Projection equipment and projection control method

Abstract

The invention provides projection equipment and a projection control method, and the method comprises the steps: obtaining a detection image in response to a projection instruction, the detection image is image data collected when the camera is controlled to project the projection content on the light emitting assembly in the first projection scene or distance dot matrix image data collected when the distance sensor is controlled to project the projection content on the light emitting assembly in the second projection scene. And identifying a target mask in the detection image based on a mask model, wherein the mask model comprises an encoder and a decoder for identifying a target contour based on pixel points. And adding a first parameter value or a second parameter value in the detection queue according to whether the target mask has an intersection with the projection area in the detection image, and finally setting the projection brightness of the light emitting assembly according to the detection queue. According to the method, image data in different scenes are acquired through coordination and cooperation of double sensors, target pixel segmentation is performed on the image data in a coding and decoding mode, and meanwhile, the problem of false detection is relieved in a queue detection mode.

Description

Technical Field

[0001] This application relates to the technical field of projection devices, and in particular, to a projection device and a projection control method. Background Art

[0002] A projection device is a display device that can project images or videos onto a screen. The projection device can project laser light of a specific color through the refraction of an optical lens assembly into the projection area to form a specific image. Based on the portability of the projection device, users can move it during the projection process to project images or videos in different directions.

[0003] When the projection device projects media data, to prevent the high-power projection light source from injuring the user's eyes, the projection device can perform target human detection on the projection area. When a target human appears in the projection area, the light source is automatically turned off or the projection brightness of the light source is reduced in a timely manner.

[0004] When detecting a target human, the projection device collects data on the projection area through a camera or a time-of-flight sensor, and then determines whether there is a target human in the projection area through image processing or a trained target detection neural network based on a preset threshold. However, the method based on threshold determination has detection errors, and the data collected by the camera or the time-of-flight sensor may also be incorrect in some special scenarios. For example, in poor ambient light conditions, it is difficult for the camera to capture the image features of the human target. When a human target at a long distance approaches the projection device, the time-of-flight sensor cannot detect the human target, etc., resulting in a low detection accuracy of the target human, which may cause problems such as non-human target interference objects triggering the operation of adjusting the projection brightness, and the operation of adjusting the projection brightness not being triggered when part of the body of the human target invades the projection area. Summary of the Invention

[0005] This application provides a projection device and a projection control method to solve the problem of low detection accuracy of the intrusion of a target human between the projection device and the projection surface.

[0006] In a first aspect, this application provides a projection device, including a light-emitting component, a target detection component, and a controller. Among them, the light-emitting component is configured to project projection content onto a projection surface. The target detection component includes a distance sensor and a camera. The camera is configured to collect image data of the projection surface according to a first acquisition period; the distance sensor is configured to collect distance dot matrix image data of the projection surface according to a second acquisition period. The controller is configured to execute the following program steps:

[0007] In response to a projection instruction, obtain a detection image, where the detection image is image data collected by controlling the camera when the light-emitting component projects projection content in a first projection scenario, or the detection image is distance dot matrix image data collected by controlling the distance sensor when the light-emitting component projects projection content in a second projection scenario;

[0008] Identify a target mask in the detection image based on a mask model, where the mask model includes an encoder and a decoder for identifying a target contour based on pixel points;

[0009] When there is an intersection between the target mask and the projection area in the detection image, add a first parameter value to a detection queue, and set the projection brightness of the light-emitting component according to the detection queue;

[0010] When there is no intersection between the target mask and the projection area in the detection image, add a second parameter value to the detection queue, and set the projection brightness of the light-emitting component according to the detection queue.

