JP2005057360A - Picture photographing apparatus and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a mobile terminal having a digital picture input means, in which a proper file name is attached to a digital picture and the resulting digital picture is stored. <P>SOLUTION: The mobile terminal recognizes characters from a received digital picture, uses a character string including a result of character recognition for a candidate of a file name to be given to the digital picture, and decides the file name on the basis of the confirmation of a user. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention belongs to a technical field relating to input means in a portable terminal having a camera.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, devices such as a digital still camera, a camera-equipped mobile phone, and a camera-equipped PDA that can digitize images stored by a camera and store them as files have been put into practical use. Hereinafter, these devices are collectively referred to as a digital camera. Many digital cameras also have a function of displaying accumulated images and a function of deleting unnecessary images. Normally, file names are given based on date, serial number, and so on. In many cases, images stored in a digital camera are transferred to a personal computer for organizing, processing, and printing.
In addition, as disclosed in Japanese Patent Application Laid-Open No. 07-072546 (Patent Document 1), there is an example in which characters are recognized from an image taken from a camera and the recognized result is recorded together with the image.
[0003]
[Patent Document 1] Japanese Patent Application Laid-Open No. 07-072546
[Non-Patent Document 1] R.A. M.M. K. Sinha, B.H. Prasada, G.M. F. Houle, M.M. Sabourin, “Hybrid Textual Text Recognition with String Matching,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 15, no. 9, December 1993
[Non-Patent Document 2] A. K. Jain, B.B. Yu, “Automatic Text Location in Images and Video Frames,” Pattern Recognition, Vol. 31, no. 12, pp. 2055-2076, 1998
[Non-patent Document 3] C.I. -L. Liu, M.M. Koga and H.K. Fujisawa, "Lexicon-drive Segmentation and Recognition of Handwriting Character Strings for Japan Address Reading," IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 24, no. 11, Nov. 2002, pp. 425-1437
[0004]
[Problems to be solved by the invention]
The problem to be solved by the present invention is to facilitate the search and use of files taken with a digital camera. When many images are stored in the apparatus, it is difficult to search for a file to be displayed or deleted using only the file name as a clue. This is because it is impossible to know what is in the image from only the date and serial number. For this reason, in many cases, after roughly registering the date and serial number, it is necessary to search for a desired image by sequentially displaying images depending on the memory. In order to support this process, a number of reduced images may be displayed on the image. However, with this method, the displayed image is small, and details of the image, for example, characters and human faces that can be seen, cannot be confirmed.
[0005]
[Means for Solving the Problems]
The first means to solve the above problem is that a digital camera is equipped with means for recognizing characters, and a file name is determined based on the result of recognizing characters from the image, or a keyword is assigned to a part of the image file. Is to embed. In many cases, there are characters in the image, and the characters are an important clue at the shooting location and at the time of shooting. For example, commemorative photos at sightseeing spots often shoot signs that show historical sites. For this reason, the characters in the image are an important clue when searching for the image. Recently, a digital camera is often used instead of a memo pad. In this case, characters in the image are an important clue for searching for the image.
[0006]
With the introduction of the above means, several technical problems arise. Therefore, they are solved by the following means.
First, there may be many character strings in the image, and it may not be obvious which is appropriate as a file name or keyword. Therefore, means for determining a file name or a keyword based on the character size, the character position, etc. is provided.
[0007]
Secondly, there are cases where an appropriate file name or keyword cannot be determined based on the size and position of characters. Therefore, means for storing a set of keywords specified in advance and means for recognizing the keyword from the image are provided so that the keyword recognized from the image is included in the file name or the keyword embedded in the image file. As means for recognizing a keyword from an image, there are a method of collating a character recognition result with a set of keywords, a method of using a set of keywords as a language dictionary in character recognition, and the like.
[0008]
Third, there are cases where the file name cannot be specified only by the character recognition result. For example, there are two or more images from which the same character recognition result can be obtained. Therefore, a means for generating a character string combining the character recognition result and the date or serial number is provided, and the file name generated thereby is set.
