Camera system, method of controlling camera system, and image evaluation unit

Abstract

A camera system includes a camera (34) configured to capture a series of successive images of a person of interest within a field of view of the camera (34), and a control unit (38), the camera system (30) configured to determine a light intensity on the person of interest, and the control unit (38) is configured to control a camera setting of the camera (34) based on the light intensity determined by the camera system (30), where the control unit (38) is configured to alter the camera setting of the camera (34) only at defined time points (tx1, tx2,..., txn) and to alter the camera setting of the camera (34) when detecting that the light intensity on an object of interest changes between two defined time points (tx1, tx2,..., txn). The current camera setting is maintained until a next defined point in time (tx1, tx2,..., txn).

Description

Technical Field

[0001] The present invention relates to a camera system, a related method for controlling a camera system, and an image evaluation unit including the camera system. Background Art

[0002] It is generally necessary to determine physiological parameters to determine the mental state of a person, that is, whether the person is drowsy, sleepy or inattentive. Such information is typically used, for example, in a vehicle, and if it is determined that a person is considered unfit to drive, the vehicle can issue a warning. In this way, the risk of (fatal) accidents can be significantly reduced. Physiological parameters can be determined in a variety of different ways, such as, for example, heart rate, pulse, blood pressure, respiratory rate, and respiratory pattern. Some methods for determining physiological parameters require attaching one or more sensors to the person to be monitored. For example, this may be inconvenient for the driver of a vehicle. A very unobtrusive way of determining physiological parameters involves capturing images of the person being monitored and determining the physiological parameters by detecting subtle changes in the pixel intensity in subsequent images. However, especially in a moving vehicle, the (rapidly) changing lighting conditions can generate unwanted noise, which can have a negative impact on the accuracy of the determination of physiological parameters.

[0003] There is a need for a camera system, a method for controlling a camera system, and an image evaluation unit including the camera system, by means of which the influence of changing lighting conditions can be significantly reduced. Summary of the Invention

[0004] A camera system includes a camera and a control unit, the camera being configured to capture a series of consecutive images of a person of interest within the field of view of the camera, the camera system being configured to determine the light intensity on the person of interest, and the control unit being configured to control the camera settings of the camera based on the light intensity determined by the camera system, wherein the control unit is configured to change the camera settings of the camera only at defined time points and to maintain the current camera settings until the next defined time point when a change in the light intensity on the object of interest is detected between two defined time points.

[0005] An image evaluation unit includes a camera system and a physiological parameter determination unit, wherein the physiological parameter determination unit is configured to determine changes in pixel intensity in at least a defined region of interest in the images captured by the camera system and to determine one or more physiological parameters of the person of interest based on the changes in pixel intensity.

[0006] A method includes: capturing a series of consecutive images of a person of interest within the field of view of a camera, determining the light intensity on the person of interest, and controlling camera settings of the camera by a control unit based on the determined light intensity, wherein the camera settings of the camera are changed only at defined time points and, when a change in the light intensity on the object of interest is detected between two defined time points, maintaining the current camera settings until the next defined time point.

[0007] Other systems, features, and advantages of the present disclosure will be or will become apparent to those skilled in the art upon review of the following detailed description and the drawings. All such additional systems, methods, features, and advantages are intended to be included within this specification, are within the scope of the invention, and are protected by the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The apparatus and method can be better understood with reference to the following description and drawings. The components in the drawings are not necessarily drawn to scale, but rather emphasis is placed upon illustrating the principles of the invention. Additionally, in the drawings, the same reference numerals indicate the same components in all different views.

[0009] Figure 1 A vehicle including an image evaluation unit according to an embodiment of the present disclosure is schematically illustrated.

[0010] Figure 2 A method for determining the mental state of a person is schematically illustrated.

[0011] Figure 3 An audio system according to an embodiment of the present disclosure is schematically illustrated.

[0012] Figure 4 A vehicle traveling along a road is schematically illustrated.

