Mobile phone camera dual-channel fusion-based body temperature trend estimation method and system

Abstract

The present application relates to the field of smart phone health monitoring, in particular to a body temperature trend estimation method and system based on a dual-channel fusion of a phone camera, wherein the method comprises the following steps: in a measurement window, synchronously collecting device thermal state data, camera local thermal response features and camera PPG signals of the phone, and respectively constructing a camera area thermal response channel and a camera PPG physiological signal channel; identifying the interference type in the measurement process; establishing a nonlinear thermal compensation parameter based on the interference type; individually calibrating a user profile; weighting and fusing the camera area thermal response channel, the camera PPG physiological signal channel, the device thermal state data, the nonlinear thermal compensation parameter and the user profile, and outputting a body temperature trend estimation value, a confidence level and an abnormality marker; and outputting a non-diagnostic risk prompt based on the confidence level and the abnormality marker. The present application realizes convenient and reliable body temperature trend estimation through dual-channel weighting fusion of a phone camera.

Description

Technical Field

[0001] This invention relates to the field of health monitoring in smartphones, specifically to a method and system for estimating body temperature trends based on dual-channel fusion of smartphone cameras. Background Technology

[0002] Body temperature detection and temperature trend monitoring play an important role in health management, disease prevention, and daily status assessment. Currently, body temperature detection or temperature trend estimation solutions mainly include traditional thermometers, wearable devices, external sensors, and various solutions based on the hardware of smartphones themselves.

[0003] Traditional thermometer solutions, including mercury thermometers, electronic thermometers, and infrared ear thermometers, require users to actively operate and maintain a specific posture during measurement. They are mostly single, instantaneous measurements, making it difficult to establish continuous temperature trends and thus failing to meet the needs of routine daily monitoring. Wearable device solutions integrate temperature sensors or PPG sensors into smart bracelets, smartwatches, or patch-type devices to continuously collect body temperature or physiological signals. However, these solutions rely on users wearing dedicated hardware, increasing usage costs and burden. Furthermore, the accuracy and algorithms of sensors from different brands and models vary, hindering widespread adoption among ordinary users. External sensor solutions connect to external temperature probes or infrared temperature measurement modules via a mobile phone interface, using the phone as a display and transmission terminal. While utilizing the phone's computing power and interactive capabilities, they still fundamentally rely on additional hardware, presenting a high barrier to entry for users and making it difficult to achieve rapid, anytime, anywhere measurement.

[0004] While smartphones typically incorporate temperature monitoring points such as battery temperature sensors, SoC / CPU / GPU temperature sensors, and system thermal zones, these sensors are primarily designed for device thermal management and system protection, not for body temperature estimation. Existing technologies have attempted to estimate ambient or human contact temperature using changes in the phone's internal temperature. However, when using only the phone's internal temperature, factors such as device self-heating, charging heat, high-load operation, heat insulation from the phone case and screen protector, heat transfer from holding the phone, and changes in ambient temperature all significantly interfere with the measurement results. Differences in heat dissipation design across different models also lead to inconsistent temperature response characteristics, making it difficult to establish a unified body temperature mapping model, resulting in insufficient accuracy and reliability in actual estimations. On the other hand, technologies that use smartphone cameras with flash to collect finger PPG signals to obtain physiological parameters such as heart rate and blood oxygen saturation are publicly available. However, these parameters are significantly affected by exercise, emotions, individual differences, and signal quality. Relying solely on heart rate to estimate body temperature lacks a clear physiological mechanism and cannot eliminate interference from other factors, making it difficult to independently provide a reliable judgment of body temperature trends in practical applications.

[0005] Therefore, there is an urgent need for a solution that can estimate body temperature trends using only the hardware of a regular smartphone, without the need for additional wearable devices or external sensors, in order to overcome the shortcomings of the existing technologies mentioned above. Summary of the Invention

[0006] To address the problem that existing technologies struggle to accurately estimate body temperature trends by relying solely on the phone's internal temperature or the camera's PPG signal, this invention provides a body temperature trend estimation method and system based on dual-channel fusion of a mobile phone camera. This method collaboratively utilizes the thermal response characteristics of the camera area and the camera's PPG physiological signal within the same measurement window, achieving body temperature trend estimation without relying on wearable devices, external sensors, or dedicated infrared temperature measurement modules through dual-channel weighted fusion.

[0007] In a first aspect, the present invention provides a method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras, specifically including the following steps: Within a single measurement window, the device thermal status data of the mobile phone, the local thermal response characteristics of the camera, and the camera PPG signal are collected simultaneously. A camera region thermal response channel is constructed based on the local thermal response characteristics of the camera, and a camera PPG physiological signal channel is constructed based on the camera PPG signal. Identify the types of interference during the measurement process, including interference from phone cases and interference from screen protectors; Based on the interference type, nonlinear thermal compensation parameters are established, and the phone case interference is converted into phone case thermal resistance parameters, phone case heat dissipation delay parameters and phone case local temperature offset parameters. The screen protector interference is converted into screen protector thermal resistance parameters, screen protector heat dissipation delay parameters and screen protector local temperature offset parameters. The user profile is individually calibrated, which includes historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individual offset, profile reliability and calibration drift score; The camera area thermal response channel, the camera PPG physiological signal channel, the device thermal status data, the nonlinear thermal compensation parameters, and the user profile are weighted and fused to output the estimated body temperature trend, confidence level, and anomaly marker. Based on the confidence level and the anomaly marker, output a non-diagnostic risk warning.