[0011] In a second aspect, the present application further provides a projection control method, which is applied to the above projection device, and the method includes:

[0012] In response to a projection instruction, obtain a detection image, where the detection image is image data collected by controlling the camera when the light-emitting component projects projection content in a first projection scenario, or the detection image is distance dot matrix image data collected by controlling the distance sensor when the light-emitting component projects projection content in a second projection scenario;

[0013] Identify a target mask in the detection image based on a mask model, where the mask model includes an encoder and a decoder for identifying a target contour based on pixel points;

[0014] When there is an intersection between the target mask and the projection area in the detection image, add a first parameter value to a detection queue, and set the projection brightness of the light-emitting component according to the detection queue;

[0015] When there is no intersection between the target mask and the projection area in the detection image, add a second parameter value to the detection queue, and set the projection brightness of the light-emitting component according to the detection queue.

[0016] As can be seen from the above technical solutions, the present application provides a projection device and a projection control method. The method can respond to a projection instruction to obtain a detection image, where the detection image is image data collected by controlling a camera when the light emitting component projects projection content in a first projection scenario or distance dot matrix image data collected by controlling a distance sensor when the light emitting component projects projection content in a second projection scenario. Based on a mask model, the target mask in the detection image is identified, where the mask model includes an encoder and a decoder for identifying the target contour based on pixel points. When there is an intersection between the target mask and the projection area in the detection image, a first parameter value is added to the detection queue; when there is no intersection between the target mask and the projection area in the detection image, a second parameter value is added to the detection queue. Finally, the projection brightness of the light emitting component is set according to the detection queue. The method coordinates the cooperation of a camera and a distance sensor, and solves the problem that the anti-eye projection function does not work in special scenarios. And it uses encoding and decoding methods to perform pixel segmentation of the target on the image data, improving the sensitivity of human recognition. At the same time, it also adopts the method of state machine transition to alleviate the problem of false detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 Schematic diagram of the projection state of the projection device in the embodiment of the present application;

[0019] Figure 2 Schematic diagram of the structure of the projection device in the embodiment of the present application;

[0020] Figure 3 Schematic diagram of the optical machine architecture of the projection device in the embodiment of the present application;

[0021] Figure 4 Schematic diagram of the optical path of the projection device in the embodiment of the present application;

[0022] Figure 5 Schematic diagram of the lens structure of the projection device in the embodiment of the present application;

[0023] Figure 6 Schematic diagram of the distance sensor and the camera in the embodiment of the present application;

[0024] Figure 7 Schematic diagram of the system framework of the projection device in the embodiment of the present application;

[0025] Figure 8It is a flowchart of the projection control method in the embodiments of the present application;

[0026] Figure 9 It is a schematic diagram of the coordinates of the projection area in the sampled image in the embodiments of the present application;

[0027] Figure 10 It is a flowchart of performing human target intrusion detection based on image data or distance dot matrix image data in the embodiments of the present application;

[0028] Figure 11 It is a schematic diagram of the structure of the first mask model in the embodiments of the present application;

[0029] Figure 12 It is a schematic diagram of the structure of the convolution operation module in the embodiments of the present application;

[0030] Figure 13 It is a schematic diagram of the structure of the second mask model in the embodiments of the present application;

[0031] Figure 14 It is a schematic diagram of the states and state transitions of the state machine in the embodiments of the present application. Detailed implementation manners

[0032] To make the purpose and implementation manners of the present application clearer, the following will clearly and completely describe the exemplary implementation manners of the present application in conjunction with the drawings in the exemplary embodiments of the present application. Obviously, the described exemplary embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0033] It should be noted that the brief description of the terms in the present application is only for the convenience of understanding the subsequent described implementation manners, rather than intending to limit the implementation manners of the present application. Unless otherwise specified, these terms should be understood in their ordinary and common meanings.

[0034] The terms "first", "second", "third", etc. in the specification, claims and the above drawings of the present application are used to distinguish similar or like objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that such terms can be interchanged under appropriate circumstances.

[0035] The terms "comprising" and "having" and any variations thereof are intended to cover but not exclude inclusion. For example, a product or device including a series of components does not necessarily have to be limited to all the clearly listed components, but may include other components that are not clearly listed or are inherent to these products or devices.

[0036] The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic or a combination of hardware or / and software code that can perform functions related to this element.

