JP6937473B2 - Image processing device, vital information acquisition system equipped with this, and image processing method - Google Patents

Description

The present invention relates to a technique for estimating a pulse from a person image without contacting the human body, and in particular, an image processing device for acquiring an image suitable for estimating the pulse, a pulse estimation system including the image processing device, and image processing. Regarding the method.

Conventionally, regarding the measurement of a person's pulse, a method in which a measurer (nurse, etc.) puts a finger on the subject's wrist to check the pulsation personally, or a measuring device dedicated to the subject's wrist or finger, etc. There is known a method of automatically detecting pulsation by attaching the. On the other hand, such a measurement method does not come into contact with the subject (human body) because the free movement of the subject is temporarily restricted or the subject needs to be equipped with a measuring device. Techniques for estimating (detecting) the pulse have been developed.

For example, regarding a technique for detecting a heart rate (usually equivalent to a pulse rate) without contacting the human body, a spectral distribution of a time-series signal is extracted from image data obtained by photographing a subject, and the spectral distribution is obtained. Therefore, a heart rate detection device that automatically detects a heart rate by specifying a peak frequency caused by a heart rate signal is known (see Patent Document 1).

Japanese Unexamined Patent Publication No. 2012-239661

By the way, since the amount of data tends to be large for images (moving images) taken by a camera, data compression technology (for example, MPEG (Moving Picture Experts Group) standard) for facilitating storage and transmission / reception of image data (for example, MPEG (Moving Picture Experts Group) standard). Data coding method based on the above) is widespread.

However, when the inventors of the present application estimate the pulse from a face image or the like taken by a camera as in Patent Document 1, the fluctuation amount of the pulse signal extracted from the image data (that is, the fluctuation of the pixel value). Since the amount) is very small, it has been found that it may be difficult to estimate the pulse or the accuracy of the estimation may be significantly lowered for the image to which the above data compression is applied. More specifically, for example, when the pulse is estimated by frequency analysis for an image that has been subjected to data compression processing using inter-frame prediction, the I picture (Intra Picture) in the GOP (Group of picture) is a P picture (P picture). Predictive Picture), or when the P picture is switched to the I picture, there is a problem that a frequency component (noise) close to the frequency component of the pulse in the spectrum distribution as described above may occur.

The present invention has been devised in view of such problems of the prior art, and is suitable for estimating the pulse while suppressing an increase in the amount of image data by appropriately executing the image data compression process. An object of the present invention is to provide an image processing device capable of acquiring various images, a pulse estimation system equipped with the image processing device, and an image processing method.

The image processing device of the present invention is an image processing device that acquires an image for estimating a pulse, and includes a data compression processing unit that executes data compression processing based on inter-frame prediction on the input captured image. Based on an instruction from the user, the data compression processing unit includes an operation mode selection unit capable of selecting either a first operation mode for normal shooting or a second operation mode for pulse estimation, and the data compression processing unit performs the second operation. When the mode is selected, the interval of the I pictures constituting the compressed image generated by the data compression process is set to a different interval from the case where the first operation mode is selected.

According to the present invention, by appropriately executing the image data compression process, it is possible to acquire an image suitable for estimating the pulse while suppressing an increase in the amount of image data.

Overall configuration diagram of the pulse estimation system according to the present invention Functional block diagram of pulse estimation system Explanatory drawing which shows the outline of processing by a noise reduction processing part of a camera Explanatory diagram of pulse wave extraction processing by pulse calculation part of pulse estimation device Explanatory drawing which shows an example of the noise which may occur in the data compression processing part of the camera of a normal shooting mode A flow chart showing a setting method for noise reduction in the noise reduction processing section of a camera. A flow chart showing a setting method for data compression processing in the data compression processing unit of the camera. Explanatory drawing which shows the process of step ST203 in FIG. 7 and the process result Explanatory drawing which shows transformation example of processing and processing result shown in FIG.

The first invention made to solve the above problems is an image processing device that acquires an image for estimating a pulse, and executes data compression processing based on inter-frame prediction on an input captured image. The data compression processing unit is provided with an operation mode selection unit capable of selecting either a first operation mode for normal shooting or a second operation mode for pulse estimation based on an instruction from the user. When the second operation mode is selected, the unit sets the interval of the I pictures constituting the compressed image generated by the data compression process to a different interval from the interval when the first operation mode is selected. It is characterized by that.

