Intelligent eye protection method and system based on projector

Abstract

The present invention relates to the field of image processing technology, and discloses an intelligent eye protection method and system based on a projector, the method comprising: identifying a dynamic area in a grayscale video frame, using the total grayscale value to identify whether there is a human eye in the dynamic area; identifying the human eye area in the dynamic area, calculating the area length of the human eye area, calculating the number of human eye blinks in the human eye area, and detecting whether the projector protects the eyes; using the area length to generate a posture correction plan for the projector within a rotation angle range and a movement range, calculating the posture correction loss of the posture correction plan, adjusting the posture correction plan using the posture correction loss, and obtaining an adjusted posture plan; judging whether the projector protects the eyes under the corrected posture; generating a lighting correction plan for the projector within the lighting range, calculating the lighting correction loss of the lighting correction plan, adjusting the lighting correction plan using the lighting correction loss, and obtaining an adjusted lighting plan. The present invention can minimize the eye protection cost of the projector.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an intelligent eye protection method and system based on a projector. Background Art

[0002] A projector refers to an enhanced version of a micro projector, which means that a wireless wifi Internet access function has been added to the micro projector and is equipped with an intelligent operating system. It is mainly used to project videos on mobile phones or computers onto other panels. In the process of video projection by the smart projector, the appearance of human faces will cause the projected video to be blocked by the human face, and the projected video will also cause damage to the human eyes. Therefore, when it is detected that the projected video causes damage to the human eyes, the smart projector needs to be corrected for eye protection.

[0003] At present, when it is detected that the video projected on the panel is blocked by a face, the angle and coordinates of the smart projector need to be moved to change the direction of the video projection of the smart projector. However, how to set the angle and coordinates of the smart projector becomes a problem. Because the panel on which the video is projected is limited, if the range of movement of the smart projector is large, the direction of the video projection of the smart projector will exceed the panel range of the projected video. Secondly, even if the video projected on the panel is blocked by a face, the human eye is illuminated by the strong light of the smart projector, but under different positions of the smart projector at different distances, the damage to the human eye caused by the smart projector may be greater or less, and in some cases, the position of the smart projector does not need to be adjusted. Finally, even if the position of the projector is adjusted immediately, the brightness of the light in the projector will still cause damage to the human eye. Therefore, a solution is urgently needed to minimize the eye protection cost of smart projectors. Summary of the invention

[0004] In order to solve the above problems, the present invention provides an intelligent eye protection method and system based on a projector, which can minimize the eye protection cost of the projector.

[0005] In a first aspect, the present invention provides an intelligent eye protection method based on a projector, comprising:

[0006] Using a projector to collect a projection video frame under a projection screen, converting the projection video frame into a grayscale video frame, identifying a dynamic area in the grayscale video frame, counting the total grayscale value in the dynamic area, and using the total grayscale value to identify whether there is a human eye in the dynamic area;

[0007] When the total grayscale value is used to identify the presence of human eyes in the dynamic area, the human eye area in the dynamic area is identified, the human eye gaze point in the human eye area is calculated, the area length of the human eye area is calculated, the number of human eye blinks in the human eye area is calculated, and whether the projector is eye-protective is detected according to the human eye gaze point and the number of human eye blinks;

[0008] When the projector does not protect the eyes, obtain the rotation angle range and the movement range of the projector, generate a posture correction plan of the projector using the region length within the rotation angle range and the movement range, calculate the posture correction loss of the posture correction plan, and adjust the posture correction plan using the posture correction loss to obtain an adjusted posture plan;

[0009] Correcting the projection posture of the projector by using the posture adjustment plan to obtain a corrected posture, and determining whether the projector is eye-protecting in the corrected posture;

[0010] When the projector does not protect the eyes in the correction posture, obtaining the illumination range of the projector, generating an illumination correction plan of the projector within the illumination range, calculating the illumination correction loss of the illumination correction plan, and adjusting the illumination correction plan using the illumination correction loss to obtain an adjusted illumination plan;

[0011] The projection illumination of the projector is corrected by using the adjusted illumination plan to obtain corrected illumination, so as to complete the intelligent eye protection of the projector through the corrected illumination and obtain the intelligent eye protection result of the projector.

[0012] In a possible implementation manner of the first aspect, identifying the dynamic area in the grayscale video frame includes:

[0013] segmenting individual regions in the grayscale video frame using a pixel value difference between a first pixel value and a second pixel value in the grayscale video frame;

[0014] Calculate the mean value of pixel coordinates between each pixel coordinate in the individual area;

[0015] Taking the mean value of the pixel coordinates as the center point of the individual region;

[0016] Identify whether the center point of the region moves in the grayscale video frames at consecutive moments;

[0017] When the coordinates of the center point of the region move in the grayscale video frames at consecutive moments, the separate individual region is regarded as a dynamic region.

[0018] In a possible implementation manner of the first aspect, the using the total grayscale value to identify whether there is a human eye in the dynamic area includes:

[0019] Obtaining a horizontal grayscale total value and a vertical grayscale total value in the grayscale total value;

[0020] Determining whether the shape of the dynamic area is a preset reference shape;

[0021] If the shape of the dynamic area is a preset reference shape, identifying whether there is a human eye in the dynamic area based on the symmetry of the vertical grayscale total value in the dynamic area, the matching of the size sequence of the horizontal grayscale total value and the size sequence of the preset facial features grayscale values;

[0022] If the shape of the dynamic area is not a preset reference shape, obtaining different horizontal grayscale total values ​​and different vertical grayscale total values ​​of grayscale video frames belonging to consecutive moments;

[0023] Extracting a first level grayscale total value and a second level grayscale total value from the different level grayscale total values;

[0024] Identify a first ordinate and a second ordinate of the first horizontal grayscale total value and the second horizontal grayscale total value in the dynamic area;

[0025] Extracting a first vertical grayscale total value and a second vertical grayscale total value from the different vertical grayscale total values;

[0026] Identify a first horizontal coordinate and a second horizontal coordinate of the first vertical grayscale total value and the second vertical grayscale total value in the dynamic area;

[0027] Acquire a first pixel point corresponding to the first ordinate and the first abscissa, and acquire a second pixel point corresponding to the second ordinate and the second abscissa;

[0028] Calculating the grayscale value similarity between the grayscale value of the first pixel and the grayscale value of the second pixel, and calculating the coordinate value similarity between the coordinate value of the first pixel and the coordinate value of the second pixel;

[0029] When the gray value similarity is greater than a preset similarity and the coordinate value similarity is not greater than a preset similarity, determining that a human eye exists in the dynamic area;

[0030] When the gray value similarity is not greater than a preset similarity and the coordinate value similarity is greater than a preset similarity, it is determined that there is no human eye in the dynamic area.