[0037] The embodiments of the present application can be applied to various types of projection devices. In the following, taking a projector as an example, the projection device and the autofocus method will be described.

[0038] A projector is a device that can project images or videos onto a screen. The projector can be connected to a computer, a radio and television network, the Internet, a VCD (Video Compact Disc), a DVD (Digital Versatile Disc Recordable), a game console, a DV, etc. through different interfaces to play corresponding video signals. Projectors are widely used in homes, offices, schools, entertainment venues, etc.

[0039] Figure 1 The placement schematic diagram of the projection device according to an embodiment of the present application is shown. Figure 2 The optical path schematic diagram of the projection device according to an embodiment of the present application is shown.

[0040] In some embodiments, referring to Figure 1-2 , a projection device provided by the present application includes a projection screen and a projection device. The projection screen is fixed at a first position, and the projection device is placed at a second position so that the projected image coincides with the projection screen. The projection device includes a laser light source 100, an optical engine 200, a lens 300, and a projection surface 400. Among them, the laser light source 100 provides illumination for the optical engine 200. The optical engine 200 modulates the light beam of the light source and outputs it to the lens 300 for imaging, and projects it onto the projection surface 400 to form a projection image. Since the laser light source 100, the optical engine 200, and the lens 300 are jointly used to emit projection light to project the projection image, in some embodiments of the present application, the laser light source 100, the optical engine 200, and the lens 300 are collectively referred to as the light-emitting components.

[0041] In some embodiments, the laser light source 100 of the projection device includes a laser component and an optical lens component. The light beam emitted by the laser component can pass through the optical lens component and then provide illumination for the optical engine. Among them, for example, the optical lens component requires a high level of environmental cleanliness and airtightness for sealing; while the chamber for installing the laser component can be sealed with a lower dust-proof level of airtightness to reduce the sealing cost.

[0042] In some embodiments, the optical engine 200 of the projection device can be implemented to include a blue optical engine, a green optical engine, a red optical engine, and can also include a heat dissipation system, a circuit control system, etc. It should be noted that in some embodiments, the light-emitting components of the projector can also be realized by LED light sources.

[0043] Figure 3The schematic diagram of the circuit architecture of a projection device according to an embodiment of the present application is shown. In some embodiments, the projection device may include a display control circuit 10, a laser light source 100, at least one laser driver component 30, and at least one brightness sensor 40. The laser light source 100 may include at least one laser corresponding one-to-one to the at least one laser driver component 30. Herein, the at least one means one or more, and the plurality means two or more.

[0044] Based on this circuit architecture, the projection device can achieve adaptive adjustment. For example, by setting the brightness sensor 40 in the light output path of the laser light source 100, the brightness sensor 40 can detect the first brightness value of the laser light source and send the first brightness value to the display control circuit 10.

[0045] The display control circuit 10 can obtain the second brightness value corresponding to the driving current of each laser, and when determining that the difference between the second brightness value of the laser and the first brightness value of the laser is greater than the difference threshold, determine that the laser has a COD fault; then the display control circuit can adjust the current control signal of the corresponding laser driver component of the laser until the difference is less than or equal to the difference threshold, thereby eliminating the COD fault of the blue laser; the projection device can timely eliminate the COD fault of the laser, reduce the damage rate of the laser, and improve the image display effect of the projection device.

[0046] Figure 4 The schematic diagram of the structure of a projection device according to an embodiment of the present application is shown.

[0047] In some embodiments, the laser light source 100 in the projection device may include an independently arranged blue laser 101, a red laser 102, and a green laser 103. The projection device may also be referred to as a three-color projection device. The blue laser 101, the red laser 102, and the green laser 103 are all Mirai Console Loader (MCL) packaged lasers, which are small in volume and conducive to the compact layout of the optical path.

[0048] In some embodiments, the controller 500 of the projection device includes at least one of a central processing unit (CPU), a video processor, an audio processor, a graphics processing unit (GPU), a RAM (Random Access Memory), a ROM (Read-Only Memory), a first interface to an nth interface for input / output, a communication bus, etc.

