JP2011070046A - Information recording and reproducing device - Google Patents

Abstract

【Task】
The drive sound of a retractable zoom lens of a digital camera (control motor sound, sound of moving lens, etc.) makes a fairly loud sound. For this reason, it may be recorded as noise during audio recording or video recording. In order to avoid this noise, in many models, the optical zoom operation may be stopped during moving image shooting.
[Solution]
By updating the noise reduction filter when the AC power supply is connected, it is possible to reduce noise with high accuracy in response to aging. For this reason, noise can be reduced during recording of a moving image, so that an optical zoom can be used even during recording of a moving image, and user convenience can be improved.
[Selection] Figure 1

Description

  The present invention relates to an information recording / reproducing apparatus. For example, the present invention relates to a technology for reducing noise generated in a small mobile device such as a digital camera or a video camera, and timing thereof.

  As a background art of the present invention, for example, there is JP-A-2006-180392. In this publication, “[Problem] During normal operation, noise is removed from the target sound picked up by one microphone, and the target sound having a good S / N ratio is reproduced. [Solution] Two sound source separation learning devices are provided in advance. The first learning data composed of the band-by-band weight values for noise removal, which is an example using the signal collected by the microphone, and the acoustic feature amount for obtaining the band-by-band weight values that most closely approximate the first learning data The sound source separation device obtains a weight value for each band for noise removal using the acoustic feature amount of the type specified by the second learning data for the signal picked up by a single microphone. It calculates and multiplies each band-divided signal obtained by dividing the band-divided signal by the signal collected by a single microphone with the weight value for each band to remove the noise component "(summary).

  Another background art is, for example, Japanese Patent Application Laid-Open No. 2004-80788. The publication provides “[Problem] To provide a camera having audio noise reduction capability. [Solution] In one embodiment, the camera 100 is desirably captured by the microphone 110 and the microphone when recording audio. And an audio noise reduction system 218 configured to reduce no noise, hi some embodiments, the audio noise reduction system facilitates the reduction of noise generated by the camera motor 206. Other embodiments The noise reduction system facilitates the reduction of audio noise from the environment in which the camera is used. ”(Summary).

JP 2006-180392 A JP 2004-80788 A

  Currently, many digital cameras have one or more microphones to enable recording of audio as well as still images. There are also many models that can shoot movies. On the other hand, a zoom operation of an image is an essential function for improving the usability of photographing, and a retractable zoom lens is often used for downsizing and high magnification. However, the drive sound of the retractable zoom lens (control motor sound, lens moving sound, etc.) is considerably loud. Therefore, it may be recorded as noise during audio recording or video recording.

  In order to avoid this noise, in many models, the optical zoom operation may be stopped during moving image shooting. Some models can be substituted by realizing electronic zoom, but optical zoom is not possible in the sense of achieving high magnification. In any case, the user's usability will be significantly reduced.

  In order to be able to use an optical zoom during shooting of a moving image, it is necessary to prevent or reduce this noise.

  For example, as a noise reduction technique, there is a noise reduction technique using a plurality of microphones (Patent Document 1). This is to reduce noise by generating learning data using two voice input means, but it is assumed that performance will be degraded by noise and aging in a complicated environment such as in a housing. Absent. Since a microphone of a digital camera is often arranged inside a small casing, noise generated inside reaches the microphone as sound including reflection. Therefore, in order to generate a filter for improving the accuracy of noise reduction, it is necessary to consider the influence of reflection in the housing and aging. For example, in order to cope with aging, a technology is disclosed that generates noise based on that information by analyzing the noise every time the power is turned on, and realizes accurate noise reduction by applying the filter. (Patent Document 2). This is considered to be possible to reduce noise accurately by analyzing the noise every time the power is turned on based on secular changes, but if you try to update the filter every time the power is turned on, the filter will be generated every time the power is turned on. Therefore, processing is added and there is a disadvantage in terms of power. For mobile products such as digital cameras, battery life is important. It is required to reduce the update frequency as much as possible.