[0009]
The second means for solving the above-mentioned problem is that a digital camera is provided with means for recognizing a face or object of a person specified in advance and means for storing a person name or object name of the face of recognition. When an object to be imaged is recognized from inside, a character string representing the face or object is set as a file name or a keyword embedded in the image.
[0010]
A third means for solving the above problems is a digital camera comprising means for extracting an attribute representing the feature of the image from the image and means for classifying the image into a predetermined category according to the feature of the image. The file name is determined according to the classification result, or the keyword is embedded in the image file. Here, the categories are related to shooting conditions such as “sunny”, “rainy”, “night”, and “indoor”. The file name or keyword assigned based on such a classification represents the content shown in the image, which is a great clue for searching for a file.
[0011]
With the introduction of the third solving means, there arises a problem that classification of images becomes difficult due to changes in image characteristics depending on image capturing conditions. For example, the camera's focus, aperture, shutter speed, white balance, sensor sensitivity, etc. are adjusted at the time of shooting. However, characteristics such as image brightness, sharpness, and color histogram change accordingly. Therefore, at the time of classification, these parameters used for adjustment at the time of shooting are also used as feature amounts.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a data flow diagram showing the flow of processing from image input to image file storage in the first embodiment of the present invention. In this embodiment, after inputting an image with a camera (104), a file name is determined based on a serial number (106) and date / time (107) as usual (108), and the image is output to a file (109). In addition to the conventional method, the file name can also be determined by using the result of the keyword recognition (105) from the image. Of the recognized keywords, those other than those used to determine the file name can be embedded in the tag portion of the image file. The embedded keyword can be used as an index when searching. The words in the keyword dictionary 103 are stored in advance in the storage device of the digital camera or registered by the keyword setting process 102. Here, the format of the input image is an RGB color image, but another color image format or a gray image may be used.
[0013]
As a keyword, for example, the name of a famous landmark is registered. In this case, the famous landmark name is given as a file name to the image taken so that the signboard of the famous landmark is reflected in a sightseeing spot or the like. Also, all commonly used words may be registered in the dictionary. In this case, the character string in the image is recognized regardless of the contents, and is reflected in the file name.
In the present embodiment, as the keyword recognition process, for example, a method as described in Non-Patent Document 1 is used. FIG. 2 shows an example of the data flow of the keyword recognition process. First, in 201, a character line is cut out from the input image. For example, a method as described in Non-Patent Document 2 is used for character line extraction. Next, in 202, individual characters are cut out from the character line. When a plurality of character lines are cut out, the plurality of character lines are used as an image of the subsequent processing. Next, what character each individual character extracted in step 203 is identified. At this time, the character identification dictionary 206 is referred to. Finally, in step 204, the character identification result is interpreted as a character string. At this time, the keyword dictionary 103 is referred to. Finally, in the keyword selection 206, an appropriate keyword is finally selected and output based on the size and position of the recognized keyword in the image, the likelihood of the character string recognition result, and the like. The input of the keyword selection 206 is a set of recognized keywords, and the output is a set of keyword recognition results arranged in an order appropriate to the file name. In the present embodiment, character extraction, character identification, and post-processing are sequentially performed. However, as described in Non-Patent Document 3, these integrated processing may be performed. In addition, after performing character recognition without using linguistic information, it is possible to collate the word recognition result with the words in the keyword dictionary using a normal text matching algorithm.
