FIELD OF THE INVENTION

Embodiments of the present invention generally relates to monitoring eye fatigue, and more particularly to a method and system for monitoring eye fatigue based on

pathophysiological response.

BACKGROUND OF THE INVENTION

Eye fatigue has nowadays attracted tremendous public attention as results of the long term eye use and the expanding implementation of visual display terminals. Although people experiencing eye fatigue can usually recover by taking rest in a timely manner, unawareness of eye fatigue and/or unwillingness to take a break till the fatigue upgrades to a severe level could lead to persistent discomfort or even functional problems such as myopia. Therefore, effective means for monitoring the stage of eye fatigue is of great interest for eye care in both clinic and home settings.

However, very limited technologies are available for monitoring the eye fatigue conditions and the assessment accuracy is still questionable. Eye fatigue is a complex syndrome modulated by multiple factors. Current methods for eye fatigue evaluation are limited at evaluating the physical response of eyes, such as critical flicker fusion threshold assessment and eye movement tracking. However, these responses are usually the consequence of physiological changes and could be subject to various interferences, such as conscious subjective reluctance to blink. Evaluations of eye fatigue based on these data are still questionable.

SUMMARY OF THE INVENTION

Therefore, it would be desirable in the art to provide solutions for monitoring eye fatigue which can accurately and easily detect the fatigue of eyes under various lighting conditions in real-time. It would also be desirable to provide flexible systems for monitoring eye fatigue which may be integrated with many appliances, such as to provide a compact wearable setup which can be used by a common user. Several pathophysiological responses associated with eye fatigue have been well observed and studied. Among them, the inventors find that periorbital edema is an easily recognized feature found in people during the process of eye fatigue development, and try to create a new eye fatigue assessment system based on it.

By real-time monitoring such pathophysiological indicator of eye fatigue, the inventors have proposed in this invention, a method and system for monitoring eye fatigue based on the alteration of periorbital edema during an eye fatigue process. An eye fatigue index (EFI) model is also provided for eye fatigue level evaluation.

To better address one or more of the above concerns, in a first aspect of the invention, a method for monitoring eye fatigue is provided. The method comprises: illuminating periorbital skin with Infrared light having a first intensity; measuring a second intensity of Infrared light reflected from the periorbital skin; and determining a level of eye fatigue based on at least the second intensity of Infrared light. Further, the determining may be further based on the first intensity of Infrared light.

In some embodiments, the measuring comprises: acquiring one or more images of the periorbital skin by using active IR imaging; extracting a region of interest (ROI) from the periorbital skin within the one or more images; and calculating the second intensity of IR light reflected from the region of interest.

In further embodiment, the second intensity I ₂ of IR light reflected from the region of interest is indicated by an intensity I _monitoring of IR light of the region of interest is calculated as I _monitoring , wherein N _R0I is the number of pixels of the region of interest, /

is an intensity of each pixel.

In some other embodiments, the illuminating comprises: illuminating a point of interest (POI) on the periorbital skin with the IR light having the first intensity; and the measuring the second intensity I ₂ comprises sensing an intensity I _monitoring of IR light reflected from the point of interest.

In further embodiments, the determining comprises calculating a relative intensity

I _r of IR light as I _r = ^momtonng _? wherein I _reference is a corresponding intensity measured with a

^reference

same measuring approach as I _monitoring but under a healthy eye condition. In some embodiments, the region of interest is one of the following: part or all of the upper eyelid, part or all of the lower eyelid, and any combinations thereof.

In some embodiments, the level of eye fatigue is determined according to a predefined Eye Fatigue Index (EFI) reference model, the EFI reference model being created based on magnitude and/or gradient of the second intensity of IR light, or based on magnitude and/or gradient of the relative intensity of the reflected IR light, or based on magnitude and/or gradient of difference of the first intensity and the second intensity of IR light.

In further embodiments, the method may further comprise at least one of:

displaying the level of eye fatigue; and alerting if the level of eye fatigue is beyond a predetermined alarm threshold, wherein the alarm threshold is determined from the EFI reference model.

In a second aspect of the invention, a system for monitoring eye fatigue is provided. The system may comprise an Infrared light source unit configured to illuminate periorbital skin with IR light having a first intensity; an IR optical sensor unit configured to measure a second intensity of Infrared (IR) light reflected from the periorbital skin; and a processor configured to determine a level of eye fatigue based on at least the second intensity of IR light.

Particular embodiments of the subject matter described in this specification can be implemented so as to realize one or more of the following advantages.

