Physiological state detection method

Abstract

The invention discloses a physiological state detection method, which is suitable for an electronic device, the electronic device comprises a display screen, a visible light camera and an infrared light camera, the visible light camera and the infrared light camera are arranged above the display screen, and the detection method comprises the following steps: receiving an infrared light image generated by the infrared light camera; receiving a visible light image generated by the visible light camera; capturing a first pixel range containing physiological features in the infrared light image, and reading gray-scale pixel values in the first pixel range; capturing a second pixel range containing physiological features in the visible light image, and reading a red pixel value in the second pixel range; and calculating the physiological parameter according to the gray-scale pixel value and the red pixel value.

Description

TECHNICAL FIELD

[0001] The present application relates to a physiological state detection method, in particular to an optical detection method of physiological state. BACKGROUND

[0002] Mobile phones have become an indispensable part of modern life, whether it is communication, information inquiry, or health management, the diversity of mobile phone functions makes it play an important role in daily life. With the continuous progress of technology, the application range of mobile phones is continuously expanding, one of the areas worth paying attention to is the use of mobile phones for physiological detection. By monitoring physiological parameters through mobile phones, not only is it convenient to achieve daily health management, but it can also collect long-term observation data to provide a more comprehensive analysis of the user's health status.

[0003] In the field of physiological monitoring, fundus image detection technology is an important physiological monitoring means, which enters the eyeball internally through optical means to capture fundus images, thereby helping doctors to diagnose various diseases. However, the optical sensors and related algorithms of existing mobile phones are not suitable for direct use in fundus image detection. This is mainly because the optical components of mobile phones are limited by size and performance, and cannot provide sufficient resolution and optical signal quality, making it difficult to reach the level of professional medical equipment.

[0004] One solution is to use external handheld lenses, which can enhance the optical capabilities of mobile phones and achieve fundus image capture. However, this method has the disadvantages of inconvenient operation and large size, and it is difficult to achieve long-term monitoring without feeling, which is far from the needs of daily health management. Therefore, how to improve the ability of mobile phones in physiological detection without relying on external devices is still a technical problem to be solved. SUMMARY

[0005] Therefore, the present application provides a physiological state detection method suitable for an electronic device, the electronic device includes a display screen and a visible light camera and an infrared light camera arranged above the display screen, the detection method includes: receiving an infrared light image generated by the infrared light camera; receiving a visible light image generated by the visible light camera; extracting a first pixel range containing a physiological feature in the infrared light image, reading a gray scale pixel value in the first pixel range; extracting a second pixel range containing the physiological feature in the visible light image, reading a red pixel value in the second pixel range; and calculating a physiological parameter according to the gray scale pixel value and the red pixel value.

[0006] The present application will be described in detail below in conjunction with the drawings and specific embodiments, but not as a limitation on the present application. BRIEF DESCRIPTION OF DRAWINGS

[0007] Figure 1A is a schematic diagram of an appearance of an electronic device according to some embodiments of the present application.

[0008] Figure 1B is a schematic diagram of an appearance of an electronic device according to some embodiments of the present application.

[0009] Figure 2 is a block diagram of an electronic device according to some embodiments of the present application.

[0010] Figure 3 is a flowchart of a physiological state detection method according to some embodiments of the present application.

[0011] Figure 4 is an absorption spectrum according to some embodiments of the present application.

[0012] Figure 5 is a flowchart of a physiological state detection method applied to eye detection according to some embodiments of the present application.

[0013] Figures 6A-6C is a schematic diagram of an infrared light image and a first pixel range according to some embodiments of the present application.

[0014] Figure 7 is a schematic diagram of a physiological state and different state degrees according to some embodiments of the present application.

