JP2009266005A - Image retrieval method, image retrieval program, music player, and article for music retrieval - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To retrieve a musical piece suitable for user's feeling, mood or condition from among a plurality of musical pieces. <P>SOLUTION: A music retrieval method for retrieving a musical piece from a predetermined music database performs an image acquisition step of acquiring an image, an image feature information calculation step of calculating image feature information representing features of the image based on the obtained image, and a music retrieval step of retrieving a musical piece from the music database based on the calculated image feature information. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a music search method, a music search program, and a music playback device for searching for music from a music database, and more specifically, a music search method for searching for music based on a photographed or scanned image, and music search The present invention relates to a program and a music playback device. The present invention also relates to a music search article to which information suitable for music search in the music playback device is attached.

  In recent years, audio devices that compress music data using a high compression rate audio compression codec such as MP3 (MPEG-1 Audio Layer-3) and store it in a large capacity HDD (Hard Disk Drive) have become widespread. In such an audio apparatus, a huge number of music pieces such as thousands to tens of thousands of songs can be stored. Therefore, the user can store desired music one after another in the audio device without paying attention to the remaining capacity of the HDD. However, while a large number of music pieces can be stored in the HDD as described above, there is a problem that it is difficult to quickly find a music piece to be listened to.

In order to solve the above problem, various music search techniques suitable for searching for a desired music from a huge number of music have been proposed. An example thereof is disclosed in Patent Document 1. According to Patent Document 1, music is ripped from a CD (Compact Disc) or the like and stored in a music database. At this time, the music is evaluated by a plurality of evaluation items and associated with the evaluation result. Each piece of music stored in the music database in this way (more precisely, a link to the music displayed as a sphere) is a virtual expression expressed in three axes corresponding to each evaluation item according to the evaluation result. Arranged in a three-dimensional space. In the three-dimensional space, in addition to the link to each music, an impression slog representing the image of the music corresponding to each evaluation item is superimposed on the vicinity of the end of the space and displayed on the display. The user can sensuously select music to be listened to by visually observing the position of each link in the three-dimensional space displayed on the display and visually capturing the image of each music.
JP 2007-11913 A

  By the way, there is a demand for the user to listen to music suitable for the emotion, mood or situation at that time. However, when using the music search function exemplified in Patent Document 1, it is difficult for the user to objectively determine his / her emotions, moods or situations, and the user always selects a song that matches his / her own emotions. It is not always possible. For this reason, even when a wide variety of music is stored in the HDD or the like, there are many cases where a music suitable for the user's emotion, mood or situation is not effectively retrieved.

  The present invention has been made in view of such circumstances, and an object of the present invention is to search for music suitable for a user's emotion, mood, or situation from a plurality of music based on objective judgment. It is an object of the present invention to provide a music search method, a music search program, and a music playback device suitable for the above. Another object of the present invention is to provide a music search article with information suitable for music search in such a music playback device.

  A music search method according to an aspect of the present invention that solves the above problem is a music search method for searching for music from a predetermined music database, and an image acquisition step for acquiring an image, and an acquired image An image feature information calculating step for calculating image feature information indicating the feature of the image based on the image feature and a music search step for searching for a song from a music database based on the calculated image feature information are performed.

  According to the method of searching for music according to the acquired image in this way, it is suitable for the user's mood etc. based on objective judgment without letting the user judge emotions, moods or situations by his / her own subjectivity. You can search for songs. For example, when a smile is photographed, image feature information indicating the smile is calculated, and music suitable for the smile (fun emotion, mood, etc.) is searched based on the calculation result.

  According to the music search method of the present invention, each music is managed in the music database in association with the music feature information indicating the characteristics of the music. In this case, in the music search step, the image feature information and each music As a result of comparison with the music feature information, the music associated with the music feature information determined to satisfy a predetermined condition may be output as a search result. Here, the image feature information and the music feature information are information in which, for example, the features of the image or the music are digitized based on the same evaluation item.