FIG. 3 shows a hardware configuration in the embodiment of the present invention. An image is picked up by an optical device 302 including a lens and a diaphragm, and then converted into an electric signal by a photoelectric conversion element 303 such as a CCD element. Further, the obtained electric signal is converted into a digital signal by an analog / digital converter 304 and further subjected to processing such as color space conversion and filter processing by a signal processing element 305 such as a DSP. The result is transferred to the RAM 309. The arithmetic unit 307 refers to the processing procedure stored in the ROM 308 and data such as a character identification dictionary, and executes the keyword recognition process 105 using the image stored in the RAM 309 as an input. Further, the arithmetic unit 307 executes the serial number counting process 106 according to the processing procedure stored in the ROM 308 and executes the file name determination process 108 with reference to the clock 312. The image file is stored in the memory card 311 via the input / output device 310. The input device 313 is used when inputting a keyword. The display device 306 is used to check the image at the time of shooting and to display the result of the file name determination process 108. The communication device 314 is used for connection with a personal computer for transferring a keyword dictionary or an image file.
[0014]
FIG. 4 shows the appearance of the apparatus in the embodiment of the present invention. The lens unit 402 of the optical device 302 is disposed on the front surface of the housing 401. In the upper part, a shutter 403 for instructing the image input 104 and a power switch 404 which are part of the input device 313 are arranged. On the side surface, a terminal 407 of a communication device 314 used for connection with a personal computer and an insertion port 408 of a memory card 311 are arranged. On the back side, a cancel button 405 and a cursor key 406, which are part of the input device 313, are arranged. Further, a display surface 409 of the display device 306 is disposed on the back surface. The cursor key 406 sends a signal indicating a direction to the input device 313 when the top, bottom, left, and right ends are pressed, and sends another signal such as confirmation to the input device 313 when the center is pressed. The above is an example of an apparatus for carrying out the present invention, and the present invention is not limited to the above arrangement. For example, instead of a conventional camera-type device, a portable terminal such as a PDA or a mobile phone may be provided with image data input means such as an imaging device. Since these mobile terminals include a wireless LAN or cellular communication wireless communication device as a communication device, the image may be downloaded / transferred by communicating with a network.
[0015]
FIG. 5 shows an operation procedure in the first embodiment of the present invention. First, the operator connects the digital camera to a personal computer and issues a download instruction (501). In response, the keyword dictionary stored on the personal computer is transferred to the digital camera (502). Next, the operator disconnects the personal computer from the digital camera and activates the keyword setting 102 using the cursor key 405 (503). In the keyword setting 102, first, a list of keywords stored in the keyword dictionary 103 is displayed (504). The operator selects necessary items from these, and registers new keywords with the cursor keys as necessary (505). Next, the operator designates a shooting mode using the cursor key 406 (506). Here, the shooting mode designates the difference between setting an image file name by a conventional method or by keyword recognition. In the figure, 507 and the subsequent figures show an operation procedure on the premise that a mode for setting an image file name by keyword recognition is selected. After the mode is designated, the image input processing 104, the keyword recognition 105, and the file name determination 108 are activated by pointing the lens 402 toward the image-to-image and pressing the shutter 403 (507). In the file name determination 108, the keyword obtained as a result of recognition is displayed (508), and the operator designates the keyword used for the file name as necessary. Using the result, the file name determination process 108 refers to information such as a serial number and date to determine the file name. The determined file name is displayed to the operator (510). Subsequently, in the file output process 109, the image is encoded, the keyword recognition result is embedded, and the file name determined in the file name determination process 108 is output. A standard method such as JPEG is used for image encoding. The keyword recognition result is stored in a tag portion of a standard image file format such as JPEG. Again, after the digital camera is connected to the personal computer, the operator issues an image file transfer instruction 511 to start processing for transferring the file from the digital camera to the personal computer (512). You may connect to a network using a communication function, and access a server that can be connected via the network for downloading keywords or transferring image files.
[0016]
FIG. 6 shows a state of the display screen 409 when the keyword input activation 503 and the mode designation 506 are operated. First, when the power is turned on by the power switch 404 and the cursor key 406 is pressed somewhere, a menu 601 is displayed. The selection item in the menu is changed by the operator pressing the upper and lower portions of the cursor key, and the center of the cursor key 406 is pressed when confirming. If neither is selected, the cancel button 405 is pressed. When starting the keyword input, the operator selects “keyword setting 604” and presses the center of the cursor key 406. When specifying the mode 506, the operator selects “shooting (automatic file name)” 603 and presses the center of the cursor key 406. When performing normal shooting, “shooting” 602 is selected.