[0013] Figure 5 A flowchart of a method according to an embodiment of the present disclosure is schematically illustrated. DETAILED DESCRIPTION

[0014] A camera system and related methods for controlling the camera system can significantly reduce the impact of (rapidly) changing lighting conditions on images captured by the camera system when determining physiological parameters of a person of interest. When, for example, analyzing images captured by the camera system by an image evaluation unit to determine physiological parameters of a person of interest (e.g., the driver of a vehicle), the determined results are very accurate.

[0015] Figure 1 A vehicle 10 including an image evaluation unit 200 according to an embodiment of the present disclosure is schematically illustrated. The image evaluation unit 200 includes a camera system 30 and a physiological parameter determination unit 32. The camera system 30 includes a camera 34 (see, for exampleFigure 3 ) and is configured to capture a series of consecutive images of a person of interest within the field of view of camera 34. When disposed in vehicle 10, camera 34 of camera system 30 can, for example, be directed towards driver's seat 102 of vehicle 10. Camera 34 has a defined field of view, and by directing camera 34 (i.e., the field of view of camera 34) towards driver's seat 102, the face of driver 40 sitting on driver's seat 102 is within the field of view of camera 34, as Figure 3 schematically shown in. Image evaluation unit 200 further includes a physiological parameter determination unit 32, which is configured to determine the pixel intensity changes in at least a defined region of interest in the images captured by camera system 30 and to determine one or more physiological parameters of the person of interest based on the pixel intensity changes.

[0016] It is generally necessary to determine physiological parameters to determine a person's mental state, i.e., whether the person is drowsy, sleepy or inattentive. Such information is typically used, for example, in a vehicle, and if it is determined that a person (i.e., driver 40 of vehicle 10) is considered unfit to drive, the vehicle can issue a warning. In this way, the risk of (fatal) accidents can be greatly reduced. Physiological parameters can include, for example, heart rate, pulse, blood pressure, respiratory rate and respiratory pattern. A very unobtrusive way of determining physiological parameters includes capturing multiple images of the person being monitored (the person of interest) and determining one or more physiological parameters by detecting subtle changes in pixel intensity in subsequent images.

[0017] Reference Figure 2 , a method for determining a person's mental state (e.g., whether driver 40 of vehicle 10 is drowsy, sleepy and / or inattentive) is schematically shown by a flow chart. The method includes determining physiological parameters of the person of interest (e.g., heart rate and / or heart rate variability) (step 201). Then, one or more physiological parameters are processed (step 202), and based on the one or more physiological parameters, the mental state of the person of interest is determined (step 203). Image evaluation unit 200 can be used to perform the first step of this method (step 201). That is, image evaluation unit 200 can be used to determine one or more physiological parameters of the person of interest. Optionally, image evaluation unit 200 can also perform the subsequent steps of the method (steps 202 and 203). However, such subsequent steps can be performed by other units separate and different from image evaluation unit 200.

[0018] Methods for determining physiological parameters of a person of interest by evaluating images captured by a camera 34 are generally known. However, such methods may be inaccurate as they involve determining one or more physiological parameters by detecting subtle changes in pixel intensities in subsequent images. Especially in a moving vehicle, (rapidly) changing lighting conditions can generate unwanted noise, which can negatively affect the accuracy of physiological parameter determination. Regarding Figure 4 This is described in more detail Figure 4 A vehicle 10 moving along a road is schematically shown. For example, as Figure 4 shown on the right, the vehicle 10 can drive through a tunnel. When driving through the tunnel (vehicle positions (A) and (B)), the lighting conditions may be poor, and camera settings matching the poor lighting conditions must be applied. For example, a relatively long exposure time and a low f-number can be set by the control unit 38. As long as the vehicle 10 drives through the tunnel, the lighting conditions may be substantially stable. However, due to irregular lighting in the tunnel, the lighting conditions in the tunnel may even change to some extent. For example, when the vehicle 10 passes a lamp, the lighting conditions may be slightly better, while when the vehicle 10 is between two subsequent lamps, the lighting conditions may be slightly worse. When the vehicle 10 leaves the tunnel, the lighting conditions may change significantly, especially when there is bright sunlight outside the tunnel. However, when the vehicle 10 is outside the tunnel, it may also pass objects such as trees or buildings, such that the lighting conditions may change rapidly between bright sunlight (vehicle position (C)) and shadow (vehicle position (D)).