  Moreover, the music search method according to the present invention may be a method of further executing a condition changing step of changing a predetermined condition according to the search result. In this case, in the music search step, the music associated with the music feature information determined to satisfy the changed condition is output as a search result.

  The music search method according to the present invention further includes a playlist creation step for creating a playlist of the searched music, and the songs in the created playlist in the order of the songs in the playlist, or randomly. It is good also as the method of performing with the music reproduction step reproduced | regenerated by a repeat. Further, a playlist display step for displaying the created playlist on a predetermined display may be further executed, and when the display is a touch panel, the touched music in the playlist may be played back.

  In the image acquisition step, a captured or scanned image may be acquired.

  Moreover, the music search program which concerns on one form of this invention which solves said subject is a program for making a computer perform the music search method described in any of the above.

  A music playback device according to an aspect of the present invention that solves the above problem is a device that searches for and plays back music from a predetermined music database, and is acquired by an image acquisition unit that acquires an image. Image feature information calculating means for calculating image feature information indicating the feature of the image based on the image, and music search means for searching for music from a music database based on the calculated image feature information. Yes.

  According to the configuration for searching for music according to the acquired image in this way, it is suitable for the user's mood etc. based on objective judgment without letting the user judge emotions, moods or situations by his / her own subjectivity. You can search for songs. For example, when a smile is photographed as a subject, image feature information indicating the smile can be calculated, and music suitable for a smile (fun emotion, mood, etc.) can be searched from the music database based on the calculation result.

  A music database is good also as a structure which links | relates and manages each music in the music characteristic information which shows the characteristic. In this case, the music search means outputs the music associated with the music feature information determined to satisfy the predetermined condition as a search result as a result of comparing the image feature information with the music feature information of each music. The image feature information and the music feature information referred to here are, for example, information obtained by quantifying the features of the image or the music based on the same evaluation item.

  In addition, the music reproducing device according to the present invention may further include a condition changing unit that changes a predetermined condition according to a search result. In this case, the music search means outputs the music associated with the music feature information determined to satisfy the changed condition as a search result.

  The music playback device according to the present invention further includes playlist creation means for creating a playlist of searched music, and the music in the created playlist is in the order of the songs in the playlist or randomly. Alternatively, it may be configured to reproduce by repeat. Further, it may further include playlist display means for displaying the created playlist on a predetermined display, and when the display is a touch panel, the touched music in the playlist may be played back.

  Here, the image acquisition means may be configured to acquire an image when a user operation is performed, immediately after power-on, or every time a predetermined time elapses.

  Further, the image acquisition means may be means for photographing a subject or scanning an image.

  In addition, a music search article according to an aspect of the present invention that solves the above-described problem is included in the music playback device in which the image feature information calculated by the image feature information calculation unit included in the music playback device is embedded. An article to which a predetermined code pattern photographed or scanned by an image acquisition means is attached. Examples of the music search article include a QR code (registered trademark) in which image feature information is embedded and a magazine on which a barcode is printed.

  According to the image search method, the image search program, and the music playback device according to the present invention, it is possible to make an objective determination based on the acquired image without causing the user to determine an emotion, a mood, a situation, or the like based on their own subjectivity. Based on this, it is possible to search for music suitable for the user's mood. Moreover, according to the music search article according to the present invention, information suitable for performing music search in the music playback device is provided.

  Hereinafter, with reference to the drawings, a car navigation device that executes music search processing for searching for music according to the present invention will be described.

  FIG. 1 is a block diagram showing a configuration of a car navigation apparatus 100 according to an embodiment of the present invention. As shown in FIG. 1, the blocks of the car navigation apparatus 100 are connected to each other via a system bus 10. Each block accesses necessary hardware via the system bus 10 after the car navigation apparatus 100 is powered on.