[0031] In a possible implementation manner of the first aspect, calculating the number of blinks of human eyes in the human eye area includes:

[0032] Obtaining a first pixel point and a second pixel point corresponding to the human eye area;

[0033] Calculating a first similarity between the first pixel and other pixel points in the grayscale video frames at consecutive moments, and calculating a second similarity between the second pixel and other pixel points in the grayscale video frames at consecutive moments;

[0034] When both the first similarity and the second similarity are not greater than a preset similarity, recording the closed-eye video frames corresponding to the human eye region in the closed-eye state in the grayscale video frames at consecutive moments;

[0035] The following formula is used to calculate the number of eye blinks in the eye area:

[0036] ;

[0037] in, Indicates the number of times a person blinks. Indicates closed-eye video frame When the moments of The number of Indicates total duration remove After closing the eyes video frame The number of occurrences.

[0038] In a possible implementation manner of the first aspect, detecting whether the projector is eye-protecting according to the gaze point of the human eye and the number of blinks of the human eye includes:

[0039] Determining whether the gaze point of the human eye falls on the projector;

[0040] When the gaze point of the human eye falls on the projector, query whether the number of blinks of the human eye is greater than a preset number of blinks;

[0041] When the number of blinks of the human eye is greater than the preset number of blinks, determining that the projector does not protect the eyes;

[0042] When the gaze point of the human eye does not fall on the projector and the number of blinks of the human eye is not greater than the preset number of blinks, it is determined that the projector is eye-protecting.

[0043] In a possible implementation manner of the first aspect, generating a posture correction plan for the projector by using the region length within the rotation angle range and the movement range includes:

[0044] Obtaining uncorrected coordinates of the projector and generating random coordinates of the projector;

[0045] determining whether the distance between the uncorrected coordinate and the random coordinate is within the moving range;

[0046] When the distance between the uncorrected coordinate and the random coordinate is within the moving range, using the random coordinate as the corrected coordinate of the projector;

[0047] Obtaining the projection direction of the projector under the corrected coordinates;

[0048] Obtaining the left boundary and the right boundary of the projection direction;

[0049] Acquire a first straight line between the corrected coordinate and the left end point of the region length, and acquire a second straight line between the corrected coordinate and the right end point of the region length;

[0050] Extracting a first angle between the right boundary and the first straight line, and a second angle between the left boundary and the second straight line;

[0051] Determining whether the first angle and the second angle are greater than the rotation angle range;

[0052] When the first angle and the second angle are not greater than the rotation angle range, generating a correction angle of the projector from the first angle and the second angle;

[0053] Determine whether there is a human eye corresponding to the human eye area in the projection direction;

[0054] When there is a human eye corresponding to the human eye area in the projection direction, the corrected coordinates and the correction angle are used to generate a posture correction plan for the projector.

[0055] In a possible implementation manner of the first aspect, determining whether the projector protects eyes in the corrected posture includes:

[0056] Acquire a human eye-projector distance between the projector and the human eye in the corrected posture, and acquire a human eye-projector angle of the projector in the corrected posture;

[0057] Collecting historical illumination, historical distance and historical angle of the projector;

[0058] The following formula is used to generate the illumination-distance-angle regression model between the historical illumination, the historical distance, and the historical angle:

[0059] ;

[0060] in, represents the illumination-distance-angle regression model, Represents historical illumination, Represents the historical distance, From a historical perspective, , , represents the regression coefficient;

[0061] Calculate the appropriate illumination for the human eye corresponding to the distance between the human eye and the projector and the angle between the human eye and the projector by using the illumination-distance-angle regression model;

[0062] Acquiring initial illumination of the projector in the corrected posture;

[0063] When the initial illumination meets the illumination suitable for human eyes;

[0064] Determining that the projector is eye-protective in the corrected posture;

[0065] When the initial illumination does not meet the illumination suitable for human eyes;

[0066] It is determined that the projector does not protect the eyes in the corrected posture.

[0067] In a possible implementation manner of the first aspect, calculating the illumination correction loss of the illumination correction plan includes:

[0068] Obtaining a pre-correction area and pre-correction illumination of the projector;

[0069] Obtaining the corrected area and corrected illumination of the illumination correction plan;

[0070] The illumination correction amount of the illumination correction plan is calculated using the following formula according to the area before correction, the illumination before correction, the area after correction, and the illumination after correction:

[0071] ;

[0072] in, represents the lighting correction cost, Indicates The pre-corrected illumination in the pre-corrected area, Indicates The corrected illumination in the corrected region, represents the number of regions before correction;

[0073] According to the corrected area and the corrected illumination, the illumination dispersion of the illumination correction plan is calculated using the following formula:

[0074] ;

[0075] in, represents the illumination dispersion, Indicates The corrected illumination in the corrected region, represents the number of regions before correction, express indivual The average light intensity of

[0076] The illumination correction amount and the illumination dispersion are used as the illumination correction loss of the illumination correction plan.

[0077] In a possible implementation manner of the first aspect, adjusting the illumination correction plan by using the illumination correction loss to obtain an adjusted illumination plan includes:

[0078] Obtaining the illumination correction amount and illumination uniformity in the illumination correction loss;

[0079] When the illumination correction amount is minimum and the illumination dispersion is less than a preset dispersion, the illumination correction plan corresponding to the illumination correction amount and the illumination uniformity is used as the adjusted illumination plan.

[0080] In a second aspect, the present invention provides an intelligent eye protection system based on a projector, the system comprising:

[0081] The human eye detection module is used to collect a projection video frame under the projection screen by using a projector, convert the projection video frame into a grayscale video frame, identify a dynamic area in the grayscale video frame, count the total grayscale value in the dynamic area, and use the total grayscale value to identify whether there is a human eye in the dynamic area;

[0082] an eye protection detection module, for identifying a human eye region in the dynamic region when the total grayscale value is used to identify the presence of a human eye in the dynamic region, calculating a human eye gaze point in the human eye region, calculating a region length of the human eye region, calculating a number of human eye blinks in the human eye region, and detecting whether the projector is eye-protecting according to the human eye gaze point and the number of human eye blinks;

[0083] A posture adjustment module is used to obtain a rotation angle range and a movement range of the projector when the projector does not protect the eyes, generate a posture correction plan for the projector using the region length within the rotation angle range and the movement range, calculate a posture correction loss of the posture correction plan, and adjust the posture correction plan using the posture correction loss to obtain an adjusted posture plan;

[0084] A posture correction module, used to correct the projection posture of the projector by using the posture adjustment plan to obtain a corrected posture, and determine whether the projector is eye-protective in the corrected posture;

[0085] an illumination adjustment module, for obtaining, when the projector does not protect the eyes in the correction posture, an illumination range of the projector, generating an illumination correction plan of the projector within the illumination range, calculating an illumination correction loss of the illumination correction plan, and adjusting the illumination correction plan using the illumination correction loss to obtain an adjusted illumination plan;

[0086] The intelligent eye protection module is used to use the adjusted illumination plan to correct the projection illumination of the projector to obtain corrected illumination, so as to complete the intelligent eye protection of the projector through the corrected illumination and obtain the intelligent eye protection result of the projector.