[0049] In some embodiments, the projection device may be configured with a camera for operating in cooperation with the projection device to achieve adjustment and control of the projection process. For example, the camera configured by the projection device may be specifically implemented as a 3D camera or a binocular camera. When the camera is implemented as a binocular camera, it specifically includes a left camera and a right camera. The binocular camera can acquire the image and playback content presented on the screen corresponding to the projection device, that is, the projection surface, and the image or playback content is projected by the optical engine built into the projection device.

[0050] When the projection device moves its position, its projection angle and the distance to the projection surface change, which will cause the projection image to be deformed, and the projection image will be displayed as a trapezoidal image or other deformed images. The controller 500 of the projection device can achieve automatic trapezoid correction based on the image captured by the camera by coupling the angle between the optical engine and the projection surface and the correct display of the projection image.

[0051] Among them, the camera can be used to capture the image displayed on the projection surface and can be a camera. The camera can include a lens assembly, and a photosensitive element and a lens are provided in the lens assembly. Through the refraction of light by multiple lenses in the lens assembly, the light of the image of the scene can be irradiated on the photosensitive element. The photosensitive element can select the detection principle based on charge-coupled device or complementary metal-oxide semiconductor according to the specifications of the camera, convert the optical signal into an electrical signal through the photosensitive material, and output the converted electrical signal as image data.

[0052] Figure 5 The schematic diagram of the lens structure of the projection device in some embodiments is shown. To support the automatic focusing process of the projection device, as Figure 5 shown, the lens 300 of the projection device may further include an optical component 310 and a driving motor 320. Among them, the optical component 310 is a lens group composed of one or more lenses, which can refract the light emitted by the optical engine 200, so that the light emitted by the optical engine 200 can be transmitted onto the projection surface to form a transmitted content image.

[0053] The optical component 310 may include a lens barrel and a plurality of lenses provided in the lens barrel. According to whether the position of the lens can move, the lenses in the optical component 310 can be divided into movable lenses 311 and fixed lenses 312. By changing the position of the movable lens 311, the distance between the movable lens 311 and the fixed lens 312 is adjusted, and the overall focal length of the optical component 310 is changed. Therefore, the driving motor 320 can drive the movable lens 311 to move its position by connecting the movable lens 311 in the optical component 310 to achieve the automatic focusing function.

[0054] It should be noted that in some embodiments of the present application, the focusing process refers to changing the position of the movable lens 311 by driving the motor 320, so as to adjust the distance between the movable lens 311 and the fixed lens 312, that is, to adjust the image plane position. Therefore, according to the imaging principle of the lens combination in the optical assembly 310, the adjustment of the focal length is actually the adjustment of the image distance. However, in terms of the overall structure of the optical assembly 310, adjusting the position of the movable lens 311 is equivalent to adjusting the overall focal length of the optical assembly 310.

[0055] When the projection device is at different distances from the projection surface, the lens of the projection device needs to adjust different focal lengths to project a clear image on the projection surface. During the projection process, the distance between the projection device and the projection surface varies depending on the user's placement position, requiring different focal lengths. Therefore, to adapt to different usage scenarios, the projection device needs to adjust the focal length of the optical assembly 310.

[0056] Figure 6 Shows a schematic diagram of the distance sensor and camera structure in some embodiments. As Figure 6 shown, the projection device may also be built-in or externally connected with a camera 700. The camera 700 can take pictures of the projected image of the projection device to obtain the projection content image. The projection device then determines whether the current lens focal length is appropriate by detecting the clarity of the projected content image, and adjusts the focal length when it is inappropriate. When performing automatic focusing based on the projection content image captured by the camera 700, the projection device can continuously adjust the lens position and take pictures, and find the focusing position by comparing the clarity of the pictures at the front and rear positions, so as to adjust the movable lens 311 in the optical assembly to an appropriate position. For example, the controller 500 can first control the driving motor 320 to gradually move the movable lens 311 from the focusing starting position to the focusing ending position, and continuously obtain the projection content image through the camera 700 during this period. Then, by detecting the clarity of multiple projection content images, determine the position with the highest clarity. Finally, control the driving motor 320 to adjust the movable lens 311 from the focusing terminal to the position with the highest clarity to complete automatic focusing.

