JP2006229401A - Electronic camera, noise reduction device, and noise reduction control program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To store noise data without requiring a quiet environmental space. <P>SOLUTION: Factory workers set a setting mode of a noise eliminator/suppressor with an operation input, connect external equipment such as a data writer to an input/output connector, and set a writing mode from the external equipment with the operation input before shipment of digital cameras after completion. Then, the judgement of a step S101 becomes YES, an input/output interface circuit is started and connected with the external equipment such as the data writer (a step S103). A previously stored filter coefficient or a noise parameter such as a noise spectrum is received from the external equipment such as the data writer (a step S104). The received noise suppression parameter is set and stored in a memory of a noise elimination/suppression circuit, namely, a noise suppression parameter memory (a step S104). <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an electronic camera that reduces noise generated with the operation of a mounted motor, a noise reduction device for reducing noise, and a noise reduction control program.

Conventionally, for example, in a camera that records ambient sound detected by a microphone together with a photographed image, it is possible to prevent noise from being mixed into the recorded sound by reducing camera operation noise that occurs with the operation of a zoom motor or the like. A device equipped with a noise reduction device has been proposed. This noise reduction device detects camera operation noise in advance and stores it in the storage means in the adjustment process at the factory. Then, at the time of shooting by the user, the camera operation noise stored in advance is removed from the ambient sound detected by the microphone, thereby preventing the camera operation noise from being recorded (see, for example, Patent Document 1). .
Japanese Patent Laid-Open No. 11-327063

  In the case of a noise reduction device that stores camera operation noise in advance and performs noise reduction processing, in order to remove noise with high accuracy, camera operation is performed in an environment where no other noise is mixed in the adjustment process at the factory. It is essential to detect noise and store it in the storage means. However, in general, miscellaneous noise associated with the operation of various manufacturing equipment is always generated in factories. Therefore, in order to detect camera operation noise and store it in the storage means in an environment where no other noise is mixed, a quiet work space must be provided in the factory, and various noises are constantly generated. Establishing a quiet work space in the factory requires a lot of equipment costs.

  In addition, the operation noise generated from the drive mechanism that is a noise generation source is detected by the microphone because it changes over time due to wear frequency, usage time, or secular change of the motor or gear constituting the drive mechanism. The operating noise included in the ambient sound also changes over time. On the other hand, the stored operating noise is constant. Therefore, even if the stored constant operation noise is removed from the ambient sound including the operation noise that changes with time, the operation noise included in the ambient sound cannot be properly removed. Therefore, the noise removal performance decreases with time.

  The present invention has been made in view of such conventional problems, and can store noise data without requiring a quiet environment space, or can maintain proper noise removal performance over time. An object of the present invention is to provide an electronic camera, a noise reduction device, and a noise reduction control program.

  In order to solve the above-mentioned problem, in the electronic camera according to the first aspect of the present invention, an image pickup means for picking up a moving image, a detection means for detecting an ambient sound at the time of image pickup by the image pickup means, and an image pickup by the image pickup means Recording means for recording a moving image and ambient sound detected by the detecting means, operating means for generating noise in accordance with the operation, and storage means for storing data of noise generated in accordance with the operation of the operating means The noise reduction processing means for performing noise reduction processing on the ambient sound detected by the detection means during the operation of the operation means based on the noise data stored in the storage means, and an external device are connected. A connection means; an acquisition means for acquiring noise data transferred from the external device connected to the connection means; and a noise data acquired by the acquisition means. The and a writing means for writing in said storage means.

  In other words, in a factory where miscellaneous noise is constantly generated, when the ambient sound detected by the detection means is stored in the storage means for each operation of the operation means, the storage means is generated along with the operation of the operation means. In addition to the operation noise, miscellaneous noise is also mixed and stored. However, by connecting to an external device, acquiring noise data transferred from the external device, and writing it in the storage means, the noise data can be stored in the storage means without requiring a quiet environment space. .

  In the electronic camera according to the second aspect of the present invention, an imaging unit that captures a moving image, a detection unit that detects ambient sound during imaging by the imaging unit, the moving image captured by the imaging unit, and the Recording means for recording the ambient sound detected by the detecting means, operating means for generating noise in accordance with the operation, storage means for storing data of noise generated in accordance with the operation of the operating means, and the storage Based on the noise data stored in the means, noise reduction processing means for performing noise reduction processing on ambient sounds detected by the detection means during operation of the operation means, and stored in the storage means Updating means for updating noise data.

  Therefore, even if the noise caused by the operation of the operating means changes over time, the noise data is updated, and noise reduction processing is executed using the noise data corresponding to the time-varying noise. can do. Therefore, it is possible to maintain proper noise removal performance over time.

  According to a third aspect of the present invention, in the electronic camera according to the first aspect, the electronic camera further includes an updating unit that updates the noise data written in the storage unit by the writing unit. Therefore, as described above, noise data can be stored in the storage means without requiring a quiet environment space, and appropriate noise removal performance can be maintained over time.

  According to a fourth aspect of the present invention, in the electronic camera according to the first aspect, the writing means writes the noise data acquired by the acquiring means into the storage means to store the memory. Updating means for updating the noise data stored in the means; Therefore, as described above, noise data can be stored in the storage means without requiring a quiet environment space, and proper noise removal performance can be maintained over time using an external device. It becomes possible.

  According to a fifth aspect of the present invention, in the electronic camera according to the second aspect, the update means includes a connection means for connecting to an external device and the external device connected to the connection means. An acquisition unit that acquires the transferred noise data, and a writing unit that writes the noise data acquired by the acquisition unit into the storage unit. Therefore, even if the generated noise changes variously due to various noise data acquired from the external device, appropriate noise removal performance can be maintained over time.

  In the electronic camera according to the sixth aspect of the present invention, the noise data updated by the updating unit differs depending on a noise reduction method of the noise reduction processing unit. Therefore, appropriate noise data can be updated according to the noise reduction method of the noise reduction processing means.

  In the electronic camera according to claim 7, the noise data updated by the update means is at least one of a filter coefficient and an adaptive update algorithm when the noise reduction processing means is an adaptive filter system. In the case of a microphone array system, including at least one of an angle or a delay time in the direction of a noise source, in the case of a linear predictor, including an adaptive coefficient corresponding to the generated noise, and in the case of a spectral subtraction system, It includes at least one of spectral data of noise sources, estimated data thereof, and subtraction gain.

  In the electronic camera according to the eighth aspect of the invention, the updating means includes means for updating the noise data when the power is turned on and / or when the operation mode is switched. Therefore, the noise data is automatically updated when the power is turned on or every time the mode is switched.

  In the electronic camera according to the ninth aspect of the present invention, the updating means updates the noise data based on ambient sounds detected by the detecting means at power-on and / or operation mode switching time. Means to do.

  Further, in the electronic camera according to the invention of claim 10, the updating means includes an operation history storage means for storing an operation history of the operation means, and an operation history stored in the operation history storage means. The noise data is updated based on the above. Therefore, it is possible to update the appropriate noise data according to the noise that changes according to the operation history of the operation means.

  Further, in the electronic camera according to the invention of claim 11, the update means has an estimation means for estimating noise data based on the operation history stored in the operation history storage means. To update the noise data estimated by.

  In the electronic camera according to the twelfth aspect of the present invention, the updating unit updates the noise data so that the amplitude of the noise data increases as the operation amount of the operation unit increases with time. Do. That is, the noise generated with the operation of the operating means tends to increase with time, so that it is possible to update the noise data corresponding to the change of the operating noise with time.

  Further, in the electronic camera according to the invention of claim 13, the updating means shifts and updates the frequency component of the noise data as the operation amount of the operating means with time increases. Do. Therefore, the noise data can be updated by a simple process for shifting the frequency component.

  In the electronic camera according to the fourteenth aspect, the frequency component is shifted to the high frequency side. That is, since the frequency of the noise generated with the operation of the operating means tends to increase with time, it is possible to update the noise data corresponding to the frequency change of the operating noise with time.

  Further, in the electronic camera according to the invention of claim 15, the updating means extracts an extracting means for extracting a sound in a specific low frequency band from the ambient sound detected by the detecting means, and the extracting means extracts the sound. Based on the sound of the specified low frequency band, the aging change detecting means for detecting the aging change of the operating means, and updated to the noise data corresponding to the aging change detected by the aging change detecting means. . Therefore, it is possible to update the noise data corresponding to the actual noise change accompanying the secular change of the operation means.