[0087] Compared with the prior art, the technical principle and beneficial effects of this solution are:

[0088] The embodiment of the present invention converts the projected video frame into a grayscale video frame so as to distinguish the image area where each independent object or independent organism in the projected video frame is located by using the difference between a grayscale value of 0 and a grayscale value of 1. Furthermore, the embodiment of the present invention identifies the dynamic area in the grayscale video frame so as to subsequently determine whether the dynamic area is the area where a human face is located. Furthermore, the embodiment of the present invention counts the total grayscale value in the dynamic area so as to first count the grayscale value of the dynamic area based on the principle that the grayscale value of the facial features is greater than the grayscale value of the skin and the grayscale value of the eye is greater than the grayscale value of other facial features, thereby distinguishing whether the dynamic area is a human eye area based on the statistical grayscale value. Furthermore, the embodiment of the present invention identifies whether there is a human eye in the dynamic area by using the total grayscale value so as to distinguish whether the dynamic area is a human eye area based on the statistical grayscale value. Furthermore, the embodiment of the present invention calculates the gaze point of the human eye in the human eye area so as to identify the gaze point of the human eye in the human eye area. Whether the point falls on the projector, since the projector itself will have a light source when projecting, when it is subsequently identified that the gaze point of the human eye in the human eye area falls on the projector, it can be determined that the light source of the projector itself causes damage to the human eye. Furthermore, the embodiment of the present invention calculates the number of blinks of the human eye in the human eye area, so that if the projected image on the white panel is blocked by the human face, the human eye is exposed to the strong light of the projector, but in the case of different projector positions at different distances, the damage to the human eye by the projector may be large or small, and in some cases, when there is no need to adjust the position of the projector, the degree of damage to the human eye by the strong light of the projector is determined by detecting the number of blinks of the human eye. Furthermore, the embodiment of the present invention detects whether the projector protects the eyes according to the gaze point of the human eye and the number of blinks of the human eye, so as to determine whether the human eye is damaged by the light of the projector when it is detected that the video projected on the white panel is blocked by the human eye. Furthermore, the embodiment of the present invention calculates the illumination correction loss of the illumination correction plan to minimize the eye protection cost. Therefore, the intelligent eye protection method and system based on a projector proposed in the embodiment of the present invention can minimize the eye protection cost of the projector. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0090] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0091] Figure 1 A schematic diagram of a flow chart of an intelligent eye protection method based on a projector provided in one embodiment of the present invention;

[0092] Figure 2 A module schematic diagram of a smart eye protection system based on a projector provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0093] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0094] The embodiment of the present invention provides an intelligent eye protection method based on a projector, and the execution subject of the intelligent eye protection method based on a projector includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present invention. In other words, the intelligent eye protection method based on a projector can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0095] See also Figure 1 FIG. 1 is a flow chart of an intelligent eye protection method based on a projector provided by an embodiment of the present invention. Figure 1 The intelligent eye protection method based on the projector described in the invention includes:

[0096] S1. Use a projector to capture a projection video frame under the projection screen, convert the projection video frame into a grayscale video frame, identify a dynamic area in the grayscale video frame, count the total grayscale value in the dynamic area, and use the total grayscale value to identify whether there is a human eye in the dynamic area.

[0097] In an embodiment of the present invention, the projector includes movable wheels, a camera and a brightness adjustment device. The projected screen refers to the screen displayed when the smart projector projects video content onto a white panel. The projected video frame refers to the video on the projection screen captured by the camera carried by the smart projector.

[0098] Furthermore, the embodiment of the present invention converts the projected video frame into a grayscale video frame so as to distinguish the image region where each independent object or independent organism in the projected video frame is located by utilizing the difference between a grayscale value of 0 and a grayscale value of 1.

[0099] The grayscale video frame refers to a video frame that only contains a grayscale value of 0 and a grayscale value of 1.

[0100] In one embodiment of the present invention, converting the projected video frame into a grayscale video frame includes: calculating a binarization threshold of the projected video frame using the following formula:

[0101] ;

[0102] in, represents the binarization threshold, represents the variance, Represents the projected video frame The probability of gray level, Represents the projected video frame The probability of gray level, Represents the projected video frame gray level probability;

[0103] When the pixel value in the projected video frame is greater than the binarization threshold, the pixel value in the projected video frame is adjusted to a first pixel value; when the pixel value in the projected video frame is not greater than the binarization threshold, the pixel value in the projected video frame is adjusted to a second pixel value; based on the first pixel value and the second pixel value, the projected video frame is converted into a grayscale video frame.

[0104] The first pixel value is 1, and the second pixel value is 0.

[0105] Furthermore, the embodiment of the present invention identifies the dynamic area in the grayscale video frame to subsequently determine whether the dynamic area is the area where a human face is located.

[0106] The dynamic area refers to an image area that only contains dynamic objects.

[0107] In one embodiment of the present invention, the identifying of the dynamic area in the grayscale video frame includes: utilizing the pixel value difference between the first pixel value and the second pixel value in the grayscale video frame to segment the separate individual areas in the grayscale video frame; calculating the pixel point coordinate mean between the coordinates of each pixel point in the separate individual area; using the pixel point coordinate mean as the area center point of the separate individual area; identifying whether the coordinates of the area center point move in the grayscale video frames at consecutive moments; when the coordinates of the area center point move in the grayscale video frames at consecutive moments, treating the separate individual area as a dynamic area.

[0108] The separate individual region refers to a region where pixels are in the same area, and each continuous region (the continuous region belongs to the same object or organism) can be determined based on the grayscale value of 0 and the grayscale value of 1 in the grayscale video frame.

[0109] Furthermore, an embodiment of the present invention counts the total grayscale value in the dynamic area, so as to first count the grayscale value of the dynamic area based on the principle that the grayscale value of the facial features is greater than the grayscale value of the skin and the grayscale value of the eyes is greater than the grayscale values ​​of other facial features, thereby subsequently distinguishing whether the dynamic area is a human eye area based on the statistical grayscale value.

[0110] The total grayscale value refers to the sum of the grayscale values ​​of each row or column of pixels in the dynamic area, but not the sum of the grayscale values ​​of all rows or columns of pixels.

[0111] In one embodiment of the present invention, the counting of the total grayscale value in the dynamic area includes: counting the sum of the first pixel values ​​of each row in the dynamic area to obtain the horizontal total grayscale value; counting the sum of the first pixel values ​​of each column in the dynamic area to obtain the vertical total grayscale value; and taking the horizontal total grayscale value and the vertical total grayscale value as the total grayscale value.

[0112] Furthermore, the embodiment of the present invention uses the total grayscale value to identify whether there is a human eye in the dynamic area, so as to distinguish whether the dynamic area is a human eye area based on the statistical grayscale value.