[0015] wherein the reference signs

[0016] 10: electronic device

[0017] 11: display screen

[0018] 121: visible light camera

[0019] 122: infrared light camera

[0020] 123: short-wavelength ambient light sensor

[0021] 124: long-wavelength ambient light sensor

[0022] 125: eye movement tracking sensor

[0023] 13: light source

[0024] 14: memory

[0025] 15: processor

[0026] 90: infrared light image

[0027] 91: first pixel range

[0028] 92: physiological feature

[0029] 921: Sclera

[0030] 922: Iris

[0031] 923: Bloodshot

[0032] 924: Between the eyebrows

[0033] R: Red light band

[0034] NIR: Near-infrared band

[0035] S101~S110: Steps

[0036] S201~S209: Steps Detailed Implementation

[0037] The following detailed description of various embodiments of the present invention, illustrated with accompanying drawings. In addition to these detailed descriptions, the present invention can be widely implemented in other embodiments, and any easy substitutions, modifications, or equivalent changes to the described embodiments are included within the scope of the present invention, as determined by the claims. In the description of the specification, many specific details and implementation examples are provided to give the reader a more complete understanding of the present invention; however, these specific details and implementation examples should not be considered as limitations on the present invention. Furthermore, well-known steps or elements are not described in the details to avoid unnecessarily limiting the present invention.

[0038] Figure 1A These are schematic diagrams of the appearance of an electronic device according to some embodiments of the present invention; Figure 2 These are block diagrams of electronic devices according to some embodiments of the present invention, please refer to them as well. Figure 1A and Figure 2 In this embodiment, the electronic device 10 includes a display screen 11, a visible light camera 121, an infrared camera 122, a short-wavelength ambient light sensor 123, a long-wavelength ambient light sensor 124, an eye-tracking sensor 125, a light source 13, a memory 14, and a processor 15. The processor 15 is coupled to the display screen 11, the visible light camera 121, the infrared camera 122, the short-wavelength ambient light sensor 123, the long-wavelength ambient light sensor 124, the eye-tracking sensor 125, the light source 13, and the memory 14. Figure 1A As shown, in this embodiment, sensing components such as a visible light camera 121, an infrared camera 122, a short-wavelength ambient light sensor 123, a long-wavelength ambient light sensor 124, and an eye-tracking sensor 125, as well as a light source 13, are disposed above the display screen 11. This ensures that when a user operates the electronic device 10, their hand will not obstruct the sensing components and affect detection. Furthermore, in some embodiments, the display screen 11 can display physiological monitoring data in real time; or, it can further provide users with parameter settings.

[0039] Figure 1B These are schematic diagrams of the appearance of an electronic device according to other embodiments of the present invention, which should also be referred to. Figure 1B and Figure 2 In this embodiment, the electronic device 10 includes a foldable display screen 11. In this embodiment, a sensing component is disposed above the display screen 11. Physiological monitoring data can be displayed on the left side of the folded portion of the display screen 11, while the right side allows the user to set parameters. The aforementioned electronic device 10 may be, but is not limited to, a personal computer, mobile phone, watch, tablet computer, or laptop computer.

[0040] Processor 15 may be, but is not limited to, a SoC chip, a central processing unit (CPU), a micro-control unit (MCU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a neural network processor, a quantum processor, or a logic circuit.

[0041] Visible light camera 121 and infrared light camera 122 are used to detect visible light images and infrared light images, respectively. In some embodiments, visible light camera 121 and infrared light camera 122 employ silicon-based sensor modules. The visible light sensor can be paired with an infrared cut-off filter with a cutoff wavelength of approximately 700-850 nanometers to improve the aesthetics of the captured image and reduce the difference between the image and human vision. The infrared light sensor can be paired with a long-pass filter of 760 nanometers or higher to achieve infrared light image detection. In some embodiments, visible light camera 121 and / or infrared light camera 122 can also be configured with a focusing module, such as a voice coil motor (VCM), to focus and track local physiological features 92.

