A method and system for controlling a projector

Abstract

This invention belongs to the field of control and discloses a projector control method and system. The method includes S1, acquiring an image (imgfir) for identifying the brightness of the hand region during the k-th control cycle. k S2, obtain imgfir k S3, calculate the brightness set of the infrared fill light; S4, illuminate the shooting area with infrared light based on the fill light brightness set; S5, obtain the set of images gstrec for gesture recognition in the k-th control cycle. k S6, from gstrec k Select target image aimimg from the middle k , for aimimg k S7. Perform gesture recognition to obtain the type of gesture; S8. Control the projector based on the type of gesture. This invention can more effectively reduce the probability of overexposure or underexposure, thereby improving the probability of correctly recognizing the speaker's hand movements.

Description

TECHNICAL FIELD

[0001] The present application relates to the control field, and particularly to a projector control method and system. BACKGROUND

[0002] The intelligent projector is a projector installed with an operating system. When a PPT needs to be displayed, the PPT file only needs to be imported into the system of the projector, and the PPT file can be demonstrated without connecting a computer or other equipment.

[0003] However, the general control method of such an intelligent projector is to control it through a remote controller. For example, when the next slide needs to be switched to, the button on the remote controller can be directly pressed. However, this method makes the speaker need to hold the remote controller all the time when giving a speech, which is not convenient. With the development of image recognition technology, a scheme of using gesture recognition to control the projector gradually appears. In this scheme, the speaker no longer needs to carry a remote controller. When control is needed, only a corresponding gesture needs to be made. However, in an environment with poor light conditions, the accuracy of gesture recognition will be greatly affected because the camera cannot capture the hand of the speaker due to insufficient light. The solution to this problem is to use the way of shooting infrared images to perform gesture recognition.

[0004] However, this method is not perfect either because the speaker will move, which causes the distance between the hand of the speaker and the lens for shooting infrared images to change. The infrared fill light in the prior art controls the brightness of the fill light according to the overall brightness of the environment. Therefore, when the overall brightness of the environment does not change, a single brightness obviously cannot make the lens obtain the optimal shooting effect for the hand of the speaker at different distances, that is, there will be a situation that the hand of the speaker is overexposed or underexposed in the infrared image obtained in some area. Obviously, this will reduce the probability of correctly recognizing the action of the hand of the speaker and result in poor user experience. SUMMARY

[0005] The present application aims to disclose a projector control method and system, which solves the problem of how to control the brightness of the infrared fill light in the process of obtaining the infrared image of the hand of the speaker, thereby improving the probability of correctly recognizing the action of the hand of the speaker.

[0006] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0007] On the one hand, the present application provides a projector control method, comprising:

[0008] S1, acquire the image imgfir used for identifying the brightness of the hand region during the k-th control cycle. k ;

[0009] S2, for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ;

[0010] S3, based on the brightness of the hand area (hanbri) k Calculate the set of infrared fill light brightness filbrg k ;

[0011] S4, based on the supplementary light brightness set filbrg k Irradiate infrared light onto the shooting area;

[0012] S5, capture images of the shooting area to obtain the set of images gstrec for gesture recognition in the k-th control cycle. k ;

[0013] S6, from gstrec k Select target image aimimg from the middle k , for aimimg k Perform gesture recognition to obtain the type of gesture;

[0014] S7 controls the projector based on the type of gesture.

[0015] Preferably, the image imgfir used for identifying the brightness of the hand region in the kth control cycle is acquired. k ,include:

[0016] S11, Obtain the set of images gstrec used for gesture recognition in the (k-1)th control cycle. k-1 ;

[0017] S12, gstrec k-1 The image captured latest in the middle is used as the image for identifying the brightness of the hand region in the k-th control cycle. k .