[0113] In one embodiment of the present invention, the use of the grayscale total value to identify whether there is a human eye in the dynamic area includes: obtaining the horizontal grayscale total value and the vertical grayscale total value in the grayscale total value; judging whether the shape of the dynamic area is a preset reference shape; if the shape of the dynamic area is the preset reference shape, based on the symmetry of the vertical grayscale total value in the dynamic area, the size sequence of the horizontal grayscale total value and the matching of the size sequence of the preset facial features grayscale values, identifying whether there is a human eye in the dynamic area; if the shape of the dynamic area is not the preset reference shape, obtaining different horizontal grayscale total values ​​and different vertical grayscale total values ​​of grayscale video frames belonging to continuous moments; extracting a first horizontal grayscale total value and a second horizontal grayscale total value from the different horizontal grayscale total values; identifying the first vertical coordinate and the second vertical coordinate of the first horizontal grayscale total value and the second horizontal grayscale total value on the dynamic area. ; Extract the first vertical grayscale total value and the second vertical grayscale total value from the different vertical grayscale total values; identify the first horizontal coordinate and the second horizontal coordinate of the first vertical grayscale total value and the second vertical grayscale total value in the dynamic area; obtain the first pixel point corresponding to the first vertical coordinate and the first horizontal coordinate, and obtain the second pixel point corresponding to the second vertical coordinate and the second horizontal coordinate; calculate the grayscale value similarity between the grayscale value of the first pixel point and the grayscale value of the second pixel point, and calculate the coordinate value similarity between the coordinate value of the first pixel point and the coordinate value of the second pixel point; when the grayscale value similarity is greater than a preset similarity and the coordinate value similarity is not greater than a preset similarity, determine that there is a human eye in the dynamic area; when the grayscale value similarity is not greater than a preset similarity and the coordinate value similarity is greater than a preset similarity, determine that there is no human eye in the dynamic area.

[0114] Among them, the preset reference shape refers to an ellipse-like shape formed by the facial edges of a human face, the first horizontal grayscale total value refers to the maximum horizontal grayscale total value existing in all video frames, the second horizontal grayscale total value refers to the horizontal grayscale total value that is second only to the first horizontal grayscale total value in the same video frame as the first horizontal grayscale total value, the ordinate refers to the row number, the abscissa refers to the column number, the first vertical grayscale total value refers to the maximum vertical grayscale total value existing in all video frames, the second vertical grayscale total value refers to the vertical grayscale total value that is second only to the first vertical grayscale total value in the same video frame as the first vertical grayscale total value. It should be noted that the preset similarity is set to 0.9, and can also be set according to actual scenarios.

[0115] Optionally, the process of identifying whether there is a human eye in the dynamic area based on the symmetry of the total vertical grayscale value in the dynamic area, the matching of the size sequence of the total horizontal grayscale values ​​and the size sequence of preset facial features grayscale values ​​means: if the total vertical grayscale value of the dynamic area is symmetrical on the left and right, that is, symmetrical on both sides of the face, and the size sequence of the total horizontal grayscale values ​​matches the size sequence of preset facial features grayscale values ​​(for example, the grayscale value of the eyes is greater than the grayscale value of the nose, and the grayscale value of the nose is greater than the grayscale value of the mouth), then there is a human eye in the dynamic area.

[0116] S2. When the total grayscale value is used to identify the presence of human eyes in the dynamic area, the human eye area in the dynamic area is identified, the human eye gaze point in the human eye area is calculated, the area length of the human eye area is calculated, the number of times the human eye blinks in the human eye area is calculated, and whether the projector is eye-protective is detected based on the human eye gaze point and the number of times the human eye blinks.

[0117] In one embodiment of the present invention, the identifying of the human eye area in the dynamic area includes: obtaining a first pixel point and a second pixel point in the dynamic area; querying a first grayscale area of ​​the first pixel point on the dynamic area and a second grayscale area of ​​the second pixel point on the dynamic area; and using the first grayscale area and the second grayscale area as the human eye area in the dynamic area.

[0118] The first grayscale area and the second grayscale area refer to areas in the same area as the pixel point, and each continuous area (this continuous area belongs to the same object or organism) can be determined based on the grayscale value of 0 and the grayscale value of 1 in the dynamic area.

[0119] Furthermore, the embodiment of the present invention calculates the gaze point of the human eye in the human eye area to identify whether the gaze point of the human eye in the human eye area falls on the projector. Since the projector itself also has a light source when projecting, when it is subsequently identified that the gaze point of the human eye in the human eye area falls on the projector, it can be determined that the light source of the projector itself causes damage to the human eye.

[0120] The human eye gaze point refers to the coordinate point of the object that the human eye gazes at in the human eye area.

[0121] In one embodiment of the present invention, the calculation of the human eye gaze point in the human eye area includes: extracting two-dimensional feature points in the human eye area; constructing reference feature points in the human eye area; calculating the rotation matrix of the two-dimensional feature points to the reference feature points; using the rotation matrix to calculate the head posture in the human eye area; using a preset pupil center corneal reflection method to analyze the initial gaze point in the human eye area; determining the gaze point range corresponding to the head posture; when the initial gaze point is not within the gaze point range, selecting a target gaze point closest to the initial gaze point from the gaze point range; and using the target gaze point as the human eye gaze point.

[0122] Among them, the reference feature point refers to the three-dimensional reference model of the head, the head posture refers to the deflection angle of the current head relative to the three-dimensional reference model on the xyz axis, and the pupil center corneal reflection method refers to the pupil-corneal reflection vector method, which is one of the current mainstream two-dimensional mapping methods. Its basic principle is that when an infrared light source irradiates the eye, the cornea will produce obvious reflection. If the infrared light source and the image acquisition device are fixed, the pupil position will change when the eyeball rotates, while the position of the corneal reflection spot will not change. Therefore, it can be used as a reference point for pupil movement, and the movement of the eyeball can be estimated based on the relative position change between the pupil center and the corneal reflection point.

[0123] Optionally, the process of extracting two-dimensional feature points in the human eye area refers to the process of extracting pixel coordinate points, the process of calculating the rotation matrix of the two-dimensional feature point to the reference feature point refers to the process of calculating the mapping relationship between two-dimensional coordinates and three-dimensional coordinates, the process of using the rotation matrix to calculate the head posture in the human eye area refers to using the rotation angle in the rotation matrix as the head posture, the process of determining the gaze point range corresponding to the head posture refers to using the head posture to determine the angle of deflection of the head relative to the xyz axis where the three-dimensional reference model is located, so that the angle of deflection of the eye gaze point relative to the xyz axis where the three-dimensional reference model is located can be determined, that is, after the head is rotated relative to the three-dimensional reference model, the eye gaze point will also rotate, and the gaze point range of the eye gaze point will follow the rotation, so that the gaze point range after following the rotation can be determined, and the process of selecting the target gaze point closest to the initial gaze point from the gaze point range refers to if the initial gaze point is not in the gaze point range corresponding to the head posture, then selecting a nearest gaze point in the gaze point range as the gaze point after correction of the initial gaze point.