  For example, immediately after the car navigation apparatus 100 is turned on, a CPU (Central Processing Unit) 20 accesses a ROM (Read-Only Memory) 22. The CPU 20 calls the firmware stored in the ROM 22 and develops it in a work area of an SDRAM (Synchronous Dynamic Random Access Memory) 24, and activates the system of the car navigation apparatus 100. As a result, the blocks perform coordinated operations as necessary under the control of the CPU 20 to execute processing, for example, navigation processing for guiding the user to the destination is performed, or music selected by user operation or the like is reproduced. Or

  An operation key (hardware key) 26 is provided on the front panel of the car navigation apparatus 100 as a device that receives a user operation. The front panel is also provided with a liquid crystal display 28 for displaying, for example, navigation information to a destination, music information being reproduced, and the like. The liquid crystal display 28 has a touch panel structure and also functions as an input device. The user can operate the car navigation apparatus 100 by performing key input to the operation keys 26 or finger touch or pen touch on a GUI (Graphical User Interface) displayed on the liquid crystal display 28. In addition, the car navigation device 100 can be operated by a remote controller (not shown).

  The car navigation apparatus 100 is provided with a GPS (Global Positioning System) receiver 50 and a sensor 52 in order to execute navigation processing. The GPS receiver 50 is a well-known device that captures and tracks GPS signals transmitted from a plurality of GPS satellites and calculates the current position and the like. The sensor 52 is a known DR (dead reckoning) sensor such as a gyro sensor or a vehicle speed sensor. The calculation result by the GPS receiver 50 and the sensor output from the sensor 52 are transferred to the CPU 20, and the CPU 20 performs position estimation based on these data.

  The CPU 20 reads out map data around the calculated position from the map database stored in the ROM 22, displays it on the liquid crystal display 28, performs map matching, and superimposes the vehicle position mark on the displayed map. To do. At this time, if a destination is set by a user operation, the CPU 20 also performs a navigation process toward the destination.

  When a user operation for music playback is performed during the operation of the car navigation device 100 including the time when the navigation process is performed, the last played music (hereinafter referred to as “last one”) during the previous use is played by the speaker 30. Is done. Processing executed in each block of the car navigation device 100 at this time will be described below.

  First, an operation signal corresponding to the user operation is generated and input to the CPU 20. The CPU 20 accesses an EEPROM (Electrically Erasable and Programmable Read Only Memory) 32 in response to the operation signal. Here, each time a new music is reproduced, information on the music (for example, a music name) is read from the music database 34a described below, and is overwritten and saved in the EEPROM 32. That is, the EEPROM 32 is in a state where the last one information is always held.

  When the CPU 20 reads the last one information from the EEPROM 32, the CPU 20 accesses the music database 34a stored in the internal storage 34 using the information as a search key. For the built-in storage 34, for example, an HDD is adopted if the capacity is given priority, and for example, a flash memory is adopted if the access speed is given priority.

  The music database 34a stored in the built-in storage 34 stores music data ripped from, for example, a CD. Here, the ripping process executed by the car navigation apparatus 100 will be described.

  The ripping process is executed by a cooperative operation between the media drive 36 and a DSP (Digital Signal Processor) 38. The media drive 36 is a drive for reproducing a recording medium such as a CD, a DVD (Digital Versatile Disc), or a BD (Blu-ray Disc). The DSP 38 is a signal processing circuit that performs codec processing on audio data, analog conversion processing of digital audio data expanded by the codec processing, and the like.

  When, for example, a music CD is inserted into the slot of the media drive 36, the media drive 36 reads the music data (PCM (pulse code modulation) format) of each track and outputs it to the DSP 38. The DSP 38 compresses the music data in the PCM format from the media drive 36 using an audio compression codec such as MP3 or AAC (Advanced Audio Coding). The compressed music data is stored in the music database 34 a of the internal storage 34 under the control of the CPU 20. In this way, music data is ripped. Note that the format of the audio compression codec executed in the DSP 38 is not limited to MP3 or AAC. Depending on the specifications of the car navigation apparatus 100, other lossy compression codecs, lossless compression codecs, and the like may be employed.

  The built-in storage 34 further stores a music management record group for managing each music data in the music database 34a. Each music management record is data associated with each music data, and is entered in each field such as “music name”, “artist name”, “album name”, “genre”, “feature (described later)”, etc. Consists of data. The user can enter text data in each field by a user operation.

