JP4376218B2 - Content storage device, content storage method, and program recording medium - Google Patents

Description

  The present invention relates to a content storage device, a content storage method, and a program recording medium that have content storage means for storing a plurality of contents and store the content in the content storage means.

  Currently, digital broadcasting such as BS digital broadcasting, CS digital broadcasting, and terrestrial digital broadcasting is being performed. Digital broadcasting services include HDTV and SDTV digital television broadcasting, data broadcasting using BML, EPG (electronic program guide), caption information, and the like.

  The means for recording broadcast programs is changing from media such as video tapes and CD / DVDs to hard disks. An advantage of storing broadcast programs (hereinafter also referred to as contents) on the hard disk is that any content can be accessed at random, so that the content can be easily found.

The hard disk recorder has a large-capacity hard disk as a feature, and further includes a plurality of tuners such as BS digital, CS digital, and terrestrial digital broadcasting tuners so that different programs can be stored simultaneously. become.
In addition, when a specific keyword is set, the hard disk recorder has a function of referring to EPG information and the like and accumulating all contents including the keyword.

  In television broadcasting, text multiplex broadcasting capable of transmitting caption information is performed. Also, in digital terrestrial broadcasting, there is a mechanism for transmitting subtitle information in addition to video and audio. In character multiplex broadcasting, subtitles and video are superimposed on a part of the screen in the range of about 1 to 2 lines.

However, a large-capacity hard disk (content storage means) also has a capacity limit, and content cannot be stored beyond the capacity of the hard disk. Also, not all digitally broadcast content is new content, and for example, the same program may be broadcast in different time zones between BS digital broadcast and terrestrial digital broadcast.
For this reason, in particular, when content is automatically selected and stored in the hard disk using a keyword or channel set in advance by a hard disk recorder (content storage device), a case where the same content is duplicated on the hard disk is assumed.
Therefore, there is a problem that the content storage device cannot effectively use the capacity of the content storage unit when the same content is stored in the content storage unit.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a content storage device, a content storage method, and a program recording medium that can effectively use the capacity of content storage means.

A first technical means according to the present invention is a content storage device that has content storage means for storing a plurality of contents, and stores the content in the content storage means, wherein the content storage means stores the content stored in the content storage means. Content evaluation means to be evaluated, and content management means for determining whether to store or delete content in the content storage means based on a content evaluation result by the content evaluation means , wherein the content evaluation means includes the content storage means Similarity evaluation means for evaluating similarity between two contents accumulated in the means, wherein the content management means is configured to determine one of the two contents evaluated as having high similarity by the similarity evaluation means. It is characterized that you erased in accordance with the rules.

A second technical means according to the present invention, Oite the first technical means, wherein the predetermined rule is characterized in that is to choose the one to be erased on the basis of the quality of the content.

A third technical means according to the present invention, Oite the first technical means, wherein the predetermined rule is characterized in that is to choose the one to be erased based on the image size of the content.

A fourth technical means according to the present invention, Oite the first technical means, wherein the predetermined rule is characterized in that is to choose the one to be erased based on the recording date and time of the content.

According to a fifth technical means of the present invention, there is provided a content storage method by a content storage device having content storage means for storing a plurality of contents, wherein the content storage device stores the content in the content storage means . A content evaluation step for evaluating the content stored in the content storage means; and a content management step for determining whether to hold or delete the content based on a result of evaluation of the content by the content evaluation step. Includes a similarity evaluation step for evaluating similarity between two contents stored in the content storage means, and the content management step includes two contents evaluated as having high similarity by the similarity evaluation step. One of the following rules It is characterized in that to erase me.

A sixth technical means according to the present invention, Oite the fifth technical means, wherein the predetermined rule is characterized in that is to choose the one to be erased on the basis of the quality of the content.

A seventh technical means of the present invention, in the fifth technical means, wherein the predetermined rule is characterized in that is to choose the one to be erased based on the image size of the content.

Eighth technical means of the present invention, in the fifth technical means, wherein the predetermined rule is characterized in that is to choose the one to be erased based on the recording date and time of the content.

A ninth technical means according to the present invention is a computer-readable program recording medium recording a program for executing the content storage method in any one of the fifth to eighth technical means .

  According to the present invention, the content evaluation unit of the content storage device evaluates the content stored in the content storage unit, and the content management unit determines the content stored in the content storage unit based on the content evaluation result by the content evaluation unit. Is stored or deleted, the capacity of the content storage means can be used effectively.