With particular embodiments of the techniques described in this specification, because one direct pathophysiological indicator of eye fatigue, i.e., the alteration of periorbital edema, is selected to indicate the eye fatigue, it is easy to detect the eye fatigue accurately and rapidly. Further, in one embodiment, it is only necessary to acquire an image of the surrounding skin of an eye, either fully open, part open, or closed, and thus the monitoring system may be made compact and can be integrated with appliances, such as a pair of glasses, or a cap, to provide a wearable and portable setup and to real-time monitor eye fatigue. In another embodiment, even no image is required, and the imaging system is replaced by a non-contact fiber-optic spectroscopic system. Thus, the monitoring system may be made even compact. The proposed system may also be integrated with other appliances, such as mobile phone, desk lamp, computer, TV or mirror, to expand the accessibility of the presented technology. Further, by using an IR light source and an IR sensor, the detection is made invisible and unaffected by the ambient lighting condition. Moreover, a quantified index of pathophysiological response in the eye fatigue process is provided for a more accurate evaluation of the severity of eye fatigue. Other features and advantages of the embodiments of the present invention will also be understood from the following description of specific embodiments when read in conjunction with the accompanying drawings, which illustrate, by way of example, the principles of embodiments of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other aspects, features, and benefits of various embodiments of the invention will become more fully apparent, by way of example, from the following detailed description and the accompanying drawings, in which:

Fig. 1 schematically illustrates an exemplary monitoring system 100 for monitoring eye fatigue according to a first embodiment of the present invention;

Fig. 2 schematically illustrates the image processing according to the first embodiment of the present invention;

Fig. 3 schematically illustrates an eye fatigue curve 300 generated according to embodiments of the present invention;

Fig. 4 illustrates an exemplary monitoring system 400 for monitoring eye fatigue according to a second embodiment of the present invention; and

Fig. 5 is a flow chart schematically illustrating an exemplary method 500 for monitoring eye fatigue according to a first embodiment of the present invention.

Like reference numbers and designations in the various drawings indicate like elements.

DETAILED DESCRIPTION OF EMBODIMENTS

Hereinafter, the principle and spirit of the present invention will be described with reference to the illustrative embodiments. It should be understood, all these embodiments are given merely for the skilled in the art to better understand and further practice the present invention, but not for limiting the scope of the present invention. For example, features illustrated or described as part of one embodiment may be used with another embodiment to yield still a further embodiment.

When people experience eye fatigue, one physiological character associated with eye fatigue that can be observed is periorbital edema. Specifically, along with the increasing loads of eye use, the local circulation associated with eyes slows down and hence induces fluid (mainly water) retention around eyes, termed as "periorbital edema," in connective tissue under skin. Since the areas surrounding eyes have the thinnest skin of human body, periorbital edema is easy to be discerned especially in severe condition that swelling around eyes happens.

Periorbital edema should be distinguished from another syndrome, eye bag, which also usually causes swelling appearance under eyes especially for aged people. The difference between these two syndromes is that periorbital edema (eye puffiness) is usually a temporary problem caused primarily by retention of fluid, while eye bags are often caused by fat accumulation under the eye, and will develop slowly over years.

Thus, embodiments of the proposed invention look directly into the alteration of the periorbital edema and try to create a new eye fatigue assessment system based on it, so as to provide a more reliable and convenient means for eye fatigue monitoring and evaluation.

Periorbital edema developed during an eye fatigue process is primarily led by the water retention under skin. Therefore, by studying the alteration of light absorption due to the change of water concentration, we can create an indicator for the severity of eye fatigue.

Infrared (IR) sensing technique will be employed to measure an intensity of IR light reflected from periorbital skin, which indirectly indicates the intensity of IR light absorbed by the water under the periorbital skin. The skilled in the art can understand that, the more severe the eye fatigue is, the higher the tissue water concentration in the periorbital skin is, and in turn the more the light is absorbed by the water, and the less the light reflected from the periorbital skin is measured. In other words, the measured intensity of the reflected IR light is an indicator of the tissue water concentration in the periorbital skin, and thus the alteration of the light

absorption/reflection indicates the severity of eye fatigue.

Compared to other tissue components, water has very strong absorption in the IR wavelength range, especially in the Near-Infrared (NIR) wavelength range. Meanwhile, the thin skin tissue layer in periorbital areas minimizes the barrier for light penetration. Both of these help to build a solid base for accurate measurement of periorbital water concentration alteration during the development of eye fatigue.