  The car navigation apparatus 100 can access the Internet by a communication modem (not shown). For example, when TOC (Table of contents) information of a CD is read at the time of ripping, the CPU 20 accesses a CDDB (CD DataBase) on the Internet via the communication modem, and stores a meta data such as a song name corresponding to the TOC information. Download information. Next, a music management record corresponding to the music data of the CD being ripped is created. Then, each text data of the downloaded meta information is entered in each field of the created music management record. In this case, the user does not need to manually input a music title or the like.

  In the car navigation apparatus 100, music data stored in an external storage 40 such as a USB (Universal Serial Bus) memory or a memory card can also be stored in the music database 34a. Specifically, the car navigation device 100 has a configuration such as a port, a PC card slot, and the like conforming to the USB and IEEE 1394 (Institute of Electrical and Electronic Engineers 1394) standards, not shown, so that the external storage 40 can be connected. . For example, when the external storage 40 is connected to the port, the CPU 20 recognizes the external storage 40 by PnP (Plug and Play) and performs resource allocation, device driver installation, and the like. As a result, the CPU 20 can access the data stored in the external storage 40. Therefore, when a user operation is performed to save the music data stored in the external storage 40 in the car navigation device 100, the CPU 20 copies the music data in the external storage 40 to the internal storage 34. When meta information such as a music name is attached to the music data, after creating a music management record corresponding to the music data, each text data of the meta information is entered in the corresponding field. Thus, the music data is stored in the music database 34a other than ripping from a CD or the like.

  Thus, music data corresponding to the search key generated by the CPU 20 is extracted from the music database 34 a in which a large number of music data is stored, and is output to the DSP 38. The DSP 38 expands the input music data, converts the expanded music data into analog music data, and outputs the music data to the audio amplifier circuit 42. The audio amplification circuit 42 adjusts the gain of the music data in the analog format with the amplification factor corresponding to the volume, and outputs it to the speaker 30. As a result, the last one, that is, the music that was last played when used last time is played by the speaker 30.

  In the above description, it has been described that the ripping process is performed by the DSP 38 or the like when a music CD is inserted into the slot of the media drive 36. At this time, the DSP 38 also executes time shift reproduction processing. That is, the DSP 38 compresses and converts the PCM format music data read by the media drive 36 into MP3 data and the like, and buffers the compressed music data in, for example, a ring buffer built in the DSP 38. The ring buffer buffers music data for a predetermined reproduction time and sequentially sweeps it out. The swept music data is converted into analog music data, output to the audio amplification circuit 42, and reproduced by the speaker 30. That is, the DSP 38 executes ripping and reproduction of music data in parallel. For this reason, in the car navigation apparatus 100, it is effectively avoided that the sound data is silent for a long time during ripping of the music data, and the user is not stressed.

  Next, a music search for searching for a desired music from the music database 34a will be described. When a predetermined user operation is performed, a music search screen for searching for music is displayed on the liquid crystal display 28. When an operation for inputting or specifying a song name, artist name, or the like is performed on the song search screen, the CPU 20 searches the song database 34a according to the operation and reproduces a song that matches the search condition.

  However, in the case of a search method for inputting or specifying a song name or the like, the user searches for a song suitable for the emotion, mood or situation at that time, except for a specific song name or the like that can easily recall the tone of the song. It ’s difficult. For this reason, even when music suitable for the user's mood or the like is actually stored in the music database 34a, such music is often not searched effectively. On the other hand, in the case of a search method that enables a sensuous music search as described in Patent Document 1 (Japanese Patent Laid-Open No. 2007-11913), music that is suitable for the user's emotions, mood, or situation is easily searched. It seems to be possible. However, as described above, unless the user can objectively determine his / her emotions, moods, or situations, the user tends to unconsciously select a song that matches his / her preference. For this reason, in the end, music suitable for the user's emotion, mood or situation is often not searched.

