blue sky pixels extracted based on the blue sky pixel extraction conditions in the detection condition table 21 is calculated, and the ratio of the calculated blue sky pixels to the total number of pixels is calculated (step 33).
- the ratio of the number of blue sky pixels to the total number of pixels, that is, the ratio of the blue sky image in the input image (area ratio) force is judged whether it is greater than a predetermined value (step 34). For this judgment, the detection condition for the object type “blue sky” listed in the detection condition table 21 is referred to.
- the detection condition for the subject type “blue sky” is the blue sky area ratio A sky ⁇ 0.0 1.
- the number of pixels in the blue sky (area ratio) force It will be determined whether this condition is met. If the blue sky area ratio A sky is smaller than the predetermined value (blue sky area ratio A sky 0 0 1), the input image represented by the input image data is treated as an image that does not include the subject type “blue sky” (NO in step 34, step 3 5). In this case, the data indicating that the blue sky image is not included in the input image is sent from the subject type detection circuit 11 to the input feature amount calculation circuit 12 for each subject type and the target feature amount / color conversion for each subject type.
- condition calculation circuit 13 It is given to the condition calculation circuit 13. In this case, the subject type input feature quantity calculation circuit 12 and the subject type target feature quantity Z color conversion condition calculation circuit 13 do not process the subject type “blue sky”. In other words, the target feature (gain) for the subject type “blue sky” is not calculated.
- blue sky area ratio A sky is greater than or equal to the specified value (blue sky area ratio A sky ⁇ 0. 0 1) (YES in step 34), the input image is treated as an image including the subject type “blue sky”. .
- the pixel extraction condition and the detection condition can be set by the learning process. For example, if you prepare multiple images (blue sky sample images) that should be judged as the subject type “blue sky J”, the feature values and their ranges (for example, hue range, (Luminance range, saturation range, area ratio, etc.) Pixel extraction conditions and detection conditions are determined based on the derived features and their ranges Of course, the conditions obtained by the learning process (pixel extraction) (Conditions and detection conditions) may be corrected manually.
- the following processing is performed in the input feature amount calculation circuit 12 for each subject type and the target feature amount color conversion condition calculation circuit 13 for each subject type.
- Input image data is given to the input feature quantity calculation circuit 12 by subject type.
- the subject type input feature quantity calculation circuit 12 refers to the feature quantity calculation table 22 and calculates the feature quantity defined for the subject type “blue sky” ( Blue sky input feature value calculation function 12a processing).
- Figure 7 shows the contents of the feature quantity calculation table 22.
- the feature quantity calculation table 22 stores definitions related to feature quantities to be calculated for each subject type.
- the input feature amount calculation circuit 12 for each subject type calculates the feature amount of each subject type based on the input image data according to the definition stored in the feature amount calculation table 22.
- the feature quantity calculation table 22 is defined to calculate the average value of R, G, and B colors of blue sky pixels as the feature quantity calculation condition. ing.
- the input feature value calculation circuit 12 for each subject type is an average value of R, G, and B colors for the blue sky pixels in the image represented by the input image data (the blue sky pixels detected by the subject type detection circuit 11 are used).
- (Rave, Gave, Bave) is calculated (step 36).
- the calculated R GB average value (R ave, G ave, B ave) of the blue sky pixels is input to the target feature-specific target feature Z color conversion condition calculation circuit 13 as the input feature value.
- the target feature-specific target feature / color conversion condition calculation circuit 13 is determined for each input feature calculated by the subject-type input feature calculation circuit 12 and for each subject type stored in the boundary function table 23.
- the target feature (gain) is calculated for each subject type based on the boundary function data. The following processing is performed.
- Saturation S and luminance Y are calculated from the input features (R ave, G ave, B ave) for the subject type “blue sky” (step 37).
- Saturation S is calculated based on Equations 3 and 4 above.
- Luminance ⁇ is calculated based on Equation 2 above.
- the target feature-by-subject type amber color conversion condition calculation circuit 13 uses the subject type “blue” stored in the boundary function table 23. Based on the boundary function data for “Sky” and the input feature values (Rave, Gave, Bave) for the subject type “Blue Sky”, the target feature value is calculated, and Calculate the gain.