[0042] Short-wavelength ambient light sensor 123 and long-wavelength ambient light sensor 124 are used to detect ambient visible light signals and ambient infrared light signals, respectively. The ambient light sensors convert the detected ambient light sources into electrical signals to facilitate image correction by the electronic device 10. The short-wavelength ambient light sensor 123 can cover the visible light wavelength range, for example, 400–700 nm, and the ambient infrared light signal can cover the infrared wavelength range, for example, 760–106 nm. In some embodiments, the short-wavelength ambient light sensor 123 and the long-wavelength ambient light sensor 124 are integrated into a single ambient light sensor module.

[0043] An eye-tracking sensor 125 is used to monitor eye movements and can determine the direction of the eyes based on physiological characteristics 92 of the eyes (e.g., pupil position) to infer the user's gaze point. In some embodiments, the eye-tracking sensor 125 may use a low-resolution visible light (or non-visible light) sensor to capture eye direction, reducing computational load and improving detection capabilities. In other embodiments, the eye-tracking sensor 125 may be integrated with an infrared camera 122 to detect eye movements via infrared light, utilizing the non-interference characteristic of infrared light for daily monitoring.

[0044] Light source 13 is used to provide supplemental lighting for the physiological characteristics 92 of the target being photographed or to assist in physiological monitoring. For example, light source 13 generates infrared light to illuminate the surface of the eyeball, and eye-tracking sensor 125 captures the direction of movement of eye features (e.g., iris 922) based on the reflected infrared light. Light source 13 can be a full-spectrum light source covering both visible and infrared light, or it can be a separately configured visible light source (e.g., white LED) and infrared light source (e.g., LEDs above 810 nanometers) to enable specific light sources 13 according to different usage or physiological monitoring scenarios. For example, when a user is taking a selfie, the electronic device 10 only needs to turn on the visible light source for facial illumination; while when performing physiological monitoring, the electronic device 10 only needs to turn on the infrared light source to avoid interfering with the user. In some embodiments, the visible light source and the infrared light source are integrated into a single light source module.

[0045] The memory 14 may be, but is not limited to, flash memory or read-only memory (ROM), such as erasable programmable read-only memory (EPROM), flash read-only memory (Flash ROM), electrically erasable programmable read-only memory (EEPROM), or field-replaceable unit (FRU). In some embodiments, the memory 14 is used to store an image database to provide the electronic device 10 with data analysis and comparison, which will be described in detail later.

[0046] Figure 3 This is a flowchart of a physiological state detection method according to some embodiments of the present invention. Please refer to it. Figure 3 In this embodiment, the electronic device 10 performs distance detection on the object to be measured (step S101). For example, the light source 13 generates infrared light and receives the reflected light through the infrared camera 122 or the long-wavelength ambient light sensor 124, and performs distance detection by means of the time difference between infrared light transmission and reception; or, the distance is estimated based on the proportion of a specific physiological feature 92 (e.g., face) in the image. The object to be measured can refer to a local feature of the body, such as the face, eyebrows, brow, eyes, and nose. When the distance between the electronic device 10 and the face of the object to be measured is less than a distance threshold, the processor 15 begins to receive the infrared light image (step S102) or the visible light image (step S103). In some embodiments, the processor 15 continuously or periodically receives infrared light images from the infrared camera 122 (step S102), and after determining from the infrared light image that the distance between the electronic device 10 and the face is less than a distance threshold, it begins to receive the visible light image from the visible light camera 121 (step S103). In this way, the electronic device 10 does not need to turn on the visible light camera 121 or the visible light source at all times, thereby saving power and avoiding interference with the user. The distance threshold can be defined based on the distance between the face and the electronic device 10 when a normal user operates the electronic device 10, such as a value within 60 centimeters; or, based on the proportion of the face in the image.