[0008] Furthermore, the device thermal status data includes: battery temperature, SoC temperature, CPU temperature, GPU temperature, charging status, device load status, and system thermal zone related data; the camera area thermal response channel includes: camera area temperature proxy value, camera area thermal response delay, local temperature offset of the camera area relative to the average internal temperature of the device, camera area thermal trend, camera area reliability, camera working status, and measurement window timestamp; the camera PPG physiological signal channel includes: heart rate, heart rate trend, resting heart rate reference, PPG signal quality, motion artifacts, finger contact quality, and physiological channel reliability.

[0009] Furthermore, the interference from the phone case is converted into phone case thermal resistance parameters, phone case heat dissipation delay parameters, and phone case local temperature deviation parameters, wherein the phone case thermal resistance parameters... The calculation formula is:

[0010] in, This indicates the basic thermal resistance of the phone case. The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal response reliability of the camera area; The heat dissipation delay parameter of the mobile phone case The calculation formula is:

[0011] in, Indicates the thermal stability of the equipment; The local temperature deviation parameter of the phone case The calculation formula is:

[0012] in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

[0013] Furthermore, the film interference is converted into film thermal resistance parameters, film heat dissipation delay parameters, and film local temperature offset parameters, wherein the film thermal resistance parameters... The calculation formula is:

[0014] in, This indicates the basic thermal resistance of the film. This represents the normalized intensity of the local temperature shift. Indicates the thermal response reliability of the camera area; The heat dissipation delay parameter of the film The calculation formula is:

[0015] in, The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal stability of the equipment; The local temperature deviation parameter of the film The calculation formula is:

[0016] in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

[0017] Furthermore, the estimated body temperature trend value The calculation formula is:

[0018] in, This represents the user's historical baseline body temperature. This indicates the trend contribution of camera area thermal response channels and device thermal status data. This indicates the trend contribution generated by the camera's PPG physiological signal channel. This indicates the trend contribution generated by the user profile. Represents the nonlinear thermal compensation parameters. Indicates the thermal response weight. Represents the weight of physiological signals. This indicates the weight of the user profile.

[0019] Furthermore, the formula for calculating the confidence level is as follows:

[0020] in, This represents the upper limit of the final confidence level determined by the high-temperature failure state. Indicates the multi-source quality-weighted basic confidence level. The penalty coefficient represents the level of severe interference. Indicates the intensity of severe interference. This represents the penalty coefficient for calibration drift. This indicates the calibration drift score in the user profile. This indicates a combined penalty resulting from high thermal failure, equipment thermal saturation, and unreliable cameras.

[0021] Furthermore, the abnormal markers include: high thermal failure marker, device thermal saturation marker, unreliable thermal response marker in camera area, low PPG signal quality marker, charging interference marker, high load interference marker, high ambient temperature marker, grip interference marker, and calibration drift marker.

[0022] Furthermore, the non-diagnostic risk warnings include: remeasurement warning, confirmation warning using a standard thermometer warning, unreliable thermal response warning for the camera area warning, low PPG signal quality warning, high thermal failure warning, and insufficient current confidence warning.

[0023] Secondly, the present invention provides a body temperature trend estimation system based on dual-channel fusion of mobile phone camera, comprising: The device thermal status acquisition module is used to acquire the device thermal status data of the mobile phone in a single measurement window; The camera area thermal response channel construction module is used to collect local thermal response characteristics of the camera within the single measurement window and construct the camera area thermal response channel. A camera PPG physiological signal channel construction module is used to acquire camera PPG signals and construct camera PPG physiological signal channels within the single measurement window. An interference identification module is used to identify the type of interference during the measurement process, including interference from phone cases and interference from screen protectors. The thermal compensation module is used to establish nonlinear thermal compensation parameters, convert the phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters and phone case local temperature offset parameters, and convert the film interference into film thermal resistance parameters, film heat dissipation delay parameters and film local temperature offset parameters. The individualized calibration module is used to perform individualized calibration on user profiles, which include historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individualized offset, profile reliability, and calibration drift score. The weighted fusion estimation module is used to perform weighted fusion of the camera area thermal response channel, the camera PPG physiological signal channel, the device thermal state data, the nonlinear thermal compensation parameters and the user profile, and output the estimated body temperature trend value, confidence level and abnormality marker. The risk alert module is used to output non-diagnostic risk alerts based on the confidence level and the anomaly marker.