The wavelength for excitation light source is selected in IR/NIR range since the tissue absorption in this range is dominant by water. One preferable choice could be 980 nm, because this is a local absorption peak for water as well as a local absorption trough for fat. By selecting this particular excitation wavelength, we can maximally screen out the absorption from the fat accumulated in eye bags and hence focus on that due to temporary water retention in periorbital edema developed during an eye fatigue development process. In order to measure the intensity of IR light reflected from the periorbital skin, a measuring system comprising an IR light source unit and an IR optical sensor unit is employed. The IR light source unit is configured to illuminate periorbital skin with IR light having a first intensity. The IR optical sensor unit is configured to measure a second intensity of IR light reflected from the periorbital skin.

Fig. 1 illustrates an exemplary monitoring system 100 for monitoring eye fatigue according to a first embodiment of the present invention.

For realizing the functions elaborated above, in this first embodiment of the present invention, the measuring system is implemented by an active IR imaging system to monitor the relative water concentration alterations under skin in periorbital areas, wherein the IR optical sensor unit may comprise an IR camera unit configured to acquire one or more images of the periorbital skin. Meanwhile, imaging processing and eye fatigue assessment algorithms are developed to identify the representative area for water concentration monitoring and define an index representing the level of eye fatigue.

As shown in Fig. 1, the monitoring system 100 may comprise a measuring system and a processor (not shown). The measuring system may comprise an IR light source unit and an IR optical sensor unit. The IR light source unit is configured to illuminate periorbital skin with IR light having a first intensity, and the IR optical sensor unit is configured to measure a second intensity of IR light reflected from the periorbital skin. As mentioned above, the second intensity of the reflected IR light may indicate a tissue water concentration in the periorbital skin, in other words, the alteration of the second intensity of the reflected IR light is an indicator of the change of the tissue water concentration in the periorbital skin. The processor may be configured to determine a level of eye fatigue based on at least the second intensity of the reflected IR light. Optionally, the level of eye fatigue may be determined further based on the first intensity of IR light, for example, the difference of the first intensity and the second intensity. The determination will be discussed later with reference to Fig. 3.

As illustrated in Fig. 1, in this embodiment, the measuring system may be implemented as an active IR imaging system, which comprises an IR light source unit 111 and an IR camera unit 112.

The IR light source unit 111 is configured to illuminate the periorbital skin of the eye 130 to be imaged with a first intensity I _x . The IR light source unit 111 may comprise an IR laser diode 113, a beam expander 115, and other components (such as a bracket for supporting the IR laser diode 113 and the beam expander 115) as required by the specific implementation. Preferably, the IR laser diode 113 may emit IR light of 980 nm. The beam expander 115 is arranged to expand the light emitted from the IR laser diode 113, so as to cover the periorbital skin to be imaged, for example for illuminating the periorbital area including upper and lower eyelids. The skilled in the art should appreciated that, the IR light source unit 111 may include any one or more of a variety of IR radiation sources, including but not limited to,

semiconductor-based sources (e.g., LED-based sources), incandescent sources (e.g., filament lamps), gaseous discharge sources (e.g., xenon high pressure lamp), and lasers.

The IR camera unit 112 is configured to acquire one or more images (an image 114 is shown in Fig. 1 by way of example) of the periorbital skin of the eye 130. The IR camera unit 112 may be a low-cost charge-coupled device (CCD) micro-camera sensitive to IR.

As shown in Fig. 1, preferably, the IR light source unit 111 is placed at an oblique angle (for example, 30 degrees) to the IR camera unit 112 which may sit on the geometric axis of the eye 130. This setting helps to avoid the specular reflection and insures that the IR camera unit 112 may capture most of the diffused/reflected light from deeper tissue underneath the skin surface. Thus, additional holder(s) may be used to implement such arrangement. The specific structure of the measuring system depends on the appliance with which the monitoring system 100 is integrated.

The skilled in the art should appreciate that, when the monitoring system 100 is integrated with various apparatuses, the imaging system may further comprise one or more of an IR reflecting unit and an optical fiber (not shown in Fig. 1) depending on the specific structures of the apparatuses. The IR reflecting unit and the optical fiber are configured to direct the IR light beam from the IR light source unit 111 to illuminate the eye 130 to be imaged, and/or direct the IR light beam reflected from the eye 130 to the IR camera unit 112 to form an image. The skilled in the art would appreciate that the imaging system may comprise more or less components as set forth in the above description.

The use of an active IR imaging system has many advantages. First and the most important, water has very strong absorption in the IR wavelength range. Second, it can acquire images under various lighting conditions, regardless at night or during daytime. Third, the IR illumination is not visible by the user and does not interfere with the normal life of the user, for example, when the user is reading, driving, and etc. Fourth, it provides images of high imaging definition and high contrast, which can advantage the subsequent imaging processing. Upon one or more images of the periorbital skin of the eye 130 has been acquired, the processor will perform further analysis on the image using pre-written digital image processing programs.