  The electronic camera according to a sixteenth aspect of the present invention further includes instruction means for instructing update of the noise data, and the update means is configured to send the noise data in response to an instruction from the instruction means. Update. Therefore, the user can update the noise data at an arbitrary time.

  Further, in the electronic camera according to the invention of claim 17, the update means obtains the noise data based on the ambient sound detected by the detection means when the update is instructed by the instruction means. Update. Therefore, it is possible to update the noise data corresponding to the noise accompanying the operation of the operating means that is actually changing at the arbitrary time point.

  In the electronic camera according to claim 18, the update means sets different noise data according to a photographing scene or photographing mode that can be set in the electronic camera. Therefore, for example, in moving image shooting of a conversation scene, it is possible to further emphasize the voice in the human voice band and suppress the high-frequency component of noise that occurs with the operation of the operating means.

  The electronic camera according to claim 19 further comprises output means for outputting noise generated by the operation of the operation means detected by the detection means to the external device, wherein the acquisition is performed. The means acquires noise data generated by the external device based on the noise output from the output means.

  The electronic camera according to the invention of claim 20 further comprises a terminal for receiving image data from outside and an image data storage means for storing the image data received via the terminal. The connecting means is also used as the terminal. Therefore, it is not necessary to provide a separate connection means.

  Further, in the invention described in claim 21, in the noise reduction apparatus disposed in a device comprising a detecting means for detecting ambient sound and an operating means for generating noise in accordance with the operation, the operation of the operating means And a noise reduction process for ambient sounds detected by the detection means during operation of the operation means based on the noise data stored in the storage means. The noise reduction processing means for executing, the connection means for connecting to the external device, the acquisition means for acquiring the noise data transferred from the external device connected to the connection means, and the noise data acquired by the acquisition means Writing means for writing to the storage means.

  That is, as described above, in a factory where miscellaneous noise is constantly generated, when the ambient sound detected by the detection means is stored in the storage means for each operation of the operation means, the storage means The miscellaneous noise is mixed and stored together with the operation noise generated along with the operation. However, by connecting to an external device, acquiring noise data transferred from the external device, and writing it in the storage means, the noise data can be stored in the storage means without requiring a quiet environment space. .

  According to a twenty-second aspect of the present invention, there is provided a noise reduction apparatus disposed in a device including detection means for detecting ambient sounds and operation means for generating noise in accordance with the operation. Storage means for storing data of noise generated by the operation, and noise reduction for ambient sounds detected by the detection means during operation of the operation means based on the noise data stored in the storage means Noise reduction processing means for executing processing, and update means for updating noise data stored in the storage means.

  Therefore, even if the noise caused by the operation of the operating means changes over time, the noise data is updated, and noise reduction processing is executed using the noise data corresponding to the time-varying noise. can do. Therefore, it is possible to maintain proper noise removal performance over time.

  In the noise reduction control program according to the twenty-third aspect of the present invention, the detecting means for detecting the ambient sound, the operating means for generating noise accompanying the operation, and generated by the operation of the operating means. Storage means for storing noise data, and connection means for connecting a computer included in a device for reducing noise generated during operation of the operation means based on the noise data stored in the storage means to an external device And acquisition means for acquiring noise data transferred from the external device connected to the connection means, and writing means for writing the noise data acquired by the acquisition means to the storage means. Therefore, when the computer executes processing according to this program, the same effect as that attained by the 21st aspect can be attained.

  In the noise reduction control program according to the twenty-fourth aspect of the present invention, detection means for detecting ambient sound, operation means for generating noise in accordance with the operation, and the operation of the operation means are generated. Storage means for storing noise data, and a computer included in a device for reducing noise generated during operation of the operation means based on the noise data stored in the storage means are stored in the storage means. It functions as an updating means for updating the noise data. Therefore, when the computer executes processing according to this program, the same effect as that attained by the 22nd aspect can be attained.

  As described above, according to the inventions according to claims 1, 2, and 23, since the noise data transferred from the external device is connected to the external device and written in the storage means, a quiet environment space can be obtained. Noise data can be stored in the storage means without need. Therefore, it is not necessary to provide a quiet work space in the factory, and noise data can be stored at a low cost without requiring a large equipment cost.

  Further, according to the inventions according to claims 2, 22, and 24, even if the noise caused by the operation of the operating means changes with time, the noise data is updated, so that the noise changing with time Noise reduction processing can be executed using noise data according to the above. Therefore, it is possible to maintain proper noise removal performance over time.

  (First embodiment)

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a circuit configuration of a digital camera 100 to which an embodiment of the present invention is applied. The digital camera 100 has a general function such as AE, AWB, AF, and the like, and has an imaging lens 101 including a zoom lens and a focus lens. On the optical axis of the imaging lens 101, an imaging unit 103 including a diaphragm 102 and a CCD is disposed. The imaging lens 101 and the diaphragm 102 are driven by a driving unit 104 having a motor and the like.

  A control circuit 105 that controls the entire digital camera 100 is composed of a CPU, a work RAM, and the like. A driver 106 is connected to the control circuit 105 together with the driving unit 104, and the driver 106 drives the imaging unit 103 based on a timing signal generated by the control circuit 105. In addition, although illustration is abbreviate | omitted, the drive mechanism etc. which have a mechanical shutter and a motor for driving them are actually provided.

  An object is imaged by the imaging lens 101 on the light receiving surface of the imaging unit 103. The imaging unit 103 is driven by the driver 103 and outputs an analog imaging signal corresponding to the optical image of the subject to the unit circuit 107. The unit circuit 107 includes a CDS circuit that removes noise included in the output signal of the imaging unit 103 by correlated double sampling, an A / D converter that converts the imaging signal from which noise has been removed, to a digital signal, and the like. The imaging signal converted into digital is output to the video signal processing unit 108.

  The video signal processing unit 108 performs processing such as pedestal clamping on the input imaging signal, converts it into a luminance (Y) signal and a color difference (UV) signal, and performs auto white balance, contour enhancement, pixel interpolation, and the like. Performs digital signal processing to improve image quality. The YUV data converted by the video signal processing unit 108 is sequentially stored in the image memory 109, and in the REC through mode, every time one frame of data (image data) is accumulated, it is converted into a video signal for display control. The image is sent to the display unit 112 via the display memory 111 of the unit 110 and displayed on the screen as a through image.

  In the still image shooting mode, the control circuit 105 causes the imaging unit 103, the driver 106, the unit circuit 107, and the video signal processing unit 108 to be triggered by a shutter key operation provided in the operation input unit 121 described later. On the other hand, switching from the through image shooting mode to the still image shooting mode is instructed, and the image data obtained by the shooting processing in the still image shooting mode and temporarily stored in the image memory 109 is compressed by the image compression / encoding block 116. After being encoded and temporarily stored in the compressed / encoded image data memory 117, it is finally recorded in the external memory 114 via the memory IF 113 as a still image file of a predetermined format. In the moving image shooting mode, a plurality of image data sequentially stored in the image memory 109 is sequentially compressed by the image compression / encoding block 116 between the first shutter key operation and the second shutter key operation. After being sequentially stored in the compressed / encoded image data memory 117, it is recorded in the external memory 114 as a moving image file. The still image file and the moving image file recorded in the external memory 114 are read out to the image decompression / decoding unit 118 and decompressed and decoded in accordance with the user's selection operation in the PLAY mode, and are displayed as YUV data. Is displayed on the display unit 112.

  In the program memory 115, various programs for causing the control circuit 105 to control the respective units, for example, a program for controlling AE, AF, and AWB, a program for data communication, and further, the control circuit 105 includes the noise reduction of the present invention. Various programs such as a program for functioning as processing means, update means, acquisition means, and the like are stored. The data memory 119 stores user history data, noise suppression parameters, and update history data.

  A power switch 120 and an operation input unit 121 are connected to the control circuit 105 via an input circuit 122, and a rechargeable battery 123 such as a nickel metal hydride battery is connected via a power control circuit 124. Yes. The operation input unit 121 is provided with a plurality of operation keys and switches such as a mode selection key, a shutter key, and a zoom key.