[0047] In this embodiment, after receiving an infrared image from the infrared camera 122 (step S102), the processor 15 receives an ambient infrared light signal from the long-wavelength ambient light sensor 124 and corrects the infrared image based on the ambient infrared light signal (step S104). On the other hand, after receiving a visible light image from the visible light camera 121 (step S103), the processor 15 receives an ambient visible light signal from the short-wavelength ambient light sensor 123 and corrects the visible light image based on the ambient visible light signal (step S105). Through the correction of the ambient light signal, the quality of the visible light image and / or the infrared image is sufficiently improved to adapt to different ambient light conditions. For example, the processor 15 can perform white balance correction on the visible light image based on the ambient visible light signal to adjust the color of the visible light image according to the color temperature of the ambient light source, avoiding the problem of white light appearing yellowish or bluish. The processor 15 can also determine the average brightness of the environment based on the intensity of the ambient light, thereby adjusting the local brightness distribution in the image to avoid the appearance of overly bright or dark areas, thereby improving the brightness uniformity of the image. Alternatively, the processor 15 can adjust the shutter speed, aperture size, or photosensitive element brightness of the visible light camera 121 and / or the infrared camera 122 according to the intensity of ambient light to avoid image overexposure. In this embodiment, the visible light image and the infrared image are dually calibrated based on signals generated by different ambient light sensors. This design eliminates the influence of different intensity distributions of ambient light at different wavelengths, avoiding signal saturation problems in other wavelength images that may occur when only the visible light image or the infrared image is uniformly calibrated.

[0048] Processor 15 captures a first pixel range 91 containing physiological feature 92 within an infrared image (step S106), and reads the grayscale pixel values ​​within the first pixel range 91 (step S108). Conversely, processor 15 captures a second pixel range containing physiological feature 92 within a visible light image (step S107), and reads the red pixel values ​​within the second pixel range (step S109). Physiological feature 92 can refer to local features of the body, such as the face, eyebrows, brow, eyes, and nose. In this embodiment, both the first pixel range 91 and the second pixel range contain the same physiological feature 92. Processor 15 obtains the parameters of physiological feature 92 by reading and statistically analyzing the pixel values ​​within these pixel ranges. For infrared images, pixel values ​​only contain grayscale values; for visible light images, pixel values ​​can contain red, green, and blue pixel values. Processor 15 can calculate the average (or median) grayscale pixel values ​​of all pixels within the first pixel range 91, and calculate the average (or median) monochrome pixel values ​​of all pixels within the second pixel range.

[0049] In this embodiment, the processor 15 extracts features from both the infrared and visible light images for a specific physiological characteristic 92, and compares local regions within the images that conform to the specific physiological characteristic 92 to define a first pixel range 91 and a second pixel range. In some embodiments, the processor 15 first extracts multiple feature points contained within the first pixel range 91 from the infrared image, and then compares the extracted feature points with the visible light image to extract a second pixel range within the visible light image that contains the same physiological characteristic 92. In this way, the processor 15 can adapt to the physiological differences of different individuals and quickly obtain the first pixel range 91 and the second pixel range containing the same physiological characteristic 92.

[0050] The processor 15 calculates physiological parameters based on grayscale pixel values ​​and red pixel values ​​(step S110). Figure 4 The absorption spectra are based on some embodiments of the present invention. Please refer to them. Figure 4 The absorption spectrum shows the relationship between the extinction coefficient of heme and the wavelength of absorbed light under different blood oxygen saturation levels. Figure 4 The horizontal axis represents the wavelength (nm) of the absorbed light, and the vertical axis represents the molar extinction coefficient. Figure 4 The red light band R (600 nm~700 nm) and the near-infrared light band NIR (800 nm~850 nm) are marked with boxes. These two regions are closely related to the spectral absorption characteristics of heme. Figure 4 Different curves represent the extinction coefficients of heme at different blood oxygen saturations, ranging from 100% oxyheme to 100% deoxyheme. These curves show that the absorption capacity of heme to light varies with blood oxygen saturation: in the red band (R), there are significant differences in the molar extinction coefficients between the curves; in the near-infrared band (NIR), the differences in the molar extinction coefficients between the curves are negligible. Therefore, the absorption spectrum in the near-infrared band (NIR) serves as a reference value for the absorption spectrum in the red band (R), and the processor 15 can estimate the blood oxygen concentration based on the molar extinction coefficient of the absorbed light in the near-infrared band (NIR). The measurement advantages of the near-infrared band (NIR) are its higher light penetration, enabling it to reflect the physiological state of deeper tissues; furthermore, it is less susceptible to energy attenuation due to melanin absorption.