[0018] Preferably, for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ,include:

[0019] S21, acquire the target image aimimg in the (k-1)th control cycle. k-1The set of pixels in the identified hand region, handcoll k-1 ;

[0020] S22, in aimimg k-1 Get handcoll k-1 The minimum bounding rectangle of the mibudret k-1 ;

[0021] S23, Obtain mibudret k-1 Distribution range of the x-coordinates of pixels in x k-1,min and x k-1,max Representing the minimum and maximum values ​​of the distribution interval of the x-axis, respectively, to obtain mibudret k-1 Distribution range of the ordinate of pixels in y k-1,min and y k-1,max These represent the minimum and maximum values ​​of the distribution range of the horizontal axis, respectively;

[0022] S24, calculate the expansion coefficient for the k-th control cycle;

[0023] S25, estimated imgfir k The hand region hanare(x,y) in the image;

[0024] S26, in imgfir k Calculate the average gray value avepixval of the pixels located in the hand region hanare(x,y);

[0025] S27, the luminance of the hand area is calculated using the following formula (hanbri). k :

[0026] hanbri k =(1+δ) avepixval

[0027] Where δ represents the brightness range control parameter, δ>0.02.

[0028] Preferably, based on the brightness of the hand area, hanbri k Calculate the set of infrared fill light brightness filbrg k ,include:

[0029] S31, calculate the brightness variation coefficient:

[0030]

[0031] Among them, posiva k Handcoll represents the brightness change coefficient during the k-th control cycle.k-2 The target image aimimg represents the target image in the (k-2)th control cycle. k-2 The set of pixels in the identified hand region, dist(handcoll k-1 ,handcoll k-2 ) indicates handcoll k-1 The average coordinates of the pixels and handcoll k-2 The distance between the average coordinates of pixels in the graph, where bydist represents the preset standard distance. and They represent aimimg respectively k-2 and aimimg k-1 The shooting time is indicated by "byshotim" which represents the preset shooting duration; hanbri k-1 This represents the brightness of the hand region calculated in the (k-1)th control cycle; α represents the distance weight, β represents the time weight, and η represents the brightness weight.

[0032] The range of variation in the supplementary light brightness is [hanbri k -posiva k ×perloc,hanbri k +posiva k [×perloc]; perloc represents the preset brightness value;

[0033] S32, [hanbri k -posiva k ×perloc,hanbri k +posiva k The integer values ​​in [×perloc] are stored in the supplementary light brightness set filbrg k .

[0034] Preferably, based on the supplementary light brightness set filbrg k Irradiate the shooting area with infrared light, including:

[0035] S41, Calculate the set filbrg k The duration of illumination for each value of supplemental light brightness;

[0036] S42 illuminates the shooting area according to the illumination duration corresponding to each supplementary light brightness.

[0037] Preferably, the shooting area is captured to obtain a set of images (gstrec) for gesture recognition in the k-th control cycle. k ,include:

[0038] S51, calculates continuous shooting interval;

[0039] S52, based on the continuous shooting interval, captures images of the shooting area using continuous shooting during the time period when infrared light is irradiated onto the shooting area, obtaining a set of images gstrec for gesture recognition in the k-th control cycle. k .

[0040] Preferably, from gstrec k Select target image aimimg from the middle k ,include:

[0041] Calculate gstrec separately k The quality score of each image in the dataset;

[0042] gstrec k The image with the highest quality score is used as the target image (aimimg). k .

[0043] Preferably, the types of gestures include extending one finger, extending two fingers, extending three fingers, extending four fingers, opening the palm, and making a fist.

[0044] Preferably, controlling the projector based on the type of gesture includes:

[0045] Get the control action corresponding to the type of gesture;

[0046] The projector is controlled based on control actions.