[0024] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention can estimate body temperature trends using only the built-in camera, flash, and system temperature interface of a regular smartphone. Users do not need to purchase or wear any wearable devices, external sensors, or dedicated infrared temperature measurement modules, which reduces the cost of use and the operational threshold. It is suitable for smartphones that can acquire camera data and thermal status data of at least one type of device. When some thermal status data is unavailable or missing, the system forms a quality score based on the available data and lowers the corresponding confidence level.

[0025] (2) This invention acquires the thermal response characteristics of the camera area and the camera PPG physiological signal within the same measurement window, and achieves the coordinated utilization of the device's thermal state and the user's physiological state through weighted fusion estimation. Compared with schemes that rely solely on temperature sensors or solely on PPG heart rate, this invention effectively suppresses estimation bias caused by interference from a single signal source through dual-channel mutual verification and compensation, thereby improving the reliability of body temperature trend estimation.

[0026] (3) This invention quantifies the interference of phone cases and screen protectors on heat conduction into thermal resistance parameters, heat dissipation delay parameters, and local temperature deviation parameters, and corrects them through nonlinear thermal compensation, rather than simply prompting users to remove the phone case or screen protector. This enables the invention to work normally in real-world usage scenarios, significantly improving the practicality and user acceptance of the solution.

[0027] (4) This invention maintains an independent individualized profile for each user, including parameters such as historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, and individualized offset, and performs bounded online updates based on the quality of the calibration records. Compared with fixed models or population statistical models, individualized calibration can adapt to the differences in physiological characteristics of different users, gradually improving the estimation accuracy with the increase in the number of uses, while the bounded update strategy effectively prevents the contamination of the profile by a single abnormal calibration.

[0028] (5) While outputting the estimated value of body temperature trend, the present invention outputs confidence level and abnormality mark, and outputs corresponding non-diagnostic risk warning according to the level of confidence level and the type of abnormality mark. This can help users understand the reliability of the current measurement results and avoid high-confidence error output when the measurement conditions are poor or the system status is abnormal. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a flowchart of a body temperature trend estimation method based on dual-channel fusion of mobile phone camera according to the present invention.

[0031] Figure 2 This is a block diagram of the overall structure of a body temperature trend estimation system based on dual-channel fusion of mobile phone camera according to the present invention.

[0032] Figure 3 This is a schematic diagram of the thermal response channel in the camera area of ​​the present invention.

[0033] Figure 4 This is a schematic diagram of the PPG physiological signal channel of the camera in this invention.

[0034] Figure 5 This is a comprehensive diagram of interference identification, thermal compensation, high thermal failure, individualized calibration, and confidence control in this invention. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0037] This invention provides the following technical solutions: like Figure 1 As shown, a method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras mainly includes the following steps: S1. In a single measurement window, simultaneously collect the device thermal status data of the mobile phone, the local thermal response characteristics of the camera, and the camera PPG signal. Construct a camera region thermal response channel based on the local thermal response characteristics of the camera, and construct a camera PPG physiological signal channel based on the camera PPG signal.

[0038] In a preferred embodiment of this application, the device thermal status data includes: battery temperature, SoC temperature, CPU temperature, GPU temperature, charging status, device load status, and system thermal zone related data; the camera area thermal response channel includes: camera area temperature proxy value, camera area thermal response delay, local temperature offset of the camera area relative to the average internal temperature of the device, camera area thermal trend, camera area reliability, camera working status, and measurement window timestamp; the camera PPG physiological signal channel includes: heart rate, heart rate trend, resting heart rate reference, PPG signal quality, motion artifacts, finger contact quality, and physiological channel reliability.

[0039] S2. Identify the types of interference during the measurement process, including interference from phone cases and interference from screen protectors.

[0040] As a preferred embodiment of this application, the interference types also include handheld interference, charging interference, high load interference, high environmental heat interference, unreliable thermal response in the camera area, and low PPG signal quality.

[0041] S3. Based on the interference type, establish nonlinear thermal compensation parameters, convert the phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters, and phone case local temperature offset parameters, and convert the film interference into film thermal resistance parameters, film heat dissipation delay parameters, and film local temperature offset parameters.

[0042] S31. Convert the phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters, and phone case local temperature deviation parameters, wherein the phone case thermal resistance parameters The calculation formula is:

[0043] in, This indicates the basic thermal resistance of the phone case. The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal response reliability of the camera area; The heat dissipation delay parameter of the mobile phone case The calculation formula is:

[0044] in, Indicates the thermal stability of the equipment; The local temperature deviation parameter of the phone case The calculation formula is:

[0045] in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

[0046] S32. Convert the film interference into film thermal resistance parameters, film heat dissipation delay parameters, and film local temperature offset parameters, wherein the film thermal resistance parameters The calculation formula is:

[0047] in, This indicates the basic thermal resistance of the film. This represents the normalized intensity of the local temperature shift. Indicates the thermal response reliability of the camera area; The heat dissipation delay parameter of the film The calculation formula is:

[0048] in, The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal stability of the equipment; The local temperature deviation parameter of the film The calculation formula is:

[0049] in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

[0050] S4. Perform individualized calibration on the user profile, which includes historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individualized offset, profile reliability, and calibration drift score.