  Therefore, it is desired to provide a music search method, a program, a device, and the like that can realize a music search suitable for the user's emotions, moods, and situations. In order to meet such a demand, the car navigation apparatus 100 according to the present invention includes an image acquisition device 60. Moreover, the music search process shown by the flowchart of FIG. 2 is performed using the image data acquired by the image acquisition device 60. As the image acquisition device 60, for example, a vehicle interior camera installed in the vehicle, a portable compact DSC (Digital Still Camera), a scanner device mounted on the liquid crystal panel 28, and the like are assumed. The vehicle interior camera is a DSC fixed at a position where the face of a driver or a passenger can be photographed, for example. A vehicle interior camera, a compact DSC, or the like is connected to a USB port or the like of the car navigation apparatus 100.

  When a user performs a shooting or scanning operation using the image acquisition device 60, the image data acquired by the image acquisition device 60 is input to the DSP 38 (step 1, hereinafter, the steps in this specification and drawings are referred to as “ Abbreviated as "S"). For example, landscapes such as forests, seas, mountains, building streets, and the like that can be seen from the passenger compartment and the vehicle window are assumed as objects to be photographed, and magazines and photographs are assumed as objects to be scanned. Here, a case where a driver's face (for example, a smile) is photographed by a vehicle interior camera will be described as an example. Note that the timing of the above shooting or the like is not limited to the timing at which the user operates the image acquisition device 60. Shooting or the like by the image acquisition device 60 may be automatically performed, for example, when an automobile engine is driven (immediately after the car navigation apparatus 100 is turned on) or every time a predetermined time elapses.

  The DSP 38 analyzes the image data from the image acquisition device 60 and calculates image feature data of the image (S2). The calculated image feature data is transferred to the CPU 20. Here, the image feature data is data in which the feature points of the image are expressed numerically for each predetermined evaluation item. The evaluation items constituting the image feature data include items related to emotions such as “joy”, “anger”, “sorrow”, “easy”, “forest”, “city”, “sea”, “mountain”, etc. There are items related to the scenery. In addition, various items such as “Energy”, “Healing”, “Cool”, “Hot”, “Tempo”, and “Brightness” are assumed. Here, for convenience of explanation, the image feature data is composed of two items, “tempo” and “brightness”, and is data obtained by evaluating the image in 100 levels from 0 to 99 for each item. Note that details of a technique for evaluating image-recognized landscapes, human faces, facial expressions, and the like and quantifying their characteristics are disclosed in, for example, Japanese Patent Application Laid-Open No. 2007-328675.

  In the process of S3, the CPU 20 accesses the music database 34a using the image feature data transferred in the process of S2 as a search key. Next, the image feature data is compared with the song feature data entered in the “feature” field of each song management record, and the degree of coincidence (%) is calculated. Then, music data associated with the upper predetermined number of music feature data having a high degree of coincidence is obtained as a search result.

  Here, the music feature data is data obtained by quantifying the music features for each predetermined evaluation item by the music analysis processing by the DSP 38. That is, when ripping music data recorded on a music CD or the like, the DSP 38 performs FFT (Fast Fourier Transform) processing on a certain frame length from a predetermined data analysis start point of the music data to obtain a power spectrum. Is calculated. Next, music feature data is calculated based on the calculated power spectrum. Each calculated music feature data is entered in the “feature” field of the corresponding music management record. In the music analysis process, the music data is evaluated by two items, “tempo” and “brightness”, as in the above-described image analysis process, and the features are quantified in 100 levels from 0 to 99 for each item. . Details of such music analysis processing are disclosed in, for example, Japanese Patent Application Laid-Open No. 2007-11913.