[0047] On the other hand, the present invention provides a projector control system, including an acquisition module, an identification module, a calculation module, an illumination module, an imaging module, a filtering module, and a control module;

[0048] The acquisition module is used to acquire the image (imgfir) for identifying the brightness of the hand region during the k-th control cycle. k ;

[0049] The recognition module is used for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ;

[0050] The calculation module is used for hanbri based on the brightness of the hand area. k Calculate the set of infrared fill light brightness filbrg k ;

[0051] The illumination module is used for illumination based on the supplementary light brightness set filbrg k Irradiate infrared light onto the shooting area;

[0052] The imaging module is used to capture images of the imaging area, obtaining a set of images (gstrec) for gesture recognition in the k-th control cycle. k ;

[0053] The filtering module is used to filter from gstrec k Select target image aimimg from the middle k , for aimimg k Perform gesture recognition to obtain the type of gesture;

[0054] The control module is used to control the projector based on the type of gesture.

[0055] Compared to existing methods that control infrared fill lights based on overall ambient brightness to illuminate the shooting area, this invention first estimates the brightness of the hand area, then calculates a set of fill light brightness levels based on that brightness, captures images at different fill light brightness levels, and finally performs gesture recognition on these images to achieve gesture control of the projection line. During projection, the user's hand position changes, and the overall brightness cannot accurately reflect the brightness of the user's hand area. This invention uses the brightness of the hand area to determine the fill light brightness; therefore, the calculated fill light brightness is more suitable for the hand area, more effectively reducing the probability of overexposure or underexposure when shooting the hand area. This improves the probability of correctly recognizing the speaker's hand movements and enhances the user experience. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a schematic diagram of a projector control method according to the present invention.

[0058] Figure 2 This is a schematic diagram of a projector control system according to the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this application or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0060] On the one hand, such as Figure 1 As shown in one embodiment, the present invention provides a projector control method, comprising:

[0061] S1, acquire the image imgfir used for identifying the brightness of the hand region during the k-th control cycle. k ;

[0062] S2, for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ;

[0063] S3, based on the brightness of the hand area (hanbri) k Calculate the set of infrared fill light brightness filbrg k ;

[0064] S4, based on the supplementary light brightness set filbrg k Irradiate infrared light onto the shooting area;

[0065] S5, capture images of the shooting area to obtain the set of images gstrec for gesture recognition in the k-th control cycle. k ;

[0066] S6, from gstrec k Select target image aimimg from the middle k , for aimimg k Perform gesture recognition to obtain the type of gesture;

[0067] S7 controls the projector based on the type of gesture.

[0068] Compared to existing methods that control infrared fill lights based on overall ambient brightness to illuminate the shooting area, this invention first estimates the brightness of the hand area, then calculates a set of fill light brightness levels based on that brightness, captures images at different fill light brightness levels, and finally performs gesture recognition on these images to achieve gesture control of the projection line. During projection, the user's hand position changes, and the overall brightness cannot accurately reflect the brightness of the user's hand area. This invention uses the brightness of the hand area to determine the fill light brightness; therefore, the calculated fill light brightness is more suitable for the hand area, more effectively reducing the probability of overexposure or underexposure when shooting the hand area. This improves the probability of correctly recognizing the speaker's hand movements and enhances the user experience.

[0069] One advantage of this invention is that it fully utilizes the gesture recognition results of the (k-1)th control cycle. During gesture recognition in the (k-1)th control cycle, the hand area is first obtained, and then image recognition is performed on the area to determine the gesture type. Therefore, the approximate location of the hand area is known, eliminating the need to re-capture the area and calculate the hand area's brightness based on the captured image. Re-capturing presents challenges: firstly, the brightness of the supplementary light is uncertain; secondly, hand recognition takes time, preventing the capture area from being captured during the recognition phase. This significantly increases the probability of failing to capture hand movements, sometimes resulting in the projector not responding even when the user makes a movement, severely impacting user experience. This invention, however, can immediately enter the next control cycle after obtaining the target image, within the gesture recognition period. While recognizing the gesture type from the previous control cycle, it can calculate the set of supplementary light intensities for the next control cycle and immediately begin capturing the area, effectively ensuring a high probability of capturing hand movements.