[0124] In one embodiment of the present invention, the calculation of the area length of the human eye area includes: obtaining a first grayscale area and a second grayscale area in the human eye area; calculating the maximum coordinate distance between the grayscale coordinates on the first grayscale area and the grayscale coordinates on the second grayscale area; and using the maximum coordinate distance as the area length.

[0125] Furthermore, the embodiment of the present invention calculates the number of times a human eye blinks in the human eye area so that when the projected image on the white panel is blocked by the human face, the human eye is exposed to the strong light of the projector, but in cases where the projector is positioned at different distances, the damage to the human eye caused by the projector may be greater or lesser. In some cases, there is even no need to adjust the position of the projector, and the extent of the damage to the human eye caused by the strong light of the projector can be determined by detecting the number of times a human eye blinks.

[0126] In one embodiment of the present invention, the method of calculating the number of eye blinks of the eye region includes: obtaining a first pixel and a second pixel corresponding to the eye region; calculating a first similarity between the first pixel and other pixel points in grayscale video frames at consecutive moments, and calculating a second similarity between the second pixel and other pixel points in grayscale video frames at consecutive moments; when both the first similarity and the second similarity are not greater than a preset similarity, recording the eye-closed video frames corresponding to the eye region in the grayscale video frames at consecutive moments that are in a closed state; and calculating the number of eye blinks of the eye region using the following formula:

[0127] ;

[0128] in, Indicates the number of times a person blinks. Indicates closed-eye video frame When the moments of The number of Indicates total duration remove After closing the eyes video frame The number of occurrences.

[0129] For example, the For example, if three consecutive frames of images are in the eye-closed state, it can be determined that the human eye has not been open during this period, and the eye-closed state of these three frames is counted as one eye-closed state. Indicates the number of times the eye is closed when multiple consecutive frames are present. For example, the eye is closed once between 3:10 and 3:20, and once between 4:10 and 4:20. The number of times the eyes were closed was once between 3:10 and 3:20, and once between 4:10 and 4:20, which is 2 times.

[0130] It should be noted that when the first similarity and the second similarity are both not greater than the preset similarity, since the first pixel point and the second pixel point are obtained by the maximum total grayscale value in all video frames, that is to say, the first pixel point and the second pixel point must correspond to the position of the human eye (if the pixel point is obtained by the maximum total grayscale value in a single video frame, the eye judgment will fail due to the problem that the grayscale value of the eye skin in the closed eye state is less than the grayscale value of the nose), it indicates that the other pixel points that are not greater than the preset similarity are the skin area, that is, the eyelid area in the closed eye state.

[0131] Furthermore, an embodiment of the present invention detects whether the projector is eye-protecting based on the gaze point of the human eye and the number of times the human eye blinks, so as to determine whether the human eye is damaged by the light of the projector when it is detected that the video projected on the white panel is blocked by the human eye.

[0132] In one embodiment of the present invention, detecting whether the projector is eye-protective based on the human eye gaze point and the number of human eye blinks includes: determining whether the human eye gaze point falls on the projector; when the human eye gaze point falls on the projector, checking whether the number of human eye blinks is greater than a preset number of blinks; when the number of human eye blinks is greater than the preset number of blinks, determining that the projector is not eye-protective; when the human eye gaze point does not fall on the projector and the number of human eye blinks is not greater than the preset number of blinks, determining that the projector is eye-protective.

[0133] S3. When the projector does not protect the eyes, obtain the rotation angle range and movement range of the projector, generate a posture correction plan of the projector using the area length within the rotation angle range and the movement range, calculate the posture correction loss of the posture correction plan, and use the posture correction loss to adjust the posture correction plan to obtain an adjusted posture plan.

[0134] In the embodiment of the present invention, the posture correction plan includes corrected coordinates and correction angles.

[0135] In one embodiment of the present invention, the posture correction plan of the projector is generated by using the area length within the rotation angle range and the movement range, including: obtaining the uncorrected coordinates of the projector and generating random coordinates of the projector; judging whether the distance between the uncorrected coordinates and the random coordinates is within the movement range; when the distance between the uncorrected coordinates and the random coordinates is within the movement range, using the random coordinates as the corrected coordinates of the projector; obtaining the projection direction of the projector under the corrected coordinates; obtaining the left and right boundaries of the projection direction; obtaining the first straight line between the corrected coordinates and the left endpoint of the area length. line, and obtain a second straight line between the corrected coordinates and the right endpoint of the area length; extract a first angle between the right boundary and the first straight line, and a second angle between the left boundary and the second straight line; determine whether the first angle and the second angle are greater than the rotation angle range; when the first angle and the second angle are not greater than the rotation angle range, generate a correction angle of the projector from the first angle and the second angle; determine whether there is a human eye corresponding to the human eye area in the projection direction; when there is a human eye corresponding to the human eye area in the projection direction, generate a posture correction plan for the projector using the corrected coordinates and the correction angle.

[0136] Among them, the uncorrected coordinates refer to the coordinates of the projector before correction, the projection direction refers to the projection light direction of the projector, for example, from south to north, the left direction boundary refers to the left interval of the direction interval corresponding to the projection direction, the right direction boundary refers to the right interval of the direction interval corresponding to the projection direction, and the correction angle refers to the rotation angle of the projector when it moves to the corrected coordinates.

[0137] Optionally, the process of generating a correction angle of the projector from the first angle and the second angle refers to taking any one of the first angle and the second angle as the correction angle.

[0138] In one embodiment of the present invention, the calculating the posture correction loss of the posture correction plan includes: calculating the posture correction loss of the posture correction plan using the following formula:

[0139] ;

[0140] in, represents the pose correction loss, Indicates the corrected coordinates in the pose correction planning. represents the uncorrected coordinates of the projector, Indicates the correction angle in posture correction planning.

[0141] In one embodiment of the present invention, the posture correction plan is adjusted by using the posture correction loss to obtain the adjusted posture plan, including: when the posture correction loss is minimum, the posture correction plan corresponding to the posture correction loss is used as the adjusted posture plan.

[0142] S4. Correcting the projection posture of the projector using the posture adjustment plan to obtain a corrected posture, and determining whether the projector is eye-protecting in the corrected posture.