  If the content evaluation means evaluates that the similarity between the two contents is high, the content management means selects and deletes one of these highly similar contents according to a predetermined rule, so that the similarity to the content storage means High content is not accumulated repeatedly.

FIG. 1 is a block diagram illustrating a configuration example of Embodiment 1 of a content storage device according to the present invention. In FIG. 1, the content storage device 10 includes a variety of types such as a terrestrial tuner, a BS tuner, and a CS tuner. Tuners 1a, 1b, 1c,..., 1n, content storage unit 2 for storing program information and content received by these tuners 1a, 1b, 1c,..., 1n, and 2 stored in content storage unit 2 Content evaluation unit 3 that evaluates the similarity between two contents, and content that determines whether the content is retained or erased (hereinafter referred to as retention / deletion) based on the similarity evaluation of the two contents by the content evaluation unit 3 Management unit 4 and output control unit for outputting content stored in content storage unit 2 to display unit 5 based on a user's viewing instruction If, ROM 7 for recording content storage program, and a control unit 8 for controlling the respective units.
The ROM 7 holds the content in the content storage unit 2 based on the content evaluation process for evaluating the similarity between the two contents stored in the content storage unit 2 and the evaluation of the similarity between the two contents by the content evaluation process. A content storage program for executing a content management process for determining whether to delete is recorded.

The content storage device 10 having the above-described configuration can be stored in the content storage unit 2 automatically using preset keywords and channels, and various types such as a terrestrial tuner, a BS tuner, and a CS tuner can be used. Tuners 1a, 1b, 1c,..., 1n of the same content, for example, a replay of a drama that was broadcast in the morning on the same channel in the morning, or terrestrial broadcasting and BS broadcasting even if the channels are different The same drama broadcasted in is stored in the content storage unit 2.
The content evaluation unit 3 evaluates the similarity between two contents stored in the content storage unit 2. The content management unit 4 deletes one of the two contents evaluated as having high similarity by the content evaluation unit 3 according to a predetermined rule.
Further, the output control unit 6 outputs the content stored in the content storage unit 2 according to a predetermined rule to the display unit 5 based on the user's viewing command (playback).

FIG. 2 is a flowchart showing the flow of content storage processing of the content storage device shown in FIG.
Here, after a certain content (hereinafter referred to as content A1) is accumulated in the content accumulation unit 2, other contents B1, B2,... Bk ... BN already accumulated in the content accumulation unit 2 (all of these contents are similar) The total number is assumed to be N of 1 to N).

In FIG. 2, first, in order to sequentially compare the content A1 and the content B1 to BN, the total number N of similar content stored in the content storage unit 2 is acquired (step S1).
First, k = 1, the content A1 and the content B1 are compared, and if the content A1 and the content B1 are not similar, the process proceeds as k = 2, and the content A1 and the content B2 are compared to sequentially obtain the similarity. (Step S2).
If there is content similar to the content A1 in the middle, one is left at that time, the other is deleted (step S3), the similarity determination for the content A1 is finished, and the next content A2 is as described above. Then, the similarity is sequentially determined with respect to the contents B1, B2,.

Note that the content A1 may be compared with all other content B1,... BN, but the contents added to the content list and stored are not highly similar to each other, and the content A1 is the comparison target content B1,. At the stage where at least one of the BNs has a high similarity, the remaining content does not have a high similarity, so the comparison operation may be aborted in the middle as described above.
If there is no similar content for all N pieces of content, N is incremented by 1 (N + 1) (step S4), the content at that time is added to the content list, The total number is set as N + 1 of B1 to BN + 1 (step S5).

Next, as a method for evaluating the similarity between two contents, a method using unique information of the two contents will be described.
For example, program titles, subtitles, time lengths, and performers included in BML data broadcasting and EPG information used in digital broadcasting are compared.

FIG. 3 is a flowchart showing a flow of processing for determining the identity of two contents by comparing program information of the two contents.
In FIG. 3, program information (step S11) of content A (represented by A as one of a number of content before similarity determination) and program information of content Bk, first, k = 1 B1 program information is acquired (step S12).
Next, the title of content A and the title of content Bk are compared (step S13).
As a result of the comparison, if the title of content A and the title of content Bk are the same (YES in step S13), the subtitle of content A and the subtitle of content Bk are compared (step S14).

If the title of content A and the title of content Bk are not the same (NO in step S13), K and N are compared (step S15).
As a result of this comparison, if K is smaller than N, K is counted up (K + 1) (step S16), and the process returns to step S12.
Next, when the subtitle of content A is the same as the subtitle of content Bk (YES in step S14), the time length of content A is compared with the time length of content Bk (step S17).