[0070] Specifically, the start time of the kth control cycle can be the time when the target image was last acquired, and the end time can be the time when the target image is acquired this time. Then, the time when the target image is acquired this time can be used as the start time of the (k+1)th control cycle, and so on.

[0071] Preferably, when k is less than or equal to the set threshold, global brightness is used to obtain the supplementary light brightness of the supplementary light lamp, and then illumination is performed. The shooting area is continuously photographed using a preset continuous shooting interval to obtain a set of images for gesture recognition. Then, the projector is controlled using the S6-S7 method.

[0072] Preferably, the infrared fill light is always on. After the infrared light is irradiated onto the shooting area according to the brightness in the fill light brightness set, the infrared fill light will maintain the last fill light brightness until a new fill light brightness set is obtained, at which point the fill light brightness will be readjusted according to the fill light brightness set.

[0073] Preferably, the image imgfir used for identifying the brightness of the hand region in the kth control cycle is acquired. k ,include:

[0074] S11, Obtain the set of images gstrec used for gesture recognition in the (k-1)th control cycle. k-1 ;

[0075] S12, gstrec k-1 The image captured latest in the middle is used as the image for identifying the brightness of the hand region in the k-th control cycle. k .

[0076] Specifically, since the speaker may move, this invention selects the image with the latest acquisition time to identify the brightness of the hand area, thereby increasing the probability that the calculated set of supplementary lighting brightness includes the true brightness of the space where the speaker's hands are located in real time.

[0077] Preferably, for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ,include:

[0078] S21, acquire the target image aimimg in the (k-1)th control cycle. k-1 The set of pixels in the identified hand region, handcoll k-1 ;

[0079] S22, in aimimg k-1 Get handcoll k-1 The minimum bounding rectangle of the mibudret k-1 ;

[0080] S23, Obtain mibudret k-1 Distribution range of the x-coordinates of pixels in x k-1,min and x k-1,max Representing the minimum and maximum values ​​of the distribution interval of the x-axis, respectively, to obtain mibudret k-1 Distribution range of the ordinate of pixels in y k-1,min and y k-1,maxThese represent the minimum and maximum values ​​of the distribution range of the horizontal axis, respectively;

[0081] S24, Calculate the expansion factor for the k-th control cycle:

[0082]

[0083] Among them, ampcoef k This represents the expansion factor for the k-th control cycle. Indicates imgfir k The filming time, It means aimimg k-1 The shooting time, Φ represents the preset standard magnification factor;

[0084] S25, estimated imgfir k The hand region hanare(x,y) in the image.

[0085]

[0086] Where chgr represents the range variation parameter;

[0087] S26, in imgfir k Calculate the average gray value avepixval of the pixels located in the hand region hanare(x,y);

[0088] S27, the luminance of the hand area is calculated using the following formula (hanbri). k :

[0089] hanbri k =(1+δ) avepixval

[0090] Where δ represents the brightness range control parameter, δ>0.02.

[0091] In this invention, the brightness of the hand area is determined by first acquiring the imgfir image. k The hand region in the image is used to calculate the average grayscale value. This is because imgfir... k This is not the target image from the previous control cycle, therefore, imgfir k The hand area is unknown, however, due to imgfir k and aimimg k-1 Since the images are from the same set used for gesture recognition, the time interval between their captures is relatively small, resulting in minimal difference in the position of the hand region between the two images. Therefore, this invention utilizes this relationship to estimate the imgfir image. kThe hand area in the image is selected, thus avoiding the need to use a full image recognition algorithm to obtain the imgfir image. k The hand area is effectively improved, thus increasing the efficiency of obtaining the hand area.

[0092] The expansion factor is mainly used to expand the range of the original minimum bounding rectangle. The expansion factor is related to imgfir. k and aimimg k-1 The magnification factor is related to the difference in shooting time. The greater the difference in shooting time, the greater the magnification factor, and the larger the range of the hand area obtained, thus effectively magnifying the imgfir image. k The region containing the hand is included in the estimated hand region, thus enabling the estimation of the imgfir using the estimated hand region. k The brightness of the area where the hand is located.