[0051] In detail, in some embodiments, the processor 15, based on the Beer-Lambert Law, can calculate the molar extinction coefficient of heme within the physiological characteristic 92 range at different wavelengths based on red pixel values ​​and grayscale pixel values. It can also calculate absorbance based on estimated heme concentration (which can be the average concentration value for a normal person or physiological records stored in the electronic device 10) and estimated optical path length (which can be the average tissue thickness value for a normal person or physiological records stored in the electronic device 10). Blood oxygen concentration is estimated by calculating the ratio of the absorbance corresponding to the red pixel value to the absorbance corresponding to the grayscale pixel value. Alternatively, tissue oxygen consumption rate or metabolic rate can be further estimated based on changes in blood oxygen concentration.

[0052] Furthermore, based on the fluctuations in grayscale pixel values ​​and / or red pixel values ​​and / or green pixel values ​​over time, the user's heart rate and respiratory rate can also be estimated. For example, the processor 15 captures and records the changes in red pixel values ​​within a second pixel range over time to obtain a time-domain signal. The resonant spectrum is then obtained using frequency-domain analysis methods (such as, but not limited to, Fast Fourier Transform (FFT) or Short-Time Fourier Transform (STFT)) to estimate the heart rate value or its variation.

[0053] The following embodiments use the eye as the target physiological feature 92 for illustration. It should be understood that the physiological state detection methods of each embodiment can also be applied to different physiological features 92.

[0054] Figure 5 This is a flowchart illustrating the application of a physiological state detection method based on some embodiments of the present invention to eye detection; Figures 6A-6C This is a schematic diagram of infrared light images and the range of the first pixel according to some embodiments of the present invention. Please refer to it first. Figure 5 In this embodiment, the processor 15 determines whether the distance between the electronic device 10 and the face is less than a distance threshold (step S201). When the processor 15 determines that the distance between the electronic device 10 and the face is greater than the distance threshold (step S201, the determination result is "No"), the physiological state detection method ends (step S206) to avoid poor measurement signal quality and wasted computing resources due to the face being too far away; when the processor 15 determines that the distance between the electronic device 10 and the face is less than the distance threshold (step S201, the determination result is "Yes"), the processor 15 respectively captures the pixel range including the left eye and the right eye (step S202). Figure 6AAs shown, the processor 15 reads the infrared image 90 captured by the infrared camera 122, the image containing the user's face. Subsequently, the processor 15 extracts a first pixel range 91 containing physiological features 92 such as the left and right eyes (step S202). The processor 15 can differentiate the physiological features 92 encompassed within the first pixel range 91, for example, processing the ranges containing the left eye, right eye, left iris 922, left sclera 921, or brow 924 separately. In other embodiments, the processor 15 can also read the visible light image captured by the visible light camera 121 and extract a second pixel range containing physiological features 92 such as the left and right eyes (step S202).

[0055] In some embodiments, when the processor 15 determines that the distance between the electronic device 10 and the face is less than a distance threshold (step S201, the determination result is "yes"), the processor 15 further receives the eye-tracking image generated by the eye-tracking sensor 125, and after determining that the physiological features 92 contained in the eye-tracking image are within a default angle range, it begins to capture the first pixel range 91 or the second pixel range. For example, the eye-tracking image contains physiological features 92 such as the pupil, iris 922, and sclera 921. The processor 15 can track the angle of one or more of the aforementioned physiological features 92 and determine whether they are within a default angle range (e.g., the angle range of the gaze towards the display screen 11 of the electronic device 10). If the angle of the physiological feature 92 is not within the preset angle range, the physiological state detection method ends (step S206) to avoid wasting computing resources; if the angle of the physiological feature 92 is within the preset angle range, the pixel range containing physiological features 92 such as the left and right eyes is further captured (step S202). Therefore, in the context of eye physiological characteristics 92, the processor 15 can quickly filter out physiologically significant image data from eye-tracking images.