If the subtitle of content A and the subtitle of content Bk are not the same (NO in step S14), K and N are compared (step S15).
As a result of this comparison, if K is smaller than N, the process returns to step S12 in the same manner as described above.
When the time length of the content A and the time length of the content Bk are the same (YES in step S17), it is determined that the content A and the content Bk are the same, and are stored as the same content A and the content Bk (step). S18).

If the time length of the content A and the content Bk are not the same (NO in step S17), K and N are compared (step S15).
As a result of this comparison, if K is smaller than N, the process returns to step S12 in the same manner as described above.
If K is equal to N (NO in step S15), it is determined that there is no content similar to content A, and the process proceeds to step S4 (step S19).

FIG. 4 is a flowchart showing a flow of processing for deleting one of two contents determined to be the same.
Here, it is assumed that the content A and the content Bk are determined to be the same.
First, program information of content A and program information of content Bk are acquired (step S21).
Next, the image quality of the content A and the image quality of the content Bk are compared (step S22).
As a result of the comparison, if the image quality of the content A and the image quality of the content Bk are the same (YES in step S22), the file size of the content A and the file size of the content Bk are compared (step S23).
If the image quality of the content A is not the same as that of the content Bk (NO in step S22), the content with the worse image quality of the content A and the content Bk is deleted (step S24).
If the file size of content A and the file size of content Bk are the same (YES in step S23), the broadcast date / time of content A and the broadcast date / time of content Bk are compared (step S25).
If the file size of the content A and the file size of the content Bk are not the same (NO in step S23), the content having a smaller file size is deleted from the file size of the content A and the file size of the content Bk (step S26).

If the broadcast date and time of content A and the broadcast date and time of content Bk are the same (YES in step S25), either one of content A or content Bk is deleted (step S27).
If the broadcast date / time of content A and the broadcast date / time of content Bk are not the same (NO in step S25), the older broadcast date / time of content A and content Bk is deleted (step S28).

  As described above, in the content storage apparatus 10 according to the first embodiment, even when content having the same content is redundantly stored in the content storage unit 2, the two contents have the title, subtitle, and time length of the program information. If they are the same, it is determined that these contents are the same, and one of the determined contents is selected based on the image quality, the file size, and the broadcast date and time, and the other is deleted. There is an effect that it can be used.

In the first embodiment described above, the titles, subtitles, and time lengths of the two contents are compared, but the same determination can be made even if performer information is added. Moreover, although the comparison was made in the order of title, subtitle, and time length, the same determination can be made even if the comparison order is changed. When there is the performer information, the identity of the two contents can be determined according to the degree of coincidence.
In the first embodiment described above, a step of comparing subtitles of two contents is provided. However, when there is no subtitle in the contents, this step can be omitted. In addition, when one of the two contents has a subtitle and the other does not have a subtitle, these contents may be determined to be non-identical.

In Embodiment 1 described above, the time lengths of two contents are compared. However, if the sponsors are different due to different broadcasting media and timing, some of the CMs differ, and the time of these contents is different. It is also assumed that the length is different.
For example, assuming that the time length of one of the two contents is 50 minutes and the time length of the other content is 52 minutes, the latter is calculated by adding four CMs having a length of 30 seconds. Since this level of error is assumed to be sufficient,
For example,
0.95 <(time length of content A) / (time length of content Bk) <1.05
If they fall within a certain range as shown in FIG.

Next, a rule for determining which of the two contents determined to be the same will be deleted will be described.
First, it is conceivable to select content based on the broadcast date and time of the content.
Either the old or new contents can be left. For example, it may be considered that the old content is the original and the new one is a rebroadcast program, and if the content includes commercials etc., the new one is more valuable than the old one. It may be judged that it is high. Furthermore, depending on the genre of the content, it may be possible to change whether the new one or the old one remains.

Second, it is conceivable to select content from the viewpoint of the image quality and sound quality of the content.
When the image quality of the content A and the image quality of the content Bk are different, the information regarding the image quality (bit rate, HD or SD) of the program information is compared, the one with the better image quality is left, and the image with the lower image quality is deleted. For example, content with a larger file size contains more additional information, and the quality of video and audio is better. Therefore, the larger content file size is stored, and the smaller content file size is set as the erasure target.