Fig. 2 schematically illustrates the image processing according to the first embodiment of the present invention.

If more than one image is acquired, the processor is configured to screen the images based on the extent or degree of the eye closed (in other words, the extent of eye opening). In one embodiment, images of fully closed eye are selected for further processing. In other embodiments, different extents of the eye closed may be identified, and the image with the same extent as that of a reference image (which will be discussed later) will be selected.

Then, as shown in Fig. 2, the processor is configured to extract a region of interest (ROI) 201 from the periorbital skin within the selected image. The region of interest (ROI) may be one of the following: part or the entire of the upper eyelid, part or the entire of the lower eyelid, and any combinations thereof.

For example, as illustrated in Fig. 2, the region of interest ROI 201 may be the whole eyelid, as indicated by a white curve 202 which profiles the whole upper eyelid and the whole lower eyelid. In such case, the maximal periorbital area can be achieved. The extraction may be implemented by using pattern recognition techniques in image processing. The skilled in the art would appreciate that the extracting may be implemented by using other techniques in computer vision and image processing. The present invention has no limitation in this aspect.

Then, the processor is further configured to calculate the second intensity I ₂ of IR light reflected from the ROI, which may be indicated by an intensity I _monitoring of IR light of the ROI. In one embodiment, I _monitorin oi ROI may be calculated as:

wherein N _R0I is the number of pixels of the ROI, / is an intensity of each pixel.

Normally, the intensity of the pixel on the image varies as the real incoming IR light intensity on the camera, in this example, the second intensity I ₂ of the reflected IR light. The relationship between the pixel intensity and the real incoming intensity may vary slightly depending on the IR camera by which the image is acquired, the light source, the reflector, and etc. Thus, the relationship may be predetermined and used to convert the pixel intensity into the real incoming intensity as required. Generally, the relationship may be linear.

As previously explained, the higher the tissue water concentration in the periorbital skin is, the more the light is absorbed by the water, and in turn the less the light reflected from the periorbital skin is measured. In other words, the measured intensity I _monitoring of IR light is an indicator of the tissue water concentration in the periorbital skin. Generally, the measured intensity I _monitoring will decrease as the eye fatigue becomes more severe.

In addition, as discussed above, the region of interest (ROI) may be one of the following: part or the entire of the upper eyelid, part or the entire of the lower eyelid, and any combinations thereof. If only part or the entire of the upper eyelid is specified as ROI, such option can minimize the absorption interference from fat usually accumulated under the eye. If only part or the entire of the lower eyelid is specified as ROI, such option can offer a better image quality due to the stable status of lower eyelid area.

The skilled in the art can understand that, when different ROIs are selected for measuring the intensity of IR light reflected from the periorbital skin, different I _monitoring will be obtained even under a same level of eye fatigue. The reason is that the extent of eye opening (or eye closed) will influence the water concentration under the periorbital skin.

Thus, to mitigate such influence, a relative intensity of IR light l _r is employed. The relative intensity I _r may be calculated as: I _r = ^Imonitoring , (Eq. 2)

^reference

wherein I _reference is a corresponding intensity measured with a same measuring approach as Imonitoring ^ ⁰ ^ under a healthy eye condition, i.e., without eye fatigue. The same measuring approach means that a same IR light source with a same first intensity and a same IR camera are used, an image of the same extent of eye opening is selected (the image may be referred as a reference image), the same ROI is specified, and the same calculation algorithm is employed. In other words, the measured intensity I _monitoring with various measuring approaches will be normalized as I _r .

I reference is obtained beforehand during a health eye condition without fatigue. In some embodiments, several I _reference corresponding to different extents of eye opening and different ROIs and even different first intensities I _x may be obtained and stored for later use. Further, by normalization of I _monitoring , the conversion of I _monitoring into 7 ₂ may be omitted, because as previously discussed, the relationship between the pixel intensity and the real incoming intensity may be linear. In the following description, unless otherwise specified, Monitoring ^an d I ₂ ^ar e interchangeably used, because 7 ₂ may be derived from I _monitoring on demand.

In some embodiments, for the purpose of intuitive understanding, an intensity I ₃ of IR light absorbed by the water under the periorbital skin may be used to determine the level of eye fatigue. Obviously, the known or fixed emitted light intensity, i.e., the first intensity Ι _γ , is equal to sum of the absorbed light intensity I and the reflected light intensity I ₂ .

Having calculated I _monitoring , I _r or I ₃ , the level of eye fatigue may be determined based on I _monitoring , I _r or I ₃ . For example, an Eye Fatigue Index (EFI) reference model may be created based on the calculated I _monitoring , I _r or I ₃ to evaluate the level of eye fatigue.