[0019] Whenever the lighting conditions change, the camera settings are (automatically) adjusted to avoid underexposure or overexposure of the image. Camera settings that can be adjusted to avoid underexposure or overexposure can include, for example, automatic variable iris adjustment, where the aperture controls the light flow from the camera's optical subsystem to the photosensitive element. Closing the aperture reduces the light flowing to the photosensitive element of the camera 34, thereby reducing the brightness of the pixels in the case of a digital camera. Additionally or alternatively, camera settings that can be adjusted to avoid underexposure or overexposure can include, for example, automatic exposure control. The exposure time, i.e., the time the photosensitive element is exposed to the incident light, determines the total amount of energy reaching the photosensitive element and thus determines the brightness of the pixels. The exposure time and the f-number are typically related to automatic exposure control. Additionally or alternatively, camera settings that can be adjusted to avoid underexposure or overexposure can include various other techniques for analog or digital amplification of the signal received from the photosensitive element in order to obtain sufficient pixel brightness values in terms of human perception or processing algorithms. Generally speaking, any of the above methods will "multiply" the pixel brightness by a factor, and the result is to amplify or attenuate the resulting signal. Generally speaking, the use of any such adjustment technique can eliminate the effects of underexposure (in the case where the pixels of the image are too dark and thus not suitable for further analysis due to low brightness or difficulty in distinguishing between each other), or overexposure (in the case where the pixels are too bright and also not suitable for further analysis due to strong brightness and difficulty in distinguishing between each other), thereby providing an acceptable dynamic range for the captured object.

[0020] Changes in pixel intensity due to lighting condition changes and resulting camera setting changes are typically significantly greater than those due to skin color changes caused by the blood pulsation of the person of interest. Automatically adjusting the camera settings causes a sharp change in the properties of all pixels or at least a subset of pixels in the captured image (e.g., an increase or decrease in brightness). The magnitude of the change is relatively large and, in many cases, far greater than one pixel brightness unit. Heart rate estimation systems typically calculate the change in skin reflectance caused by the blood pulsation in blood vessels. The pulsation causes a change in the "color" and pixel intensity of the portion of the image representing the skin. Such systems can be based on both digital signal processing algorithms and systems using machine learning and artificial neural networks. For standard levels of camera technology, such pixel intensity changes due to skin color changes caused by the blood pulsation of the person of interest are typically difficult to match the dynamic range of the camera 34. In other words, pixel intensity changes related to physiological parameter determination typically have a magnitude comparable to or even smaller than a single unit of pixel brightness change. For example, for a standard camera where the illuminance component is encoded in 8 bits, corresponding to an illuminance range of 0 – 255, such changes are typically less than 1 / 255 of the illuminance range defined by the camera, or 1 unit. Averaging and using information from all relevant pixels (or at least a subset of pixels) in the image typically provides the possibility of a system with components having sufficient characteristics for useful applications. When the lighting conditions change rapidly and the camera settings are adjusted accordingly quickly, algorithms for evaluating pixel intensity to determine the physiological parameters of the person of interest can no longer provide reliable results.

[0021] Pixel brightness changes caused by adjustments to the camera settings rather than by the monitored physiological parameters themselves are typically regarded as noise. As described above, the magnitude of such noise can be significantly greater than the changes caused by the monitored physiological parameters. Noise caused by adjustments to the camera settings is typically transient and thus cannot (easily) be eliminated by appropriate processing techniques (e.g., by averaging the pixel brightness of a single image).