[0143] In one embodiment of the present invention, the determining whether the projector is eye-protecting in the corrected posture includes: obtaining the human eye-projector distance between the projector and the human eye in the corrected posture, and obtaining the human eye-projector angle of the projector in the corrected posture; collecting the historical illumination, historical distance and historical angle of the projector; and generating an illumination-distance-angle regression model between the historical illumination and the historical distance and the historical angle using the following formula:

[0144] ;

[0145] in, represents the illumination-distance-angle regression model, Represents historical illumination, Represents the historical distance, From a historical perspective, , , represents the regression coefficient;

[0146] The suitable lighting for the human eye corresponding to the human eye-projector distance and the human eye-projector angle is calculated through the lighting-distance-angle regression model; the initial lighting of the projector in the corrected position is obtained; when the initial lighting meets the suitable lighting for the human eye, it is determined that the projector protects the eyes in the corrected position; when the initial lighting does not meet the suitable lighting for the human eye, it is determined that the projector does not protect the eyes in the corrected position.

[0147] Among them, the human eye-projector angle refers to the deviation angle of the human eye relative to the center direction of the projection direction of the projector, the historical illumination refers to the historical light brightness, the historical distance refers to the distance between the human eye and the projector in the historical period, the historical angle refers to the deviation angle of the human eye relative to the center direction of the projection direction of the projector in the historical period, and the initial illumination refers to the illumination of the projector before correction.

[0148] S5. When the projector does not protect the eyes in the corrected posture, obtain the illumination range of the projector, generate an illumination correction plan of the projector within the illumination range, calculate the illumination correction loss of the illumination correction plan, and adjust the illumination correction plan using the illumination correction loss to obtain an adjusted illumination plan.

[0149] In the embodiment of the present invention, the illumination range refers to the light brightness range of the projector.

[0150] In one embodiment of the present invention, generating the illumination correction plan of the projector within the illumination range includes: generating a corrected area and corrected illumination of the projector within the illumination range; calculating an illumination mean of the corrected illumination in the corrected area; and when the illumination mean meets the appropriate illumination for the human eye; using the corrected area and the corrected illumination as the illumination correction plan of the projector.

[0151] Optionally, the process of generating the corrected area and the corrected illumination of the projector within the illumination range is randomly generated.

[0152] Furthermore, the embodiment of the present invention calculates the illumination correction loss of the illumination correction plan to minimize the eye protection cost.

[0153] In one embodiment of the present invention, the calculating of the illumination correction loss of the illumination correction plan includes: obtaining a pre-correction area and pre-correction illumination of the projector; obtaining a post-correction area and post-correction illumination of the illumination correction plan; and calculating the illumination correction amount of the illumination correction plan according to the pre-correction area, the pre-correction illumination, the post-correction area, and the post-correction illumination using the following formula:

[0154] ;

[0155] in, represents the lighting correction cost, Indicates The pre-corrected illumination in the pre-corrected area, Indicates The corrected illumination in the corrected region, represents the number of regions before correction;

[0156] According to the corrected area and the corrected illumination, the illumination dispersion of the illumination correction plan is calculated using the following formula:

[0157] ;

[0158] in, represents the illumination dispersion, Indicates The corrected illumination in the corrected region, represents the number of regions before correction, express indivual The average light intensity of

[0159] The illumination correction amount and the illumination dispersion are used as the illumination correction loss of the illumination correction plan.

[0160] In one embodiment of the present invention, the illumination correction plan is adjusted by using the illumination correction loss to obtain an adjusted illumination plan, including: obtaining the illumination correction amount and illumination uniformity in the illumination correction loss; when the illumination correction amount is minimum and the illumination discreteness is less than a preset discreteness, using the illumination correction plan corresponding to the illumination correction amount and the illumination uniformity as the adjusted illumination plan.

[0161] The preset discreteness is set according to the actual scenario and will not be further elaborated here.

[0162] S6. Use the adjusted illumination plan to correct the projection illumination of the projector to obtain corrected illumination, so as to complete the intelligent eye protection of the projector through the corrected illumination and obtain the intelligent eye protection result of the projector.

[0163] Optionally, the process of correcting the projection light of the projector by adjusting the illumination plan to obtain the corrected illumination refers to adjusting the luminance of the projection light of the projector to a form in the illumination correction plan.

[0164] It can be seen that the embodiment of the present invention converts the projected video frame into a grayscale video frame so as to use the difference between a grayscale value of 0 and a grayscale value of 1 to distinguish the image area where each independent object or independent organism in the projected video frame is located. Furthermore, the embodiment of the present invention identifies the dynamic area in the grayscale video frame so as to judge whether the dynamic area is the area where a human face is located in the subsequent process. Furthermore, the embodiment of the present invention counts the total grayscale value in the dynamic area so as to first count the grayscale value of the dynamic area based on the principle that the grayscale value of the facial features is greater than the grayscale value of the skin and the grayscale value of the eye is greater than the grayscale value of other facial features, so as to distinguish whether the dynamic area is a human eye area based on the statistical grayscale value in the subsequent process. Furthermore, the embodiment of the present invention identifies whether there is a human eye in the dynamic area by using the total grayscale value so as to distinguish whether the dynamic area is a human eye area based on the statistical grayscale value. Furthermore, the embodiment of the present invention calculates the gaze point of the human eye in the human eye area so as to identify the human eye in the human eye area. Whether the gaze point falls on the projector. Since the projector itself also has a light source when projecting, when it is subsequently identified that the gaze point of the human eye in the human eye area falls on the projector, it can be determined that the light source of the projector itself causes damage to the human eye. Further, the embodiment of the present invention calculates the number of human eye blinks in the human eye area, so that if the projected image projected on the white panel is blocked by the human face, the human eye is exposed to the strong light of the projector, but in the case of different projector positions at different distances, the damage to the human eye by the projector may be large or small. In some cases, when there is no need to adjust the position of the projector, the damage to the human eye by the strong light of the projector is determined by detecting the number of human eye blinks. Further, the embodiment of the present invention detects whether the projector protects the eyes according to the human eye gaze point and the number of human eye blinks, so as to determine whether the human eye is damaged by the light of the projector when it is detected that the video projected on the white panel is blocked by the human eye. Further, the embodiment of the present invention calculates the illumination correction loss of the illumination correction plan to minimize the eye protection cost. Therefore, the intelligent eye protection method and system based on a projector proposed in the embodiment of the present invention can minimize the eye protection cost of the projector.

[0165] like Figure 2 The figure shows a functional module diagram of the intelligent eye protection system based on a projector according to the present invention.