- Figure 8 shows the boundary function represented by the boundary function data for the subject type “blue sky” on a graph with the vertical axis representing luminance ⁇ and the horizontal axis representing saturation S.
- the boundary function F sky (s) for the subject type “blue sky” was experimentally obtained for the subject type “blue sky”.
- the plot point according to the saturation S and luminance Y obtained from the input features (Rave, Gave, Bave) is located at the position of the point A (Sa, Ya) in the function graph shown in Fig. 8.
- the plot point A (S a, Y a) is located in the outer region defined by the boundary function F sky (S).
- the subject type “blue sky” has sufficient brightness even if it is not corrected. Therefore, it is judged that there is no need for special correction processing for the subject type “blue sky”.
- the target feature value is equal to the input feature value.
- the gain G sky for the subject type “blue sky” is “1” (NO in step 38).
- the plot point according to the saturation S and luminance Y obtained from the input features (Rave, Gave, Bave) is the position of the point B (Sb, Yb) in the function graph shown in Fig. 8.
- Plot point B (S b, Yb) is located in the inner region defined by the boundary function F sky (S).
- the input image containing the subject type “blue sky” will be dark.
- the boundary function F sky (s) is used to calculate the gain G sky for the subject type “blue sky”. It can be said that the boundary function F sky (s) represents a condition that specifies whether or not the object type “blue sky” should be corrected. It can also be called a function that specifies the correction value (gain) if it should be corrected.
- the subject type “blue sky” is the common gain for each of the gains G sky, R, G, and B obtained by Equation 5 above.
- the target feature value for the subject type “blue sky” is (G sky * Rave, G sky-G ave, G sky ⁇ B ave).
- the target feature value may be determined in advance for each subject type.
- the correction value is calculated using the target feature value that has been determined in advance and the calculated input feature value.
- FIG. 9 shows how the boundary function F sky (S) for the subject type “blue sky” is changed.
- S boundary function F sky
- data is input from the input device 2 to “increase the brightness of the subject type“ blue sky ””.
- the degree of brightness correction is input by sliding the slide bar displayed on the display screen of the display device 3 with a mouse or the like.
- the boundary function F sky force for the blue sky stored in the boundary function table 23 is shown by the alternate long and short dash line in FIG. To be changed. Saturation S and luminance Y obtained from the input features are larger than the boundary shown by the dashed line. If it is located inside, the gain Gsky (> 1) is calculated in which the plot point is corrected on the boundary function indicated by the alternate long and short dash line. Conversely, if the operator tries to finish the subject type “blue sky” darker than the standard,
- the input device 2 is used to input “decrease the brightness of the subject type“ blue sky ””.
- the boundary function Fsky is changed to the one shown by the two-dot chain line.
- Figure 10 shows the correction function (gain line) to which the gain Gsky is applied.
- the correction function when the chain line is gain ⁇ 1 ” is shown, and the correction function with the gain Gsky (> 1) is shown as the solid line.
- the gain (target feature) may be calculated separately for each of R, G, and B, taking into account the white balance.
- the ratio of R, G, and B is the predetermined ratio).
- the gain (Gri, Ggi, Gbi) i means each detected object type for correcting the input feature-for each of R, G, B to the target feature ) Force Calculated for each detected subject type.
- the color conversion condition composition circuit 14 is a target feature value by subject type Z color conversion condition calculation circuit 13 for each subject type given by the subject (Gri, Ggi, Gbi) and for each subject type input from the input device 2.
- a circuit that uses the importance Wi to calculate a new gain (“applied gain”) for which the input feature value for each subject type approaches the target feature value for each subject type as a whole. is there.
- Color conversion condition composition circuit 14 calculates target feature color conversion condition by subject type Based on the gain for each subject type (Gri, Ggi, Gbi) given from the circuit 13 and the importance Wi for each subject type input from the input device 2 by the operator, the applied gain (Gr, G g, Gb) is calculated.