In one embodiment, the EFI reference model may be created based on a plurality of I _monitoring , I _r or I ₃ as monitored during a progress of eye fatigue of a user. The progress of eye fatigue means a progress from a normal state to a highest-grade fatigued state of the user. For example, the user is in a normal state and starts reading, and the monitoring system 100 is initiated to monitor the eye fatigue of the user. The reading lasts a long time until the user feels very fatigued. During that period, the monitoring system 100 continues monitoring the eye fatigue and records the calculated I _monitoring , I _r or I ₃ periodically. For example, the monitoring system 100 may take a sample of the eye image of the user every 5 minutes and calculate the

I monitoring ' ^or · Of course, the time interval may be set as other values. Finally, a plurality of

I monitoring ' ^{or ma} Y b ^e obtained during the entire progress of eye fatigue of the user. Those values may be used to generate an eye fatigue curve.

Fig. 3 schematically illustrates an eye fatigue curve 300 generated according to an embodiment of the present invention. The horizontal axis 310 represents the time passed from the beginning of the monitoring, in other words, the time spent on working (e.g., reading, watching TV, etc.) with the eyes. For purpose of intuitive understanding, the calculated reflected light intensity I _monitoring or I ₂ may be converted into an intensity I ₃ of IR light absorbed by the water under the periorbital skin, wherein the known emitted light intensity, i.e., the first intensity Ι _γ , is equal to sum of the absorbed light intensity I and the reflected light intensity I ₂ . Thus, the vertical axis 320 may represent the converted absorbed light intensity I ₃ , which can indicate the level or degree of the eye fatigue directly. The skilled in the art could appreciate that, the vertical axis 320 may represent other parameters associated with the eye fatigue, such ^a s I _m o„itorin _g ' ' ^0Γ 1 " ' ^{s0 mat} respective eye fatigue curve can be generated. In an

embodiment where l _r or \-l _r is used to generate the eye fatigue curve, the conversion from ¹ monitors ^int o I ₂ may be avoided.

As illustrated in Fig. 3, the eye fatigue curve 300 shows that the magnitude of absorbed light intensity I ₃ increases as the time spent working with the eyes becomes longer. The magnitude of I ₃ finally goes to a constant (a maximum value of I ₃ , theoretically, the upper limit of is I _x ), which indicates the highest-grade fatigued state of the user. Thus, the EFI reference model may be created based on the magnitude of I . For example, a fatigue threshold of magnitude may be determined as a percentage of the maximum magnitude of I , such as 50%. Specifically, when the value of I ₃ is under the fatigue threshold, it indicates a normal state, and when the value of I ₃ is over the fatigue threshold, it indicates a fatigued state. The skilled in the art can easily envisage other ways to define the threshold.

Alternatively, a plurality of fatigue thresholds may be set to hierarchically indicate the levels of eye fatigue. That is, the EFI reference model may be divided into several levels, each level indicating different degree of eye fatigue. For example, as shown in Fig. 3, two fatigue thresholds (304 and 306) are selected to indicate the levels of eye fatigue. In particular, the fatigue threshold 304 may be selected as 50% of the maximum magnitude of I , and the fatigue threshold 306 may be selected as 75% of the maximum magnitude of I , respectively. Thus, when the value of I ₃ is under the fatigue threshold 304, it indicates a normal state 301, when the value of I ₃ is between the fatigue threshold 304 and the fatigue threshold 306, it indicates a low-grade fatigued state 302, and when the value of I ₃ is over the fatigue threshold 306, it indicates a high-grade fatigued state 303.

The above embodiment needs to monitor the entire progress of eye fatigue, which may take a long time, such as 3 or 4 hours. Additionally, the maximum magnitude of I may vary from user to user, and thus the EFI reference model based on magnitude of I may need calibration (will be discussed below). The inventors also find that, I changes rapidly when the eyes begin to become fatigued, and changes slowly when the eyes have been heavily fatigued. The above changes may be obvious from the eye fatigue curve 300 as shown in Fig. 3.

Thus, in another embodiment, the EFI reference model may be created based on the rate of changes in I ₃ (i.e., the gradient of the eye fatigue curve 300) as monitored during a time period. Similarly, during that period, the monitoring system 100 continues monitoring the eye fatigue and records the calculated I _monitoring , I _r , or I ₃ periodically. However, the time period may be less than the period of an entire progress of eye fatigue, because when the gradient of the eye fatigue curve 300 varies slowly but the magnitude of the I _monitoring , I _r , or I ₃ has not yet reached its limit, the monitoring may be stopped and an EFI reference model may be created therefrom. Further, such change trend in I _monitoring , I _r , or I ₃ is almost same for most people, and thus the EFI reference model based on gradient of I _monitoring , I _r , or I ₃ is not necessarily calibrated.