  FIG. 3 shows a graph for explaining the process of S3. The vertical axis in FIG. 3 corresponds to the evaluation item “tempo”, and the horizontal axis corresponds to the evaluation item “brightness”. In FIG. 3, each piece of music data (marked with ● in FIG. 3) stored in the music database 34a is arranged in accordance with the music feature data on a virtual two-dimensional coordinate area centered on each evaluation item. Further, the image feature data (mark A in FIG. 3) calculated in the process of S2 is arranged on the two-dimensional coordinate area according to the image feature data. In the process of S2, image feature data having high evaluation items “tempo” and “brightness” are calculated based on the fact that a smile was photographed in the process of S1. In the two-dimensional coordinate region, the mark ● that is closer to the mark A indicates that the music data has a higher degree of coincidence with the image feature data calculated in the process of S2. Accordingly, in the process of S3, a predetermined number of high-ranked music data of ● marks closer to the mark A on the two-dimensional coordinate area are obtained as search results.

  Next, the CPU 20 creates a playlist based on the search result obtained in the process of S3 (S4). Next, the created playlist is interrupted and displayed on the forefront of the liquid crystal display 28 (S5), and the music search process in FIG. 2 is terminated. In the play list, the music searched in the process of S3 is listed in descending order of coincidence. After the play list is displayed, the music reproduction is started automatically in accordance with the order of the music in the play list or when a user operation is accepted. When the music in the playlist is selected by finger touch or pen touch, the music to be played is changed. When the random playback function or the repeat playback function is set by the user operation, the music in the playlist is played back randomly or played back repeatedly.

  In the above, the car navigation device 100 determines that the driver is smiling (fun emotions, moods, etc.) by the image analysis process, and the evaluation items “tempo” and “brightness” are both high and fun. Music suitable for emotions and moods is searched. That is, according to the music search process which is a characteristic process of the present invention, the driver's emotions and moods are objectively determined through facial expressions, and according to the determination results, that is, the driver's emotions and moods. Music suitable for is searched.

  Further, consider a case where the object to be photographed by the image acquisition device 60 is a landscape. For example, when a forest is photographed while traveling on a mountain road, an image in which feature points of the forest are digitized based on the photographed forest image (for example, evaluation items “tempo” and “brightness” are both low) Feature data is calculated. As a result, music suitable for the forest image is searched for which both the evaluation items “tempo” and “brightness” are low. That is, according to the music search process, which is a characteristic process of the present invention, the situation of the travel location is determined based on the landscape image, and the music according to the determination result, that is, the situation of the travel location, etc. Searched.

  In this way, the car navigation device 100 can quickly search for music suitable for the user's mood and the like based on the objective determination without causing the user to determine the emotion, mood, or situation based on his / her own subjectivity. Further, in the music search process according to the present invention, since the user's emotion, mood, or situation changes from time to time, there is an advantage that the searched music is not biased to some genres. While listening to various music pieces, new music that suits the user's taste may be discovered.

  The present invention is not limited to the above embodiment, and various modifications are possible within the scope of the technical idea of the present invention. For example, various types of apparatuses are assumed as apparatuses for executing the music search process according to the present invention. Examples of this include in-car audio equipment that is mounted on a moving body, as well as car navigation devices, stationary audio set components installed in the home, commercial karaoke devices, and portable music players that are supposed to be carried around. .

  In addition to the driver's facial expression, landscape, and the like, a QR code (registered trademark), a barcode, or the like printed on a magazine or the like is assumed as an object to be photographed or scanned by the image acquisition device 60. In this case, numerical data corresponding to the evaluation items “tempo” and “brightness” are embedded in QR code (registered trademark), barcode, or the like as an alternative to the image feature data. Such a code pattern may be a more multidimensional pattern.

  The search method executed in the process of S3 in FIG. 2 is not limited to the above method, and various other methods are assumed. For example, instead of the upper predetermined number of music feature data having a high degree of coincidence, all music data associated with music feature data having a predetermined degree of coincidence or higher may be obtained as a search result. When the music data satisfying the search condition is one music, the effectiveness of displaying the playlist on the liquid crystal display 28 is poor. Therefore, in this case, the playlist creation process and the display process need not be executed. Further, when there is no music data satisfying the search condition or the predetermined number has not been reached, the search process of S3 in FIG. 2 may be executed again after relaxing the search conditions. Thus, the search condition may be automatically changed according to the search result, or may be arbitrarily set by a user operation.