  Accordingly, an object of the present invention is to improve user convenience. For example, an apparatus capable of suppressing noise consumption during normal use as much as possible and reducing noise with high accuracy by performing filter update timing when charging or using an AC power supply is provided.

In order to realize the above device, for example,
In an information recording / reproducing apparatus capable of shooting video and audio,
Optical zoom lens means for varying the magnification of an optical image, a plurality of microphones, noise reduction means for reducing only noise from sound mixed with sound from outside the housing and noise generated inside the housing, and power supply It is composed of a power source determination means for determining the origin and a secular change detection means capable of detecting a secular change,
An apparatus is provided for checking whether or not there has been a secular change when the power source is an AC power source, and updating the information necessary for noise reduction in the housing if the secular change has occurred.

  According to the present invention, for example, in a digital camera, noise generated by a retractable zoom lens can be reduced while suppressing power consumption. For example, since noise can be further reduced, optical zoom can be used even during moving image recording, and user convenience can be improved.

Configuration diagram of this system Noise reduction block diagram Processing flow at power on Filter update processing flow Example of a display prompting the user

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a system configuration diagram for realizing the present invention. A system in which the present invention is applied to a digital camera equipped with a moving image shooting function will be described.

  In FIG. 1, reference numeral 100 denotes a retractable zoom lens. Reference numeral 101 denotes a sensor such as a CCD or CMOS that converts an optical signal obtained through the retractable zoom lens 100 into an electric signal. Reference numeral 102 denotes an A / D converter that converts an analog electric signal converted by the sensor 101 into a digital signal, and reference numeral 103 denotes a format that can compress the Mpeg of the digital signal converted by the A / D converter 102. A signal processing unit 104 that converts the format to NTSC or PAL at the time of reproduction, and 104 converts the video signal converted by the signal processing unit 103 into Mpeg, H.P. This is a video compression / decompression unit that performs H.264 compression. At the time of reproduction, the compressed signal is decompressed and sent to the signal processing unit 103. Reference numeral 110 denotes a microphone, 111 denotes a speaker, 112 denotes an AMP unit that amplifies an audio signal from the microphone 110, and 113 denotes A / D and D / D that convert the audio signal amplified by the AMP unit 112 into a digital signal. The A converter 150 is an NR unit that reduces noise from the audio signal digitized by the A / D and D / A converter 113, and 114 is a digital audio signal whose noise is reduced by the NR unit 150. This is an audio compression / decompression unit that performs Dolby compression and also decompresses compressed data. Reference numeral 105 denotes an LCD panel such as a liquid crystal monitor, and 120 multiplexes data output from the video compression / decompression unit 104 and the audio compression / decompression unit 114 into a system stream conforming to a standard such as Mpeg or AVCHD, or an element from the system stream. A demultiplexing unit 121 that performs separation into streams is a recording medium, and an SD card or a built-in semiconductor memory is assumed. There may be a plurality of media.

  Reference numeral 130 denotes a system control unit that controls the initialization of the system, the start and end timing of recording, and controls the entire system including noise control. Reference numeral 131 denotes an operation unit, which represents a zoom variable button, a button for instructing the start of a moving image or a still image, or an input unit required when the user gives an instruction. Reference numeral 132 denotes a power source determination unit that determines a power supply source connected to the main body, 133 denotes a battery, and 134 denotes an AC power source. Reference numeral 140 denotes an aging change detection unit that detects aging changes.

FIG. 2 is a block diagram of the inside of the NR unit 150.
200 and 204 are input and output connection terminals from the A / D and D / A converter 113, 202 and 203 are output and input connection terminals to the audio compression / decompression unit 114, and 201 is a selector. 210 is a noise reduction processing unit for reducing noise generated in the housing, 220 is a learning data generation unit for generating an appropriate filter, and 221 is a filter generation unit for generating an appropriate filter for noise reduction. It is.