- Gg ⁇ (Wi-Ggi) / ⁇ W i
- Gb ⁇ (WiGbi) / ⁇ W i
- Figure 1a is a graph showing the input / output relationship (gain function) for the R component of the input image that includes the subject type “night view” and the subject type “face”. These input / output relationships are expressed by the gain for the R component obtained for each of the subject types “night view” and “face” calculated by the target feature type target feature / color conversion condition calculation circuit 13. For the subject night scene, the input feature value and the target feature value are obtained for each of the shadow and highlight (see Fig. 7). In the graph shown in Fig.
- the input feature value and target feature value for the subject type “night view” (shadow) are black circles
- the input feature value for the subject type “night view” (highlight) The target feature amount is indicated by a white circle.
- the input features and target features for the subject type “face” are indicated by triangles.
- the input / output relationship (gain function) force determined by applying gain Gr for the R component obtained by applying Equation 6 above is shown in Fig. 1 lb.
- the weighted average of the gains (Gri, Ggi, Gbi) obtained for each subject type is taken, and the applied gain (Gr, Gg, Gb) to be applied to the input image data )
- Each input feature value for each detected subject type is Applicable gains (G r, G g, G b) that approximate the corresponding target feature as a whole are obtained.
- the weighted average is based on the importance W i input by the operator, the applied gain more strongly reflects the gain for the subject type that the user places importance on.
- the applied gain (G r, G g, G b) calculated by the color conversion condition synthesis circuit 14 is given to the image correction circuit 15.
- the image correction circuit 15 corrects the RGB value of each pixel constituting the image represented by the image data read from the storage device 4 using the applied gain (G r, G g, G b). .
- the corrected image data is output from the image processing device 1 and given to the display device 3 or the printer 5.
- Figure 12 shows an example of a screen for inputting user's requests displayed on the display screen of display device 3.
- Input device 2 is used to select one of “for printing and productivity”, “for printing and quality”, “for photo and productivity” or “for photo and quality”.
- the processing in the image processing device 1 changes according to the selection (request) by the user.
- the image processing device 1 When the user selects “For printing one quality priority” or “For photograph one quality priority J”, the user selects “For printing one productivity priority” or “For photo one productivity priority”. Processing in the image processor 1 is performed so that high-quality correction results can be obtained. For example, the area ratio (No. 4) used to determine whether the object type “face” is included in the input image. Is corrected to a small value, so that even if a face image with a smaller area is included, it is determined that the subject type “face” is included in the input image. In other words, even small-sized faces can be detected. Conversely, when “For printing one productivity” or “For photo one productivity” is selected, the image processing device 1 performs processing with an emphasis on processing speed.
- the gain function is compared to the case where “print-product-oriented” or “print-quality-oriented” is selected. Increase the slope of (Fig. Lib)
- FIG. 13 shows another modification of the first embodiment (Modification 2), and is a block diagram showing the electrical configuration of the image processing apparatus 1 corresponding to FIG. 2 described above together with the storage device 4. It is.
- subject information is input from the input device 2 and provided to the subject type detection circuit 11.
- the subject information input from the input device 2 is, for example, information on the presence / absence of presence in the input image for each subject type, data indicating the location of the subject type in the input image, for example, the input image to be processed It is assumed that the subject type “face” does not exist in advance (ascertained by the operator of the image processing system).
- the subject type detection circuit 11 detects the subject type “face”
- the subject feature type input feature value calculation circuit 12 inputs the subject feature type “face”
- the object type detection circuit 1 1 If the object type information given from the input device 2 to the object type detection circuit 1 1 includes data indicating that the object type “ ⁇ J does not exist, the object type detection circuit 1 1 The process for the subject type “ ⁇ ” in the subject type input feature quantity calculation circuit 12 and the subject type target feature quantity Z color conversion condition calculation circuit 13 is not executed. As a result, the processing speed of the image processing device 1 can be increased, and the accuracy of the finally obtained applied gain can be increased.
- the existence area in the input image of the subject type “face” may be included in the subject information input from the input device 2. Even in this case, the object type detection circuit 11 does not execute the detection process for the object type “face”.
- the input feature value calculation circuit 12 for each subject type uses the pixels included in the existence area (designated area) for the subject type “face” given by the input device 2 as the pixel ( (Face pixels) are extracted from the input image (step 32 in FIG. 4 is skipped).