  Further, the sound quality (5.1 channel, stereo, monaural) of contents may be compared. Furthermore, both the image quality and the sound quality of these contents may be compared. If you decide in advance whether to prioritize the image quality of the content or the sound quality of the content, even if one of these contents has good image quality and poor sound quality, the other has good sound quality and poor image quality , You can decide what content to erase.

Next, the timing for deleting one of the two contents determined to be identical will be described.
In the first embodiment described above, the comparison operation between the content A stored in the content storage unit 2 and the other content stored in the content storage unit 2 is performed every time the content A is stored in the content storage unit 2. Alternatively, it may be performed in a predetermined condition, for example, during a time period when the user is not viewing (nighttime or daytime when going out).
Further, if the same content exists by performing a comparison operation between the content A and other content, the content to be deleted may be deleted immediately, or the free space of the content storage unit 2 becomes insufficient The contents may be deleted. In addition, when erasing content when the free space of the content storage unit 2 becomes insufficient, this content does not need to be compared.

  Also, when a user searches for content using the search means, even if one similar content is deleted, the other content remains, so the deleted content can be excluded from the search target. In addition, as a display method of the result of searching for content, it is possible to display content that remains without being deleted in the interface (similar content that remains with respect to content that is scheduled to be deleted), and can also be selected as similar content. If this is done, the user can select such contents, which makes the interface easier to use.

Next, an example of the content storage apparatus, content storage method, and program recording medium according to Embodiment 2 of the present invention will be described with reference to the drawings.
In the first embodiment described above, the program information of two contents is compared to determine the identity of these contents, but instead, the contents of the two contents are compared to compare these contents. The identity of is determined.
In addition, the same code | symbol is attached | subjected to the component corresponding to the component of Embodiment 1 mentioned above, description is abbreviate | omitted about a common part, and only a characteristic part is demonstrated.
Note that if images are compared in all frames of the content, the amount of calculation becomes enormous, so the images of the content are sampled and compared.
When comparing digital broadcast images, if the content is an HD image, the comparison is 1920 × 1080 pixels, and the amount of calculation is enormous. Therefore, by resizing the HD image of the content to the SD size or extracting pixel information at a predetermined position of the HD image, the number of pixels for comparing (differing) the content can be reduced, and the calculation is performed. Speed can be increased.

FIG. 5 is a flowchart showing the flow of processing for determining the identity of two contents by comparing the images of the two contents.
The image frames of the content A and the content Bk are taken at a predetermined interval (the number of steps is S ′), and these image frames are compared.
Here, the difference between the I-th image of the content A and the I-th image (Image1, Image2) of the content Bk is a local distance LI, and the accumulated distance (added difference amount) is L.
Note that the local distance LI can be, for example, the sum of squares of differences for each pixel.

Here, i is the number of the pixel to be compared, and n is the number of pixels to be compared.
In FIG. 5, the content length N ′ of the content A and the content Bk is acquired (step S31). Here, the content length N ′ of the content A and the content Bk is the same. First, k = 1 and content A and content B1 are compared.
Next, the step number S ′ of the frame images of the content A and the content Bk is determined (step S32). Here, the step number S ′ is a positive integer, and when the step number S ′ is 1, it corresponds to image comparison of all frames.

Next, in order to calculate the local distance L1 of the first images of the content A and the content Bk, I = 1 and the cumulative distance L = 0 (step S33).
Next, the local distance LI is calculated (step S34), and the local distance LI is added to the cumulative distance L (step S35).
Next, the cumulative distance L is compared with a predetermined threshold value X (first threshold value) (step S36).
As a result of the comparison, if the accumulated distance L is not greater than the predetermined threshold value X (NO in step S36), I and N ′ are compared (step S37).

If the cumulative distance L is greater than the predetermined threshold value X (YES in step S36), K and N are compared (step S38).
As a result of the comparison, if K is smaller than N, K is counted up (K + 1) (step S39), and the process returns to step S33 to compare the content A with the content Bk + 1.
If I is not greater than N ′ (NO in step S37), step number S ′ is added to I, and the process returns to step S34 (step S40).
If I is greater than N ′ (YES in step S37), it is determined that content A and content Bk are the same, and are stored as the same content A and content Bk (step S41).
If K is not smaller than N (NO in step S38), it is determined that there is no content similar to content A, and the process proceeds to step S4 (step S42).

  If the difference between the frame images of content A and content Bk is taken and the difference amount is added and becomes larger than a predetermined threshold (second threshold), content A and content Bk are not identical. It can be judged and the calculation can be aborted.