[0093] The greater the difference in shooting time, the greater the probability that the user's hand position will move. However, since the current position of the user's hand cannot be known based solely on previously obtained images, the amplification factor is controlled by the difference in shooting time. This effectively avoids the problem of obtaining an estimated hand area that is too large when the difference in shooting time is small, thus reducing the processing pressure.

[0094] Specifically, the grayscale value and brightness value of the hand area show a positive correlation. Therefore, this invention sets up a function to obtain the brightness, thereby realizing the final brightness calculation.

[0095] Preferably, the range variation parameter is the minimum bounding rectangle (mibudret). k-1 The smaller of the length and brightness.

[0096] Preferably, based on the brightness of the hand area, hanbri k Calculate the set of infrared fill light brightness filbrg k ,include:

[0097] S31, calculate the brightness variation coefficient:

[0098]

[0099] Among them, posiva k Handcoll represents the brightness change coefficient during the k-th control cycle. k-2 The target image aimimg represents the target image in the (k-2)th control cycle. k-2 The set of pixels in the identified hand region, dist(handcoll k-1 ,handcoll k-2 ) indicates handcoll k-1The average coordinates of the pixels and handcoll k-2 The distance between the average coordinates of pixels in the graph, where bydist represents the preset standard distance. and They represent aimimg respectively k-2 and aimimg k-1 The shooting time is indicated by "byshotim" which represents the preset shooting duration; hanbri k-1 This represents the brightness of the hand region calculated in the (k-1)th control cycle; bybri represents the preset standard brightness; α represents the distance weight; β represents the time weight; and η represents the brightness weight.

[0100] The range of variation in the supplementary light brightness is [hanbri k -posiva k ×perloc,hanbri k +posiva k [×perloc];

[0101] S32, [hanbri k -posiva k ×perloc,hanbri k +posiva k The integer values ​​in [×perloc] are stored in the supplementary light brightness set filbrg k .

[0102] Due to imgfir k The estimated hand region location is not the actual hand region location. Therefore, if Hanbri is used directly... k As a final fill light, if the position of the hand area doesn't change significantly, accurate fill light can indeed be obtained. However, if the hand position changes, then using Hanbri... k As for fill light brightness, in real-world situations, overexposure may occur when the hand is close to the shooting device, while underexposure may occur when the hand is far away from the shooting device. Therefore, this invention first calculates the brightness variation coefficient, and then uses the brightness variation coefficient to calculate the set of fill light brightness corresponding to all possible areas where the hand might be positioned, thereby improving gstrec k This increases the probability of including images with neither overexposed nor underexposed hand areas, thereby improving the probability of correctly identifying the type of hand gesture.

[0103] The brightness variation coefficient is mainly used to assess the probability of a change in hand position. In this invention, the greater the distance between the average coordinates, the greater the difference in shooting time, and the greater the difference in brightness between the hand areas in two adjacent target images, the greater the probability of a change in hand position. The larger the brightness variation coefficient, the greater the range of variation in the supplementary lighting brightness, thereby increasing the probability of including the correct supplementary lighting brightness in the supplementary lighting brightness set.

[0104] Preferably, based on the supplementary light brightness set filbrg k Irradiate the shooting area with infrared light, including:

[0105] S41, Calculate the set filbrg k The duration of illumination for each value of supplemental light brightness;

[0106] For the supplemental light intensity d, the illumination duration of d is:

[0107]

[0108] Among them, exodur d The duration of illumination, representing the supplementary light intensity d, is mxbri(filbrg) k ) and mibri (filbrg k ) represent the maximum and minimum values ​​of the supplementary light intensity in filbrgk, respectively, and basedur represents the base illumination duration;

[0109] S42 illuminates the shooting area according to the illumination duration corresponding to each supplementary light brightness.