[0056] Please refer to the above as well. Figure 5 , Figure 6B and Figure 6C In this embodiment, the processor 15 extracts the pixel range containing each iris 922 (step S203) and determines whether the feature points match the template feature points (step S204). For example, in some embodiments, the memory 14 of the electronic device 10 stores an image database for recording template images. The template image can be a visible light image or a non-visible light image, which contains multiple template feature points, such as the outline, corners, color, and color depth of the physiological feature 92. The template image can be generated through a registration process (e.g., the electronic device 10 instructs the user to take a picture of the eye features and stores the picture as a template image) or through long-term recording of the user by the electronic device 10. The processor 15 extracts the multiple feature points contained in the pixel range and determines whether the similarity between the multiple feature points and the multiple template feature points is greater than a similarity threshold. Figure 6B andFigure 6C As shown, the processor 15 distinguishes the features of the iris 922 of the left and right eyes through image recognition to obtain a first pixel range 91 containing the iris 922 and a first pixel range 91 containing the sclera 921. Among them, the feature points such as fiber bundle radial patterns, concave points, spots, and pigment distribution contained in the iris 922 can be used for similarity comparison.

[0057] When the processor 15 determines that the similarity is less than the similarity threshold, it means that the feature point does not match the template feature point (step S204, the determination result is "No"), and the physiological state detection method ends (step S206) to avoid unauthorized operations; when the similarity is greater than the similarity threshold, it means that the feature point matches the template feature point (step S204, the determination result is "Yes"), and the electronic device 10 is unlocked (step S205). In some embodiments, the processor 15 extracts multiple feature points contained in the first pixel range 91 from the infrared light image 90 captured by the infrared light camera 122. In this way, the electronic device 10 can perform authorization verification without interfering with the user, to determine whether to unlock the electronic device 10 and further perform other operations (e.g., execute steps S207 to S209); in this embodiment, the electronic device 10 can only turn on the visible light camera 121 after the user verification is completed by the infrared light camera 122, to avoid interference with the user by supplementary lighting and power consumption issues. By leveraging the characteristic that infrared light image 90 can clearly display images without being affected by visible light, and the default template image, electronic device 10 can quickly perform feature comparison to achieve user identification function.

[0058] In some embodiments, the image database includes template images and user tags corresponding to each template image. User tags may refer to a user's name, nickname, alias, registered email address, registered account, or other parameters sufficient to distinguish individuals. The processor 15 can store physiological parameters measured for each user in the image database and correspond to the user's user tag. For example, after unlocking the electronic device 10 through verification (step S205), the electronic device 10 confirms the user tag and performs physiological measurements and parameter recording. However, in other embodiments, the electronic device 10 can also record physiological parameters without unlocking to achieve real-time (continuous monitoring or camera activation under specific conditions, such as when the face distance is less than a distance threshold) or timed (e.g., camera activation every half hour) physiological monitoring.

[0059] Rereference Figure 5In this embodiment, the processor 15 captures the pixel range including each sclera 921 (step S207) and calculates physiological parameters based on the green pixel values ​​and red pixel values ​​(step S208). For example, in some embodiments, the processor 15 reads the red pixel values ​​and green pixel values ​​within a second pixel range from the visible light image captured by the visible light camera 121. Then, the processor 15 determines the number of pixels within the second pixel range with green pixel values ​​less than a preset green level threshold (i.e., the number of green pixels) and the number of pixels within the second pixel range with red pixel values ​​greater than a preset red level threshold (i.e., the number of red pixels). The green level threshold and the red level threshold can be preset absolute values, for example, defined as half the upper and lower limits of the green pixel value range; or they can be relative values ​​calculated based on the pixel values ​​within the second pixel range, for example, defined as a multiple of the average, median, or standard deviation of all green pixel values. The processor 15 calculates the pixel ratio of the number of green pixels to the number of red pixels and determines the physiological state based on the pixel ratio (step S209). The physiological state can be represented by indicators such as fatigue level, intraocular pressure abnormality, eye health level, and mental stress.