FIG. 3 is a processing flow of the present invention.
S300 is a power activation process, S301 is a system initialization process for performing necessary initialization, S302 is a power supply determination process for determining a power supply source, and S303 undergoes secular change. If it is determined that it is, the process is a secular change process for updating the appropriate filter, and S304 indicates that the process is changed to the normal activation process.

FIG. 4 is a detailed process flow of the secular change process S303.
S400 is a secular change determination process for determining whether or not there is a secular change, S401 is a confirmation process for a user that prompts the user to check, and S402 is a secular change execution determination process for determining whether to perform a secular change, S403 represents a learning data generation process for generating a noise feature amount for generating an appropriate filter, and S404 represents a filter update process for updating an appropriate filter.

  FIG. 5 is an example of a GUI displayed on the LCD panel 105 of the main body in the confirmation process S401 to the user. Although not shown in the figure, it goes without saying that if an external monitor is connected, the display is also made there.

  Reference numeral 500 denotes a screen for displaying thumbnails of captured moving images and still images, 501 is a tab indicating thumbnails of the captured moving images, 502 is a tab indicating thumbnails of the captured still images, and 503 is , 504 is an example of a thumbnail, and 505 is an example of a GUI for checking with a user.

  First, an operation of recording noise of a retractable zoom lens at the time of moving image shooting with a digital camera will be described with reference to FIG.

  An optical signal input through the retractable zoom lens 100 is converted into an electric signal by the sensor 101 and becomes a signal digitized by the A / D converter 102. This digital signal is subjected to camera processing by the signal processing unit 103. When the user gives an instruction to record a moving image from the operation unit 131, the video compression / decompression unit converts the video signal processed by the signal processing unit to MPEG or H.264. H.264 compression is performed. On the other hand, the voice input from the microphone 110 is amplified by the AMP unit 112 that amplifies the voice signal, and converted into a digital signal by the A / D and D / A converter 113. The audio signal is compressed by Mpeg or Dolby in the audio compression / decompression unit 114 through the NR unit 150. Here, it is assumed that the NR unit 150 does not perform any particular processing.

  The multiplexed streams generated by the video compression / decompression unit 104 and the audio compression / decompression unit 114 are multiplexed by the multiplexing / separating unit 120 in accordance with the specifications of the recording medium 121. For example, an SD card may be compliant with AVCHD Lite. If a medium such as a semiconductor memory cannot be taken out, unique multiplexing may be used. This multiplexed system stream is recorded on the recording medium 121. What has been described above is a series of recording flows.

  In the case of still image shooting, focusing and exposure control are automatically adjusted when the shutter button on the operation unit is pressed halfway, and it is common to press the shutter button after the zoom position has already been determined. However, in the case of moving image shooting, it is possible to operate the zoom during recording. Here, if the zoom is operated during the moving image, specifically, if the user instructs the system to zoom in or zoom out through the operation unit 131, the microphone 110 operates the retractable zoom lens 100 together with the voice of the subject. Sound, so-called noise will be picked up together. That is, extra noise is recorded only while the zoom operation is instructed.

  As described above, since the noise of the retractable zoom lens is recorded together with the microphone during moving image shooting, in many digital cameras, the retractable zoom lens is not operated during moving image recording.

  Therefore, in order to remove this inconvenience, noise in the moving image is reduced.