[0110] In this invention, the irradiation time is not the same for different supplementary light brightness levels; the closer to gstrec... k The longer the illumination time, the higher the probability of obtaining a clear image of the hand.

[0111] During the illumination process, the supplementary light needs to take the brightness of the supplementary light as the overall ambient brightness, then calculate the brightness of the infrared light to be emitted, and then conduct the illumination according to the corresponding duration.

[0112] Preferably, the shooting area is captured to obtain a set of images (gstrec) for gesture recognition in the k-th control cycle. k ,include:

[0113] S51, calculates continuous shooting interval;

[0114] S52, based on the continuous shooting interval, captures images of the shooting area using continuous shooting during the time period when infrared light is irradiated onto the shooting area, obtaining a set of images gstrec for gesture recognition in the k-th control cycle. k .

[0115] Specifically, the formula for calculating the continuous shooting interval is:

[0116]

[0117] Among them, ctnsht k This indicates the burst interval for the kth control cycle, tpbri indicates the maximum preset brightness of the hand area, and timbs indicates the preset standard value for the burst interval.

[0118] In this invention, the greater the brightness of the hand area, the smaller the continuous shooting interval, and vice versa. This can effectively increase the probability of obtaining an image with sufficient light intake when the brightness is low, thereby increasing the probability of obtaining a high-quality image when the brightness of the area where the hand is located is low.

[0119] Preferably, from gstrec k Select target image aimimg from the middle k ,include:

[0120] Calculate gstrec separately k The quality score of each image in the dataset;

[0121] gstrec k The image with the highest quality score is used as the target image (aimimg). k .

[0122] The calculation method for the quality score is existing technology. For example, patents with publication numbers CN115830002A and CN112036277A disclose the relevant calculation method, which will not be repeated here.

[0123] Filtering by quality score allows gstrec to... k The highest quality images were used for gesture recognition.

[0124] Preferably, the types of gestures include extending one finger, extending two fingers, extending three fingers, extending four fingers, opening the palm, and making a fist.

[0125] Preferably, controlling the projector based on the type of gesture includes:

[0126] Get the control action corresponding to the type of gesture;

[0127] The projector is controlled based on control actions.

[0128] Specifically, different gesture types can correspond to different control actions. For example, extending one finger can be set as the control action to switch to the next slide. This can be configured as needed by those skilled in the art.

[0129] On the other hand, such as Figure 2 As shown in one embodiment, the present invention provides a projector control system, characterized in that it includes an acquisition module, an identification module, a calculation module, an illumination module, an imaging module, a screening module, and a control module;

[0130] The acquisition module is used to acquire the image (imgfir) for identifying the brightness of the hand region during the k-th control cycle. k ;

[0131] The recognition module is used for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ;

[0132] The calculation module is used for hanbri based on the brightness of the hand area. k Calculate the set of infrared fill light brightness filbrg k ;

[0133] The illumination module is used for illumination based on the supplementary light brightness set filbrg k Irradiate infrared light onto the shooting area;

[0134] The imaging module is used to capture images of the imaging area, obtaining a set of images (gstrec) for gesture recognition in the k-th control cycle. k ;

[0135] The filtering module is used to filter from gstrec k Select target image aimimg from the middle k , for aimimg k Perform gesture recognition to obtain the type of gesture;