[0051] S5. The camera area thermal response channel, the camera PPG physiological signal channel, the device thermal status data, the nonlinear thermal compensation parameters and the user profile are weighted and fused to output the estimated body temperature trend, confidence level and abnormality marker.

[0052] S51, the estimated body temperature trend The calculation formula is:

[0053] in, This represents the user's historical baseline body temperature. This indicates the trend contribution of camera area thermal response channels and device thermal status data. This indicates the trend contribution generated by the camera's PPG physiological signal channel. This indicates the trend contribution generated by the user profile. Represents the nonlinear thermal compensation parameters. Indicates the thermal response weight. Represents the weight of physiological signals. This indicates the weight of the user profile.

[0054] S52, The formula for calculating the confidence level is:

[0055] in, This represents the upper limit of the final confidence level determined by the high-temperature failure state. Indicates the multi-source quality-weighted basic confidence level. The penalty coefficient represents the level of severe interference. Indicates the intensity of severe interference. This represents the penalty coefficient for calibration drift. This indicates the calibration drift score in the user profile. This indicates a combined penalty resulting from high thermal failure, equipment thermal saturation, and unreliable cameras.

[0056] S53, the abnormal markers include: high thermal failure marker, device thermal saturation marker, unreliable thermal response marker in camera area, low PPG signal quality marker, charging interference marker, high load interference marker, high environmental heat marker, grip interference marker, and calibration drift marker.

[0057] S6. Output a non-diagnostic risk warning based on the confidence level and the anomaly marker.

[0058] As a preferred embodiment of this application, the non-diagnostic risk warnings include: remeasurement warning, confirmation warning using a standard thermometer warning, unreliable thermal response warning for the camera area warning, low PPG signal quality warning, high thermal failure warning, and insufficient current confidence warning.

[0059] This invention discloses a body temperature trend estimation system based on dual-channel fusion of mobile phone camera, comprising: The device thermal status acquisition module is used to acquire the device thermal status data of the mobile phone in a single measurement window; The camera area thermal response channel construction module is used to collect local thermal response characteristics of the camera within the single measurement window and construct the camera area thermal response channel. A camera PPG physiological signal channel construction module is used to acquire camera PPG signals and construct camera PPG physiological signal channels within the single measurement window. An interference identification module is used to identify the type of interference during the measurement process, including interference from phone cases and interference from screen protectors. The thermal compensation module is used to establish nonlinear thermal compensation parameters, convert the phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters and phone case local temperature offset parameters, and convert the film interference into film thermal resistance parameters, film heat dissipation delay parameters and film local temperature offset parameters. The individualized calibration module is used to perform individualized calibration on user profiles, which include historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individualized offset, profile reliability, and calibration drift score. The weighted fusion estimation module is used to perform weighted fusion of the camera area thermal response channel, the camera PPG physiological signal channel, the device thermal state data, the nonlinear thermal compensation parameters and the user profile, and output the estimated body temperature trend value, confidence level and abnormality marker. The risk alert module is used to output non-diagnostic risk alerts based on the confidence level and the anomaly marker.

[0060] Example like Figure 2 As shown, the system described in this invention runs on a smartphone. The system includes a device thermal status acquisition module, a camera area thermal response channel construction module, a camera PPG physiological signal channel construction module, an interference identification module, a thermal compensation module, an individualized calibration module, a weighted fusion estimation module, and a risk warning module.

[0061] The smartphone has a camera, flash, processor, battery, system thermal zone interface, and storage module. After the user initiates the measurement, covering the camera and flash area with their finger will initiate a measurement window.

[0062] In this invention, a single measurement window refers to the continuous time period from the start of measurement to the output of measurement results. Specifically, it includes the complete process from the user initiating the measurement operation, through data acquisition, dual-channel construction, interference identification, thermal compensation, individualized calibration, and weighted fusion, to the system outputting the estimated body temperature trend, confidence level, and risk warning. The duration of the single measurement window is dynamically determined based on PPG signal quality, camera area thermal response stability, device thermal state change rate, and user operation consistency, without the need for a preset fixed duration. The system can set a maximum acquisition time limit; if data meeting quality requirements is not obtained before reaching the limit, the current measurement window is automatically terminated and a remeasurement prompt is displayed.

[0063] The synchronous or quasi-synchronous formation of the camera area thermal response channel and the camera PPG physiological signal channel within the same measurement window means that the data acquisition, generation, or update of both channels occurs within the time range of that measurement window. They do not necessarily need to be strictly aligned to the same sampling time; as long as the timestamps corresponding to each channel fall within the single measurement window and the timestamp differences are within a preset deviation range, they can be considered synchronous or quasi-synchronous. For cases where timestamp discrepancies exist, the system can establish a correspondence between the two channels' data using linear interpolation, window averaging, or timestamp alignment algorithms before inputting the data into the weighted fusion estimation module.