[0022] The camera system 30 according to an embodiment of the present disclosure (see, for example Figure 3) includes a camera 34 and a control unit 38. The camera 34 is configured to capture a series of consecutive images of a person of interest within the field of view of the camera 34. The camera system 30 is configured to determine the light intensity on the person of interest, and the control unit 38 is configured to control the camera settings of the camera 34 based on the light intensity determined by the camera system 30. To reduce the impact of noise caused by the automatic adjustment of the camera settings, the control unit 38 is configured to change the camera settings of the camera 34 only at defined time points tx1, tx2, ..., txn. When it is detected that the light intensity on the person of interest changes between two defined time points tx1, tx2, ..., txn, the current camera settings are maintained until the next defined time point tx1, tx2, ..., txn. The light intensity on the person of interest can be determined in any suitable manner. Most current cameras can determine the light intensity on the person (or object) of interest, for example, through a suitable sensor and evaluation circuit. For example, the light intensity on the person of interest can be determined by appropriately evaluating camera parameters such as, for example, pixel brightness, exposure time, variable aperture, magnification, etc. According to another example, the camera 34 included in the camera system 30 can have an integrated exposure meter, light meter, or photometer, which is configured to determine the light intensity on the person of interest. However, the camera system 30 can also include an exposure meter, light meter, or luminance meter 36 that is separate from the camera 34.

[0023] Reference Figure 4 , for example, when the vehicle 10 leaves the tunnel and the lighting conditions change, the automatic adjustment of the camera settings will be performed at time t1 in a conventional system. However, the time t1 in this example does not correspond to any defined time point tx1, tx2, tx3, tx4, etc. To avoid excessive changes to the camera settings within a short time range (for example, changing the camera settings every 1 to 4 seconds), regardless of whether the lighting conditions change, the camera settings remain unchanged until the next defined time point tx1, tx2, … txn. That is, in Figure 4 the example shown, the camera settings will only change at time tx3. Further changes at time points t2 and t3 due to changes in the lighting conditions will also be suppressed. The next change in the camera settings will occur at time point tx4. Therefore, between time point tx3 and time point tx4, there may be underexposed images (due to the shadows caused by the trees in this example). In other cases, there may also be overexposed images until the camera settings are changed again at the corresponding next defined time point tx1, tx2, …, txn.

[0024] The physiological parameter determination unit 32 of the image evaluation unit 200 knows the defined time points tx1, tx2, ... txn at which a camera setting change can be performed. For example, the evaluation algorithm for determining physiological parameters based on the captured images can be temporarily paused at the defined time points tx1, tx2, ..., txn, or the sensitivity of the evaluation algorithm can be reduced at the defined time points tx1, tx2, ..., txn. In this way, any impact caused by the change of the camera settings is significantly reduced. Between two subsequent defined time points tx1, tx2, ... txn, there is an image parameter stable interval (the image parameters / camera settings are stable), which significantly improves the accuracy of determining physiological parameters based on the captured images. The interval between two directly consecutive defined time points tx1, tx2, ..., txn can be, for example, at least 10 seconds, at least 15 seconds or at least 30 seconds. Generally speaking, increasing the time between two directly consecutive defined time points tx1, tx2, …, txn simplifies the internal structure of the noise. This allows for more effective filtering of unwanted noise.

[0025] As described above, the evaluation algorithm can rely on the fact that the camera settings are fixed during the interval between two consecutive defined time points tx1, tx2, ..., txn. However, it is generally also possible for the evaluation algorithm to compensate for any camera setting changes made at the defined time points tx1, tx2, ..., txn. For example, the control unit 38 can provide the physiological parameter determination unit 32 with information about any changes it has made to the camera settings. Then, the physiological parameter determination unit 32 can consider this information when determining the physiological parameters. That is to say, in addition to the images captured by the camera 34, the camera system 30 can also provide the physiological parameter determination unit 32 with additional information about the camera settings for further consideration when determining the physiological parameters.