  The digital camera 100 also has a recording function for recording ambient sounds in the moving image shooting mode, the recording mode for recording only sound, and the recording mode with sound (still image). A microphone 125 is provided. The audio signal from the microphone 125 is amplified by the amplifier 126 between the first shutter key operation and the second shutter key operation in the moving image shooting mode, sample-held and digitally converted by the S & H and ADC circuit 127, Noise suppression processing is performed by the noise removal / suppression circuit 128 and stored in the input voice memory 129. The audio data stored in the input audio memory 129 is sequentially compressed by the audio compression / encoding block 130 and sequentially stored in the compressed / encoded audio data memory 131. The control circuit 105 generates an audio-added moving image file including the compressed audio data and the compressed moving image data, and records it in the external memory 114. The audio data of the moving image file recorded in the external memory 114 is read out to the audio expansion / decoding unit 132 and expanded and decoded according to the user's selection operation in the PLAY mode. The decompressed and decoded audio data is temporarily stored in the output audio memory 133, converted to an analog signal by the D / A converter 134, and supplied to the speaker 137 via the LPF 135 and the amplifier 136. As played. Note that the timing of performing the audio recording is not limited to the time of moving image shooting, and may be during the recording operation in the still image shooting mode with audio, or during the recording operation in the recording mode or the after-recording mode.

  The digital camera 100 also includes an input / output connector 139 connected to the control circuit 105 via the input / output interface circuit 138. An external device 150 such as a data writer can be connected to the input / output connector 139. The control circuit 105 acquires a noise suppression parameter or the like input from the external device 150 such as the data writer, and the noise suppression parameter control block 140 stores the acquired noise suppression parameter or the like in the noise suppression parameter memory 141. Write. The noise removal / suppression circuit 128 includes a noise component calculation unit 142 and a subtraction circuit 143. The noise component calculation unit 142 calculates a noise component in the digital camera 100 and stores it in the noise suppression parameter memory 141 as shown in a flowchart described later. The subtraction circuit 143 subtracts the data stored in the noise suppression parameter memory 141 from the audio data input from the microphone 125 via the amplifier 136, S & H, and ADC circuit 127 and outputs the result to the input audio memory 129. .

  In the present embodiment having the above configuration, the control circuit 105 executes processing as shown in the flowchart of FIG. 2 based on the program. That is, the factory worker sets the setting mode of the noise removal / suppression unit by the operation input unit 121 before completion of the digital camera 100 and before shipping, and sets the input / output connector 139 to the external device 150 such as a data writer. Further, the operation input unit 121 sets the writing mode from the external device. Then, the determination in step S101 is YES, and the input / output interface circuit 138 is activated to connect to the external device 150 such as a data writer (step S103). Further, a noise coefficient such as a filter coefficient or a noise spectrum stored in advance from the external device 150 such as the data writer is received (step S104), and the received noise suppression parameter is received by the noise removal / suppression circuit 128. It is set and stored in the memory, that is, the noise suppression parameter memory 141 (step S104).

  The setting of the noise suppression parameter in the noise suppression parameter memory 141 may be performed by the user who purchased the digital camera 100 without being performed by a factory worker.

  When the user purchases after the digital camera 100 is shipped, if the user desires automatic update of the noise suppression parameter, the operation input unit 121 sets the noise removal / suppression unit setting mode. Keep it. When a mode other than the writing mode from the external device is set by the user, the determination in step S101 is NO. Accordingly, the process proceeds from step S101 to step S105, and it is determined whether or not the power is turned on or the mode is switched (step S105). If the power is turned on or the mode is switched, the drive unit 104 is stopped and ambient sound at the time of non-driving (quiet) is input from the microphone 125 (step S106). Ambient audio is temporarily stored in the RAM included in the control circuit 105 (step S107). Next, a noise source such as a lens driving motor included in the driving unit 104 is driven (step S108), and an operation sound during driving is input from the microphone 125 (step S109). Then, the quiet voice data is subtracted from the driving voice data and temporarily stored in the RAM as noise data (step S110). Further, a noise suppression parameter such as a filter coefficient or a noise spectrum is calculated according to the noise data (step S111), and the calculated noise suppression parameter is updated and set in the noise removal / suppression circuit 128, that is, the noise suppression parameter memory 141. (Step S112).

  Therefore, in the present embodiment, when the setting mode of the noise removal / suppression unit is set, the processing of steps S106 to S112 is executed when the power is turned on or when the mode is switched. Therefore, the noise removal / suppression unit When the setting mode is set, it is necessary to start the power supply or change the mode in a quiet environment where noise from the surroundings is not input to the microphone 125.

  On the other hand, if the result of determination in step S105 is not when the power is turned on or when the mode is not switched, it is determined whether or not voice is being input from the microphone 125 in the moving image recording / recording mode or the like (step S113). If the voice is being input, the voice is input from the microphone 125 (step S114), and the noise is removed or suppressed by the noise removal / suppression circuit 128 using the parameters set in the noise suppression parameter memory 141. (Step S115). Further, the input voice data whose noise is removed or suppressed is compressed / coded voice data into a voice data memory or a voice encoder, that is, the input voice memory 129 in this embodiment via a voice compression / coding block 130. The data is output to the memory 131 (step S116).

  In this embodiment, in step S105, it is determined whether the power is turned on or the mode is switched. However, only one of them may be determined. Further, an operation input unit 121 noise data update instruction switch may be provided, and it may be determined whether or not the update instruction switch has been operated in step S105.

  (Second Embodiment)

  FIG. 3 is a flowchart showing a processing procedure in the second embodiment of the present invention. In this flowchart, the processing in steps S201 to S204 is the same as the processing in steps S101 to S104 in FIG. When the user sets a mode other than the writing mode from the external device, the determination in step S201 is NO. Therefore, the process proceeds from step S201 to step S205, and it is determined whether or not the power is turned on or the mode is switched (step S205). If the power is turned on or the mode is switched, it is determined whether or not the usage history data of the drive unit 104 is equal to or greater than a predetermined value A (step S206).

  That is, every time the driving unit 104 is driven, the control circuit 105 stores the number of times of driving as “usage history data” in the data memory 119 as shown in FIG. Therefore, in step S206, it is determined whether or not the number of driving times indicated by the usage history data stored in the data memory 119 is equal to or greater than a predetermined value A. If the use history data is less than the predetermined value A as a result of the determination in step S206, the set noise removal parameter is continuously used (step S207). That is, the parameter stored in the noise suppression parameter memory 141 is not updated.

  However, if the drive count indicated by the usage history data is equal to or greater than the predetermined value A, it is further determined whether or not the drive count indicated by the usage history data is equal to or greater than the predetermined value B (where A <B) (step S208). ). As a result of this determination, if the number of times of driving is less than the predetermined value B, that is, greater than or equal to A and less than B, the noise suppression parameter memory 141 is set so as to increase the noise subtraction gain or the noise removal / suppression gain according to the usage history data. The stored data is corrected and set (step S209). As a result of the determination in step S208, if the number of driving times indicated by the usage history data is equal to or greater than the predetermined value B, the high frequency side determines the frequency band to be removed / suppressed by the noise removal / suppression circuit 128 according to the usage history data. The data stored in the noise suppression parameter memory 141 is corrected and set so as to shift (step S210).

  On the other hand, if the result of determination in step S205 is not when the power is turned on or mode is switched, processing similar to that in steps S113 to 116 in FIG. 2 is executed in steps S211 to S214.

  (Third embodiment)

  FIG. 4 is a flowchart showing a processing procedure in the third embodiment of the present invention. In this flowchart, the processing in steps S301 to S304 is the same as the processing in steps S101 to S104 in FIG. If the user sets a mode other than the writing mode from the external device, the determination in step S301 is NO. Therefore, the process proceeds from step S301 to step S305, and it is determined whether the power is turned on or the recording / shooting mode is switched (step S305). If it is during these switching times, a noise source such as a lens driving motor included in the driving unit 104 is driven (step S306), and an operation sound at the time of driving is input from the microphone 125 (step S307). Then, an audio signal in a predetermined frequency band is extracted from the input audio signal at the time of driving by a BPF (band pass filter) included in the control circuit 105 (step S308). Further, based on the extracted audio signal in the band, the usage time or wear frequency of the drive unit 104 such as a motor or gear, the degree of secular change, etc. are estimated (step S309). Further, a noise suppression parameter such as a filter coefficient or a noise spectrum is calculated according to the estimated usage time or wear frequency, aging, etc. (step S310), and the calculated noise suppression parameter is used as a noise removal / suppression circuit. Update setting to 128, that is, update setting to the noise suppression parameter memory 141 (step S311).