[0017]
FIG. 7 shows the state of the display surface 409 when the keyword display 504 and the keyword designation 505 are performed. First, when the operator designates the keyword setting 604 using the menu 601, a menu such as 701 in FIG. 7A is displayed. Reference numeral 701 is used to select a keyword dictionary to be used for keyword recognition from a plurality of keyword dictionaries previously transferred from a personal computer. In this example, “famous place / historic site” and “place name” are valid. Here, the operator changes the selected item by pressing the cursor key 406 up and down, and switches the keyword dictionary between valid and invalid by pressing the left side of the cursor key 406. When the operator presses the center of the cursor key 406, the change content is confirmed and the original state is restored. Furthermore, when the operator presses the right side of the cursor key 604 with an item selected, the keywords registered in the keyword dictionary are listed.
When the operator selects the item 702 in FIG. 7A and presses the right of the cursor key 604, a menu 703 as shown in FIG. 7B appears. In this state, the operator can change the selection item by pressing the cursor key 406 up and down, and further, by pressing the left side of the cursor key 406, the selected keyword is switched between valid and invalid. Further, by pressing the right side of the cursor key 406, a mode is entered in which the operator inputs a new keyword using the cursor key 406. In this mode, characters are entered by displaying a virtual keyboard on the screen. Further, when the operator presses the center of the cursor key 406 in the state of FIG. 7B, the change content is confirmed and the original state is restored. For example, “xx canyon” 704 is selected in the state of FIG. Here, when the operator presses the left side of the cursor key 406, the keyword “XX valley” becomes invalid.
[0018]
FIG. 8 shows an example of the input image. In this example, two character strings of “ΔΔ mountaintop” 801 written on a signboard and “□□ wear” 802 printed on clothes are shown. “ΔΔ mountaintop” 801 is a clue to the situation in which this photograph was taken, and is appropriate as an image file name. On the other hand, “□□ ware” 802 is not necessarily a clue to the situation in which the image was taken, and is inappropriate as an image file name.
[0019]
FIG. 9 shows a processing procedure of the keyword selection processing 205. The input of this process is a set of recognized keywords, and the output is a set of keyword recognition results arranged in the order appropriate to the file name. First, in a loop 901, a feature amount calculation 902 and a certainty factor calculation 903 are executed for each keyword recognition result. The feature quantity calculation 902 includes n feature quantities necessary for checking the appropriateness of each keyword recognition result as a file name.
F = (f1, f2, f3,..., Fn)
Is a process for obtaining. Here, n = 5 and the following feature amounts are used.
f1: X coordinate (pixel) on the image of the center of the recognized keyword
f2: Y coordinate (pixel) on the image of the center of the recognized keyword
f3: width of the recognized keyword center on the image (pixels)
f4: Height of the recognized keyword center on the image (pixels)
f5: Likelihood as a character string obtained by post-processing
The certainty factor calculation 903 is a process of calculating a value indicating the appropriateness of the keyword as a file name, that is, a certainty factor, based on the obtained feature amount. Next, in step 904, keyword recognition results whose certainty values are equal to or smaller than a predetermined threshold value θ are deleted. This is to prevent incorrect file names from being used with inappropriate keywords. Next, in step 905, the remaining keyword recognition results are rearranged in order of certainty.
[0020]
FIG. 10 shows a state of the display surface 409 when the recognition result display 508 is executed. Here, the result of recognizing keywords such as 1001 and 1002 is displayed immediately below the recognized character string. Further, the keyword with the highest certainty factor is highlighted and displayed as 1001. Here, pressing down the cursor key 406 highlights keywords with lower certainty, and pressing up highlights keywords with higher certainty in order. Finally, the keyword highlighted when the center of the cursor key 406 is pressed is used to determine the file name.