Next, a method for reducing this noise will be described with reference to FIGS.
As described above, if a zoom noise occurs during video recording, the noise is also taken in from the microphone 110. Therefore, a sound in which the sound from the outside (of course, there may be no sound) and the noise are mixed. Is input to the NR unit 150 through the AMP 112 and the A / D and D / A converter 113. The mixed sound input to the NR unit 150 is input from the input terminal 200 of FIG. 2 and is input to the noise reduction processing unit 210 by connecting the selector 201 to the noise reduction processing unit 210 side. The selector 201 is controlled by an instruction from the system control unit 130. During normal moving image recording, the noise reduction processing unit 210 side is selected in advance. However, even in this case, since the operation timing of the zoom operation is known from the operation unit, the reduction processing is performed in the noise reduction processing unit 210 only when the zoom is operating, and the reduction processing is not performed at other times. Anyway. The reduction processing here is not particularly limited as long as it is an algorithm that reduces only noise from the mixed sound by generating a filter based on the characteristic amount of noise learned in advance. This learning data is a feature amount of noise (frequency information, information on how it is transmitted, etc.). Specifically, the selector 201 is switched to the learning data generation unit 220 at a timing other than the time of moving image recording, and only the retractable zoom lens is operated in the anechoic sound chamber in that state, and only the noise is captured from the microphone. Then, the learning data generation unit 220 generates a voice feature amount. A filter that reduces noise from the feature amount is generated by the filter generation unit 221. When the moving image is recorded using the filter generated in this way, the selector 201 is placed on the noise reduction processing unit 210 side, and the noise reduction processing unit 210 reduces only the noise. The learning data and the filter need only be generated once before shipment.

  As explained above, it is possible to reduce noise accurately by analyzing the noise in advance and calculating the feature amount, and it is only necessary to create the learning data and filter once. In some cases, the characteristic amount of noise may change due to the secular change of the retractable zoom lens. Therefore, it is good to be able to always learn noise data, etc., to cope with aging, but the calculation of feature values generally involves frequency conversion and complex matrix operations. There are many. For this reason, the processing time is increased and the power consumption is increased. If noise is learned when the power is turned on or turned off while the battery is driven, there is a problem that the retractable zoom cannot be moved during the learning, such as when the remaining battery level is low.

  In order to solve the above-described problem, the frequency of learning data generation is reduced as much as possible, and aging is handled when an AC power source is connected so that learning data can be generated safely. The connection of the AC power source here may be not only when driving the AC power source but also when charging. Specifically, when the main body is placed in the cradle, it is driven by AC power and may be charged at the same time, but it may be executed in either case. In particular, the charging method is not shown, but there is no particular concern as long as the charging method is within the common sense of the developers.

Next, aging processing for solving the above problem will be described with reference to FIGS.
When the power is turned on as in S300 of FIG. 3, initialization of internal blocks is executed (S301). The system controller 130, the retractable zoom lens 100, and other necessary initialization are performed. Thereafter, the power source determination unit determines which of the battery 133 and the AC power source 134 is connected (S302). If the battery is connected, it is normally activated (S304). If the AC power source is connected, an aging process is executed (S303). FIG. 4 is a detailed processing flow of S303. First, it is checked whether there is a secular change. This is checked by the secular change detection unit 140. For example, it is assumed that there is a secular change when a specified date has been set since shipment or when there is a clear physical change due to an impact. The specific detection means is not particular, and other methods may be used. (However, if the frequency of the learning data is not particularly concerned, the check is not particularly performed, and in the case of an AC power source, the processing after S401 may be executed.) From the result of the secular change detection unit 140, S400 If the determination is No, learning data is not generated and the process is exited. If there is a secular change (YES), the process does not immediately shift to the next process, and it is confirmed whether or not the user can cope with the secular change. In order to confirm whether the user really wants to execute it now, the GUI shown in FIG. 5 is displayed to ask the user for confirmation. If the user selects YES as shown in the display of 505, the generation of learning data is started (you may display that it is recommended to execute in a quiet environment as possible), but if If NO, the process exits (S402). In the case of YES in S402, the system control unit operates the retractable zoom lens 100 to capture the noise sound from the microphone, and connects the selector 201 of the NR unit 150 to the learning data generation unit 220 side. Then, the noise feature amount is calculated from the sound input to the learning data generation unit 220 (S403), and the filter generation unit 221 generates a noise reduction filter based on the feature amount. This filter is written in the memory inside the noise reduction processing unit 210 (S404). By doing so, the noise reduction processing unit 210 can use the filter corresponding to the secular change from the time of the next moving image recording to accurately reduce the noise.