  On the other hand, if the result of determination in step S305 is NO, the same processing as in steps S113 to 116 in FIG. 2 is executed in steps S312 to S315.

  (Fourth embodiment)

  FIG. 5A is a flowchart showing a processing procedure of the external device 150 such as a data writer in the fourth embodiment of the present invention. That is, it is determined whether or not data writing to the digital camera 100 has been instructed (step A401). If not instructed, other processing is executed (step A402). If there is a writing instruction, the digital camera 100 is connected (step A403), and it is determined whether or not the connection is OK (step A404). If it is OK, as will be described later, an audio signal and a noise signal transmitted from the digital camera 100 side are measured or data is input (step A405), and a noise suppression parameter such as a noise spectrum or a filter coefficient is calculated. (Step A406). Then, the calculated noise spectrum, noise parameter, and the like are transmitted to the digital camera 100 (step A407), and after performing other writing processes (step A408), the connection is terminated (step A409).

  FIG. 5B is a flowchart showing a processing procedure on the digital camera 100 side in the embodiment. That is, it is determined whether or not the writing mode from the external device is set (step B401), and if it is not set, the process proceeds to control (2) described later. When the write mode is set, the external device 150 such as a data writer is connected (step B402), and it is determined whether or not the connection is OK (step B403). If OK, a noise source such as a lens driving motor included in the driving unit 104 is driven (step B404), and the operation sound of the built-in noise source is input from the microphone 125 (step B405).

  Further, the input voice or reference noise signal, the voice signal after noise suppression, and data such as the set parameters are transmitted to the external device 150 such as a data writer (step B406). In addition, the noise spectrum and noise suppression parameters transmitted from the external device 150 such as a data writer in step A407 are received, and the received noise spectrum and noise suppression parameters are set in the memory of the noise subtraction unit, that is, It is set and stored in the noise suppression parameter memory 141 (step B408). Further, the number of driving times stored in the data memory 119 (“use history data” in FIG. 1) and the update date and time (“noise suppression parameter update history” in FIG. 1) are updated (step B409), and other write data After performing the setting process (step B410), the connection is terminated.

  FIG. 6 is a flowchart showing a processing procedure of the control (2) on the digital camera 100 side. It is determined whether the power is turned on or the mode is switched (step B421). If the power is turned on or the mode is switched, the activation process and initial setting process of each unit are executed (step B422), and sound is input from the microphone 125 (step B423), and the ambient sound level is less than a predetermined level. (Step B424). When the ambient sound level becomes less than a predetermined level, a noise source such as a lens driving motor included in the drive unit 104 is driven (step B425), and the sound during driving is input from the microphone 125 (step B426). Then, the input speech is analyzed to calculate a noise spectrum and a noise suppression parameter (step B427), and the noise suppression parameters such as the calculated filter coefficient and noise spectrum are updated and set in the noise removal / suppression circuit 128, that is, noise. Update setting is made in the suppression parameter memory 141 (step B428).

  Next, it is determined whether or not the number of times of driving ("usage history data" in FIG. 1) or time from the previous update date and time stored in the data memory 119 is greater than or equal to a predetermined value (step B430). If it is greater than or equal to the predetermined value, a noise suppression parameter such as a noise spectrum stored in the noise suppression parameter memory 141 is corrected and set in accordance with usage history information such as the number of driving / shooting times and time (step B430). Further, a usage history such as the number of driving times and an update history such as the update date and time are recorded in the data memory 119 (step B431).

  If the determination in step B421 is NO, the operation input unit 121 determines whether manual updating of the noise suppression parameter is set (step B432). Steps B426 to B431 thus performed are executed. If not set, the voice is input from the microphone 125 (step B434), and noise is removed or suppressed from the input voice by the noise removal / suppression circuit 128 using the parameters set in the noise suppression parameter memory 141 (step B434). Step B435). Further, the input voice data whose noise is removed or suppressed is compressed / coded voice data into a voice data memory or a voice encoder, that is, the input voice memory 129 in this embodiment via a voice compression / coding block 130. The data is output to the memory 131 (step B436).

  Further, after performing other processing such as shooting, playback, and setting mode (step B437), the number of times of driving, shooting, and time are counted (step B438), and the data memory 119 is driven, the number of times of shooting, the time, etc. Usage history information is updated (step B439).

  (Fifth embodiment)

  FIG. 7 is a waveform diagram of operation sounds in the zoom operation of the drive unit 104. That is, when the drive unit 104 performs zoom driving, the operation sound in the time zone near the start of driving (A1 group), the operation sound from the start of driving until the end of driving (A2 group), and near the end of driving. As shown in the waveform diagrams (a) to (c), the operation sound is different from the operation sound (A3 group). Therefore, in the fifth embodiment, the noise spectrum and the noise suppression parameter set in the noise suppression parameter memory 141 are changed according to the time zone in the zoom operation of the drive unit 104.

  FIG. 8 is a flowchart showing a processing procedure in the present embodiment. In addition, although A1 group-A3 group used in this flowchart was shown in the said FIG. 7, illustration was abbreviate | omitted about B group and C group. That is, it is determined whether or not a voice input mode such as moving image shooting or recording mode has been set (step S501). If it has not been set, other mode processing is executed (step S502). If the voice input mode is set, the driving unit 104 determines whether or not the lens driving motor is operating (step S503). If it is not operating, the filter of the noise suppression parameter memory 141 is determined. A noise suppression parameter such as a coefficient or a noise spectrum is set to a parameter (group C) corresponding to the non-operation time (step S504).

  If the result of determination in step S503 is that the camera is operating, it is determined whether or not the zoom operation is in operation mode A (step S505). If the zoom operation is not being performed, the noise suppression parameter such as the filter coefficient or noise spectrum of the noise suppression parameter memory 141 is set to a parameter (group B) corresponding to an operation other than the zoom operation (step S506).

  If the result of determination in step S505 is that zooming is in progress, it is determined whether or not it is immediately after starting, that is, whether or not it is within a predetermined time after the start of driving (step S507). If it is not within the predetermined time after the start of driving, it is determined whether or not the stop process is being performed, that is, whether or not it is within the predetermined time before the driving is stopped (step S508). If the determinations in these steps S507 and S508 are both NO, it is a predetermined time after the start of driving and a predetermined time before the stop of driving. In this case, the noise suppression parameter such as the filter coefficient of the noise suppression parameter memory 141 or the noise spectrum is set to the corresponding parameter (group A2) (step S509).

  If the determination in step S508 is YES and the stop process is in progress and within a predetermined time before the drive is stopped, the noise suppression parameter such as the filter coefficient of the noise suppression parameter memory 141 or the noise spectrum is supported. Is set to the parameter (group A3) to be executed (step S510). Further, when the determination in step S507 is YES, and immediately after the start and within a predetermined time after the start of driving, the parameter corresponding to the noise suppression parameter such as the filter coefficient of the noise suppression parameter memory 141 or the noise spectrum. Set to (A1 group) (step S511).

  Thereafter, the voice is input from the microphone 125 (step S512), and noise is removed or suppressed from the input voice by the noise removal / suppression circuit 128 using the parameters set in the noise suppression parameter memory 141 (step S513). . Further, the input voice data whose noise is removed or suppressed is compressed / coded voice data into a voice data memory or a voice encoder, that is, the input voice memory 129 in this embodiment via a voice compression / coding block 130. The data is output to the memory 131 (step S514). Furthermore, the process proceeds to other processing in each mode (step S515).

  (Sixth embodiment)

  FIG. 9 is a flowchart showing a processing procedure in the sixth embodiment of the present invention. That is, it is determined whether or not a voice input mode such as moving image shooting or recording mode has been set (step S601), and if not, other mode processing is executed (step S602). If the voice input mode is set, the driving unit 104 determines whether or not the lens driving motor is operating (step S603). If it is not operating, the filter of the noise suppression parameter memory 141 is determined. A noise suppression parameter such as a coefficient or a noise spectrum is set to a parameter (group C) corresponding to non-operation (step S604).