[0021]
FIG. 11 shows a processing procedure for the file name determination 108. The input of this processing is a set of keyword recognition results arranged in an order appropriate to the file name as a result of keyword recognition 105, date information obtained from date / time management 107, and a set of names of saved image files. . First, in step 1101, the keyword recognition result is displayed as shown in FIG. Next, in step 1102, the keyword used for determining the file name is selected by the keyword selection by the operator by the method described with reference to FIG. Next, in step 1103, the name of the stored image file is checked. The file name uses the same keyword as the keyword determined in step 1102, and the date is the same as the date of the newly captured image. Explore. If the corresponding image file is found, the corresponding image file name having the largest serial number is selected. Next, in step 1104, the serial number obtained in step 1103 plus 1 is set as a new serial number. If no corresponding image file is found in step 1103, the serial number is set to 1 or no serial number is assigned. Finally, in step 1105, the file name is synthesized from the date, keyword, and serial number.
[0022]
FIG. 12 shows the state of the display surface 409 when the file name determination display 510 is executed. In this state, the determined file name and the keyword embedded in the image file are displayed on the window 1201 that appears on the screen. In this example, it is assumed that this image is the second image taken at the summit of △△ on December 17, 2002, and the file name is “200221217 △△ summit 02” combining date, keyword and serial number. It has been determined. Information such as date and time is obtained from a clock built in the apparatus.
[0023]
FIG. 13 is a data flow diagram illustrating the flow of processing from image input to image file output in the second embodiment of the present invention. First, an image is input according to the image input setting 1311 (1302). Here, the image input setting is a setting of parameters necessary for image input, such as camera focus, aperture, shutter speed, white balance, sensor sensitivity, and contrast. Next, a face is detected from the input image, and a face registered in advance in the face dictionary 1303 is recognized. Further, the information registered in the shooting situation dictionary 1304 and the output of the image input setting 1311 are referred to recognize the situation in which the input image was shot (1306). Next, as a result of face recognition 1305, shooting situation recognition 1306, date / time management 1307, the image file 1310 already stored is referred to, and the image file name is determined by the same procedure as 108 (1308). Further, based on the determined file name, an image file is output (1309).
[0024]
FIG. 14 shows the processing procedure of face recognition 1305. First, in step 1401, a face area is detected from the input image. Next, in a loop 1402, the steps 1403 and 1404 are repeated for all the face regions obtained in the step 1401. Step 1403 is a process of calculating the feature amount of the face area. A plurality of feature quantities are calculated, and the output of this step is a feature vector. In the loop 1404, the likelihood of the feature vector obtained in step 1403 is calculated for all the faces registered in the face dictionary 1303 (1405). The face dictionary 1303 stores pairs of likelihood functions and names (for example, names) corresponding to faces as many as the number of registered faces. Using this likelihood function, the likelihood is calculated in step 1405. After the end of the loop 1402, the face with the highest likelihood value is detected in step 1406, and the name corresponding to the face is output in step 1407. This output is used to determine the file name.
[0025]
FIG. 15 shows a processing procedure of the shooting situation recognition 1305. First, in step 1501, a feature value of the shooting situation is calculated from the input image. A plurality of feature quantities are calculated, and the output of this step is a feature vector. For example, as the feature quantity, a histogram value, autocorrelation coefficient, moment, etc. for each color component are used. Next, in the loop 1502, the likelihood of the feature vector obtained in step 1501 is calculated for all shooting situations registered in the shooting situation dictionary 1304 (1503). In the shooting situation dictionary 1304, as many as the number of registered shooting situations, pairs of likelihood functions and names corresponding to the shooting situations (for example, “sunny weather”, “rainy weather”, “night”, “indoor”, etc.) are stored. Using this likelihood function, the likelihood is calculated in step 1503. Next, at step 1504, the one with the highest likelihood is selected. In step 1505, the name of the shooting situation with the highest likelihood is output.
[0026]
【The invention's effect】
It is possible to automatically assign an image file name that can be easily inferred from the contents of an image, which has been difficult in the past.