According to the image processing apparatus according to the first invention, in the first operation mode for normal shooting, the frequency component of the pulse is caused by the interval between the I pictures (that is, the switching from the I picture to another picture). Even when noise having a similar frequency is generated, the image data compression process is appropriately executed in the second operation mode for pulse estimation (that is, the I-picture interval is set to the case where the first operation mode is selected. By setting different intervals), it is possible to acquire an image suitable for estimating the pulse while suppressing an increase in the amount of image data.

Further, in the second invention, in the first invention, when the second operation mode is selected, the data compression processing unit sets the interval between the I pictures and the first operation mode is selected. It is characterized by setting the interval larger than.

According to the image processing apparatus according to the second invention, in the second operation mode for pulse estimation, the interval between the I pictures is set to a larger interval than when the first operation mode is selected. The noise generated due to the interval between the pictures can be displaced to a lower frequency side than the frequency component of the pulse, and as a result, an image suitable for estimating the pulse can be obtained.

Further, in the third invention, in the first invention, when the first operation mode is selected, the data compression processing unit comprises at least an I picture and a P picture, while the first operation mode is selected. When the two operation modes are selected, the compressed image is composed of only I-pictures.

According to the image processing apparatus according to the third invention, in the second operation mode for pulse estimation, the compressed image is composed only of the I-pictures, which is caused by the interval between the I-pictures (switching to the P-picture or the like). It is possible to surely prevent the generation of noise, and as a result, it is possible to acquire an image suitable for estimating the pulse rate.

Further, in the fourth invention, in any one of the first to third inventions, the noise reduction processing unit for executing the noise reduction processing of the captured image is further provided, and the noise reduction processing unit is the second operation mode. When is selected, the noise reduction processing is reduced as compared with the case where the first operation mode is selected.

According to the image processing apparatus according to the fourth invention, in the second operation mode for pulse estimation, the noise reduction processing is reduced to obtain noise from the captured image regarding information useful for pulse estimation included in the captured image. It is possible to suppress the amount of information that is removed together with the above, and as a result, it is possible to acquire an image suitable for estimating the pulse rate.

Further, in the fifth invention, in any one of the first to third inventions, the noise reduction processing unit for executing the noise reduction processing of the captured image is further provided, and the noise reduction processing unit is the first operation mode. When is selected, the noise reduction process is executed, while when the second operation mode is selected, the noise reduction process is not executed.

According to the image processing apparatus according to the fifth invention, in the second operation mode for pulse estimation, by not executing the noise reduction processing, information useful for pulse estimation included in the captured image is noise from the captured image. It is possible to prevent the image from being removed together with the image, and as a result, it is possible to acquire an image suitable for estimating the pulse rate.

Further, in the sixth invention, in the fourth or fifth invention, when the face region is extracted from the captured image, does the noise reduction processing unit reduce the noise reduction processing only in the face region? , Or the noise reduction process is not executed only in the face region.

According to the image processing apparatus according to the sixth invention, in the second operation mode for pulse estimation, the noise reduction processing is reduced only in the face region, or the noise reduction processing is not executed only in the face region. While appropriately reducing noise other than the face region, it is possible to suppress or prevent the information for pulse estimation included in the captured image from being removed from the captured image together with the noise.

Further, the seventh invention is a pulse estimation that estimates a pulse based on the image processing apparatus according to any one of the first to sixth inventions and a compressed image in which the data compression processing is executed in the image processing apparatus. It is a pulse estimation system characterized by being equipped with a device.

According to the pulse estimation system according to the seventh invention, in the first operation mode for normal photographing in the image processing apparatus, when noise having a frequency close to the frequency component of the pulse is generated due to the interval of the I-pictures. However, in the second operation mode for pulse estimation, by appropriately executing the image data compression process, it is possible to acquire an image suitable for pulse estimation in the pulse estimation device while suppressing an increase in the amount of image data. Is possible.

The eighth invention is an image processing method for acquiring an image for estimating a pulse, which comprises a data compression processing step of executing data compression processing based on inter-frame prediction on an input captured image. In the data compression processing step, the second operation includes an operation mode selection step in which one of a first operation mode for normal shooting and a second operation mode for pulse estimation can be selected based on an instruction from the user. When the mode is selected, the interval of the I pictures constituting the compressed image generated by the data compression process is set to a different interval from the case where the first operation mode is selected.