[0026] The intervals between two consecutive defined time points tx1, tx2, …, txn may typically be chosen to be very long. For example, the interval between two directly consecutive defined time points tx1, tx2, …, txn may be several minutes, or even several hours. For example, during the entire driving process of a vehicle, the camera settings may not change at all. The camera settings can be set once at the start of the driving process and can remain unchanged for the rest of the journey. In this case, the camera settings can be regarded as constant. Automatic camera setting adjustment can be substantially disabled. In this way, the noise caused by camera setting adjustment is completely eliminated. However, when the interval between two consecutive defined time points tx1, tx2, … txn is set too long, when the lighting conditions change frequently, this may result in many images being significantly underexposed and / or overexposed. Then, it may no longer be possible to accurately detect the head position of the person of interest, and it may no longer be possible to identify the person of interest through a suitable face recognition algorithm, etc. That is to say, if the time interval between two subsequent defined time points tx1, tx2, ..., txn chosen is long, certain drawbacks will have to be accepted.

[0027] In some cases, the control unit 38 may not be able to provide information about the camera setting changes to the physiological parameter determination unit 32. In such a case, the image evaluation unit 200 may be configured to estimate the changes made to the camera settings by means of a suitable algorithm. The defined time points tx1, tx2, … txn are still known. Thus, the image evaluation unit 200 can estimate the changes made to the camera settings by appropriately evaluating the images captured directly before and directly after the defined time points tx1, tx2, ..., txn. According to one example, the image evaluation unit 200 may include or may be coupled to a near-infrared camera configured to produce a near-infrared flash. Such a flash is invisible to the person of interest. However, the image evaluation unit 200 knows the characteristics of the artificial light generated to produce the flash. A reference object can be arranged close to the person of interest. For example, a reference object of a defined color (e.g., white) can be attached to or integrated into the headrest of the vehicle 10. Then, the head of the driver 40 of the vehicle 10 will be directly adjacent to such a reference object (e.g., a simple white circle on the headrest or a fully white headrest), and both the head of the person of interest and the reference object (or at least one reference object) will be captured in the image. The image evaluation unit 200 knows the characteristics of the reference object (e.g., a specific color). Based on the known characteristics of the near-infrared flash, the known characteristics of the reference object, and the captured image, the image evaluation unit 200 can then determine the camera setting changes by means of a suitable algorithm. In other words, the image evaluation unit 200 estimates the camera settings based on how the reference object looks in the captured image (e.g., bright or dark). In addition to the headrest, the reference object can be arranged on any other suitable element within the vehicle (e.g., on a part of the steering wheel visible in the image, or on a part of the seat backrest visible in the image, etc.).

[0028] As described above, the algorithm for determining the physiological parameters may compensate for any camera setting changes made at the defined time points tx1, tx2, ..., txn. Examples of how such compensation can be implemented will be provided below. The effects of camera setting changes are generally predictable enough, at least in a first approximation. For example, doubling the exposure results in doubling the total energy received by the photosensitive element, which in turn results in doubling its average brightness. The same holds for gain control and variable aperture opening.

[0029] If the exposure value of a certain frame f is E(f), the aperture is D(f), and the gain is G(f), then their changes between frames f1 and f2 are respectively:

[0030]

[0031] The total coefficient of variation of the pixel brightness due to the adjustment (optimization) will then be:

[0032]

[0033] If this arrangement allows the acquisition of exposure, aperture, gain, and other parameters that affect the brightness variation of the pixel under discussion, these parameters can be used to calculate the brightness correction factor. For an image at coordinates (x, y) in frame f from which any pixel is taken, a normalized brightness can be obtained, for example, normalized to the first frame in the video stream:

[0034]

[0035] As can be seen from the above formulas (1), (2), and (3), this normalization can compensate for the pixel brightness variation caused by parameter adjustment and keep the image stable with respect to such variation.