- the input image data contains tag information, and this tag information contains information about the presence or absence of the subject type (for example, the presence or absence of the subject type “face”)
- the input device 2 The image processing apparatus 1 may be controlled as described above based on the tag information instead of the subject information input from. Second embodiment
- FIG. 14 shows the electrical configuration of the image processing apparatus 1A in the second embodiment.
- 3 is a block diagram shown together with a storage device 4.
- FIG. It differs from the image processing device 1 in the first embodiment (Fig. 2) in that a subject type existence probability calculation circuit 51 is provided instead of the subject type detection circuit 11. Also, the existence probability calculation table 52 is provided in place of the detection condition table 21, which is different from the image processing apparatus 1 in the first embodiment. '
- Fig. 15 is a functional block diagram showing the functions of the existence probability calculation circuit 51 by subject type, the input feature amount calculation circuit 12 by subject type, and the target feature amount Z color conversion condition calculation circuit 13 by subject type. .
- the existence probability detection circuit 51 for each subject type calculates a presence probability (probability, accuracy, or certainty that is assumed to exist) included in the input image for each predetermined subject type. It is.
- the existence probability calculation circuit 51 by subject type represents the seven subject types in the input image: “blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key”, and “night scene”.
- the power of whether an image is included and the probability (accuracy, certainty) if included are calculated (function blocks 51a to 51g in Fig. 15) .
- the subject type existence probability calculation circuit 51 confirms that none of the above seven subject types is included in the input image (hereinafter referred to as “subject type“ not subject to detection ”) and its probability (accuracy, (Confidence) is also calculated (function block 51h in Fig. 15).
- Figure 16 shows the contents of the existence probability calculation table 52 used for the existence probability calculation by the existence probability calculation circuit 51.
- the existence probability calculation table 52 corresponds to each of the seven subject types of “blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key”, and “night scene”. Stores the types of feature quantities and identification points that should be used to calculate the existence probability.
- the data stored in the existence probability calculation table 52 (types of features and identification points to be used for calculating the existence probability) is obtained by the learning process.
- Fig. 17 for example, in the learning process for the subject type “underwater”, first, multiple images including the subject type “underwater” (called underwater sample images) and the subject type “underwater” are included. Multiple images (non-sea sample images) are prepared (left side of Fig. 17). Then, one feature type, for example, R-value average, is selected, and the selected feature type (R-value average) is calculated using multiple underwater sample images, and a frequency histogram is created. To do. Similarly, the selected feature type (R-value average) is calculated using multiple non-sea sample images and the frequency histogram is calculated (center of Fig. 17).
- R-value average for example, R-value average
- the selected feature quantity type is a feature quantity type that is suitable for discriminating between subsea sample images and non-sea sample images.
- the logarithmic value of the ratio of the frequency values for each feature value corresponding to the two generated frequency histograms is calculated.
- the calculated logarithmic value is the “identification point” (right side of Fig. 17).
- the existence probability calculation circuit 51 calculates the feature quantity for the feature quantity type stored in the existence probability calculation table 52 for the input image data, and obtains an identification point corresponding to the calculated feature quantity. If the acquired identification point is a positive identification point, the input image is likely to be an underwater image. The larger the absolute value, the higher the possibility. Conversely, if it is a negative identification point, the input image is likely not to contain an underwater image, and the higher the absolute value, the higher the possibility. 05018576
- the existence probability calculation circuit 51 uses the data stored in the existence probability calculation table 52 to select the subject types “blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key” and The existence probability (the identification point described above or a value between 0 and 1 based on the identification point) for each of the seven subject types of “Night Scene” is calculated.
- all seven subject types of subject types “blue sky”, “evening scene”, “”, “high saturation”, “underwater”, “high key”, and “night scene” are used. Is calculated as the probability that the image is not included in the input image (existence probability of subject type “not detected”). The existence probability of the subject type “not detected” is calculated by the following equation.
- ⁇ ! , ⁇ 2 ,..., ⁇ 7 are subject types “Blue Sky”, “Evening Scenery”, “Face”, “High Saturation”, “Underwater”, “High Key” and “Night Scenery”. Indicates the existence probability.
- the subject types calculated by the subject type existence probability calculation circuit 51 are seven types of subjects: “blue sky”, “evening scene”, “face”, “high saturation”, “underwater”, “high key”, and “night scene”.