[0064] In a preferred embodiment of the present invention, the method operates on a single smartphone, allowing users to estimate body temperature trends using only their smartphone without needing to wear additional devices or connect external sensors. It should be noted that although the present invention is described using a single smartphone as an example in the preferred embodiment, those skilled in the art should understand that the method is equally applicable to other mobile terminal devices with cameras and system temperature interfaces. The technical principles and implementation methods of these mobile terminal devices are completely identical to those of smartphones, and all fall within the scope of protection of this invention.

[0065] S1. In a single measurement window, simultaneously collect the device thermal status data of the mobile phone, the local thermal response characteristics of the camera, and the camera PPG signal. Construct a camera region thermal response channel based on the local thermal response characteristics of the camera, and construct a camera PPG physiological signal channel based on the camera PPG signal.

[0066] S11. Collect equipment thermal status data The system continuously collects or periodically reads the device thermal status data of the mobile phone through the device thermal status acquisition module at a preset sampling frequency.

[0067] In a preferred embodiment, the device thermal status data includes: battery temperature, SoC temperature, CPU temperature, GPU temperature, charging status, device load status, and system thermal zone related data. The device thermal status data can be read through operating system open interfaces, hardware abstraction layers, vendor interfaces, or accessible system thermal zones. When some thermal status data is unavailable or missing, the system generates a device thermal status quality score based on the remaining available data and reduces the corresponding channel weights and final confidence levels.

[0068] S12, Constructing the camera area thermal response channel like Figure 3As shown, the system obtains the local thermal response characteristics of the area surrounding the camera module or the area near the camera through the camera area thermal response channel construction module. The area surrounding the camera module is the area of ​​the mobile phone casing adjacent to the camera module, and is less likely to be completely covered by the phone case, screen protector, or lens film compared to the back of the mobile phone or the screen area, and is used to provide a local external heat exchange reference.

[0069] The sources of the camera area thermal response characteristics include system thermal zone readings, changes in the device's local thermal response before and after the camera is turned on, changes in camera image noise, exposure changes, brightness changes, frame stability-related thermal characteristics, or surrogate values ​​estimated based on the device's internal temperature gradient and the camera's operating state. It should be clarified that the camera area thermal response characteristics are not limited to readings from the camera's internal temperature sensor.

[0070] In a preferred embodiment, the camera area thermal response channel includes: camera area temperature proxy value, camera area thermal response delay, local temperature offset of the camera area relative to the average internal temperature of the device, camera area thermal trend, camera area reliability, camera operating status, and measurement window timestamp.

[0071] S13. Constructing the PPG physiological signal channel for the camera. like Figure 4 As shown, the system uses the camera's PPG physiological signal channel construction module to collect PPG signals and extract heart rate features within the measurement window.

[0072] Within the same measurement window, the user's finger covers the camera or flash area, and the smartphone camera captures video frames and extracts the PPG signal. The system extracts the current heart rate, heart rate trend, resting heart rate reference, PPG quality score, motion artifacts, finger contact quality, heart rate and body temperature sensitivity coefficients, and physiological channel reliability from the PPG signal.

[0073] It should be noted that the heart rate trend in the camera's PPG physiological signal channel is not used alone for body temperature measurement, but is combined with the camera area thermal response channel, device thermal status data, and user individual profiles to serve as a physiological aid for estimating body temperature trend.

[0074] S2. Identify interference types like Figure 5As shown, the system identifies factors such as phone cases, screen protectors, lens protectors, hand grip, charging, high load, high ambient heat, unreliable thermal response in the camera area, and low PPG quality. The system converts phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters, and phone case local temperature offset parameters. Similarly, it converts screen protector interference into screen protector thermal resistance parameters, screen protector heat dissipation delay parameters, and screen protector local temperature offset parameters. For charging, high load, high ambient heat, and hand grip, the system establishes corresponding thermal interference intensities.

[0075] The system can perform nonlinear thermal compensation based on device self-heating, ambient heat, phone case thermal resistance, screen protector thermal resistance, grip heat, and camera area thermal response reliability. When high-temperature environments, high device loads, charging heat generation, and device thermal instability occur simultaneously, the system can enter a high-thermal-failure mode. This high-thermal-failure mode is not a simple warning; rather, it reduces the weight of the camera area thermal response channel, the PPG physiological signal channel, the individualized calibration weight, and the final confidence level, thereby preventing high-confidence erroneous outputs caused by abnormal device heating, PPG quality degradation, or incorrect calibration records.

[0076] S3. Establish nonlinear thermal compensation parameters S31. Convert the phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters, and phone case local temperature deviation parameters, wherein the phone case thermal resistance parameters The calculation formula is:

[0077] in, This indicates the basic thermal resistance of the phone case. The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal response reliability of the camera area; In a preferred embodiment, when a phone case is detected... Take 0.22, otherwise take 0.04.