  If the result of determination in step S603 is that it is operating, it is determined whether or not it is in drive mode 1, that is, whether or not it is within a predetermined time after the start of driving (step S605). If it is not within the predetermined time after the start of driving, it is determined whether or not the driving mode 3 is set, that is, whether or not within the predetermined time before the driving is stopped (step S606). If both the determinations in steps S605 and S606 are NO, it is a predetermined time after the start of driving and a predetermined time before the stop of driving. In this case, a predetermined parameter (group A) is set as a noise suppression parameter such as a filter coefficient of the noise suppression parameter memory 141 or a noise spectrum (step S607).

  If the determination in step S606 is YES and the stop process is being performed and within a predetermined time before the drive is stopped, the noise suppression parameter such as the filter coefficient of the noise suppression parameter memory 141 or the noise spectrum is The removal / suppression band of the setting parameter (group A) is set by shifting to the high frequency side (step S608). If the determination in step S607 is YES and it is immediately after the start and within a predetermined time after the start of driving, the setting is made as a noise suppression parameter such as a filter coefficient of the noise suppression parameter memory 141 or a noise spectrum. The parameter (group A) removal / suppression band is shifted to the low frequency side and set (step S609).

  Thereafter, the voice is input from the microphone 125 (step S610), and noise is removed or suppressed from the input voice by the noise removal / suppression circuit 128 using the parameters set in the noise suppression parameter memory 141 (step S611). . Further, the input voice data whose noise is removed or suppressed is compressed / coded voice data into a voice data memory or a voice encoder, that is, the input voice memory 129 in this embodiment via a voice compression / coding block 130. The data is output to the memory 131 (step S612). Further, the process proceeds to other processing in each mode (step S613).

  (Seventh embodiment)

  FIG. 10 is a flowchart showing a processing procedure according to the seventh embodiment of the present invention, in which shooting is performed by setting shooting conditions corresponding to shooting scenes according to shooting scene-specific programs and shooting mode selection operations. In addition, the noise suppression characteristics, the noise spectrum, the filter characteristics, and the like are changed according to the shooting scene. That is, it is determined whether or not the recording mode is set (step S701), and if not set, it is further determined whether or not the shooting mode with sound is set (step S702), and the shooting mode with sound is set. Is not set, the process in the other mode is executed (step S703).

  If the shooting mode with sound is set, it is determined whether a person shooting scene or a conversation shooting scene is selected (step S704). When a person photographing scene or a conversation photographing scene is not selected, a normal parameter group for suppressing operation sound is set as a noise suppression parameter such as a filter coefficient of the noise suppression parameter memory 141 or a noise spectrum. (Step S705).

  If the recording mode is set as a result of the determination in step S701, and if the person shooting scene or the conversation shooting scene is selected as a result of the determination in step S704, the filter of the noise suppression parameter memory 141 is used. As a noise suppression parameter such as a coefficient or a noise spectrum, the human voice band is set to be emphasized more than usual, and the high frequency band of the operation sound is set to be suppressed more than usual (step S706).

  Thereafter, the voice is input from the microphone 125 (step S707), and noise is removed or suppressed from the input voice by the noise removal / suppression circuit 128 using the parameters set in the noise suppression parameter memory 141 (step S708). . Further, the input voice data whose noise is removed or suppressed is compressed / coded voice data into a voice data memory or a voice encoder, that is, the input voice memory 129 in this embodiment via a voice compression / coding block 130. The data is output to the memory 131 (step S709). Furthermore, the process proceeds to other processing in each mode (step S710).

  (Eighth embodiment)

  FIGS. 11 to 19 show an embodiment showing a specific configuration of the noise removal / suppression means. That is, in the eighth embodiment shown in FIG. 11, the digital camera 200 is provided with a camera control unit 201 that controls each unit. The camera control unit 201 includes a CPU, a ROM storing a program, a work RAM, and the like. The camera control unit 201 stores key usage history data and the above-described update history data provided in the power switch 202 and other operation unit 203. It has a function to memorize.

The digital camera 200 also includes a main microphone 204 and a reference microphone 205. The main microphone 204 detects a sound s (n) to be recorded in a moving image shooting mode or the like. The sound s (n) is passed through an amplifier 206, an S & H, and an ADC circuit 207, and is a noise removal / suppression circuit. 222 is input. On the other hand, the reference microphone 205 detects noise W _{s (n)} from the noise source 209 that generates noise by the operation of the drive circuit 208 that drives the lens unit under the control of the camera control unit 201. The noise W _{s (n)} is input to the noise removal / suppression circuit 222 via the amplifier 210, the S & H, and the ADC circuit 211.

  The digital camera 200 also includes a USB or external input / output terminal 223. The USB or external input / output terminal 223 is connected to the input / output interface circuit 224, and the input / output interface circuit 224 is connected to the camera control unit 201 and the control unit 225 of the noise removal / suppression circuit. The control unit 225 of the noise removal / suppression circuit includes a voice / noise data transmission unit 226 and a noise removal / suppression parameter setting control unit 227. From the noise removal / suppression parameter setting control unit 227, the noise estimation parameter 228, the filter coefficient update signal 229, and the gain control signal 230 are output to the noise removal / suppression circuit 222.

  Further, the noise removal / suppression circuit 222 is provided with a noise analysis / noise estimation unit 231, a digital filter 232, a subtraction gain unit 233, a multiplier 234, and a subtraction circuit 235, and the output from the subtraction circuit 235 is an audio memory. Alternatively, it is configured to be output to the speech encoder 236. The noise analysis / noise estimation unit 231 analyzes and estimates noise based on the noise estimation parameter 228, S & H, and the observed speech X (n) = s (n) + W (n) input from the ADC circuit 207, This is output to the digital filter 232. The digital filter 232 extracts a noise signal component based on the output from the noise analysis / noise estimation unit 231, the observed speech X (n) = s (n) + W (n), and the filter coefficient update signal 229. The multiplier 234 multiplies the noise signal component extracted by the digital filter 232 by a gain based on the gain control signal 230 and outputs the noise signal y (n) to the subtraction circuit 235. The subtraction circuit 235 subtracts y (n) from the observed speech X (n) = s (n) + W (n) and outputs the result to the speech memory or speech encoder 236.

  On the other hand, an external device 250 such as a data writer connected to the external input / output terminal 223 is provided with an input / output interface circuit 251, a communication connection control unit 252, an audio / noise data receiving unit 253, A voice / noise data analysis unit 254 is provided. The communication connection control unit 252 controls the input / output interface circuit 251 and the voice / noise data reception unit 253, and the voice / noise data reception unit 253 is transmitted from the voice / noise data transmission unit 226 of the digital camera 200. Receiving voice / noise data. The voice / noise data analysis unit 254 analyzes the received voice / noise data to generate a noise estimation parameter 255, filter coefficient setting data 256, and subtraction gain setting data 257, and these parameters and data are input. It is configured to be transmitted to the noise removal / suppression parameter setting control unit 227 of the digital camera 200 via the output interface circuit 251 and the USB or external input / output terminal 223.

  In such an embodiment, in the final manufacturing process of the digital camera 200, the digital camera 200 is connected to an external device 250 such as a data writer, and voice / noise data is transmitted from the voice / noise data transmission unit 226 of the digital camera 200. The external device 250 is input. Then, noise estimation parameter 255, filter coefficient setting data 256, and subtraction gain setting data 257 are obtained by analysis of voice / noise data analysis unit 254 of external device 250, and these are obtained as noise removal / suppression parameters of digital camera 200. The setting control unit 227 is initially set and shipped. Data in the noise removal / suppression parameter setting control unit 227 is automatically updated by executing the processing shown in each flowchart described above in the digital camera 200 that is being used by the user purchased after shipment. .

  In the manufacturing process, since an inspection device or an adjustment device having a high processing capacity can be used as the external device 250, initial settings of a large number of digital cameras 200 can be performed at high speed and with high accuracy. In addition, since the operation sound, vibration, acoustic characteristics, etc. of each digital camera 200 and five can be measured with high accuracy, the noise characteristics can be analyzed in detail, and the optimum noise suppression parameters can be calculated and set automatically. It is possible to easily cope with variations in the digital camera 200 and even when the same module is changed to a housing or mounting method having a different shape or acoustic characteristic.