  As described above, according to the present invention, it is possible to reduce noise generated by a retractable zoom lens in a digital camera while suppressing power consumption. Therefore, optical zoom can be used even during video recording. User convenience can be improved.

  As mentioned above, although preferred embodiment concerning this invention was described referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to this example. It is obvious for those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea described in the claims. It is understood that it belongs to. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.

DESCRIPTION OF SYMBOLS 100 ... Retractable zoom lens 101 ... Sensor 102 ... A / D converter 103 ... Signal processing part 104 ... Video compression / decompression part 105 ... LCD panel 110 ... Microphone 111- ..Speaker 112 ... AMP
113 ... A / D converter and D / A converter 114 ... audio compression / decompression unit 120 ... multiplexing / separation unit 121 ... recording medium 130 ... system control unit 131 ... operation unit 132: Power determination unit 133: Battery 134: AC power supply 140: Aging detection unit 150: NR unit 200: Connection terminal 201: Selector 202 ... Connection terminal 203 ... Connection terminal 204 ... Connection terminal 210 ... Noise reduction processing unit 220 ... Learning data generation unit 221 ... Filter generation unit S300 ... Power activation process S301 ... System initialization process S302 ... AC power supply determination process S303 ... Aging process S304 ... Normal activation process S400 ... Aging change determination process S401 ... Confirmation process to user S402 ... Aging change Implementation determination processing S403 ... Learning data generation processing S404 ... Filter update processing 500 ... Thumbnail display screen 501 ... Movie tab 502 ... Still image tab 503 ... Title 504 ... Thumbnail 505 ... Example of display prompting update

Claims (8)

In an information recording / reproducing apparatus capable of shooting video and audio,
An optical zoom lens device for varying the magnification of the optical image;
Multiple microphones,
A noise reduction device for reducing noise from sound mixed with sound from outside the housing and noise generated inside the housing;
A power supply determination device for determining a power supply source;
A secular change detection device capable of detecting secular change;
Is configured with
When the power supply source is an AC power source, it is determined whether or not there has been a secular change, and if there is a secular change, the noise reduction device updates information necessary for noise reduction inside the housing. An information recording / reproducing apparatus characterized by the above. In claim 1,
The information recording / reproducing apparatus characterized in that the necessary information is a characteristic amount of noise and is information necessary for generating a filter for realizing noise reduction. In claim 1 or 2,
The noise reduction device includes: a learning data generation device that generates the noise feature amount; a filter generation unit that generates a filter that achieves noise reduction; and a noise reduction processing unit that reduces noise from the mixed sound. An information recording / reproducing apparatus characterized by comprising. In any of claims 1 to 3,
The update of the necessary information is executed according to a user instruction, and a display for further prompting the instruction is performed. In any of claims 1 to 4,
The information recording / reproducing apparatus characterized in that the secular change detection device determines that a secular change has occurred when a specified time has passed or a physical impact has occurred. In any of claims 1 to 5,
An information recording / reproducing apparatus characterized in that, in the power determination apparatus, when the power source is a battery, the secular change process is not performed. In any one of Claims 1 thru | or 6.
An information recording / reproducing apparatus characterized in that the case where the power source is an AC power source includes a case where the apparatus main body is charged. In a digital camera capable of shooting not only still images but also moving images, a retractable zoom lens, a plurality of microphones, and a noise reduction device capable of reducing noise of the retractable zoom lens, It comprises a device that determines the power source and a device that can detect secular changes,
When it is determined that the power source is an AC power source at the time of power activation, it is determined whether a secular change has occurred, and if it is determined that a secular change has occurred, the user is prompted to give instructions. An information recording / reproducing apparatus that updates a filter necessary for noise reduction when instructed to perform the operation.

Images