According to the image processing method according to the eighth invention, even when noise having a frequency close to the frequency component of the pulse is generated due to the interval of the I-picture in the first operation mode for normal shooting, the pulse is estimated. By appropriately executing the image data compression process in the second operation mode for the above, it is possible to acquire an image suitable for estimating the pulse while suppressing an increase in the amount of image data.

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

1 and 2 are an overall configuration diagram and a functional block diagram of the pulse estimation system 1 according to the present invention, respectively, and FIG. 3 is an explanatory diagram showing an outline of processing by the noise reduction processing unit 15 of the camera 2. FIG. 4 is an explanatory diagram of pulse wave extraction processing by the pulse calculation unit 23 of the pulse estimation device 3, and FIG. 5 is an explanation showing an example of noise that can be generated in the data compression processing unit 16 of the camera 2 in the normal shooting mode. It is a figure.

The pulse estimation system 1 estimates the pulse (usually the same as the heartbeat) from the information (image taken) obtained without contacting the human body, and as shown in FIG. 1, the person H who is the subject A camera (image processing device) 2 that captures at least a part of the subject, and a pulse estimation device 3 that estimates the pulse (pulse rate and pulse wave) of the person H from the captured image (moving image) obtained by the image taken by the camera 2. I have. Further, in the pulse estimation system 1, the camera 2 and the pulse estimation device 3 are connected to each other so as to be able to communicate with each other via a network 4 such as the Internet or a LAN (Local Area Network). However, the present invention is not limited to this, and the camera 2 and the pulse estimation device 3 may be directly connected so as to be communicable by a known communication cable.

The camera 2 is a video camera having a known photographing function, and as shown in FIG. 2, an imaging unit 11 for forming an image of light from a subject on an image sensor (CCD, CMOS, etc.) through a lens mechanism (not shown), and the imaging unit 11 thereof. It includes an image processing unit 12 that performs predetermined image processing on the captured image (digital video signal) input from the image capturing unit 11. The image processing unit 12 includes a noise reduction processing unit 15 that executes noise reduction (noise reduction processing) for reducing or suppressing noise that deteriorates the image quality of the captured image, and a process (data) for compressing the data of the captured image. Data compression processing unit 16 that executes compression processing), and an image processing control unit (operation mode selection unit) that comprehensively controls the operation of each unit related to various image processing including the processing of the noise reduction processing unit 15 and the data compression processing unit 16. It includes 17 and an image storage unit 18 that stores data of captured images that have undergone image processing (here, noise reduction and data compression).

As will be described in detail later, the camera 2 has two operation modes, a normal shooting mode (first operation mode) for executing image processing for normal shooting (for example, for portrait shooting, landscape shooting, etc.) and a pulse. It has a vital information acquisition mode (second operation mode) for executing image processing for estimation. The image processing control unit 17 can select either the normal shooting mode or the vital information acquisition mode based on an instruction from the user (for example, button operation, touch panel input, etc.). The image processing unit 12 is not limited to the one shown here, and can appropriately perform other known signal processing (for example, color tone correction, contour enhancement, etc.). Further, the image processing unit 12 can be configured to further include other operation modes in addition to the above-mentioned normal shooting mode and vital information acquisition mode.

As noise reduction, the noise reduction processing unit 15 can perform three-dimensional noise reduction (3DNR) that reduces noise in a captured image based on the correlation between a plurality of frames that are adjacent in time. More specifically, in the noise reduction processing unit 15, as shown in FIG. 3, in the adder 19, the image of the previous frame is added at a predetermined ratio to the image (pixel value) of the current frame (current frame). And the previous frame is synthesized based on a predetermined filter coefficient (M: N)), so that the current frame processed by 3DNR with an improved S / N ratio is generated. For example, the image processing control unit 17 sets the filter coefficient to M: N = 0.5: 0.5 in the normal shooting mode, and M: N = 0.9: 0.1 in the vital information acquisition mode. Can be set to. The captured image subjected to noise reduction is input to the data compression processing unit 16.

The setting (or change) of the filter coefficient is not limited to the method of setting one filter coefficient for the entire image, and a method of setting each divided area provided in the image may be adopted. .. For example, in the vital information acquisition mode, the skin color area (or face area) is extracted from the image, the filter coefficient for the area other than the skin color area is set to be the same as the normal shooting mode, and the filter coefficient for the skin color area is set to the normal shooting mode. It is possible to set a value different from.