[0060] Figure 7 This is a schematic diagram illustrating physiological states and different degrees of these states according to some embodiments of the present invention. Please refer to it. Figure 7 Considering the positive correlation between the degree of eye bleeding and the degree of fatigue, for example, in state level 3, the lower left corner of the eye shows localized blood vessels (923) due to blood leakage from microvessels, while in state level 4, the right side of the eye shows extensive blood vessels (923). The pixel count ratio represents the ratio of the number of pixels with bleeding points to the number of pixels without bleeding points. Specifically, the processor 15 captures a second pixel range within the visible light image that only includes the sclera (921) and obtains the pixel count ratio of the number of green pixels to the number of red pixels. The applicant found in the study that in state level 1, the pixel count ratio range can be defined as below 5%, which represents the physiological state of the eye under normal conditions; in state level 2, the pixel count ratio range can be defined as 5% to 20%, which represents the physiological state of the eye under mild fatigue conditions; in state level 3, the pixel count ratio range can be defined as 20% to 50%, which represents the physiological state of the eye under moderate fatigue conditions; and in state level 4, the pixel count ratio range can be defined as above 50%, which represents the physiological state of the eye under severe fatigue conditions. In some embodiments, when the state level exceeds the default upper limit, the electronic device 10 can generate an alarm notification to the user via the display screen 11, a speaker, or a vibration motor to suggest a rest.

[0061] In some embodiments, the processor 15 simultaneously defines the physiological state based on the pixel count ratios obtained from the left and right eyes. For example, when the pixel count ratios corresponding to the second pixel range including the left sclera 921 and the second pixel range including the right sclera 921 are both greater than a specific state level, the physiological state is determined by the specific state level. For example, when the processor 15 determines that the pixel count ratio of the left eye is 25% (state level 3) and the pixel count ratio of the right eye is 10% (state level 2), the user's physiological state is state level 2. This avoids external factors causing unilateral eye injury and bleeding (e.g., rubbing the eyes) from affecting the accuracy of the physiological state determination. In some embodiments, when the processor 15 determines that a specific physiological characteristic 92 (e.g., left or right eye) is at a specific state level, the specific state level is stored in the memory 14 and associated with a user tag. Subsequently, the processor 15 can extract features from the infrared image 90 and the visible light image respectively according to the specific state level, and compare the local regions in the images that conform to the specific physiological feature 92 to define the first pixel range 91 and the second pixel range. In this way, the processor 15 can adapt to the differences in individual physiological states and quickly obtain the first pixel range 91 and the second pixel range containing the specific physiological feature 92.

[0062] like Figure 6A As shown, in some embodiments, the processor 15 captures a pixel range including the center of the eyebrows 924 and calculates physiological parameters based on green and red pixel values. For example, in some embodiments, the processor 15 simultaneously monitors physiological features 92 such as the eyes and the center of the eyebrows 924 to avoid interference from factors such as glasses or bangs affecting physiological monitoring. In this embodiment, the processor 15 captures multiple first pixel ranges 91 within the infrared image 90, including the center of the eyebrows 924, the left eye, and the right eye, and reads the grayscale pixel values ​​within each first pixel range 91; furthermore, it captures multiple second pixel ranges within the visible light image, including the center of the eyebrows 924, the left eye, and the right eye, and reads the red pixel values ​​within each second pixel range. Furthermore, by calculating the ratio of the absorbance corresponding to the red pixel value to the absorbance corresponding to the grayscale pixel value, the blood oxygen concentration is estimated. Alternatively, the tissue oxygen consumption rate or metabolic rate is further estimated based on changes in blood oxygen concentration.