- Existence probability and subject type ⁇ excluding detection target ” are used in the weighting process (Equation 6) described above.
- the following integrated importance W a is used instead of the importance Wi of the first embodiment (see Equation 6).
- W pi is a numerical value (existence probability-dependent importance) corresponding to each of the existence probabilities pi for each detected object type obtained by the existence probability calculation circuit 51 for each object type.
- Graph shown in figure (Conversion Calculated based on (graph).
- the gains for the object types that the user places importance on are more strongly reflected in the applied gains (G r, G g, G b) applied to the correction of the input image data.
- the gain for a subject type that has a high probability of being present in the input image is more strongly reflected.
- W si is a value (area-dependent importance) based on the area occupied in the input image for each detected subject type.
- the gain for the subject type that occupies a larger area in the input image is more strongly reflected in the applied gain.
- the processing in the image processing apparatus 1 may be changed in accordance with the user's request for image correction.
- the processing in the image processing apparatus 1A may be supported based on external subject information (such as information about the presence or absence of presence in the input image for each subject type).
- the image processing apparatuses 1 and 1A are configured by hardware circuits, but some or all of them may be realized by processing by software.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
- Image Analysis (AREA)
- Color Image Communication Systems (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP05790447.6A EP1802095B1 (en) | 2004-09-30 | 2005-09-30 | Image correction device, method, and image correction program |
JP2006537860A JP4478155B2 (ja) | 2004-09-30 | 2005-09-30 | 画像補正装置および方法,ならびに画像補正プログラム |
US11/664,117 US8111940B2 (en) | 2004-09-30 | 2005-09-30 | Image correction apparatus and method, and image correction program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004285682 | 2004-09-30 | ||
JP2004-285682 | 2004-09-30 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006036026A1 true WO2006036026A1 (ja) | 2006-04-06 |
Family
ID=36119138
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2005/018576 WO2006036026A1 (ja) | 2004-09-30 | 2005-09-30 | 画像補正装置および方法、ならびに画像補正プログラム |
Country Status (5)
Country | Link |
---|---|
US (1) | US8111940B2 (ja) |
EP (1) | EP1802095B1 (ja) |
JP (1) | JP4478155B2 (ja) |
CN (1) | CN100499733C (ja) |
WO (1) | WO2006036026A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010093341A (ja) * | 2008-10-03 | 2010-04-22 | Canon Inc | 画像処理装置及び画像種別特定方法 |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8687230B2 (en) * | 2008-03-31 | 2014-04-01 | Brother Kogyo Kabushiki Kaisha | Image processor |
JP4623388B2 (ja) * | 2008-09-08 | 2011-02-02 | ソニー株式会社 | 画像処理装置および方法、並びにプログラム |
JP5045619B2 (ja) * | 2008-09-08 | 2012-10-10 | ソニー株式会社 | 画像処理装置および方法、並びにプログラム |
EP2350975A4 (en) * | 2008-11-25 | 2012-04-11 | Hewlett Packard Development Co | MODIFYING STORAGE COLORS IN DIGITAL IMAGES |
JP2011109245A (ja) * | 2009-11-13 | 2011-06-02 | Renesas Electronics Corp | 画像処理装置及び画像処理方法 |
KR101896386B1 (ko) * | 2011-11-22 | 2018-09-11 | 삼성전자주식회사 | 화이트 밸런스 조절장치 및 방법 |
CN105450923A (zh) * | 2014-09-25 | 2016-03-30 | 索尼公司 | 图像处理方法、图像处理装置以及电子设备 |
WO2016046664A1 (en) * | 2014-09-25 | 2016-03-31 | Sony Corporation | Image processing method, image processing apparatus and electronic device |