[0057] The distance sensor 600 can be a sensor device based on the Time of Flight (TOF) principle, such as a lidar or an infrared radar that can detect the target distance. The distance sensor 600 can be set at the position of the optical machine 200, including a signal transmitting end and a receiving end. During the process of detecting the interval distance, the transmitting end of the distance sensor 600 can emit a wireless signal in the direction of the projection surface. After the wireless signal contacts the projection surface, it will be reflected back to the receiving end of the distance sensor 600. Thus, according to the time when the signal is emitted by the transmitting end and the time when the signal is received by the receiving end, calculate the signal flight time, and then combine the flight speed to obtain the actual flight distance of the wireless signal.

[0058] Figure 7 The figure shows a schematic diagram of the system framework for a projection device to achieve display control according to an embodiment of the present application.

[0059] In some embodiments, the projection device has the characteristics of long - focus micro - projection. Its controller 500 can perform display control on the projected light image through a preset algorithm to achieve functions such as automatic trapezoid correction of the display screen, automatic screen entry, automatic obstacle avoidance, automatic focusing, and anti - eye - radiation.

[0060] In some embodiments, the projection device is configured with a gyroscope sensor. During the movement of the device, the gyroscope sensor can sense the position movement and actively collect movement data. Then, the collected data is sent to the application service layer through the system framework layer to support the application data required during user interface interaction and application program interaction. The collected data can also be used for data calls in the algorithm service implementation of the controller 500.

[0061] In some embodiments, the projection device is configured with a time - of - flight sensor. After the time - of - flight sensor collects the corresponding data, the data will be sent to the corresponding time - of - flight service in the service layer. After the above - mentioned time - of - flight service obtains the data, the collected data is sent to the application service layer through the process communication framework, and the data will be used for data calls of the controller 500, user interface, program application, etc. for interaction.

[0062] In some embodiments, the camera 700 configured in the projection device can be a binocular camera, a depth camera, or a 3D camera, etc. The data collected by the camera 700 will be sent to the camera service, and then the camera service will send the collected image data to the process communication framework and / or the projection device calibration service. The projection device calibration service can receive the camera - collected data sent by the camera service, and the controller 500 can call the corresponding control algorithm in the algorithm library for different functions to be achieved.

[0063] In some embodiments, data interaction is performed with the application service through the process communication framework, and then the calculation result is fed back to the calibration service through the process communication framework. The calibration service sends the obtained calculation result to the projection device operating system to generate a control signal, and sends the control signal to the optical engine 200 to control the operating condition of the optical engine 200 and achieve automatic correction of the displayed image.

[0064] In some embodiments, when an image correction instruction is detected, the projection device can correct the projected image. For the correction of the projected image, the correlation between the distance, the horizontal included angle, and the offset angle can be created in advance. Then, the controller 500 in the projection device determines the included angle between the optical engine 200 and the projection surface at that moment by obtaining the current distance from the optical engine 200 to the projection surface and combining the associated relationship, so as to realize the correction of the projected image. Among them, the included angle is specifically implemented as the included angle between the central axis of the optical engine 200 and the projection surface.

[0065] In some embodiments, after the projection device automatically completes the correction and refocuses, the controller 500 will detect whether the autofocus function is enabled; when the autofocus function is not enabled, the controller 500 will end the autofocus service; when the autofocus function is enabled, the projection device will calculate by obtaining the detection distance of the time-of-flight sensor through the middleware.