[0036] Possible extensions or generalizations of the compensation are:

[0037] - Instead of normalizing to the first frame, normalize to another frame, which is selected through the internal logic of the physiological parameter determination algorithm, for example, bound to the moment of algorithm calibration;

[0038] - Periodically select a frame for normalization (for example, select a reference frame for normalization every 10 seconds);

[0039] - As an extreme case of the previous extension, inter-frame compensation, when compensating for the difference between two consecutive frames:

[0040]

[0041] - Consider the non-linear effects of the influence of parameter changes on pixel brightness variation. Generally and depending on the implementation, the dependence of the influence of parameter values on brightness will have the following form:

[0042]

[0043] where P(x, y, f) is the brightness of the pixel (x, y) in frame f, e(x, y, f) is the amount of energy that enters the camera during the capture of frame f and is attributed to the pixel (x, y), D(e, D) is the energy reduction factor e at aperture D, E(e, E) is the energy reduction factor e at exposure value E, and G is a similar factor for gain. Since e(x, y, f) can be regarded as the normalized pixel brightness and it does not depend on the camera parameters:

[0044]

[0045] The camera system 30 and the image evaluation unit 200 have been described above as being arranged in the vehicle 10. In the vehicle 10, physiological parameters of the driver 40 or any other passenger can be determined for different applications, for example, in order to determine the mental state of the driver 40 (i.e., whether the driver 40 is fit to drive). In the vehicle 10, the lighting conditions can change rapidly. However, it is generally possible to use the camera system 30 and the image evaluation unit 200 in any other environment outside the vehicle.

[0046] As Figure 5 schematically shown in, a method according to an embodiment of the present disclosure includes: capturing a series of consecutive images of a person of interest in the field of view of the camera 34 by the camera 34 (step 501); determining the light intensity on the person of interest (step 502); and controlling the camera settings of the camera 34 by the control unit 38 based on the determined light intensity (step 503), wherein the camera settings of the camera 34 are changed only at defined time points tx1, tx2, …, txn, and when it is detected that the light intensity on the object of interest changes between two defined time points tx1, tx2, …, txn, the current camera settings are maintained until the next defined time point tx1, tx2, …, txn.

[0047] The method may further include determining a change in pixel intensity in at least a defined region of interest in the captured images and determining one or more physiological parameters of the person of interest based on the change in pixel intensity. Subsequently, the one or more physiological parameters can be processed, and the mental state of the person of interest can be determined based on the physiological parameters. For example, the one or more physiological parameters may include at least one of heart rate and heart rate variability.

[0048] It will be understood that the systems and methods shown are only examples. Although various embodiments of the present invention have been described, it will be apparent to those of ordinary skill in the art that there can be more embodiments and implementations within the scope of the present invention. Specifically, those skilled in the art will recognize the interchangeability of various features from different embodiments. Although these technologies and systems have been disclosed in the context of certain embodiments and examples, it should be understood that these technologies and systems can be extended beyond the specifically disclosed embodiments to other embodiments and / or their uses and obvious modifications. Therefore, the present invention is not limited except in accordance with the appended claims and their equivalents.

[0049] The description of the embodiments has been presented for purposes of illustration and description. Appropriate modifications and variations of the embodiments can be effected in accordance with the above description or can be obtained through practice of the methods. The arrangement is exemplary in nature and can include additional elements and / or omit elements. As used in this application, an element recited in the singular and preceded by the word "a" or "an" should be understood as not excluding a plurality of such elements, unless such exclusion is stated. Further, a reference to "an embodiment" or "an example" of the present disclosure is not intended to be construed as excluding the existence of additional embodiments that also incorporate the recited features. The terms "first," "second," and "third," etc. are used merely as labels and are not intended to impose numerical requirements or a particular positional order on their objects. The subject matter of the present disclosure includes all novel and non-obvious combinations and sub-combinations of the various systems and configurations and other features, functions, and / or properties disclosed. The appended claims particularly point out the subject matter from the foregoing disclosure that is regarded as novel and non-obvious.