Continuing with Fig. 3, the eye fatigue curve 300 shows that the gradient of I ₃ decreases as the time spent with the eyes becomes longer. The gradient of the curve 300 is finally approaching to zero, which indicates the highest-grade fatigued state of the user. Thus, the EFI reference model may be created based on the gradient of I . Normally, the monitoring would continue for a period of time, and thus the gradient of I may be calculated from the samples taken in that period. The skilled in the art would understand that, the calculating of the gradient may be implemented by any known technique, and thus the detailed description thereof is omitted herein.

Also, the EFI reference model based on the gradient of I may be divided into several levels, each level indicating different degree of eye fatigue. The detailed division may reference the above discussion with respect to the EFI reference model based on the magnitude of I ₃ , and thus the description thereof is omitted here.

It should be noted that, the above embodiments may be implemented separately or in combination, such that the monitoring system 100 can be adapted to various scenarios. For example, when the monitoring system 100 is started up for the first time or at some point in time, the EFI reference model based on the magnitude of I may be used to evaluate the level of eye fatigue. When the monitoring system 100 continuously monitors the eye fatigue, the EFI reference model based on the gradient of I ₃ may be used.

As previously noted, in some embodiments, the EFI reference model may be calibrated as the use of the monitoring system 100. The EFI reference model may be initialized according to the statistics of pathophysiological response (in this context, the periorbital edema) to eye fatigue of human beings. However, the pathophysiological response associated with eye fatigue may vary from user to user, and sometimes it may vary from time to time. Thus, calibration may be used to adapt a specific user during a specific period.

For example, when a specific user has continuously used the monitoring system 100 for a time period, the plurality of I _monitoring , I _r , or I ₃ as monitored during this period may be used to calibrate the EFI reference model. Various techniques, such as curve fitting,

interpolation, etc., may be utilized to implement the calibration. Thus, the monitoring system 100 can determine the level of eye fatigue accurately for its user.

In some embodiments, the monitoring system 100 may further comprise at least one of a display unit and an alerting unit (not shown).

The display unit may be configured to display the level of eye fatigue in real time. In one embodiment, the display unit may be a mini screen, which can display the calculated I _monitoring , I _r , or I ₃ in real time in a form of eye fatigue curve. Alternatively, the display unit can only display the level of the eye fatigue, such as Level 1 fatigue, Level 2 fatigue, etc. In another embodiment, the display unit may be one or more LED lights, which can indicate the level of eye fatigue by the number or color of the LED light(s). For example, red light can indicate a high-grade eye fatigued state, orange light can indicate a low-grade eye fatigued state, and green light can indicate a normal state. The skilled in the art would easily conceive other display manners to indicate the eye fatigue.

The alerting unit may be configured to alert if the monitored level of eye fatigue is beyond a predetermined alarm threshold. The alert can be signals in any form, such as audible, visible, or sensible signals. The predetermined alarm threshold may be determined from the EFI reference model. For example, the alarm threshold may be selected as one of the fatigue threshold(s) in the EFI reference model.

Alternatively, multiple alarm thresholds may be selected to indicate different levels of eye fatigue, and different alarm signals may be associated with different alarm thresholds. Although the display unit and the alerting unit are described in separate components, they can be embodied as a single component.

Fig. 4 illustrates another exemplary monitoring system 400 for monitoring eye fatigue according to a second embodiment of the present invention.

In this second embodiment of the present invention, different from the first embodiment, the measuring system of the monitoring system 400 is implemented by a non- contact fiber-optic spectroscopic system to monitor the water concentration alterations under skin in periorbital areas. Instead of looking at the image from a region of periorbital area, the fiber-optic system investigates the tissue water concentration in local tissue block. Thus, no imaging processing is needed. The EFI reference model may be created similarly as that in the first embodiment.

Likewise, the monitoring system 400 may comprise a measuring system and a processor (not shown). The measuring system may comprise an IR light source unit and an IR optical sensor unit. The IR light source unit is configured to illuminate periorbital skin with IR light having a first intensity, and the IR optical sensor unit is configured to measure a second intensity of IR light reflected from the periorbital skin. The processor may be configured to determine a level of eye fatigue based on at least the second intensity of IR light.

As illustrated in Fig. 4, the measuring system may comprise an IR laser diode light source 410 with source fiber coupling lens, a source fiber 412, an IR optical sensor 420, a detector fiber 422, and a fiber holder 414 for holding the two fibers 412 and 422.