US10530995B2 (en) | 2015-06-12 | 2020-01-07 | Gopro, Inc. | Global tone mapping |
CN108198142B (zh) * | 2017-12-28 | 2020-06-26 | 北京数科网维技术有限责任公司 | 色彩调节方法、终端及计算机可读存储介质 |
US10885613B2 (en) * | 2019-01-10 | 2021-01-05 | International Business Machines Corporation | Real-time alteration of underwater images |
JP2022172787A (ja) * | 2021-05-07 | 2022-11-17 | 富士フイルムビジネスイノベーション株式会社 | 情報処理装置及びプログラム |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0832838A (ja) * | 1994-07-21 | 1996-02-02 | Matsushita Electric Ind Co Ltd | 階調補正方法および階調補正装置 |
US5940530A (en) | 1994-07-21 | 1999-08-17 | Matsushita Electric Industrial Co., Ltd. | Backlit scene and people scene detecting method and apparatus and a gradation correction apparatus |
JP2002218480A (ja) * | 2001-01-23 | 2002-08-02 | Minolta Co Ltd | 画像撮影装置 |
US20030053689A1 (en) | 2001-08-27 | 2003-03-20 | Fujitsu Limited | Image processing method and systems |
JP2004240622A (ja) * | 2003-02-05 | 2004-08-26 | Konica Minolta Holdings Inc | 画像処理方法、画像処理装置及び画像処理プログラム |
Family Cites Families (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5687303A (en) * | 1994-05-18 | 1997-11-11 | Xerox Corporation | Printer controller for object optimized printing |
JP3962496B2 (ja) * | 1998-01-29 | 2007-08-22 | キヤノン株式会社 | 画像処理方法、装置および記録媒体 |
US6496607B1 (en) * | 1998-06-26 | 2002-12-17 | Sarnoff Corporation | Method and apparatus for region-based allocation of processing resources and control of input image formation |
US7019778B1 (en) * | 1999-06-02 | 2006-03-28 | Eastman Kodak Company | Customizing a digital camera |
JP2001076149A (ja) | 1999-09-06 | 2001-03-23 | Fuji Photo Film Co Ltd | 主要被写体抽出方法および装置 |
JP2001310520A (ja) * | 2000-04-27 | 2001-11-06 | Canon Inc | 画像処理方法、装置および記録媒体 |
US7088390B2 (en) * | 2000-06-19 | 2006-08-08 | Olympus Optical Co., Ltd. | Imaging apparatus in which exposure control is performed to suppress changes in sensitivity due to changes in gradation mode |
US6788813B2 (en) * | 2000-10-27 | 2004-09-07 | Sony Corporation | System and method for effectively performing a white balance operation |
US7136508B2 (en) * | 2000-11-09 | 2006-11-14 | Minolta Co., Ltd. | Image processing apparatus, method, and program for processing a moving image |
US20020181801A1 (en) * | 2001-06-01 | 2002-12-05 | Needham Bradford H. | Feature-based image correction |
US7092573B2 (en) * | 2001-12-10 | 2006-08-15 | Eastman Kodak Company | Method and system for selectively applying enhancement to an image |
US20040151376A1 (en) | 2003-02-05 | 2004-08-05 | Konica Minolta Holdings, Inc. | Image processing method, image processing apparatus and image processing program |
US8948468B2 (en) * | 2003-06-26 | 2015-02-03 | Fotonation Limited | Modification of viewing parameters for digital images using face detection information |
US7315630B2 (en) * | 2003-06-26 | 2008-01-01 | Fotonation Vision Limited | Perfecting of digital image rendering parameters within rendering devices using face detection |
US7423674B2 (en) * | 2003-12-08 | 2008-09-09 | Nikon Corporation | Electronic camera having color adjustment function and program therefor |
WO2006036027A1 (ja) * | 2004-09-30 | 2006-04-06 | Fujifilm Corporation | 画像処理装置および方法、ならびに画像処理プログラム |
US7715649B2 (en) * | 2005-02-14 | 2010-05-11 | Fujifilm Corporation | Generation and adjustment of a luminance correction curve to prevent saturation of the image during contrast enhancement |
JP2007295478A (ja) * | 2006-04-27 | 2007-11-08 | Sony Corp | 撮影装置および撮影方法、並びにプログラム |
-
2005
- 2005-09-30 US US11/664,117 patent/US8111940B2/en not_active Expired - Fee Related