[0136] The control module is used to control the projector based on the type of gesture.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A projector control method, characterized in that, include: S1, acquire the image imgfir used for identifying the brightness of the hand region during the k-th control cycle. k ; S2, for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ; S3, based on the brightness of the hand area (hanbri) k Calculate the set of infrared fill light brightness filbrg k ; S4, based on the supplementary light brightness set filbrg k Irradiate infrared light onto the shooting area; S5, capture images of the shooting area to obtain the set of images gstrec for gesture recognition in the k-th control cycle. k ; S6, from gstrec k Select target image aimimg from the middle k , for aimimg k Perform gesture recognition to obtain the type of gesture; S7 controls the projector based on the type of gesture; Obtain the image (imgfir) used for identifying the brightness of the hand region during the k-th control cycle. k ,include: S11, Obtain the set of images gstrec used for gesture recognition in the (k-1)th control cycle. k-1 ; S12, gstrec k-1 The image captured latest in the middle is used as the image for identifying the brightness of the hand region in the k-th control cycle. k ; For imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ,include: S21, acquire the target image aimimg in the (k-1)th control cycle. k-1 The set of pixels in the identified hand region, handcoll k-1 ; S22, in aimimg k-1 Get handcoll k-1 The minimum bounding rectangle of the mibudret k-1 ; S23, Obtain mibudret k-1 The distribution range of the x-coordinates of the pixels in [x] k-1,min ,x k-1,max ], x k-1,min and x k-1,max Representing the minimum and maximum values ​​of the distribution interval of the x-axis, respectively, to obtain mibudret k-1 The distribution range of the y-coordinate of the pixels in [y] k-1,min ,y k-1,max ], y k-1,min and y k-1,max These represent the minimum and maximum values ​​of the distribution range of the horizontal axis, respectively; S24, calculate the expansion coefficient for the k-th control cycle; S25, estimated imgfir k The hand region hanare(x,y) in the image; S26, in imgfir k Calculate the average gray value avepixval of the pixels located in the hand region hanare(x,y); S27, the luminance of the hand area is calculated using the following formula (hanbri). k : hanbri k =(1+δ) avepixval Where δ represents the brightness range control parameter, δ>0.02; Hanbri based on the brightness of the hand area k Calculate the set of infrared fill light brightness filbrg k ,include: S31, calculate the brightness variation coefficient: Among them, posiva k Handcoll represents the brightness change coefficient during the k-th control cycle. k-2 The target image aimimg represents the target image in the (k-2)th control cycle. k-2 The set of pixels in the identified hand region, dist(handcoll k-1 ,handcoll k-2 ) indicates handcoll k-1 The average coordinates of the pixels and handcoll k-2 The distance between the average coordinates of pixels in the graph, where bydist represents the preset standard distance. and They represent aimimg respectively k-2 and aimimg k-1 The shooting time is indicated by "byshotim" which represents the preset shooting duration; hanbri k-1 This represents the brightness of the hand region calculated in the (k-1)th control cycle; α represents the distance weight, β represents the time weight, and η represents the brightness weight. The range of variation in the supplementary light brightness is [hanbri k -posiva k ×perloc,hanbri k +posiva k [×perloc]; perloc represents the preset brightness value; S32, [hanbri k -posiva k ×perloc,hanbri k +posiva k The integer values ​​in [×perloc] are stored in the supplementary light brightness set filbrg k ; Based on the complement light brightness set filbrg k Irradiate the shooting area with infrared light, including: S41, Calculate the set filbrg k The duration of illumination for each value of supplemental light brightness; S42 illuminates the shooting area according to the illumination duration corresponding to each supplementary light brightness.

2. The projector control method according to claim 1, characterized in that, The shooting area is captured to obtain a set of images (gstrec) for gesture recognition in the k-th control cycle. k ,include: S51, calculates continuous shooting interval; S52, based on the continuous shooting interval, captures images of the shooting area using continuous shooting during the time period when infrared light is irradiated onto the shooting area, obtaining a set of images gstrec for gesture recognition in the k-th control cycle. k .

3. The projector control method according to claim 1, characterized in that, From gstrec k Select target image aimimg from the middle k ,include: Calculate gstrec respectively k The quality score of each image in the dataset; gstrec k The image with the highest quality score is used as the target image (aimimg). k .

4. The projector control method according to claim 1, characterized in that, The types of gestures include extending one finger, extending two fingers, extending three fingers, extending four fingers, opening the palm, and making a fist.

5. The projector control method according to claim 1, characterized in that, Projector control based on gesture type includes: Get the control action corresponding to the type of gesture; The projector is controlled based on control actions.