[Brief description of the drawings]
FIG. 1 is a data flow diagram showing a flow of processing from image input to image file output in the first embodiment.
FIG. 2 is a data flow diagram showing the flow of keyword recognition processing in the first embodiment of the present invention.
FIG. 3 is a hardware configuration diagram according to the embodiment of the present invention.
FIG. 4 is an external view of an apparatus according to an embodiment of the present invention.
FIG. 5 is a diagram showing an operation procedure in the first embodiment of the present invention.
FIG. 6 is a diagram showing a state of a display screen when specifying a keyword setting and a shooting mode.
FIG. 7 is a diagram showing a state of a display surface when displaying a keyword dictionary and specifying a keyword.
FIG. 8 is a schematic diagram of an input image.
FIG. 9 is a diagram showing a processing procedure for keyword selection processing;
FIG. 10 is a display screen when a keyword recognition result is displayed.
FIG. 11 is a diagram showing a processing procedure for determining a file name in the first embodiment of the present invention.
FIG. 12 is a display screen when a file name is displayed in the first embodiment of the present invention.
FIG. 13 is a data flow diagram showing a flow of processing from image input to image file output in the second embodiment of the present invention.
FIG. 14 is a diagram showing a face recognition processing procedure in the second embodiment of the present invention.
FIG. 15 is a diagram showing a processing procedure for photographing state recognition in the second embodiment of the present invention.
[Explanation of symbols]
101: Processing from image input to image file output in the first embodiment, 102 ... Keyword setting, 103 ... Keyword dictionary, 104 ... Image input, 105 ... Keyword recognition, 106 ..Sequence number counting, 107 ... Date / time management, 108 ... File name determination, 109 ... File output, 110 ... Image file, 201 ... Character line extraction, 202 ... Character extraction , 203 ... Character identification, 204 ... Post-processing, 205 ... Keyword selection, 206 ... Character identification dictionary, 301 ... Digital camera, 302 ... Optical device, 303 ... Photoelectric conversion Elements 304, analog-digital converter, 305 signal processing element, 306 display, 307 arithmetic unit, 308 ROM, 309,. · RAM, 310 ··· input / output device, 311 ··· memory card, 312 ··· clock, 313 ··· input device, 314 ··· communication device, 401 ··· housing, 402 ··· lens 403 ... Shutter 404 ... Power switch 405 ... Cancel button 406 ... Cursor key 407 ... Communication device terminal 408 ... Memory card insertion slot 409 ... Display Surface, 501 ... Download support, 502 ... Keyword transfer, 503 ... Keyword input activation, 504 ... Keyword display, 505 ... Keyword specification, 506 ... Mode specification, 507 ... Shooting 508... Recognition result display 509... File name keyword specification 510... File name display 511. ..Image file transfer, 601... Mode designation menu, 602 .. shoot, 603 .. shoot (automatic file name), 604... Keyword designation, 701. Selected keyword dictionary, 703... Keyword list window, 704... Selected keyword, 801..., Recognized keyword ".DELTA .. Summit", 802. 901... Loop relating to each keyword recognition result 902... Feature amount calculating step 903 .. certainty factor calculating step 904 .. certainty factor threshold value processing 905. Steps for replacement, 1001, 1002... Keyword recognition result, 1101. 1102 ... Selecting a keyword 1103 ... Searching for an existing maximum serial number, 1104 ... Adding 1 to the serial number, 1105 ... Synthesizing a file name, 1201... Window for displaying file name, 1301... Processing from image input to image file output in the second embodiment of the present invention, 1302... Image input, 1303. ..Shooting situation dictionary, 1305... Face recognition, 1306 .. Shooting situation recognition, 1307... Date / time management, 1308... File name determination, 1309. File, 1401... Step for detecting face area from input image, 1402... Loop for all face areas, 1403. ..Step of calculating feature amount, 1404... Loop relating to faces registered in all face dictionaries, 1405... Step of calculating likelihood, 1406... Step 1407: Outputting the name of the face with the maximum likelihood value, 1501 ... Step for calculating the feature value, 1502 ... Step for all shooting situations, 1503 ... Calculation of likelihood 1504... Detecting the shooting situation with the highest likelihood value, 1505... Outputting the name of the shooting situation with the highest likelihood value.