  On the other hand, the update in the digital camera 200 after shipment involves voice input when the user turns on the power and starts using the digital camera 200 or by switching the mode from the process according to the flowchart. Since the noise suppression parameter is updated and set every time video recording, recording, still image shooting with sound, or the like is used, it is possible to cope with changes in the characteristics of the operating sound due to wear of the drive unit or aging. Further, by using a personal computer as the external device 250, the user can make adjustments and update settings using application software for adjustment setting of the noise suppression circuit.

  (Ninth embodiment)

  FIG. 12 shows a ninth embodiment of the present invention, which is a specific configuration of the noise removal / suppression means, and uses an adaptive filter type noise canceller. That is, the digital camera 300 is provided with a camera control unit 301 that controls each unit. The camera control unit 301 includes a CPU, a ROM storing a program, a work RAM, and the like. The usage history data of the keys provided in the power switch 302 and the other operation unit 303 and the update history data described above are stored. It has a function to memorize.

The digital camera 300 also includes a main microphone 304 and a reference microphone 305. The main microphone 304 detects the sound s (n) to be recorded in the moving image shooting mode or the like. The sound s (n) is passed through the amplifier 306, the S & H, and the ADC circuit 307, and the noise removal circuit unit 322. Is input. On the other hand, the reference microphone 305 detects noise W _{s (n)} from the noise source 309 that generates noise by the operation of the drive circuit 308 that is controlled by the camera control unit 301 and drives the lens unit. The noise W _{s (n)} is input as a reference signal to the noise removal circuit unit 322 via the amplifier 310, the S & H, and the ADC circuit 311.

  The digital camera 300 also includes a USB or external input / output terminal 323. The USB or external input / output terminal 323 is connected to the input / output interface circuit 324, and the input / output interface circuit 324 is connected to the camera control unit 301 and the adaptive filter control unit 335. The adaptive filter control unit 335 includes a noise data analysis unit 326 and a filter coefficient update unit 327, and the filter coefficient is output from the filter coefficient update unit 327 to the noise removal circuit unit 322. Yes.

  The noise removal circuit unit 322 is provided with a delay unit 331, an adaptive filter 332, and a subtraction circuit 335, and is configured such that the emphasized speech from the subtraction circuit 335 is output to a speech memory or speech encoder 336. ing. The adaptive filter 332 filters the reference signal based on the filter coefficient, and the subtraction circuit 335 subtracts the adaptive filter 332 from the output from the delay unit 331 and outputs the result to the speech memory or speech encoder 336.

  On the other hand, an external device 350 such as a data writer connected to the external input / output terminal 323 is provided with an input / output interface circuit 351, a communication connection control unit 352, and an audio / noise data analysis unit 354. Is provided. The communication connection control unit 352 controls the input / output interface circuit 351. The voice / noise data analysis unit 354 analyzes the received voice / noise data to generate filter coefficient setting data 356, and the filter coefficient setting data 356 is input to the input / output interface circuit 351 and the USB or external input. The filter coefficient update unit 327 of the digital camera 300 is transmitted to the digital camera 300 via the output terminal 323.

In this embodiment, the main microphone 304 receives the desired audio signal s (n) and the path h _k (m) from the noise source 309, and is s (n) + w (n ) Is entered. The noise signal W (n) is expressed by the following equation using the impulse response {h _k (m)} (m = 1, 2,..., P−1) of the noise path.
_{w (n) = Σ m h} k (m) w s (n-m),

Further, the output y (n) of the adaptive filter 332 is expressed by the following equation when the impulse response of the adaptive filter 332 is {h _f (m)} (m = 1, 2,..., P−1).
_{y (n) = Σ m h} f (m) w s (n-m),
At this time, the output s ^ (n) of the noise canceller is
s ^ (n) = s ( n) + w (n) -y (n) = s (n) + Σ m {h k (m) -h f (m)} w s (n-m) (m = 1 , 2 ... P-1)
Therefore, if h _f (m) = h _k (m) can be obtained, s (n) = s (n), and the noise signal can be removed to extract only the audio signal.
Normally, to determine the unknown noise path h _k (m), the adaptive filter coefficient h _f (m) is updated to statistically minimize the square value of the estimation error s ^ (n). . Since statistic E [s ⁻² (n)] = E [{s (n) + w (n) −Σ _m {h _f (m)} w _s (n−m)} ² ] The point that minimizes this is the point at which the slope of E [s ⁻² (n)] becomes zero. However, in order to obtain the optimal amount of h _f (k) where δE [s ⁻² (n)] / δh _f (k) = 0, it is necessary to solve the P-element simultaneous equations, and the statistic of the signal is It is necessary and usually difficult to get in real time. For this reason, the adaptive filter 332 needs an adaptive algorithm such as the LSM (least mean square) method or the NLMS (normalized least mean square) method in order to learn statistics and search for an optimal solution sequentially.

However, in the case of the operation sound generated by the lens drive motor, the initial value of the optimum value of h _f (k) can be determined by measuring the statistic of the operation sound in advance. In such a noise canceller, even if the path from the noise source to the main microphone 304 is unknown, if the impulse response of the noise path can be satisfactorily estimated by the adaptive filter 332, noise removal can be performed and the noise characteristics fluctuate. There is an advantage that can follow. However, noise removal becomes difficult if crosstalk occurs in which the input sound flows into the reference microphone 305. In a small device such as a small camera, only a noise signal is input to the reference microphone 305, and the mounting method of the reference microphone 305 is devised so that there is no audio input, or when the crosstalk is large, a delay device or output After that, it is necessary to add a distortion correction filter or the like.

FIG. 13 shows an example of an adaptive filter used in the noise removal circuit unit 322. An adaptive IIR (infinite length impulse response) filter and an adaptive algorithm (adaptive update equation) {α ₁ (k + 1) = α ₁ ( An example of k) −μy (k) ψ (k)} is shown. What is necessary is just to obtain | require an initial value and update values, such as (alpha) 0 of this filter coefficient, (alpha) ₁ (k + 1), and update and set it as a noise suppression parameter.

  Further, there is an adaptive notch filter as a circuit that removes narrow band noise having a relatively narrow frequency band component such as motor noise. FIG. 14A shows an example in which the adaptive IIR filter of FIG. 13 is configured with an adaptive notch filter having a transfer function of H (z), and FIG. 14B shows the transfer function of G (z). An example of an adaptive notch filter is shown. Of course, you may comprise by the adaptive filter of other structures, such as a FIR filter and a lattice type filter.

  (Tenth embodiment)

  FIG. 15 shows a tenth embodiment of the present invention, which is a specific configuration of noise removal / suppression means, and is an example using a microphone array as an example of a noise suppression circuit using a plurality of microphones. . That is, the digital camera 400 is provided with a camera control unit 401 that controls each unit. The camera control unit 401 includes a CPU, a ROM storing programs, a work RAM, and the like. The camera control unit 401 stores the key usage history data provided in the power switch 402 and other operation units 403 and the above-described update history data. It has a function to memorize.

  In addition, the digital camera 400 includes a main microphone 404 and a reference microphone 405 that are arranged at an interval d. The main microphone 404 detects the sound s (n) to be recorded in the moving image shooting mode or the like, and this sound s (n) is input to the delay device 421 via the amplifier 406, the S & H, and the ADC circuit 407. Is done. On the other hand, the output from the reference microphone 405 is input to the multiplier 422 via the amplifier 410, the S & H, and the ADC circuit 411. The output of the multiplier 422 and the output of the delay unit 431 are input to the subtracter 440, and the output of the subtracter 440 is

  The digital camera 400 also includes a USB or external input / output terminal 423. The USB or external input / output terminal 423 is connected to the input / output interface circuit 424, and the input / output interface circuit 424 is connected to the camera control unit 401, the delay amount control unit 437, and the subtraction gain control unit 430. The delay amount control unit 437 and the subtraction gain control unit 430 are connected to the noise data analysis unit 426. Further, a subtracter 435 is provided at the subsequent stage of the multiplier 422, and an output from the subtracter 435 is provided to a speech memory or speech encoder 436.

  On the other hand, an external device 450 such as a data writer connected to the external input / output terminal 423 is provided with an input / output interface circuit 451, a communication connection control unit 452, a noise analysis unit 454, a thickness sound source direction. A θ estimation unit 455, a delay amount setting unit 456, and a subtraction gain setting unit 457 are provided.