[0063] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A method for detecting physiological states, characterized in that, An electronic device is applicable, the electronic device comprising a display screen and a visible light camera and an infrared camera disposed above the display screen, the detection method comprising: Receive an infrared image generated by the infrared camera; Receive a visible light image generated by the visible light camera; Extract a first pixel range containing a physiological feature from the infrared image, and read a grayscale pixel value within the first pixel range; Extract a second pixel range containing the physiological feature within the visible light image, and read a red pixel value within the second pixel range; and A physiological parameter is calculated based on the grayscale pixel values ​​and the red pixel values.

2. The physiological state detection method as described in claim 1, characterized in that, The electronic device further includes a short-wavelength ambient light sensor and a long-wavelength ambient light sensor above the display screen, and the detection method includes: Receive an ambient visible light signal generated by the short-wavelength ambient light sensor, and correct the visible light image based on the ambient visible light signal; and The system receives an ambient infrared light signal generated by the long-wavelength ambient light sensor and corrects the infrared light image based on the ambient infrared light signal.

3. The physiological state detection method as described in claim 1, characterized in that, It also includes: Extract multiple feature points contained within the first pixel range; and The visible light image is compared based on the plurality of feature points to extract the second pixel range within the visible light image that contains the physiological feature.

4. The physiological state detection method as described in claim 1, characterized in that, It also includes: Read the value of a green pixel within the second pixel range; Determine the number of green pixels within the second pixel range that have a green pixel value less than a green level threshold; Determine the number of red pixels within the second pixel range that have a red pixel value greater than a red level threshold; Calculate the pixel ratio of the number of green pixels to the number of red pixels; as well as A physiological state is determined based on the pixel count ratio.

5. The physiological state detection method as described in claim 4, characterized in that, The physiological characteristics include the left and right eyes, and the physiological state detection method includes: Separately capture one second pixel range containing the left eye and another second pixel range containing the right eye within the visible light image, and read the red pixel value within each second pixel range; and The physiological state is determined by determining that the ratio of the number of pixels corresponding to each of the second pixel ranges is greater than a state degree.

6. The physiological state detection method as described in claim 5, characterized in that, The physiological feature includes the area between the eyebrows, and the physiological state detection method includes: Extract the first pixel range containing the center of the eyebrows within the infrared image, and read the grayscale pixel values ​​within the first pixel range; and The second pixel range containing the center of the eyebrows is captured within the visible light image, and the red pixel value within the second pixel range is read.

7. The physiological state detection method as described in claim 1, characterized in that, After receiving the infrared light image generated by the infrared light camera and before receiving the visible light image generated by the visible light camera, the method further includes determining, based on the infrared light image, that the distance between the electronic device and the face is less than a distance threshold before receiving the visible light image.

8. The physiological state detection method as described in claim 7, characterized in that, The electronic device further includes an eye tracking sensor disposed above the display screen, and the physiological state detection method further includes: Receive an eye-tracking image generated by the eye-tracking sensor; and After determining that the physiological features contained in the eye-tracking image are within a default angle range, and after determining that the distance between the electronic device and the face is less than the distance threshold based on the infrared light image, the first pixel range or the second pixel range is captured.

9. The physiological state detection method as described in claim 1, characterized in that, The electronic device further includes an image database, and after the step of reading the grayscale pixel values ​​within the first pixel range, it further includes: Extract multiple feature points contained within the first pixel range; Read a template image contained in the image database, the template image containing multiple template feature points; and If the similarity between the plurality of feature points and the plurality of template feature points is greater than a similarity threshold, the electronic device is unlocked.

10. The physiological state detection method as described in claim 9, characterized in that, The image database further includes a user tag corresponding to the template image, and the physiological state detection method further includes storing the physiological parameters in the image database and corresponding to the user tag.

Images