  As described above, in the content storage apparatus 10 according to the second embodiment, the content information is determined by comparing the two content videos, so the program information of the content, for example, the title, the subtitle, and the like are the same. However, there is an effect that the video of content can be compared to determine the identity of these content.

In Embodiment 2 described above, frame images are taken from the content A and the content Bk at predetermined intervals. Instead, a reference image is selected from the content A, and the reference image is included in the content Bk. You can also search for it.
First, a reference image is selected from the content A. Next, it is searched whether the reference image for the content A is included in the content Bk.
When the similarity between the content A and the content Bk becomes higher than a predetermined threshold, it is determined that the content Bk includes a similar image, and a search is performed to determine whether the next reference image of the content A is included in the content Bk.
Similarly, if all the reference images of the content A are included in the content Bk and all the reference images of the content A are included in the content Bk, the content A and the content Bk are the same. judge.

When a reference image is selected from the content A, the more reference images, the more accurate the determination can be made.
In addition, when the reference images for the content A are searched in time-series order, it is only necessary to search for frames after the frame determined to include the similar image in the content Bk, so that the reference image can be efficiently searched. .

Next, Modification 1 of Embodiment 2 described above will be described.
In the second embodiment described above, as an evaluation of the similarity between two contents, the videos of the two contents are compared. Instead, the audio information of the two contents is compared.
As in the case of comparing the videos of the two contents described above, all or some of the audio information included in the contents is compared to determine the identity of these contents. In addition, the calculation may be terminated in the same way as in the case of comparing two content videos. In addition, as in the case of comparing the videos of the two contents, the comparison may be made for all the frames of the two contents, or the audio information for a certain period of time may be compared at several locations of these contents. Good.

Next, a second modification of the above-described second embodiment will be described.
In Embodiment 2 described above, the video of two contents is compared as an evaluation of the similarity of the contents. Instead, the subtitle texts of the two contents are compared.
In BS digital broadcasting and terrestrial digital broadcasting, subtitles are given to many contents, and subtitle texts in which utterance contents and narration in the contents are converted into text are distributed together with the contents. At present, for example, when content of an HD program for one hour is stored in the content storage unit 2, the HD program has a capacity of several TB, whereas the subtitle text of the HD program has a capacity of several tens of KB. .
Since subtitle text is extracted for each packet, the subtitle text is compared in units of packets.

  First, subtitle text is extracted from content A. Next, the caption text of the content Bk is extracted. Here, for example, when the comparison result in the packet unit of the content is the same, 0 is calculated when the result is completely different, and when the result is completely different, the similarity is expressed as a distance, and the accumulated distance is added to the end of the content. When the added cumulative distance is smaller than a predetermined threshold, it is determined that they are the same.

If the two contents are essentially the same, the accumulated distance between these contents is 0. For example, if different CMs (commercials) are inserted in these contents, the accumulated contents of these contents The distance is not necessarily zero in practice.
In addition, the cumulative distance of content may be added to the end, but when the cumulative distance of content exceeds the second threshold, it may be determined that they are not the same and the calculation may be aborted.

Next, with reference to the drawings about the content storage equipment according to a reference example of the present invention.
In the first embodiment, when the content evaluation unit 3 evaluates the similarity between two contents stored in the content storage unit 2, the evaluation can be performed based on the viewing history of the user.
In addition, the same code | symbol is attached | subjected to the component corresponding to the component of Embodiment 1 mentioned above, description is abbreviate | omitted about a common part, and only a characteristic part is demonstrated.

For example, when content to be stored in the content storage device 20 with the keyword “Hideki Matsui” is determined, the content stored in the content storage device 20 commonly includes the keyword “Hideki Matsui”. However, the content including these common keywords includes a portion that differs depending on other attributes of the content between the content that was actually viewed and the content that was not viewed. Therefore, regarding the content attributes, the difference between the content that is actually viewed and the content that is not viewed is compared.
Here, the content attributes include broadcast station, genre, sub-genre, performer, content outline, content size (HD / SD: 1080i, 720p, 525i, etc.) and the like.