[0166] The intelligent eye protection system 200 based on the projector of the present invention can be installed in an electronic device. According to the functions implemented, the intelligent eye protection system based on the projector can include a human eye detection module 201, an eye protection detection module 202, a posture adjustment module 203, a posture correction module 204, an illumination adjustment module 205 and an intelligent eye protection module 206. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0167] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0168] The human eye detection module 201 is used to collect projection video frames under the projection screen by using a projector, convert the projection video frames into grayscale video frames, identify the dynamic area in the grayscale video frames, count the total grayscale value in the dynamic area, and use the total grayscale value to identify whether there is a human eye in the dynamic area;

[0169] The eye protection detection module 202 is used to identify the human eye area in the dynamic area when the total grayscale value is used to identify the presence of human eyes in the dynamic area, calculate the human eye gaze point in the human eye area, calculate the area length of the human eye area, calculate the number of human eye blinks in the human eye area, and detect whether the projector is eye-protecting according to the human eye gaze point and the number of human eye blinks;

[0170] The posture adjustment module 203 is used to obtain the rotation angle range and movement range of the projector when the projector does not protect the eyes, generate a posture correction plan for the projector using the region length within the rotation angle range and the movement range, calculate the posture correction loss of the posture correction plan, and adjust the posture correction plan using the posture correction loss to obtain an adjusted posture plan;

[0171] The posture correction module 204 is used to correct the projection posture of the projector by using the posture adjustment plan to obtain a corrected posture, and determine whether the projector is eye-protective in the corrected posture;

[0172] The illumination adjustment module 205 is used to obtain the illumination range of the projector when the projector does not protect the eyes in the correction posture, generate an illumination correction plan for the projector within the illumination range, calculate the illumination correction loss of the illumination correction plan, and adjust the illumination correction plan using the illumination correction loss to obtain an adjusted illumination plan;

[0173] The intelligent eye protection module 206 is used to correct the projection illumination of the projector by using the adjusted illumination plan to obtain corrected illumination, so as to complete the intelligent eye protection of the projector through the corrected illumination and obtain the intelligent eye protection result of the projector.

[0174] In detail, the modules in the projector-based intelligent eye protection system 200 in the embodiment of the present invention are used in the same manner as described above. Figure 1 The intelligent eye protection method based on the projector is the same technical means as described in the text and can produce the same technical effects, so I will not go into details here.

[0175] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0176] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0177] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0178] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0179] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An intelligent eye protection method based on a projector, characterized in that: The method comprises: Using a projector to collect a projection video frame under a projection screen, converting the projection video frame into a grayscale video frame, identifying a dynamic area in the grayscale video frame, counting the total grayscale value in the dynamic area, and using the total grayscale value to identify whether there is a human eye in the dynamic area; When the total grayscale value is used to identify the presence of human eyes in the dynamic area, the human eye area in the dynamic area is identified, the human eye gaze point in the human eye area is calculated, the area length of the human eye area is calculated, the number of human eye blinks in the human eye area is calculated, and whether the projector is eye-protective is detected according to the human eye gaze point and the number of human eye blinks; When the projector does not protect the eyes, the rotation angle range and the movement range of the projector are obtained, the posture correction plan of the projector is generated by using the region length within the rotation angle range and the movement range, the posture correction loss of the posture correction plan is calculated, and the posture correction plan is adjusted by using the posture correction loss to obtain an adjusted posture plan, wherein the posture correction plan of the projector is generated by using the region length within the rotation angle range and the movement range, including: Obtaining uncorrected coordinates of the projector and generating random coordinates of the projector; determining whether the distance between the uncorrected coordinate and the random coordinate is within the moving range; When the distance between the uncorrected coordinate and the random coordinate is within the moving range, using the random coordinate as the corrected coordinate of the projector; Obtaining the projection direction of the projector under the corrected coordinates; Obtaining the left boundary and the right boundary of the projection direction; Acquire a first straight line between the corrected coordinate and the left end point of the region length, and acquire a second straight line between the corrected coordinate and the right end point of the region length; Extracting a first angle between the right boundary and the first straight line, and a second angle between the left boundary and the second straight line; Determining whether the first angle and the second angle are greater than the rotation angle range; When the first angle and the second angle are not greater than the rotation angle range, generating a correction angle of the projector from the first angle and the second angle; Determine whether there is a human eye corresponding to the human eye area in the projection direction; When there is a human eye corresponding to the human eye area in the projection direction, generating a posture correction plan of the projector using the corrected coordinates and the correction angle; Correcting the projection posture of the projector by using the posture adjustment plan to obtain a corrected posture, and determining whether the projector is eye-protecting in the corrected posture; When the projector does not protect the eyes in the corrected posture, the illumination range of the projector is obtained, an illumination correction plan of the projector is generated within the illumination range, an illumination correction loss of the illumination correction plan is calculated, and the illumination correction plan is adjusted using the illumination correction loss to obtain an adjusted illumination plan, wherein the illumination correction plan of the projector is generated within the illumination range, including: generating a corrected area and corrected illumination of the projector within the illumination range; calculating an illumination mean of the corrected illumination in the corrected area; and when the illumination mean meets the appropriate illumination for the human eye, using the corrected area and the corrected illumination as the illumination correction plan of the projector; The projection illumination of the projector is corrected by using the adjusted illumination plan to obtain corrected illumination, so as to complete the intelligent eye protection of the projector through the corrected illumination and obtain the intelligent eye protection result of the projector.

2. The method according to claim 1, characterized in that The identifying a dynamic area in the grayscale video frame comprises: segmenting individual regions in the grayscale video frame using a pixel value difference between a first pixel value and a second pixel value in the grayscale video frame; Calculate the mean value of pixel coordinates between each pixel coordinate in the individual area; Taking the mean value of the pixel coordinates as the center point of the individual region; Identify whether the center point of the region moves in the grayscale video frames at consecutive moments; When the coordinates of the center point of the region move in the grayscale video frames at consecutive moments, the separate individual region is regarded as a dynamic region.

3. The method according to claim 1, characterized in that The identifying whether there is a human eye in the dynamic area by using the total grayscale value includes: Obtaining a horizontal grayscale total value and a vertical grayscale total value in the grayscale total value; Determining whether the shape of the dynamic area is a preset reference shape; If the shape of the dynamic area is a preset reference shape, identifying whether there is a human eye in the dynamic area based on the symmetry of the vertical grayscale total value in the dynamic area, the matching of the size sequence of the horizontal grayscale total value and the size sequence of the preset facial features grayscale values; If the shape of the dynamic area is not a preset reference shape, obtaining different horizontal grayscale total values ​​and different vertical grayscale total values ​​of grayscale video frames belonging to consecutive moments; Extracting a first level grayscale total value and a second level grayscale total value from the different level grayscale total values; Identify a first ordinate and a second ordinate of the first horizontal grayscale total value and the second horizontal grayscale total value in the dynamic area; Extracting a first vertical grayscale total value and a second vertical grayscale total value from the different vertical grayscale total values; Identify a first horizontal coordinate and a second horizontal coordinate of the first vertical grayscale total value and the second vertical grayscale total value in the dynamic area; Acquire a first pixel point corresponding to the first ordinate and the first abscissa, and acquire a second pixel point corresponding to the second ordinate and the second abscissa; Calculating the grayscale value similarity between the grayscale value of the first pixel and the grayscale value of the second pixel, and calculating the coordinate value similarity between the coordinate value of the first pixel and the coordinate value of the second pixel; When the gray value similarity is greater than a preset similarity and the coordinate value similarity is not greater than a preset similarity, determining that a human eye exists in the dynamic area; When the gray value similarity is not greater than a preset similarity and the coordinate value similarity is greater than a preset similarity, it is determined that there is no human eye in the dynamic area.