In this way, in the vital information acquisition mode, by intentionally reducing noise reduction (reducing the noise removal effect), information contained in the captured image that is useful for pulse estimation is removed from the captured image together with noise. Suppress the amount.

Alternatively, in the vital information acquisition mode, it is possible to configure the noise reduction not to be executed (the noise reduction is executed only in the normal shooting mode). According to this, in the normal shooting mode, the noise of the captured image is appropriately reduced, and in the vital information acquisition mode, the information useful for pulse estimation contained in the captured image is removed from the captured image together with the noise. Can be prevented.

Further, when the noise reduction processing unit 15 has a function of extracting a face region in the captured image as in the region extraction unit 22 described later, when the face region is extracted from the captured image, the above-mentioned is described only in the face region. As described above, the noise reduction processing may be reduced, or the noise reduction processing may not be executed only in the face region. As a result, the same effect as described above can be obtained while appropriately reducing noise other than the face region.

The data compression processing unit 16 can execute the data compression processing of the captured image based on the data coding method conforming to the MPEG standard. The captured image (compressed image) that has undergone data compression processing is stored in the image storage unit 18 and, if necessary, transmitted to the pulse estimation device 3 (when the vital information acquisition mode is executed).

Although not shown, the camera 2 as described above is volatile, for example, as a processor that comprehensively executes various image processing and control of peripheral devices based on a predetermined shooting control program, a work area of the processor, and the like. It is equipped with a RAM (Random Access Memory) as a memory, a ROM (Read Only Memory) as a non-volatile memory for storing control programs and data executed by a processor, an auxiliary storage device, and the like. The functions of the above-mentioned parts of the image processing unit 12 as described above are realized by the hardware thereof and the shooting control program executed by the processor.

The pulse estimation system 1 may be capable of performing at least one of noise reduction and data compression processing. Therefore, at least one of the noise reduction processing unit 15 and the data compression processing unit 16 in the image processing unit 12 is omitted. Is possible. Further, at least a part of the functions (noise reduction, data compression processing, etc.) of the image processing unit 12 in the camera 2 may be executed by another device (for example, the pulse estimation device 3) of the pulse estimation system 1.

The pulse estimation device 3 has an image input unit 21 in which an image (video signal) captured from the camera 2 is input as a time-continuous captured image including at least a part of the person H, and the captured image of the person H. An area extraction unit 22 that extracts a skin color region (here, a face region), and a pulse calculation unit (pulse estimation unit) 23 that calculates (estimates) the pulse of the person H based on the extracted skin color region of the person H. The display unit 24 is composed of a known display device capable of displaying various information including the estimation result of the pulse to the user. The skin color region extracted by the region extraction unit 22 is a region where the skin is exposed in the human body, and the pulse can be estimated from the captured image data of that region. Further, the captured image input to the image input unit 21 is not limited to the one transmitted from the camera 2, and may be a captured image stored in a known memory or the like after shooting.

The region extraction unit 22 executes a known face detection process for recognizing the feature amount of the face on each captured image (frame image), and extracts the detected face region as the skin color region of the person H. Track. Further, the region extraction unit 22 sends the data of the captured image relating to the extracted face region to the pulse calculation unit 23.

In the region extraction unit 22, not limited to the above method, a preset skin color component (for example, a preset ratio for each pixel value of RGB, which is a value different depending on race or the like) is extracted from the captured image. The pixels may be extracted, and the region from which the pixels are extracted may be used as a skin color region. In this case, a portion other than the face where the skin is exposed (for example, a hand or an arm) can also be extracted as a skin color region. However, as described above, by extracting the face region of the person H as the skin color region, there is an advantage that the skin color region can be easily extracted. Although only one person H is shown in FIG. 1, when a plurality of people are included in the captured image, the area extraction unit 22 can extract a plurality of face areas.

The pulse calculation unit 23 calculates, for example, the pixel values (0-255 gradations) of each component of RGB for each pixel constituting the skin color region extracted in the time-continuous captured images, and represents the representative value (here). Then, the time-series data of (the average value of each pixel) is generated as a pulse signal. Here, the average value of each pixel is used as a representative value, and the average value is a value with decimal point accuracy. Further, in this case, time series data can be generated based on the pixel value of only the green component (G) whose fluctuation due to pulsation is particularly large.