[0066] The controller 500 queries a preset mapping table according to the obtained distance to obtain the focal length of the projection device; then the middleware sets the obtained focal length to the optical engine 200 of the projection device; among them, the middleware is a series of application programs for the focusing control process. After the optical engine 200 emits laser light with the above-mentioned focal length, the camera will execute the photographing instruction; the controller 500 determines whether the focusing process of the projection device is completed according to the obtained captured image and the evaluation function.

[0067] If the determination result meets the preset completion condition, the control autofocus process will end; if the determination result does not meet the preset completion condition, the middleware will finely adjust the focal length parameter of the optical engine 200 of the projection device. For example, the focal length can be gradually finely adjusted with a preset step size, and the adjusted focal length parameter is set to the optical engine 200 again; thus, the steps of repeatedly taking pictures and evaluating the clarity are realized, and finally the optimal focal length is found through the clarity comparison to complete the autofocus.

[0068] In some embodiments, in order to prevent the high-power projection light source from injuring the user's eyes, the projection device further includes a target detection component for detecting a human target between the projection device and the projection surface 400. The projection device can determine whether there is a human target intrusion between the projection device and the projection surface 400 according to the data detected by the target detection component. When there is a human target intrusion, the projection device is controlled to turn on the anti-eye-shooting function, and when there is no human target intrusion, the projection device is controlled to turn off the anti-eye-shooting function. Among them, the anti-eye-shooting function refers to the operation of triggering the projection device to automatically adjust the projection brightness of the light-emitting component, including at least one of reducing the projection brightness of the light-emitting component and adjusting the projection brightness of the light-emitting component to a preset projection brightness.

[0069] In some embodiments, the target detection component may be the camera 700 in the above embodiments, which is used to collect image data facing the projection surface 400. The controller 500 of the projection device may perform human target detection based on the image data collected by the camera 700 to determine whether there is a human target intrusion between the projection device and the projection surface 400. The target detection component may also be the distance sensor 600, such as a Time Of Flight (TOF) sensor, which is used to collect distance dot matrix image data facing the projection surface 400. The controller 500 of the projection device may perform human target detection based on the distance dot matrix image data collected by the distance sensor 600 to determine whether there is a human target intrusion between the projection device and the projection surface 400.

[0070] However, the detection effect of the camera 700 or the distance sensor 600 is poor in some scenarios. For example, in the case of poor ambient light conditions, the ambient light around the projection device is dim, and it is difficult for the camera 700 to capture the image features of the human target. When a human target at a long distance approaches the projection device, since the position of the human target is not within the detection range of the distance sensor 600, the human target cannot be detected.

[0071] In some embodiments, after receiving the data detected by the target detection component, the projection device may determine whether there is a target human in the projection area through an image processing method or a trained target detection neural network, and based on a preset threshold. However, the method of determining based on the threshold has detection errors, which may result in the situation where there is no human target but it is determined that a human target is detected, or there is a human target but it is determined that no human target is detected, thus causing the problem of false triggering of the anti-eye irradiation function.

[0072] Moreover, both the image processing method and the trained target detection neural network detect the overall features of the human target, which may cause interference objects that are not human targets to trigger the anti-eye irradiation function to be turned on, or the anti-eye irradiation function cannot be triggered when only part of the body of the human target intrudes into the projection area.

[0073] Therefore, to solve the above problems, some embodiments of the present application provide a projection control method, which can be applied to the projection device provided in the above embodiments. To implement the projection control method, the projection device should at least include a light-emitting component, a target detection component, and a controller 500. Among them, the light-emitting component is configured to project projection content onto the projection surface, and the projection surface may be a wall or a screen. The target detection component includes a distance sensor 600 and a camera 700. The camera 700 is configured to collect image data of the projection surface 400; the distance sensor 600 is configured to collect distance dot matrix image data of the projection surface 400. The controller 500 is configured to execute the program steps corresponding to the projection control method, such asFigure 8 As shown in the figure, it is a schematic flowchart of the projection control method provided by the embodiment of the present application, which specifically includes the following content:

[0074] S100: In response to a projection instruction, obtain a detection image.