At the near tissue end of the fibers, a micro auto-zoom lens unit 416 coupled with an optical distance detector 418 will be implemented in order to deliver focused excitation light on a point of interest (POI) 430 on the tissue surface (via the source fiber 412) as well as to collect the diffuse reflected light from the POI 430 (via the detector fiber 422). The fiber holder 414 may be further arranged to hold the micro lens unit 416 and the distance detector 418.

As shown in Fig. 4, light emitted from the diode light source 410 will be delivered on the source fiber 412 onto the point of interest POI 430, and the other fiber (i.e., the detector fiber 422) will detect the diffuse-reflected light at certain distance away from the POI 430. The volume of the interrogated tissue block is decided by the distance 424 between the source fiber 412 and the detector fiber 422. Thus, the IR optical sensor 420 may be configured to measure the second intensity I ₂ of IR light reflected from the POI 430, which may be indicated by an intensity

^monitoring of IR light detected on the IR optical sensor 420.

Similar to the first embodiment, a relative intensity of IR light I _r may be calculated as:

^monitoring 2)

^reference

wherein I _reference is a corresponding intensity measured with a same measuring approach as I _monitoring but under a healthy eye condition, i.e., without eye fatigue. In this embodiment, the same measuring approach means that a same IR light source with a same first intensity is used, and the same POI is detected. In other words, the measured intensity I _monitoring with various measuring approaches will be normalized as I _r .

Again, the level of eye fatigue may be determined based on I _monitoring or l _r .

Specifically, an EFI reference model may be created based on the calculated I _monitoring or l _r to evaluate the level of eye fatigue. The creation of EFI reference model is similar to that in the first embodiment, and thus the description thereof is omitted here.

The monitoring system 100/400 thus has been described in detail with reference to Figs. 1-4. Such monitoring system 100/400 may be made wearable by selecting compact units and building them upon a pair of eye glasses. Although installation on the glasses is a preferred setup for this system which could provide a wearable and real-time monitoring solution for eye fatigue, integration with other apparatus such as a mobile phone, a desk lamp, an ebook, a computer, a TV, a cap or a mirror could also expand the accessibility of such technology.

Moreover, the monitoring system 100/400 may further configured to evaluate the level of eye fatigue based on image(s) of both eyes. The concrete structure of the monitoring system for both eyes may be different due to the apparatus with which the monitoring system integrates.

Further, with the monitoring system 100, it is only necessary to acquire an image of the surrounding skin of an eye, either fully open, part open, or closed, and thus the monitoring system 100 may be made compact and can be integrated with appliances, such as a pair of glasses, or a cap, to provide a wearable and portable setup and to real-time monitor eye fatigue. The monitoring system 100 with imaging setup provides an area based EFI calculation relying on the changes of water concentration in regional tissue. As the EFI is averaged over an area, the result will be more stable with minimal artifacts incurred if there is slight movement during the monitoring period.

With the monitoring system 400, even no image is required, and the imaging system is replaced by a non-contact fiber-optic spectroscopic system. Thus, the monitoring system 400 may be made even compact. The monitoring system 400 with the fiber-optic probe setup, provides high measurement sensitivity via the point detection manner, is simple and cheap.

Fig. 5 is a flow chart schematically illustrating an exemplary method 500 for monitoring eye fatigue according to a first embodiment of the present invention.

As illustrated in Fig. 5, the method 500 may start at the step S501 at which IR light with a first intensity is used to illuminate the periorbital skin to be imaged. Then the method 500 proceeds to the step S502 at which one or more images of the periorbital skin of an eye are acquired.

As previously described, the images of the periorbital eye may be acquired by using an active IR imaging. For example, an IR light source unit 111 (in Fig. 1) illuminates the periorbital skin to be imaged, and an IR camera unit 112 captures the light beam reflected from the periorbital skin to form an image.

Then, the acquired one or more images may be transferred to the processor, and the one or more images are processed to obtain a second intensity of IR light reflected from the periorbital skin, wherein the second intensity of IR light indicates a tissue water concentration in the periorbital skin.

At the step S503, if more than one image is acquired, the processor is configured to screen the images based on the extent or degree of the eye opening (or eye closed). In one embodiment, images of fully closed eye are selected for further processing. In other

embodiments, different extents of the eye closed may be identified, and the image with the same extent as that of a reference image will be selected.

Then, at the step S504, the processor is configured to extract a region of interest (ROI) from the periorbital skin within the selected image. The region of interest (ROI) may be one of the following: part or the entire of the upper eyelid, part or the entire of the lower eyelid, and any combinations thereof.