[0078] The heat dissipation delay parameter of the mobile phone case The calculation formula is:

[0079] in, Indicates the thermal stability of the equipment; The local temperature deviation parameter of the phone case The calculation formula is:

[0080] in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

[0081] S32. Convert the film interference into film thermal resistance parameters, film heat dissipation delay parameters, and film local temperature offset parameters, wherein the film thermal resistance parameters The calculation formula is:

[0082] in, This indicates the basic thermal resistance of the film. This represents the normalized intensity of the local temperature shift. Indicates the thermal response reliability of the camera area; In a preferred embodiment, when the film or lens coating is detected... Take 0.14, otherwise take 0.03.

[0083] The heat dissipation delay parameter of the film The calculation formula is:

[0084] in, The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal stability of the equipment; The local temperature deviation parameter of the film The calculation formula is:

[0085] in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

[0086] In a preferred embodiment, when the camera area is higher than the device's internal average, the compensation term is typically negative to suppress localized gripping or device self-heating; when the camera area is lower, the compensation term can be a small positive value. Upper and lower limits prevent exaggerated temperature corrections when local proxy values ​​are abnormal.

[0087] S33. The nonlinear thermal compensation parameters also include grip thermal correction parameters, charging thermal correction parameters, high load thermal correction parameters, environmental high thermal correction parameters, and equipment self-heating nonlinear correction parameters.

[0088] S331, Grip thermal correction parameters The calculation formula is:

[0089] in, This indicates the intensity of thermal interference during gripping.

[0090] S332, Charging Thermal Correction Parameters The calculation formula is:

[0091] in, This indicates the intensity of thermal interference during charging.

[0092] S333, High Load Thermal Correction Parameters The calculation formula is:

[0093] in, This indicates the intensity of thermal interference under high load.

[0094] S334, Environmental High-Temperature Correction Parameters The calculation formula is:

[0095] in, This indicates the intensity of high thermal interference in the environment.

[0096] S335, Equipment self-heating nonlinear correction parameters The calculation formula is:

[0097] in, Indicates the nonlinear compensation strength. This indicates the level of self-heating of the equipment.

[0098] S34, Total Heat Compensation The calculation formula is:

[0099] Average internal temperature of the equipment after compensation The calculation formula is:

[0100] in, This indicates the average effective temperature inside the equipment. This represents the effective compensation amount after conservative gating due to high-temperature failure.

[0101] S4. Perform individualized calibration on user profiles. The user profile includes historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individualized offset, profile reliability, and calibration drift score.

[0102] S41. The formula for online updating of the heart rate and body temperature sensitivity coefficients is:

[0103] in, This indicates the updated heart rate and body temperature sensitivity coefficient. This indicates the heart rate and body temperature sensitivity coefficients before the update. This indicates the step size for updating the sensitivity coefficient.

[0104] S42. The formula for online updating of historical body temperature baseline is:

[0105] in, This represents the updated individual's historical baseline body temperature. This represents the individual's historical baseline body temperature before the update. Indicates the first The actual weight of each calibration record in user profile updates This indicates the actual body temperature calibration point obtained from reference methods such as standard thermometers and input by the user.

[0106] S43. The formula for online updating of resting heart rate is:

[0107] in, This indicates the updated individual resting heart rate. This indicates the individual's resting heart rate before the update. Indicates the first The actual weight of each calibration record in user profile updates Indicates the first The current heart rate recorded during the next calibration.

[0108] S44. The formula for online updating of individualized offsets is:

[0109] in, This indicates the updated individual offset. This indicates the individual offset before the update. Indicates the first The actual weight of each calibration record in user profile updates This indicates the actual body temperature calibration point obtained through reference methods such as standard thermometers and input by the user. Indicates the first The body temperature value predicted by the user profile during the second calibration.

[0110] S45. The formula for online updating of user profile credibility is:

[0111] in, This indicates the credibility of the updated user profile. This indicates the credibility of the user profile before the update. Indicates the first The actual weight of each calibration record in user profile updates This represents the resting coefficient.

[0112] S46. The formula for online updating of calibration drift score is:

[0113] in, This indicates the updated calibration drift score. This indicates the calibration drift score before the update. This indicates the drift sample in this calibration.

[0114] S47. When the calibration record quality is low, the user is not at rest, the user is within a preset time after exercise, the device is charging, the device is under high load, or a high thermal failure mode is triggered, reduce the weight of the calibration record for user profile updates.

[0115] S5. The camera area thermal response channel, the camera PPG physiological signal channel, the device thermal status data, the nonlinear thermal compensation parameters and the user profile are weighted and fused to output the estimated body temperature trend, confidence level and abnormality marker.