[0075] When the projection device is running, the user can input a projection instruction in various ways such as a remote control, voice instruction, control panel, gesture, etc. to control the projection device to project projection content onto the projection surface 400 to form a projection image. To prevent the high-power projection light source from injuring the user's eyes, the projection device can, in response to the projection instruction, periodically control the target detection component to perform human target detection on the projection area, and when a human target appears in the projection area, automatically turn off the light source or reduce the projection brightness of the light source in a timely manner.

[0076] It should be noted that in the embodiment of the present application, the technical solution of the present application is described by taking a human target as an example. It should be understood that the technical solution provided by the embodiment of the present application is not limited to human targets, and other targets can also be detected according to actual needs.

[0077] In order to adapt to and improve the accuracy of data collection in different projection scenarios, the target detection component includes a distance sensor 600 and a camera 700. The projection device can coordinate the distance sensor 600 and the camera 700 in different projection scenarios. That is, the projection device, in response to a projection instruction, detects the current projection scenario. If the current projection scenario is the first projection scenario, control the camera 700 to collect image data of the projection surface 400 according to the first collection period. If the current projection scenario is the second projection scenario, control the distance sensor 600 to collect distance dot matrix image data of the projection surface 400 according to the second collection period.

[0078] That is to say, if the current projection scenario is the first projection scenario, obtain the image data captured by the camera 700 from the camera 700 thread stream for subsequent processing. Among them, the image data captured by the camera 700 can be RGB image data, but it is not limited to other data formats such as YUV, nor is it limited to the resolution size. If the current projection scenario is the second projection scenario, obtain the distance dot matrix image data captured by the distance sensor 600 from the distance sensor 600 thread stream for subsequent processing. Among them, the distance dot matrix image data obtained by the distance sensor 600 can be a grayscale image with a color value in the range of [0 - 255], but it is not limited to other data formats, nor is it limited to its resolution size.

[0079] Among them, the first projection scenario can be a scenario when the ambient light around the projection device is relatively high, and the second projection scenario can be a scenario when the ambient light around the projection device is relatively low. That is, when the projection device is in a scenario with relatively high ambient light, human target detection is performed based on the camera 700. When the projection device is in a scenario with relatively low ambient light, human target detection is performed based on the distance sensor 600. Through the coordinated cooperation of the camera 700 and the distance sensor 600, the problem that the detection effect of the camera 700 or the distance sensor 600 is poor in special projection scenarios is solved. In the daytime or in the presence of light sources, regardless of the distance between the user and the projection device, whether the user is close to the projection surface or not, as long as the user is within the projection area, the anti-eye irradiation function can be accurately triggered. At the same time, the anti-eye irradiation function can also be accurately triggered under poor ambient light conditions.

[0080] It should be noted that the embodiments of the present application do not specifically limit the first acquisition period and the second acquisition period, and the first acquisition period and the second acquisition period can be the same. For example: both the first acquisition period and the second acquisition period are 3s, that is, the camera 700 performs image data acquisition every 3s, and the distance sensor 600 performs distance dot matrix image data acquisition every 3s. The first acquisition period and the second acquisition period can be different. For example: the first acquisition period is 3s, and the second acquisition period is 4s, that is, the camera 700 performs image data acquisition every 3s, and the distance sensor 600 performs distance dot matrix image data acquisition every 4s.

[0081] To distinguish different projection scenarios, as Figure 10 shown, the projection device can perform image shooting based on the camera 700 when the light output component projects the projection content to obtain a sampled image. Then, a non-projection area that does not include the projection content is extracted from the sampled image, the color attribute value of the non-projection area is detected, and the current projection scenario is marked as the first projection scenario or the second projection scenario based on the color attribute value. It can be understood that the sampled image includes the projection screen content projected by the projection device and the image of the environment around the projection device. According to the image color distribution information outside the projection screen content in the sampled image, the brightness of the environment around the projection device can be detected, so as to determine the current projection scenario.