4. The method according to claim 1, characterized in that: The calculating the number of eye blinks of the eye region includes: Obtaining a first pixel point and a second pixel point corresponding to the human eye area; Calculating a first similarity between the first pixel and other pixel points in the grayscale video frames at consecutive moments, and calculating a second similarity between the second pixel and other pixel points in the grayscale video frames at consecutive moments; When both the first similarity and the second similarity are not greater than a preset similarity, recording the closed-eye video frames corresponding to the human eye region in the closed-eye state in the grayscale video frames at consecutive moments; The number of eye blinks in the eye area is calculated using the following formula: ； in, Indicates the number of times a person blinks. Indicates closed-eye video frame When the moments of The number of Indicates total duration remove After closing the eyes video frame The number of occurrences.

5. The method according to claim 1, characterized in that The detecting whether the projector is eye-protecting according to the gaze point of the human eye and the number of times the human eye blinks comprises: Determining whether the gaze point of the human eye falls on the projector; When the gaze point of the human eye falls on the projector, query whether the number of blinks of the human eye is greater than a preset number of blinks; When the number of blinks of the human eye is greater than the preset number of blinks, determining that the projector does not protect the eyes; When the gaze point of the human eye does not fall on the projector and the number of blinks of the human eye is not greater than the preset number of blinks, it is determined that the projector is eye-protecting.

6. The method according to claim 1, characterized in that The determining whether the projector is eye-protecting in the correction posture includes: Acquire a human eye-projector distance between the projector and the human eye in the corrected posture, and acquire a human eye-projector angle of the projector in the corrected posture; Collecting historical illumination, historical distance and historical angle of the projector; The following formula is used to generate the illumination-distance-angle regression model between the historical illumination, the historical distance, and the historical angle: ； in, represents the illumination-distance-angle regression model, Represents historical illumination, Represents the historical distance, From a historical perspective, , , represents the regression coefficient; Calculate the appropriate illumination for the human eye corresponding to the distance between the human eye and the projector and the angle between the human eye and the projector by using the illumination-distance-angle regression model; Acquiring initial illumination of the projector in the corrected posture; When the initial illumination meets the illumination suitable for human eyes; Determining that the projector is eye-protective in the corrected posture; When the initial illumination does not meet the illumination suitable for human eyes; It is determined that the projector does not protect the eyes in the corrected posture.

7. The method according to claim 1, characterized in that The calculating the illumination correction loss of the illumination correction plan includes: Obtaining a pre-correction area and pre-correction illumination of the projector; Obtaining the corrected area and corrected illumination of the illumination correction plan; The illumination correction amount of the illumination correction plan is calculated using the following formula according to the area before correction, the illumination before correction, the area after correction, and the illumination after correction: ； in, represents the lighting correction cost, Indicates The pre-corrected illumination in the pre-corrected area, Indicates The corrected illumination in the corrected region, represents the number of regions before correction; According to the corrected area and the corrected illumination, the illumination dispersion of the illumination correction plan is calculated using the following formula: ； in, represents the illumination dispersion, Indicates The corrected illumination in the corrected region, represents the number of regions before correction, express indivual The average light intensity of The illumination correction amount and the illumination dispersion are used as the illumination correction loss of the illumination correction plan.

8. The method according to claim 1, characterized in that The step of adjusting the illumination correction plan by using the illumination correction loss to obtain an adjusted illumination plan includes: Obtaining the illumination correction amount and illumination uniformity in the illumination correction loss; When the illumination correction amount is minimum and the illumination dispersion is less than a preset dispersion, the illumination correction plan corresponding to the illumination correction amount and the illumination uniformity is used as the adjusted illumination plan.

9. An intelligent eye protection system based on a projector, characterized in that: The system comprises: The human eye detection module is used to collect a projection video frame under the projection screen by using a projector, convert the projection video frame into a grayscale video frame, identify a dynamic area in the grayscale video frame, count the total grayscale value in the dynamic area, and use the total grayscale value to identify whether there is a human eye in the dynamic area; an eye protection detection module, for identifying a human eye region in the dynamic region when the total grayscale value is used to identify the presence of a human eye in the dynamic region, calculating a human eye gaze point in the human eye region, calculating a region length of the human eye region, calculating a number of human eye blinks in the human eye region, and detecting whether the projector is eye-protecting according to the human eye gaze point and the number of human eye blinks; The posture adjustment module is used to obtain the rotation angle range and movement range of the projector when the projector does not protect the eyes, generate the posture correction plan of the projector using the area length within the rotation angle range and the movement range, calculate the posture correction loss of the posture correction plan, and adjust the posture correction plan using the posture correction loss to obtain an adjusted posture plan, wherein the generation of the posture correction plan of the projector using the area length within the rotation angle range and the movement range includes: Obtaining uncorrected coordinates of the projector and generating random coordinates of the projector; determining whether the distance between the uncorrected coordinate and the random coordinate is within the moving range; When the distance between the uncorrected coordinate and the random coordinate is within the moving range, using the random coordinate as the corrected coordinate of the projector; Obtaining the projection direction of the projector under the corrected coordinates; Obtaining the left boundary and the right boundary of the projection direction; Acquire a first straight line between the corrected coordinate and the left end point of the region length, and acquire a second straight line between the corrected coordinate and the right end point of the region length; Extracting a first angle between the right boundary and the first straight line, and a second angle between the left boundary and the second straight line; Determining whether the first angle and the second angle are greater than the rotation angle range; When the first angle and the second angle are not greater than the rotation angle range, generating a correction angle of the projector from the first angle and the second angle; Determine whether there is a human eye corresponding to the human eye area in the projection direction; When there is a human eye corresponding to the human eye area in the projection direction, generating a posture correction plan of the projector using the corrected coordinates and the correction angle; A posture correction module, used to correct the projection posture of the projector by using the posture adjustment plan to obtain a corrected posture, and determine whether the projector is eye-protective in the corrected posture; An illumination adjustment module, for obtaining the illumination range of the projector when the projector does not protect the eyes in the correction posture, generating an illumination correction plan for the projector within the illumination range, calculating the illumination correction loss of the illumination correction plan, and adjusting the illumination correction plan using the illumination correction loss to obtain an adjusted illumination plan, wherein generating the illumination correction plan for the projector within the illumination range comprises: generating a corrected area and corrected illumination of the projector within the illumination range; calculating an illumination mean of the corrected illumination in the corrected area; and when the illumination mean meets the appropriate illumination for the human eye, using the corrected area and the corrected illumination as the illumination correction plan for the projector; The intelligent eye protection module is used to use the adjusted illumination plan to correct the projection illumination of the projector to obtain corrected illumination, so as to complete the intelligent eye protection of the projector through the corrected illumination and obtain the intelligent eye protection result of the projector.

Images