Claims (16)

An image pickup device that photoelectrically converts an image and takes it in as a digital signal, and a storage device that encodes the digital signal, assigns a name, and stores it.
A means for recognizing characters from the input image and a candidate for an image name including a character string obtained as a result of recognizing the character are created, and when the candidate is determined to be valid, the candidate is used as the name of the image. And an image photographing device. 2. The image photographing apparatus according to claim 1, further comprising means for storing a set of words, wherein the word is recognized from the image and used to create a candidate for the name of the image. 2. The image processing apparatus according to claim 1, further comprising means for acquiring a date or time, wherein the character string obtained as a result of the character recognition and a character string representing the date or time are combined to form a part of an image name candidate. Image shooting device. 4. The image photographing apparatus according to claim 3, wherein a character string obtained as a result of character recognition and a serial number are combined to form a part of an image name candidate. 2. The image photographing apparatus according to claim 1, wherein when two or more character strings are recognized from the image, a character string used for an image name candidate is determined based on the position of the recognized character string. . 2. The image photographing according to claim 1, wherein when two or more character strings are recognized from the image, a character string to be used as a candidate for an image name is determined based on the size of the recognized character string. apparatus. A process for recognizing characters from images stored in a file and creating a new image file name candidate that includes the character string obtained as a result of character recognition. A program for describing a procedure for determining the candidate as the name of the image and storing an image with a newly determined file name. 8. The program according to claim 7, wherein a procedure for accessing a means for storing a set of words, recognizing the word from the image, and using the name of the image for determination is described. Describe the processing procedure to access the date or time acquisition means and synthesize the character string obtained as a result of the character recognition and the character string representing the date or time to become part of the image name candidate. The program according to claim 7, wherein: 10. The program according to claim 9, wherein a processing procedure is described in which a character string obtained as a result of character recognition and a serial number are combined to form part of the name of the image. When two or more character strings are recognized from the image, a processing procedure for determining a character string to be used as a candidate for the image name based on the position of the recognized character string is described. The program according to claim 7. When two or more character strings are recognized from the image, a processing procedure for determining a character string to be used as a candidate for the image name based on the size of the recognized character string is described. The program according to claim 7. An image pickup device that photoelectrically converts an image and takes it in as a digital signal, and a storage device that encodes the digital signal, assigns a name, and stores it.
A means for detecting a face area from the input digital signal, a means for calculating a feature quantity of the detected face area, and a likelihood function and face name pair for calculating likelihood from the feature quantity of the face area. And a means for determining the name of the image so as to include the name of the face having the highest likelihood of the detected face area. Stores a process for detecting a face area from an image, a process for calculating a feature value of the detected face area, and a set of likelihood function and face name pairs for calculating likelihood from the feature value of the face area. A process for accessing a means for performing the process, a process for determining a name of a new image file so as to include a face name having the highest likelihood of the detected face area, and storing the image with the newly determined file name A program characterized by describing processing. An image pickup device that photoelectrically converts an image and takes it in as a digital signal, and a storage device that encodes the digital signal, assigns a name, and stores it.
Means for calculating a feature value of a shooting situation from the input digital signal; means for storing a set of a likelihood function and a shooting situation name pair for calculating likelihood from the feature quantity of the shooting situation; Means for determining the name of the image so as to include a name of a high photographing situation. Means for calculating the feature value of the shooting situation from the image, means for storing a set of a pair of a likelihood function and a shooting situation name from the feature quantity of the shooting situation, and a shooting situation with the highest likelihood A program for describing a procedure for determining a name of a new image file so as to include the name of the image and a process for storing the image with the newly determined name.

Images