  In such a configuration, in the microphone array according to the present embodiment, unlike the above-described embodiment, it is assumed that both the desired voice and the noise enter each microphone 404, 405, and the two microphones 404, 405 are used. By using the time difference of arrival of the signal to the noise, a blind spot in the arrival direction of the noise is formed to remove the noise. In order to determine the delay amount of the delay unit 431, it is necessary that the direction θ of the noise source is known. When the noise source is fixed in the digital camera 400 such as a lens driving motor, the noise source direction θ or the delay amount D (= dsin θ / c, d: microphone interval, C: sound velocity) is set as a noise parameter. do it. In addition, in the case of a configuration in which the direction of the motor drive unit or noise source changes, the shape design of the housing, or the model that changes the position of the microphone 404 or the motor, these noise source directions or delay amounts are updated as appropriate. Can be set. Alternatively, an adaptive filter may be provided for the outputs of the microphones 404 and 405, and all the outputs may be added to form an adaptive microphone array that adaptively sets the noise source direction.

  (Eleventh embodiment)

  FIG. 16 shows an eleventh embodiment of the present invention, which is a specific configuration of the noise removal / suppression means, and is an example configured with a noise suppression circuit using a linear predictor (LPC). That is, the digital camera 500 is provided with a camera control unit 501 that controls each unit. The camera control unit 501 includes a CPU, a ROM storing a program, a work RAM, and the like. The usage history data of the keys provided in the power switch 502 and other operation unit 503 and the above-described update history data are stored in the camera control unit 501. It has a function to memorize. A driving unit 508 serving as a noise source 509 is connected to the camera control unit 501.

  The digital camera 500 includes a microphone 504. The microphone 504 detects sound s (n) to be recorded in the moving image shooting mode or the like, and this sound s (n) is input to the adder 521 via the amplifier 506, the S & H, and the ADC circuit 507. And input to a linear predictor (LPC) 522. The linear predictor 522 empty prediction signal y (n) is input to the adder 527, and the prediction error signal e (n) = x (n) −y (n) from the adder 521 is also a multiplier. An adder 527 is provided via 528, and an output from the adder 527 is provided to a speech memory or speech encoder 536.

  The digital camera 500 also includes a USB or external input / output terminal 523. The USB or external input / output terminal 523 is connected to an input / output interface circuit 524, and the input / output interface circuit 524 is connected to a camera control unit 501 and an adaptive count setting unit 530. A prediction error signal analysis unit 526 is connected.

  On the other hand, an external device 550 such as a data writer connected to the external input / output terminal 523 is provided with an input / output interface circuit 551, a communication connection control unit 552, an audio / noise data analysis unit 554, In addition, an initial setting data section 557 for adaptive counting is provided.

  In such a configuration, the linear predictor (LPC) 522 is a circuit that examines the relationship between the past signal and the current signal and expresses the characteristics of the input signal. Therefore, if the characteristic of noise varies relatively slowly and the linear predictor (LPC) 522 is also operated slowly, the linear predictor that does not react to speech with a large characteristic variation and expresses only the characteristic of noise. (LPC) 522, which can be used for noise removal and noise suppression.

FIG. 17A shows an example of the configuration of a linear predictor (LPC) 522, and this linear predictor (LPC) 522 includes a plurality of delay units 537, multipliers 538, and adders 539, respectively. ing. Then, the adaptive coefficients and initial values of the linear predictor 522 such as h ₁ (n), h ₂ (n)... H _M (n) shown in the figure are set or updated as noise suppression parameters. Thereby, the audio signal x (n) including noise illustrated in FIG. 18 can be output as the audio signal y (x) in which the noise illustrated in FIG. 18B is suppressed. As shown in FIG. 17B, linear predictors 522 may be cascade-connected.

  (Twelfth embodiment)

  FIG. 19 shows a twelfth embodiment of the present invention, which is a specific configuration of the noise removal / suppression means, and an example using a spectral subtraction method (spectral subtraction method) (hereinafter referred to as SS method). It is. That is, the digital camera 600 is provided with a camera control unit 601 that controls each unit. The camera control unit 601 includes a CPU, a ROM storing a program, a work RAM, and the like. The usage history data of the keys provided in the power switch 602 and other operation units 603 and the update history data described above are stored. It has a function to memorize. The camera control unit 601 is connected to a drive unit 508 serving as an in-device noise source 609.

  The digital camera 600 also includes a microphone 604. The microphone 604 detects the sound s (n) to be recorded in the moving image shooting mode or the like. The sound s (n) is input to the spectrum subtraction unit 622 via the amplifier 606, the S & H, and the ADC circuit 607. Is done.

  The digital camera 600 also includes a USB or external input / output terminal 623. The USB or external input / output terminal 623 is connected to the input / output interface circuit 624, and the input / output interface circuit 624 is connected to the spectrum subtraction unit 622. A spectrum subtraction unit 622, a parameter setting control unit 626, a subtraction gain setting unit 627, a noise spectrum estimation unit 628, a multiplier 629, and a subtractor 630 are provided. The spectrum subtraction unit 622 includes a window function unit 631, a Fourier transform unit 632, a phase unit 633, an inverse Fourier transform unit 634, and a superposition addition unit 635, and an output from the superposition addition unit 635 is a speech memory or a speech encoder 636. It is configured to be output to.

  On the other hand, an external device 650 such as a data writer connected to the external input / output terminal 623 is provided with an input / output interface circuit 651, a communication connection control unit 652, and an audio / noise data analysis unit 654. Is provided. The communication connection control unit 652 controls the input / output interface circuit 651 and the voice / noise data analysis unit 654. The voice / noise data analysis unit 654 analyzes the received voice / noise data to generate noise power spectrum estimation data 655 and subtraction gain setting data 657, and these parameters and data are input / output interface circuit 651 and USB or It is configured to be transmitted to the subtraction gain setting unit 627 of the digital camera 600 via the external input / output terminal 623.

In such an embodiment, in the SS method, the signal x (n) = s (n) + w (n) of the input audio signal including the audio signal s (n) and the noise signal w (n) is obtained every predetermined sampling. The frame is divided and subjected to windowing with a window function such as a Hanning window or a trapezoidal window, and then converted from the time domain to the frequency domain by Fourier transform (FFT). The power spectrum of estimated noise | X ^ (ω) | is subtracted from the amplitude power spectrum | X (ω) | of one input signal (| S ^ (ω) | = | X (ω) |-| X ^ ( ω) |) and ω _x of the input signal are added thereto, and the obtained S ^ (ω) = | S ^ (ω) | exp (jω _x ) is converted into the time domain by inverse Fourier transform (inverse EET). Thus, an enhanced speech signal s ^ (n) from which noise such as operation sound is removed is obtained.

Considering noise removal by the SS method as a filter of the transfer function H (ω), the transfer function H (ω) is
H (ω) = S ^ (ω) / X (ω) {| X ^ (ω) | − | X ^ (ω) |} exp (jω _x ) X (ω),
H (ω) = 1− {| X ^ (ω) | / | X (ω) |}.

  In the SS method, processing is not performed on phase information that is not very important for human hearing, and processing is simple because processing is performed mainly with amplitude information. Moreover, noise can be suppressed with only one microphone, and it is not necessary to know the number of noises in advance, but at least a processing delay of one frame occurs. In addition, prior information on the noise power vector is required. In cellular phones and the like, the SNR (SNR) for each subband of the signal converted to the frequency domain is calculated, non-adaptive noise estimation is performed, and spectral subtraction (difference) and suppression (multiplication) by spectral gain are performed. Or a method of performing noise estimation adaptively by weighting the power vector of the input signal so that it is inversely proportional to the SNR estimate, and performing noise suppression only by adjusting (multiplying) the spectrum gain, etc. However, complicated noise estimation methods are being studied, but in order to remove the motor operating noise in the equipment, the noise spectrum data | W ^ (ω) | There is an advantage that it becomes easy to use.

  In the embodiment, the present invention is applied to a digital camera. However, the present invention is not limited to the camera, but is applied to various devices having a hard disk driven by a drive source, various recording devices, and the like. You can also.