6. A projector control system, characterized in that, It includes an acquisition module, an identification module, a calculation module, an illumination module, an imaging module, a filtering module, and a control module; The acquisition module is used to acquire the image (imgfir) for identifying the brightness of the hand region during the k-th control cycle. k ; The recognition module is used for imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ; The calculation module is used for hanbri based on the brightness of the hand area. k Calculate the set of infrared fill light brightness filbrg k ; The illumination module is used for illumination based on the supplementary light brightness set filbrg k Irradiate infrared light onto the shooting area; The imaging module is used to capture images of the imaging area, obtaining a set of images (gstrec) for gesture recognition in the k-th control cycle. k ; The filtering module is used to filter from gstrec k Select target image aimimg from the middle k , for aimimg k Perform gesture recognition to obtain the type of gesture; The control module is used to control the projector based on the type of gesture; Obtain the image (imgfir) used for identifying the brightness of the hand region during the k-th control cycle. k ,include: S11, Obtain the set of images gstrec used for gesture recognition in the (k-1)th control cycle. k-1 ; S12, gstrec k-1 The image captured latest in the middle is used as the image for identifying the brightness of the hand region in the k-th control cycle. k ; For imgfir k Perform image recognition to obtain imgfir k Brightness of the hand area in hanbri k ,include: S21, acquire the target image aimimg in the (k-1)th control cycle. k-1 The set of pixels in the identified hand region, handcoll k-1 ; S22, in aimimg k-1 Get handcoll k-1 The minimum bounding rectangle of the mibudret k-1 ; S23, Obtain mibudret k-1 The distribution range of the x-coordinates of the pixels in [x] k-1,min ,x k-1,max ], x k-1,min and x k-1,max Representing the minimum and maximum values ​​of the distribution interval of the x-axis, respectively, to obtain mibudret k-1 The distribution range of the y-coordinate of the pixels in [y] k-1,min ,y k-1,max ], y k-1,min and y k-1,max These represent the minimum and maximum values ​​of the distribution range of the horizontal axis, respectively; S24, calculate the expansion coefficient for the k-th control cycle; S25, estimated imgfir k The hand region hanare(x,y) in the image; S26, in imgfir k Calculate the average gray value avepixval of the pixels located in the hand region hanare(x,y); S27, the luminance of the hand area is calculated using the following formula (hanbri). k : hanbri k =(1+δ) avepixval Where δ represents the brightness range control parameter, δ>0.02; Hanbri based on the brightness of the hand area k Calculate the set of infrared fill light brightness filbrg k ,include: S31, calculate the brightness variation coefficient: Among them, posiva k Handcoll represents the brightness change coefficient during the k-th control cycle. k-2 The target image aimimg represents the target image in the (k-2)th control cycle. k-2 The set of pixels in the identified hand region, dist(handcoll k-1 ,handcoll k-2 ) indicates handcoll k-1 The average coordinates of the pixels and handcoll k-2 The distance between the average coordinates of pixels in the graph, where bydist represents the preset standard distance. and They represent aimimg respectively k-2 and aimimg k-1 The shooting time is indicated by "byshotim" which represents the preset shooting duration; hanbri k-1 This represents the brightness of the hand region calculated in the (k-1)th control cycle; α represents the distance weight, β represents the time weight, and η represents the brightness weight. The range of variation in the supplementary light brightness is [hanbri k -posiva k ×perloc,hanbri k +posiva k [×perloc]; perloc represents the preset brightness value; S32, [hanbri k -posiva k ×perloc,hanbri k +posiva k The integer values ​​in [×perloc] are stored in the supplementary light brightness set filbrg k ; Based on the complement light brightness set filbrg k Irradiate the shooting area with infrared light, including: S41, Calculate the set filbrg k The duration of illumination for each value of supplementary light brightness; S42 illuminates the shooting area according to the illumination duration corresponding to each supplementary light brightness.

Images