  Note that when the user operation is set to eliminate the search condition, a playlist in which all the songs in the song database 34a are listed in descending order of coincidence is created by the song search process of FIG.

It is a block diagram which shows the structure of the car navigation apparatus of embodiment of this invention. It is a flowchart figure of the music search process of embodiment of this invention. It is a graph for demonstrating the process of S3 of FIG.

Explanation of symbols

10 System bus 20 CPU
22 ROM
24 RAM
26 Operation Keys 28 Liquid Crystal Display 30 Speaker 32 EEPROM
34 Internal storage 34a Music database 36 Media drive 38 DSP
42 audio amplification circuit 44 image processing circuit 50 GPS receiver 52 sensor 60 image acquisition device 100 car navigation device

Claims (17)

In a music search method for searching for music from a predetermined music database,
An image acquisition step of acquiring an image;
An image feature information calculating step for calculating image feature information indicating the feature of the image based on the acquired image;
A music search step for searching for music from the music database based on the calculated image feature information;
The music search method characterized by performing. Each song is managed in the song database in association with song feature information indicating the feature of the song,
In the music search step, as a result of comparing the image feature information with the music feature information of each music, outputting a music associated with the music feature information determined to satisfy a predetermined condition as a search result. The music search method according to claim 1, wherein the music search method is characterized.   The music search method according to claim 2, wherein the image feature information and the music feature information are information in which the image or the music features are expressed numerically based on the same evaluation item. . Further executing a condition changing step of changing the predetermined condition according to the search result;
The music search step outputs the music associated with the music feature information determined to satisfy the changed condition as a search result in the music search step. Music search method. A playlist creating step for creating a playlist of the searched music;
A music playback step of playing back the music in the created playlist in the order of the songs in the playlist, randomly or repeatedly;
The music search method according to claim 1, further comprising: Further executing a playlist display step of displaying the created playlist on a predetermined display;
The music search method according to claim 5, wherein when the display is a touch panel, the touched music in the playlist is played.   The music search method according to claim 1, wherein a captured or scanned image is acquired in the image acquisition step.   A music search program for causing a computer to execute the music search method according to any one of claims 1 to 7. In a music playback device for searching and playing music from a predetermined music database,
Image acquisition means for acquiring images;
Image feature information calculating means for calculating image feature information indicating the features of the image based on the acquired image;
Music search means for searching for music from the music database based on the calculated image feature information;
A music reproducing apparatus comprising: The music database manages each music piece in association with music feature information indicating its characteristics,
The music search means outputs the music associated with the music feature information determined to satisfy a predetermined condition as a search result as a result of comparing the image feature information with the music feature information of each music. The music reproducing apparatus according to claim 9, wherein the music reproducing apparatus is characterized.   11. The music playback device according to claim 10, wherein the image feature information and the music feature information are information in which the image or the music features are expressed numerically based on the same evaluation item. . It further has condition changing means for changing the predetermined condition according to the search result,
The music search means outputs a music associated with the music feature information determined to satisfy the changed condition as a search result. Music playback device. A playlist creating means for creating a playlist of the searched music;
The music reproducing device according to any one of claims 9 to 12, wherein the music in the created playlist is reproduced in the order of the songs in the playlist, randomly, or repeatedly. . A playlist display means for displaying the created playlist on a predetermined display;
The music playback device according to claim 13, wherein when the display is a touch panel, the touched music in the playlist is played back.   The music reproduction according to any one of claims 9 to 14, wherein the image acquisition means acquires an image when a user operation is performed, immediately after power-on, or whenever a predetermined time elapses. apparatus.   The music reproducing device according to any one of claims 9 to 15, wherein the image acquisition means is means for photographing a subject or scanning an image.   17. A predetermined code pattern that is photographed or scanned by the image acquisition means of the music reproduction device in which the image feature information calculated by the image feature information calculation means of the music reproduction device according to claim 16 is embedded. Article for music search characterized by having been attached.

Images