- 2005-09-30 EP EP05790447.6A patent/EP1802095B1/en not_active Expired - Fee Related
- 2005-09-30 WO PCT/JP2005/018576 patent/WO2006036026A1/ja active Application Filing
- 2005-09-30 CN CNB2005800333672A patent/CN100499733C/zh active Active
- 2005-09-30 JP JP2006537860A patent/JP4478155B2/ja active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0832838A (ja) * | 1994-07-21 | 1996-02-02 | Matsushita Electric Ind Co Ltd | 階調補正方法および階調補正装置 |
US5940530A (en) | 1994-07-21 | 1999-08-17 | Matsushita Electric Industrial Co., Ltd. | Backlit scene and people scene detecting method and apparatus and a gradation correction apparatus |
JP2002218480A (ja) * | 2001-01-23 | 2002-08-02 | Minolta Co Ltd | 画像撮影装置 |
US20030053689A1 (en) | 2001-08-27 | 2003-03-20 | Fujitsu Limited | Image processing method and systems |
JP2004240622A (ja) * | 2003-02-05 | 2004-08-26 | Konica Minolta Holdings Inc | 画像処理方法、画像処理装置及び画像処理プログラム |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010093341A (ja) * | 2008-10-03 | 2010-04-22 | Canon Inc | 画像処理装置及び画像種別特定方法 |
Also Published As
Publication number | Publication date |
---|---|
EP1802095A1 (en) | 2007-06-27 |
CN101032160A (zh) | 2007-09-05 |
EP1802095B1 (en) | 2013-07-10 |
JP4478155B2 (ja) | 2010-06-09 |
US20070280538A1 (en) | 2007-12-06 |
CN100499733C (zh) | 2009-06-10 |
EP1802095A4 (en) | 2009-06-17 |
JPWO2006036026A1 (ja) | 2008-05-15 |
US8111940B2 (en) | 2012-02-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2006036026A1 (ja) | 画像補正装置および方法、ならびに画像補正プログラム | |
US7652717B2 (en) | White balance correction in digital camera images | |
JP5377691B2 (ja) | オートホワイトバランスを備える画像処理装置 | |
JP4549352B2 (ja) | 画像処理装置および方法,ならびに画像処理プログラム | |
US7916963B2 (en) | Method and apparatus for an intuitive digital image processing system that enhances digital images | |
US7680326B2 (en) | Method, system, and program for correcting the image quality of a moving image | |
US8009905B2 (en) | System, medium, and method with noise reducing adaptive saturation adjustment | |
JP2003169231A (ja) | 画像処理装置、コンピュータ・プログラム | |
JP2003189235A (ja) | 被写体情報を用いた画像処理 | |
JP2002232728A (ja) | 画像処理プログラム、画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体、画像処理装置および画像処理方法 | |
KR20110016505A (ko) | 컬러 조정 | |
US7885458B1 (en) | Illuminant estimation using gamut mapping and scene classification | |
US9786076B2 (en) | Image combining apparatus, image combining method and non-transitory computer readable medium for storing image combining program | |
US7796827B2 (en) | Face enhancement in a digital video | |
JP3673092B2 (ja) | 画質調整装置及び画質調整方法、並びに画像調整用プログラムを記録した記録媒体 | |
JP6611543B2 (ja) | 画像処理装置、画像処理方法、およびプログラム | |
JP4219577B2 (ja) | 画像処理装置と画像出力装置と画像処理方法及び記憶媒体 | |
JP4659354B2 (ja) | 画像処理方法および装置 | |
JP4522229B2 (ja) | 画像処理方法及びその装置 | |
JP3817371B2 (ja) | 画像処理方法、装置および記録媒体 | |
JP3801177B2 (ja) | 画像処理装置、画像処理プログラム、および画像処理方法 | |
JP2008147714A (ja) | 画像処理装置およびその方法 | |
JP2007318320A (ja) | 画像処理装置、撮影装置、画像処理方法及び画像処理プログラム | |
JP2010199844A (ja) | 画像処理装置、画像処理方法、プログラムおよび記憶媒体 | |
JP2019101552A (ja) | 画像検索システム、画像検索プログラム、画像検索端末および画像検索端末制御プログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KP KR KZ LC LK LR LS LT LU LV LY MA MD MG MK MN MW MX MZ NA NG NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2006537860 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11664117 Country of ref document: US Ref document number: 200580033367.2 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2005790447 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2005790447 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 11664117 Country of ref document: US |