[0116] S51, the estimated body temperature trend The calculation formula is:

[0117] in, This represents the user's historical baseline body temperature. This indicates the trend contribution of camera area thermal response channels and device thermal status data. This indicates the trend contribution generated by the camera's PPG physiological signal channel. This indicates the trend contribution generated by the user profile. Represents the nonlinear thermal compensation parameters. Indicates the thermal response weight. Represents the weight of physiological signals. This indicates the weight of the user profile.

[0118] S52, The formula for calculating the confidence level is:

[0119] in, This represents the upper limit of the final confidence level determined by the high-temperature failure state. This represents the multi-source quality-weighted basic confidence level. The penalty coefficient represents the level of severe interference. Indicates the intensity of severe interference. This represents the penalty coefficient for calibration drift. This indicates the calibration drift score in the user profile. This indicates a combined penalty resulting from high thermal failure, equipment thermal saturation, and unreliable cameras. This indicates a high-temperature failure state with a value of 0 or 1.

[0120] Multi-source quality-weighted basic confidence The calculation formula is:

[0121] in, This refers to the following: camera thermal response reliability, PPG quality, physiological channel reliability, device thermal condition quality, interference condition confidence, thermal compensation confidence, or user profile confidence. Quality indicators This indicates the positive weight of the corresponding quality indicator, and >0.

[0122] Furthermore, the intensity of severe interference The calculation formula is:

[0123] in, , , , , and These represent the intensity of interference caused by charging, high load, high ambient heat, gripping, motion artifacts, and device thermal saturation, respectively. It is a continuous intensity in the range of 0 to 1.

[0124] S53, the abnormal markers include: high thermal failure marker, device thermal saturation marker, unreliable thermal response marker in camera area, low PPG signal quality marker, charging interference marker, high load interference marker, high environmental heat marker, grip interference marker, and calibration drift marker.

[0125] In a preferred embodiment, a high-temperature failure flag is triggered when the overall high thermal pressure exceeds 0.46 and the number of active heat sources exceeds 2; a device thermal saturation flag is triggered when the device thermal saturation intensity exceeds 0.58; a camera area thermal response reliability flag is triggered when the camera area thermal response reliability is lower than a preset threshold; a PPG signal quality flag is triggered when the PPG signal quality score is lower than a preset threshold; a charging interference flag is triggered when the charging thermal interference intensity exceeds 0.45; a high load interference flag is triggered when the high load thermal interference intensity exceeds 0.45; an ambient high temperature flag is triggered when the ambient temperature exceeds 34°C; a grip thermal interference flag is triggered when the grip interference intensity exceeds a preset threshold; and a calibration drift flag is triggered when the calibration drift score exceeds a preset threshold.

[0126] S54. Determine if the high-temperature failure mode has been triggered. The high-temperature failure mode is triggered when at least two of the following conditions are met: high ambient temperature, high equipment load, charging heat generation, unstable equipment thermal state, equipment thermal saturation, or unreliable thermal response in the camera area.

[0127] When a high thermal failure mode is triggered, reduce the weight of the camera region thermal response channel, the weight of the camera PPG physiological signal channel, the weight of individualized calibration, and the final confidence level.

[0128] S6. Output a non-diagnostic risk warning based on the confidence level and the anomaly marker.

[0129] In a preferred embodiment, the non-diagnostic risk warnings include: a remeasurement warning, a confirmation warning using a standard thermometer, a warning of unreliable thermal response in the camera area, a warning of low PPG signal quality, a warning of high thermal failure, and a warning of insufficient current confidence.

[0130] This invention achieves body temperature trend estimation using only the built-in camera, flash, and system temperature interface of a regular smartphone, without requiring additional hardware. By simultaneously acquiring thermal response characteristics of the camera area and camera PPG physiological signals within the same measurement window and performing weighted fusion, it achieves coordinated utilization of the device's thermal state and the user's physiological state, effectively suppressing estimation bias from a single signal source and improving the reliability of body temperature trend estimation. By quantifying interference from phone cases and screen protectors into thermal resistance, heat dissipation delay, and local temperature shift parameters and performing nonlinear thermal compensation, the solution can adapt to real-world usage scenarios. By maintaining an independent, individualized profile for each user, it adapts to individual differences and prevents abnormal calibration from contaminating the profile, further ensuring the estimation accuracy and output reliability for long-term use.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating body temperature trends based on dual-channel fusion of mobile phone camera footage, characterized in that, Includes the following steps: Within a single measurement window, the device thermal status data of the mobile phone, the local thermal response characteristics of the camera, and the camera PPG signal are collected simultaneously. A camera region thermal response channel is constructed based on the local thermal response characteristics of the camera, and a camera PPG physiological signal channel is constructed based on the camera PPG signal. Identify the types of interference during the measurement process, including interference from phone cases and interference from screen protectors; Based on the interference type, nonlinear thermal compensation parameters are established, and the phone case interference is converted into phone case thermal resistance parameters, phone case heat dissipation delay parameters and phone case local temperature offset parameters. The screen protector interference is converted into screen protector thermal resistance parameters, screen protector heat dissipation delay parameters and screen protector local temperature offset parameters. The user profile is individually calibrated, which includes historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individual offset, profile reliability and calibration drift score; The camera area thermal response channel, the camera PPG physiological signal channel, the device thermal status data, the nonlinear thermal compensation parameters, and the user profile are weighted and fused to output the estimated body temperature trend, confidence level, and anomaly marker. Based on the confidence level and the anomaly marker, output a non-diagnostic risk warning.

2. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The device thermal status data includes: battery temperature, SoC temperature, CPU temperature, GPU temperature, charging status, device load status, and system thermal zone related data; the camera area thermal response channel includes: camera area temperature proxy value, camera area thermal response delay, local temperature offset of the camera area relative to the average internal temperature of the device, camera area thermal trend, camera area reliability, camera working status, and measurement window timestamp; the camera PPG physiological signal channel includes: heart rate, heart rate trend, resting heart rate reference, PPG signal quality, motion artifacts, finger contact quality, and physiological channel reliability.

3. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The interference from the phone case is converted into phone case thermal resistance parameters, phone case heat dissipation delay parameters, and phone case local temperature deviation parameters. The phone case thermal resistance parameters... The calculation formula is: in, This indicates the basic thermal resistance of the phone case. The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal response reliability of the camera area; The heat dissipation delay parameter of the mobile phone case The calculation formula is: in, Indicates the thermal stability of the equipment; The local temperature deviation parameter of the phone case The calculation formula is: in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

4. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The interference from the film application is converted into film thermal resistance parameters, film heat dissipation delay parameters, and film local temperature offset parameters. The film thermal resistance parameters... The calculation formula is: in, This indicates the basic thermal resistance of the film. This represents the normalized intensity of the local temperature shift. Indicates the thermal response reliability of the camera area; The heat dissipation delay parameter of the film The calculation formula is: in, The normalized intensity represents the thermal response delay of the camera area. Indicates the thermal stability of the equipment; The local temperature deviation parameter of the film The calculation formula is: in, This represents the difference between the temperature proxy value of the camera area and the average value inside the device.

5. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The estimated body temperature trend The calculation formula is: in, This represents the user's historical baseline body temperature. This indicates the trend contribution of camera area thermal response channels and device thermal status data. This indicates the trend contribution generated by the camera's PPG physiological signal channel. This indicates the trend contribution generated by the user profile. Represents the nonlinear thermal compensation parameters. Indicates the thermal response weight. Represents the weight of physiological signals. This indicates the weight of the user profile.

6. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The confidence level is calculated using the following formula: in, This represents the upper limit of the final confidence level determined by the high-temperature failure state. This represents the multi-source quality-weighted basic confidence level. The penalty coefficient represents the level of severe interference. Indicates the intensity of severe interference. This represents the penalty coefficient for calibration drift. This indicates the calibration drift score in the user profile. This indicates a combined penalty resulting from high thermal failure, equipment thermal saturation, and unreliable cameras.

7. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The anomaly markers include: high thermal failure marker, device thermal saturation marker, unreliable thermal response marker in camera area, low PPG signal quality marker, charging interference marker, high load interference marker, high environmental heat marker, grip interference marker, and calibration drift marker.

8. The method for estimating body temperature trends based on dual-channel fusion of mobile phone cameras according to claim 1, characterized in that, The non-diagnostic risk warnings include: remeasurement warning, confirmation warning using a standard thermometer warning, unreliable thermal response warning for the camera area warning, low PPG signal quality warning, high thermal failure warning, and insufficient current confidence warning.

9. A body temperature trend estimation system based on dual-channel fusion of mobile phone camera, used to implement the method described in any one of claims 1-8, characterized in that, include: The device thermal status acquisition module is used to acquire the device thermal status data of the mobile phone in a single measurement window; The camera area thermal response channel construction module is used to collect local thermal response characteristics of the camera within the single measurement window and construct the camera area thermal response channel. A camera PPG physiological signal channel construction module is used to acquire camera PPG signals and construct camera PPG physiological signal channels within the single measurement window. An interference identification module is used to identify the type of interference during the measurement process, including interference from phone cases and interference from screen protectors. The thermal compensation module is used to establish nonlinear thermal compensation parameters, convert the phone case interference into phone case thermal resistance parameters, phone case heat dissipation delay parameters and phone case local temperature offset parameters, and convert the film interference into film thermal resistance parameters, film heat dissipation delay parameters and film local temperature offset parameters. The individualized calibration module is used to perform individualized calibration on user profiles, which include historical body temperature baseline, resting heart rate, heart rate and body temperature sensitivity coefficient, individualized offset, profile reliability, and calibration drift score. The weighted fusion estimation module is used to perform weighted fusion of the camera area thermal response channel, the camera PPG physiological signal channel, the device thermal state data, the nonlinear thermal compensation parameters and the user profile, and output the estimated body temperature trend value, confidence level and abnormality marker. The risk alert module is used to output non-diagnostic risk alerts based on the confidence level and the anomaly marker.

Images