The time-series data of the generated pixel value (average value) is, for example, as shown in FIG. 4 (A), a slight fluctuation (for example, less than one gradation of the pixel value) based on the change in the hemoglobin concentration in the blood. Fluctuation). Therefore, the pulse calculation unit 23 extracts the power spectrum as shown in FIG. 4B by executing the frequency analysis process by FFT (Fast Fourier Transform) on the time series data based on the pixel value. Can be done. In this power spectrum, the pulse calculation unit 23 can detect a high-power frequency component (for example, the maximum value of the spectrum) and estimate the pulse (number) based on the frequency component.

The pulse estimation device 3 as described above can be composed of, for example, an information processing device such as a PC (Personal Computer). Although not shown, the pulse estimation device 3 includes a processor that comprehensively executes various information processing and control of peripheral devices based on a predetermined control program, a RAM as a volatile memory that functions as a work area of the processor, and the like. ROM as a non-volatile memory that stores control programs and data executed by the processor, network interface that executes communication processing via the network, monitor (image output device), speaker, input device, HDD (Hard Disk Drive), etc. It has a hardware configuration including the above, and at least a part of the functions of each part of the pulse estimation device 3 shown in FIG. 2 can be realized by the processor executing a predetermined control program. At least a part of the function of the pulse estimation device 3 may be replaced by processing by other known hardware.

Here, with reference to FIG. 5, noise that may occur in the data compression processing in the data compression processing unit 16 will be described. FIG. 5A shows the frequency analysis result when the captured image is not subjected to the data compression processing for comparison. Further, in FIG. 5A, for convenience of explanation, only the spectrum based on the pulse is schematically shown. On the other hand, in FIG. 5B, with respect to the captured image having a frame rate of 30 fps, the I picture 35 is periodically inserted at the head of the GOP (Group of Picture) composed of the I picture 35 and the P picture 36 at intervals of 30 frames. The frequency analysis result when it is done is shown. Further, in FIG. 5B, for convenience of explanation, only the spectrum based on the pulse and the noise based on the data compression process are schematically shown.

As shown in FIG. 5A, when the data compression process is not performed, for example, when the pulse rate is about 60 beats / minute, a spectrum based on the pulse appears at a frequency of about 1 Hz, and based on this spectrum. The pulse can be estimated.

On the other hand, as shown in FIG. 5B, when the I picture 35 is inserted at intervals of 30 frames in the captured image having a frame rate of 30 fps, the pulse is caused by the switching from the I picture 35 to the subsequent P picture 36. Noise is generated in the frequency range close to the frequency of (about 1 Hz). It is considered that this is because the difference in the amount of information lost when switching between the I picture 35 and the P picture 36 appears as a spectrum in the frequency analysis. This makes it difficult to distinguish between the pulse-based spectrum and noise, and as a result, it becomes difficult to estimate the pulse based on the spectrum as shown in FIG. 4 (B) above.

Therefore, in the pulse estimation system 1, as will be described in detail later, the image processing control unit 17 is set to a GOP composed of a plurality of pictures (I picture, P picture, B picture) according to the operation mode of the camera 2. By changing the temporal interval of the inserted I-picture (controlling the GOP), it is possible to acquire an captured image suitable for estimating the pulse rate.

In the present embodiment, the frame rate of the captured image is 30 fps, but the frame rate is not limited to this, and the I picture is inserted in the vicinity of the spectrum based on the pulse depending on the insertion interval of the I picture in the GOP at other frame rates as well. Noise due to can occur. Further, here, attention is paid to the noise generated by the switching from the I picture to the P picture, but the same applies to the noise generated by the switching from the I picture to the B picture.

FIG. 6 is a flow chart showing a setting method for noise reduction in the noise reduction processing unit 15 of the camera 2. When the camera 2 is activated, the image processing control unit 17 executes initial settings related to noise reduction of the noise reduction processing unit 15 (ST101). In the present embodiment, the normal shooting mode is selected (ON) when the camera 2 is activated, and the image processing control unit 17 sets the filter coefficient of 3DNR to an initial value (here, M: N = 0.) With respect to noise reduction. Set to 5: 0.5). As a result, the user can acquire an captured image in which noise is appropriately reduced in normal shooting by the camera 2.