The method 500 in turn goes to the step S505, at which, the second intensity of IR light reflected from the ROI may be calculated. In one embodiment, the second intensity I ₂ of IR light reflected from the ROI may be indicated by an intensity I _monitoring of IR light of the ROI, which may be calculated as:

wherein N _R0I is the number of pixels of the ROI, / is an intensity of each pixel.

As previously explained, a relative intensity of the reflected IR light I _r may be further calculated as:

2) wherein I _reference is a corresponding intensity measured with a same measuring approach as I _monitoring but under a healthy eye condition, i.e., without eye fatigue. The same measuring approach means that a same IR light source with a same first intensity and a same IR camera are used, an image of the same extent of eye opening is selected, the same ROI is specified, and the same calculation algorithm is employed. In other words, the measured intensity I _monitoring with various measuring approaches will be normalized as I _r .

Also, the absorbed light intensity I ₃ may be calculated based on the following equation:

/ ₃ = / ₁ - / ₂ , (Eq. 3)

wherein I _x is the known emitted light intensity, i.e., the first intensity; I ₂ may be derived from described above.

Having calculated I _monitoring , I _r , or I ₃ , then at the step S506, the level of eye fatigue may be determined based on I _monitoring , I _r , or I .

As described above, an Eye Fatigue Index (EFI) reference model may be created based on the calculated I _monitoring , I _r , or I ₃ to evaluate the level of eye fatigue. The EFI reference model may be created based on the magnitudes and/or gradient of I _monitoring , I _r , or I ₃ .

Thus, the level of eye fatigue may be determined according to the EFI reference model.

In some embodiments, the method 500 may further comprise a self calibration process. As shown in the step S510, the monitored I _monitoring , I _r , or I ₃ may be recorded along the process of eye fatigue and used to generate an eye fatigue curve, just like the creation of EFI reference model. Then, at the step S512, the EFI reference model may be calibrated or updated according to the newly generated curve, in other words, one or more fatigue thresholds in the EFI reference model may be calibrated. In this way, the monitoring system may be adapted to a specific user.

The skilled in the art would appreciate that, the self calibration process may be omitted when the EFI reference model is created based on the gradient of I _monitoring , I _r , or I ₃ , because normally most people have a same trend in eye fatigue, i.e., first changing rapidly, and then slowly as time goes by. The skilled should also appreciate that, the self calibration process may be activated even when the EFI reference model is created based on the gradient of

^monitoring ' ^ r ' ^3 '

In some embodiments, the method 500 may further comprise the step S507, at which it is determined whether the EFI hits a predetermined alarm threshold. If not, the method 500 may return to the step of S502 of acquiring an image of the periorbital skin of an eye.

Otherwise, if so, then at the step S508, an alerting system may be activated. The alert can be signals in any form, such as audible, visible, or sensible signals. The predetermined alarm threshold may be determined from the EFI reference model. For example, the alarm threshold may be selected as one of the fatigue threshold(s) in the EFI reference model. Alternatively, multiple alarm thresholds may be selected to indicate different levels of eye fatigue, and different alarm signals may be associated with different alarm thresholds.

In other embodiments, the method 500 may further comprise the step of displaying the level of eye fatigue (not shown in Fig. 5). Various display techniques may be used to indicate the level of eye fatigue, including but not limited to, graphical display, data display, light display, etc.

The skilled in the art can easily conceive the processing flow of the second embodiment of the present invention based on the flow chart as illustrated in Fig. 5. The mere difference is in the measuring approach of the intensity of IR light. Thus, the description thereof is omitted here.

The monitoring system and method are mainly designed for eye care applications. The technology as presented herein has offered a new solution for eye fatigue evaluation by quantifying the severity of periorbital edema induced by eye fatigue. It can be applied in the field of self health-monitoring devices for eye care. The data collected not only provides personalized guidance on healthy eye use but also has the potential of introducing a new diagnostic parameter for clinical eye fatigue evaluation.

While this specification contains many specific implementation details, these should not be construed as limitations on the scope of any implementation or of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments of particular implementations. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination.

Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a sub-combination.

It should also be noted that the above described embodiments are given for describing rather than limiting the invention, and it is to be understood that modifications and variations may be resorted to without departing from the spirit and scope of the invention as those skilled in the art readily understand. Such modifications and variations are considered to be within the scope of the invention and the appended claims. The protection scope of the invention is defined by the accompanying claims. In addition, any of the reference numerals in the claims should not be interpreted as a limitation to the claims. Use of the verb "comprise" and its conjugations does not exclude the presence of elements or steps other than those stated in a claim. The indefinite article "a" or "an" preceding an element or step does not exclude the presence of a plurality of such elements or steps.