Claims

1. A camera system (30), comprising a camera (34) and a control unit (38), wherein the camera (34) is configured to capture a series of consecutive images of a person of interest within the field of view of the camera (34); the camera system (30) is configured to determine the light intensity on the person of interest; and the control unit (38) is configured to control the camera settings of the camera (34) based on the light intensity determined by the camera system (30), wherein the control unit (38) is configured to change the camera settings of the camera (34) only at defined time points (tx1, tx2,..., txn), and to maintain the current camera settings until the next defined time point (tx1, tx2,..., txn) when it is detected that the light intensity on the object of interest changes between two defined time points (tx1, tx2,..., txn).

2. The camera system (30) according to claim 1, wherein the interval between two directly consecutive defined time points (tx1, tx2,..., txn) is at least 10 seconds, at least 15 seconds or at least 30 seconds.

3. The camera system (30) according to claim 1 or 2, wherein the camera settings controlled by the control unit (38) include at least one of an exposure time and an f-number.

4. An image evaluation unit (200), comprising: the camera system (30) according to any one of claims 1 to 3; and a physiological parameter determination unit (32), wherein the physiological parameter determination unit (32) is configured to determine the pixel intensity change in at least a defined region of interest in the image captured by the camera system (30), and to determine one or more physiological parameters of the person of interest based on the pixel intensity change.

5. The image evaluation unit (200) according to claim 4, wherein the one or more physiological parameters determined by the physiological parameter determination unit (32) include at least one of a heart rate and a heart rate variability.

6. The image evaluation unit (200) according to claim 4 or 5, wherein the physiological parameter determination unit (32) is configured to determine the one or more physiological parameters by an evaluation algorithm, wherein the algorithm is temporarily suspended at the defined time points (tx1, tx2,..., txn), or the sensitivity of the algorithm is reduced at the defined time points (tx1, tx2,..., txn).

7. The image evaluation unit (200) according to claim 6, wherein the physiological parameter determination unit (32) receives the image captured by the camera (34) of the camera system (30); the physiological parameter determination unit (32) also receives information from the control unit (38) of the camera system (30) regarding the changes made to the camera settings by the control unit (38) at the defined time points (tx1, tx2,..., txn); and The evaluation algorithm compensates for changes in the camera settings made at the defined time points (tx1, tx2, …, txn).

8. The image evaluation unit (200) according to claim 6, wherein the physiological parameter determination unit (32) receives the images captured by the camera (34) of the camera system (30); the image evaluation unit (200) is configured to estimate the changes made by the control unit (38) to the camera settings at the defined time points (tx1, tx2, …, txn); and the evaluation algorithm compensates for changes in the camera settings made at the defined time points (tx1, tx2, …, txn).

9. The image evaluation unit (200) according to claim 8, wherein the image evaluation unit (200) is configured to estimate the changes made to the camera settings by evaluating the images captured immediately before and immediately after the defined time points (tx1, tx2, …, txn).

10. The image evaluation unit (200) according to any one of claims 4 to 9, wherein the image evaluation unit (200) is arranged in a vehicle (10), and the person of interest is the driver (40) of the vehicle (10).

11. A method, comprising: capturing, by a camera (34), a series of consecutive images of a person of interest within the field of view of the camera (34); determining the light intensity on the person of interest; and controlling, by a control unit (38), the camera settings of the camera (34) based on the determined light intensity, wherein the camera settings of the camera (34) are changed only at defined time points (tx1, tx2, …, txn), and when it is detected that the light intensity on the object of interest changes between two defined time points (tx1, tx2, …, txn), the current camera settings are maintained until the next defined time point (tx1, tx2, …, txn).

12. The method according to claim 11, further comprising: determining a change in pixel intensity in at least a defined region of interest in the captured images; and determining one or more physiological parameters of the person of interest based on the change in pixel intensity.

13. The method according to claim 12, further comprising: processing the one or more physiological parameters; and determining the mental state of the person of interest based on the physiological parameters.

14. The method according to any one of claims 11 to 13, wherein the one or more physiological parameters include at least one of heart rate and heart rate variability.

Images