Subsequently, when the vital information acquisition mode is selected (ON) based on the user's instruction (ST102: Yes), the image processing control unit 17 reduces the ratio of the image of the previous frame to be combined with the image of the current frame. The filter coefficient of 3DNR is changed (here, changed to M: N = 0.9: 0.1) so as to be (ST103). At this time, the camera 2 outputs an image captured to the pulse estimation device 3 with a slight noise reduction that does not interfere with the pulse estimation process in the pulse estimation device 3. This makes it possible to acquire a captured image suitable for the pulse estimation process while reducing the noise of the captured image to some extent (that is, maintaining the image quality of the captured image as much as possible).

Alternatively, in step ST103, the image processing control unit 17 can turn off the three-dimensional noise reduction (set it to M: N = 1.0: 0.0). In that case, the captured image without noise reduction is output from the camera 2 to the pulse estimation device 3. As a result, although the noise reduction effect cannot be obtained, it is possible to acquire a more suitable captured image for the pulse estimation process.

When the vital information acquisition mode is finally turned off (here, the normal shooting mode is selected) (ST104: Yes), the process returns to step ST101 again, and the same steps as described above are repeatedly executed. As a noise reduction setting method, a configuration in which the vital information acquisition mode is selected (ON) (started from step ST103) when the camera 2 is activated is also possible.

FIG. 7 is a flow chart showing a setting method regarding data compression processing in the data compression processing unit 16 of the camera 2, and FIG. 8 is an explanatory diagram showing the processing of step ST203 in FIG. 7 and the processing result thereof. 9 is an explanatory diagram showing a modified example of the processing and the processing result shown in FIG.

When the camera 2 is activated, the image processing control unit 17 executes the initial settings related to the data compression processing of the data compression processing unit 16 (ST201). In the present embodiment, the normal shooting mode is selected (ON) when the camera 2 is activated, and the image processing control unit 17 sets 30 frames of I-pictures in the captured image (compressed image) at a frame rate of 30 fps in relation to the data compression process. Set to insert at intervals. As a result, the user can acquire an image captured with an appropriately compressed data capacity in normal shooting.

Subsequently, when the vital information acquisition mode is selected (ON) based on the user's instruction (ST202: Yes), the image processing control unit 17 changes the insertion interval of the I picture regarding the data compression process (ST203). .. At this time, as shown in FIG. 8, the image processing control unit 17 is set so that the I picture 35 is inserted between the P picture 36 groups at a frame interval of, for example, about 90 to 120. As a result, the noise generated in the frequency range close to the pulse frequency (see FIG. 5B) is displaced to a lower frequency side than the pulse frequency component (here, about 1 Hz), so that the pulse is estimated. In the pulse estimation process in the device 3, it becomes easy to distinguish between the spectrum based on the pulse and the noise.

Alternatively, the image processing control unit 17 sets all the frames to be composed of the I-picture 35, for example, as shown in FIG. 9, instead of changing the insertion interval of the I-picture in step ST203. It is also possible. In this case, although the effect of data compression is reduced, it is possible to reliably prevent the generation of noise due to the switching from the I picture to the P picture.

When the vital information acquisition mode is finally turned off (here, the normal shooting mode is selected) (ST204: Yes), the process returns to step ST201 again, and the same steps as described above are repeatedly executed. As a setting method related to the data compression process, a configuration in which the vital information acquisition mode is selected (ON) (started from step ST203) when the camera 2 is started is also possible.

Although the present invention has been described above based on specific embodiments, these embodiments are merely examples, and the present invention is not limited to these embodiments. For example, in the above embodiment, an example in which the image processing device according to the present invention is realized as a part of the function of the camera 2 is shown, but the present invention is not limited to this, and the data compression processing (or the image processing) of the captured image acquired from the camera 2 is not limited to this. It may be a device that executes noise reduction). Further, the captured image processed by the camera 2 does not necessarily have to be directly transmitted to the pulse estimation device 3, and the captured image stored in the image storage unit 18 may be separately used for pulse estimation by another device. It is possible. The image processing apparatus, the pulse estimation system provided with the image processing apparatus, and the image processing method are not necessarily all indispensable, and can be appropriately selected as long as they do not deviate from the scope of the present invention.

The image processing apparatus according to the present invention, the pulse estimation system provided with the image processing apparatus, and the image processing method are suitable for estimating the pulse while suppressing an increase in the amount of image data by appropriately executing the image data compression process. It is useful as an image processing device for making an image available and an image suitable for pulse estimation, a pulse estimation system provided with the image processing device, an image processing method, and the like.