Figure 6 is a block diagram showing a configuration example of a content storage equipment according to a reference example of the present invention.
In FIG. 6, the content storage device 20 is provided with a viewing history management unit 11 that records a viewing history. Based on the user's viewing instruction (playback) for the content stored in the content storage unit 2 according to a predetermined rule, the output control unit 6 outputs the content to the display unit 5 and records the viewing history in the viewing history management unit 11. It is like that.
The content evaluation unit 3 compares the content stored in the content storage unit 2 at a predetermined timing with the viewing history recorded in the viewing history management unit 11, and the content management unit 4 performs the content storage unit 2 by the content evaluation unit 3. Based on the comparison result between the content stored in the viewing history and the viewing history recorded in the viewing history management unit 11, deletion candidates for the content are determined.
For example, even if the content is the same performer, the feature of the viewing history such as the content genre is different or only high-definition (HD) is watched is extracted, and the content that does not match this feature is set as an erasure candidate .

In addition, when the user automatically accumulates the content in the content accumulation device 20 according to the predetermined rule, the consistency is enhanced if the above features are extracted for each “predetermined rule for accumulating the content” set by the user.
In addition, regarding the content that is explicitly deleted after viewing by the user, the viewing history management unit 11 can perform the same content evaluation even if the main part of the content is deleted by leaving only the content attribute as the viewed content. .

As described above, the content storage device 20 of the present reference example includes the viewing history management unit 11 that records the viewing history, and even when the content storage unit 2 stores the same content in duplicate, the content evaluation unit 3 The content erasure candidate can be determined based on the comparison result between the content accumulated in the content accumulation unit 2 and the viewing history recorded in the viewing history management unit 11.

  The content storage method, content storage device, and program recording medium of the present invention are not limited to the above-described embodiments, and various changes can be made without departing from the scope of the present invention. is there.

It is a block diagram which shows the structural example of Embodiment 1 of the content storage apparatus which concerns on this invention. It is a flowchart which shows the flow of the content storage process of the content storage apparatus shown in FIG. It is a flowchart which shows the flow of the process which determines the identity of these contents by the comparison of the program information of two contents in Embodiment 1 of the content storage apparatus which concerns on this invention. It is a flowchart which shows the flow of the process which erases one of the two content determined to be the same in Embodiment 1 of the content storage apparatus which concerns on this invention. It is a flowchart which shows the flow of the process which determines the identity of these contents by the comparison of the image of two contents in Embodiment 2 of the content storage apparatus which concerns on this invention. A configuration example of a content storage equipment according to a reference example of the present invention is a block diagram showing.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Tuner, 2 ... Content storage part, 3 ... Content evaluation part, 4 ... Content management part, 5 ... Display part, 6 ... Output control part, 7 ... ROM, 8 ... Control part, 11 ... Viewing history management part, 10 , 20... Content storage device.

Claims (9)

A content storage device having content storage means for storing a plurality of contents and storing content in the content storage means,
Comprises a content evaluation means for evaluating the stored content to the content storage means, based on the evaluation result of the content by the content evaluation unit, and a content management unit for determining whether the holding or erasing the content to the content storage means ,
The content evaluation means includes similarity evaluation means for evaluating the similarity between two contents stored in the content storage means,
It said controller (35) to one of two contents that is evaluated to be high similarity by the similarity evaluating means, content storage device which is characterized that you erase according to a predetermined rule. In the content storage device according to claim 1,
Said predetermined rule, content storage device which is characterized in that by selecting those to be erased on the basis of the quality of the content. In the content storage device according to claim 1,
Said predetermined rule, content storage device which is characterized in that by selecting those to be erased based on the image size of the content. In the content storage device according to claim 1,
Said predetermined rule, content storage device which is characterized in that by selecting those to be erased based on the recording date and time of the content. Has a content storage means for storing a plurality of contents, a content storage method according to the content storage apparatus for storing contents to the content storage means,
The content storage apparatus, a content evaluation step of evaluating the stored content to the content storage means, based on the result of the evaluation of the content by the content evaluation step, and a content management step of determining a hold or erase the content Prepared,
The content evaluation step includes a similarity evaluation step of evaluating the similarity between two contents stored in the content storage means,
In the content management step, one of the two contents evaluated as having high similarity by the similarity evaluation step is deleted according to a predetermined rule . The content storage method according to claim 5 , wherein
It said predetermined rule, a content storage method, characterized in that is to choose the one to be erased on the basis of the quality of the content. The content storage method according to claim 5 , wherein
It said predetermined rule, a content storage method, characterized in that is to choose the one to be erased based on the image size of the content. The content storage method according to claim 5 , wherein
It said predetermined rule, a content storage method, characterized in that is to choose the one to be erased based on the recording date and time of the content. A computer-readable program recording medium storing a program for executing a content storage method according to any one of claims 5 to 8.

Images