1 is a block diagram illustrating an electrical configuration of a digital camera to which an embodiment of the present invention is applied. It is a flowchart which shows the process sequence in the 1st Embodiment of this invention. It is a flowchart which shows the process sequence in the 2nd Embodiment of this invention. It is a flowchart which shows the process sequence in the 3rd Embodiment of this invention. It is a flowchart which shows the process sequence in the 4th Embodiment of this invention. It is a flowchart which shows the process sequence of control (2) in the embodiment. It is a wave form diagram of the operation sound in the zoom operation of a drive part. It is a flowchart which shows the process sequence in the 5th Embodiment of this invention. It is a flowchart which shows the process sequence in the 6th Embodiment of this invention. It is a flowchart which shows the process sequence in the 7th Embodiment of this invention. It is a principal part block diagram which shows the 8th Embodiment of this invention. It is a principal part block diagram which shows the 9th Embodiment of this invention. It is a circuit diagram which shows the example of the adaptive filter in the same embodiment. It is a figure which shows an example of an adaptive notch filter. It is a principal part block diagram which shows the 10th Embodiment of this invention. It is a principal part block diagram which shows the 11th Embodiment of this invention. It is a circuit diagram which shows the structural example of a linear predictor. It is a wave form diagram which shows the audio | voice signal containing noise and the audio | voice signal which suppressed noise. It is a principal part block diagram which shows the 12th Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Digital camera 101 Imaging lens 103 Driver 103 Imaging part 104 Drive part 105 Control circuit 106 Driver 107 Unit circuit 108 Video signal processing part 109 Image memory 109 Sequential image memory 110 Display control part 111 Display memory 112 Display part 113 Memory IF
114 External memory 115 Program memory 116 Image compression / coding block 117 Compression / encoding image data memory 118 Image expansion / decoding unit 119 Data memory 120 Power switch 121 Operation input unit 122 Input circuit 125 Microphone 126 Amplifier 127 ADC circuit 128 Noise Removal / suppression circuit 129 Input audio memory 130 Audio compression / encoding block 131 Compression / encoding audio data memory 132 Audio expansion / decoding unit 133 Output audio memory 134 D / A converter 135 LPF
136 Amplifier 137 Speaker 138 Input / Output Interface Circuit 139 Input / Output Connector 140 Noise Suppression Parameter Control Block 141 Noise Suppression Parameter Memory 142 Noise Component Calculation Unit 143 Subtraction Circuit 150 External Device

Claims (24)

An imaging means for capturing a video;
Detecting means for detecting ambient sound during imaging by the imaging means;
Recording means for recording the moving image picked up by the image pickup means and the ambient sound detected by the detection means;
An operation means for generating noise along with the operation;
Storage means for storing data of noise generated along with the operation of the operation means;
Noise reduction processing means for performing noise reduction processing on ambient sound detected by the detection means during operation of the operation means based on the noise data stored in the storage means;
A connection means for connecting to an external device;
Acquisition means for acquiring noise data transferred from the external device connected to the connection means;
An electronic camera comprising: writing means for writing noise data acquired by the acquisition means into the storage means. An imaging means for capturing a video;
Detecting means for detecting ambient sound during imaging by the imaging means;
Recording means for recording the moving image picked up by the image pickup means and the ambient sound detected by the detection means;
An operation means for generating noise along with the operation;
Storage means for storing data of noise generated along with the operation of the operation means;
Noise reduction processing means for performing noise reduction processing on ambient sound detected by the detection means during operation of the operation means based on the noise data stored in the storage means;
An electronic camera comprising: update means for updating noise data stored in the storage means.   2. The electronic camera according to claim 1, further comprising update means for updating noise data written to the storage means by the writing means. The writing means includes
The electronic camera according to claim 1, further comprising an update unit that updates the noise data stored in the storage unit by writing the noise data acquired by the acquisition unit into the storage unit. The update means includes connection means for connecting to an external device;
Acquisition means for acquiring noise data transferred from the external device connected to the connection means;
3. The electronic camera according to claim 2, further comprising a writing unit that writes the noise data acquired by the acquiring unit into the storage unit.   6. The electronic camera according to claim 2, wherein the noise data updated by the updating unit varies depending on a noise reduction method of the noise reduction processing unit. The noise data updated by the updating means is
When the noise reduction processing means is an adaptive filter method, it includes at least one of a filter coefficient and an adaptive update algorithm,
In the case of a microphone array system, it includes at least one of the angle of the noise source direction or the delay time,
If it is a linear predictor, it includes an adaptive coefficient according to the generated noise,
The electronic camera according to claim 6, wherein in the case of the spectrum subtraction method, at least one of noise source spectrum data, estimated data thereof, and subtraction gain is included.   The electronic camera according to claim 2, wherein the updating unit includes a unit that updates the noise data when the power is turned on and / or when the operation mode is switched.   9. The electronic camera according to claim 8, wherein the updating means includes means for updating the noise data based on the ambient sound detected by the detecting means when the power is turned on and / or when the operation mode is switched. The update unit includes an operation history storage unit that stores an operation history of the operation unit;
10. The electronic camera according to claim 2, further comprising a unit that updates the noise data based on the operation history stored in the operation history storage unit.   The update means includes estimation means for estimating noise data based on an operation history stored in the operation history storage means, and means for updating to noise data estimated by the estimation means. Item 11. The electronic camera according to Item 10. The update means includes means for updating in a direction in which the amplitude of the noise data increases as the operation amount of the operation means with time increases. Electronic camera.   The update means includes means for shifting and updating the frequency component of the noise data as the operation amount of the operation means with time increases. Electronic camera.   14. The electronic camera according to claim 13, wherein the frequency component is shifted to a high frequency side. The updating means is an extracting means for extracting a sound of a specific low frequency band from the ambient sound detected by the detecting means;
Based on the sound of the specific low frequency band extracted by the extracting means, the secular change detecting means for detecting the secular change of the operating means,
15. The electronic camera according to claim 2, further comprising means for updating to noise data corresponding to the secular change detected by the secular change detecting means. Further comprising instruction means for instructing update of the noise data;
The electronic camera according to claim 2, wherein the update unit updates the noise data in response to an instruction from the instruction unit.   17. The electronic camera according to claim 16, wherein the updating means includes means for updating the noise data based on the ambient sound detected by the detecting means when the updating is instructed by the instructing means. .   The electronic camera according to any one of claims 2 to 17, wherein the updating means includes means for setting different noise data in accordance with a photographing scene or photographing mode that can be set in the electronic camera. Further comprising output means for outputting to the external device noise generated with the operation of the operation means detected by the detection means;
The electronic camera according to claim 1, wherein the acquisition unit acquires noise data generated by the external device based on the noise output from the output unit. A terminal for receiving image data from the outside;
Image data storage means for storing the image data received via this terminal,
6. The electronic camera according to claim 1, wherein the connecting means is also used as the terminal. In a noise reduction device arranged in a device comprising detection means for detecting ambient sound and operation means for generating noise in accordance with the operation,
Storage means for storing data of noise generated along with the operation of the operation means;
Noise reduction processing means for performing noise reduction processing on ambient sound detected by the detection means during operation of the operation means based on the noise data stored in the storage means;
A connection means for connecting to an external device;
Acquisition means for acquiring noise data transferred from the external device connected to the connection means;
A noise reduction apparatus comprising: writing means for writing noise data acquired by the acquisition means into the storage means. In a noise reduction device arranged in a device comprising detection means for detecting ambient sound and operation means for generating noise in accordance with operation,
Storage means for storing data of noise generated along with the operation of the operation means;
Noise reduction processing means for performing noise reduction processing on ambient sounds detected by the detection means during operation of the operation means based on the noise data stored in the storage means;
A noise reduction apparatus comprising: update means for updating noise data stored in the storage means. Detection means for detecting ambient sound, operation means for generating noise in accordance with the operation, storage means for storing noise data generated in accordance with the operation of the operation means, and noise stored in the storage means Based on the data, a computer included in the device for reducing the noise generated during the operation of the operating means,
A connection means for connecting to an external device;
Acquisition means for acquiring noise data transferred from the external device connected to the connection means;
A noise reduction control program that functions as a writing unit that writes noise data acquired by the acquiring unit to the storage unit. Detection means for detecting ambient sound, operation means for generating noise in accordance with the operation, storage means for storing noise data generated in accordance with the operation of the operation means, and noise stored in the storage means Based on the data, a computer included in the device for reducing the noise generated during the operation of the operating means,
A noise reduction control program that functions as update means for updating noise data stored in the storage means.

Images