1 Pulse estimation system 2 Camera (image processing device)
3 Pulse estimation device 11 Imaging unit 12 Image processing unit 15 Noise reduction processing unit 16 Data compression processing unit 17 Image processing control unit (operation mode selection unit)
18 Storage unit 21 Image input unit 22 Area extraction unit 23 Pulse calculation unit

Claims (13)

An image processing device that acquires images
A data compression processing unit that executes data compression processing based on inter-frame prediction on the input captured image and generates a compressed image,
It is provided with an operation mode selection unit capable of selecting either a first operation mode for normal shooting or a second operation mode for pulse estimation based on an instruction from a user.
The data compression processing unit
When the first operation mode is selected, the compressed image configured to include an I picture and a P picture is generated.
An image processing device that switches the configuration of the compressed image and generates the compressed image composed of only I pictures when the first operation mode is switched to the second operation mode. The image processing apparatus according to claim 1, wherein the compressed image generated by the data compression processing unit when the first operation mode is selected further includes a B picture. The image processing apparatus according to claim 2, wherein in the compressed image generated by the data compression processing unit when the first operation mode is selected, the P picture or the B picture is next to the I picture. A noise reduction processing unit that executes noise reduction processing of the captured image is further provided.
The noise reduction processing unit according to any one of claims 1 to 3, wherein when the second operation mode is selected, the noise reduction processing is reduced as compared with the case where the first operation mode is selected. Image processing equipment. A noise reduction processing unit that executes noise reduction processing of the captured image is further provided.
The noise reduction processing unit executes the noise reduction processing when the first operation mode is selected, but does not execute the noise reduction processing when the second operation mode is selected, according to claim 1. The image processing apparatus according to any one of claims 3. 4. The noise reduction processing unit reduces the noise reduction processing only in the face region or does not execute the noise reduction processing only in the face region when the face region is extracted from the captured image. Alternatively, the image processing apparatus according to claim 5. The image processing apparatus according to any one of claims 1 to 6, which executes the data compression process based on a data coding method conforming to the MPEG standard. The image processing apparatus according to any one of claims 1 to 7, and a vital information acquisition apparatus for acquiring vital information based on a compressed image in which the data compression processing is executed in the image processing apparatus. Vital information acquisition system. The vital information acquisition system according to claim 8, wherein the image processing device transmits the compressed image to the vital information acquisition device via a network. The image processing device further includes an image storage unit that stores the compressed image.
The vital information acquisition system according to claim 8 or 9, wherein the compressed image stored in the image storage unit is transmitted to the vital information acquisition device only when the second operation mode is selected. .. An image processing device that acquires images
A data compression processing unit that executes data compression processing based on inter-frame prediction on the input captured image and generates a compressed image,
It is provided with an operation mode selection unit capable of selecting either a first operation mode for normal shooting or a second operation mode for pulse estimation based on an instruction from a user.
The data compression processing unit
When the first operating mode is selected, the compressed image configured to include a series of I-pictures and P-pictures is generated.
An image processing device that, when switched from the first operation mode to the second operation mode , switches the configuration of the compressed image and generates the compressed image configured to include a plurality of consecutive I pictures. It is an image processing method to acquire an image.
A data compression processing step that executes data compression processing based on inter-frame prediction on the input captured image and generates a compressed image, and
It includes an operation mode selection step that can select either a first operation mode for normal shooting or a second operation mode for pulse estimation based on an instruction from the user.
In the data compression processing step,
When the first operation mode is selected, the compressed image configured to include an I picture and a P picture is generated.
An image processing method in which when the first operation mode is switched to the second operation mode, the configuration of the compressed image is switched and the compressed image is generated so as to be composed of only I pictures. It is an image processing method to acquire an image.
A data compression processing step that executes data compression processing based on inter-frame prediction on the input captured image and generates a compressed image, and
It includes an operation mode selection step that can select either a first operation mode for normal shooting or a second operation mode for pulse estimation based on an instruction from the user.
In the data compression processing step,
When the first operating mode is selected, the compressed image configured to include successive I-pictures and P-pictures is generated.
An image processing method that, when switched from the first operation mode to the second operation mode , switches the configuration of the compressed image and generates the compressed image configured to include a plurality of consecutive I pictures.

Images