[0082] To facilitate the analysis of the image color distribution in the sampled image, the projection device can convert the image format of the sampled image to a target image format, where the color attribute values in the target image format include hue, saturation, and brightness. After converting the image format, the saturation and brightness of the non-projection area in the sampled image can be obtained. If the saturation is within the first threshold interval and the brightness is within the second threshold interval, the current projection scenario is marked as the second projection scenario. If the saturation is not within the first threshold interval, or the brightness is not within the second threshold interval, the current projection scenario is marked as the first projection scenario.

[0083] For example, the projection device can obtain a frame of RGB - formatted image data, i.e., the sampled image, from the thread stream of the camera 700. Calculate the coordinates of the projection area in the sampled image according to the calibrated parameters between the projection device and the camera 700, such as Figure 9 the coordinates of the four corner points (coordinate 1, coordinate 2, coordinate 3, coordinate 4) of the projection area shown. Among them, the coordinate positions are values that have been calculated during the production calibration of the projection device and are stored in the calibration information of the projection device, and can be directly obtained.

[0084] Traverse the pixel points in the sampled image, and determine whether the pixel points are within the projection area coordinates. If not, convert the RGB values of the pixel points to HSV values. Specifically, it can be calculated based on the following formula:

[0085] Let R, G, and B be the three - channel information of the sampled image, representing the brightness values of red, green, and blue respectively, and normalize them to the range of [0 - 1].

[0086] Calculate the maximum value max among the RGB three - channel values, and the minimum value min among the RGB three - channel values. Then, calculate the HSV three - channel values according to the RGB three - channel values:

[0087]

[0088]

[0089] v = max

[0090] Thus, the hue H, saturation S, and value V are obtained. Then, determine the projection scene according to the preset threshold interval. For example, when the saturation S is less than 35 and the value V is less than 20, it indicates that the ambient brightness around the projection device is low, and then mark the current projection scene as the second projection scene. Otherwise, it indicates that the ambient brightness around the projection device is high, and then mark the current projection scene as the first projection scene.

[0091] It should be noted that the method for determining the projection scene is not limited to the above method, and other methods of color - gamut conversion or directly using RGB image information for processing can also be used to judge the environmental scene.

[0092] S200: Identify the target mask in the detection image based on the mask model.

[0093] After the projection device obtains the detection image, it can input the detection image into the mask model for target detection. Among them, in the first projection scenario, the detection image is the image data collected by the camera 700 when the light-emitting component projects the projection content. In the second projection scenario, the detection image is the distance dot matrix image data collected by the distance sensor 600 when the light-emitting component projects the projection content. The mask model is an encoder and a decoder that recognize the target contour based on pixel points. The encoder and decoder refer to an image processing process, that is, the original image is encoded into an array of vectors with abstract features through the encoder, and then the vector array is decoded into an image in another form through the decoder.

[0094] In the embodiments of the present application, the input of the encoder and decoder is the image data obtained based on the camera 700 or the distance dot matrix image data obtained based on the distance sensor 600, and the output is the mask image of the target. That is, the projection device inputs the detection image into the mask model to identify the target contour in the detection image based on the mask model, so as to obtain the mask image output by the mask model. Among them, the mask image includes a first pixel value and a second pixel value. The first pixel value is used to represent the target, and the second pixel value is used to represent non-targets. The area of the first pixel value is extracted in the mask image to obtain the target mask. For example. The mask image can be a grayscale image marking the human target. The image matrix of this grayscale image includes two kinds of data 0 / 1 or 0 / 255. 0 represents non-human targets (background), and 1 or 255 represents human targets. The area of data 1 or 255 is the target mask.

[0095] The encoder and decoder can identify and segment the target mask. The advantage of the above segmentation method compared with methods such as detecting the target according to the color gamut, passing through the target detection convolutional neural network or other interference detection methods is that it can judge whether each pixel in the image data or the distance dot matrix image data is a human target, so that when interference objects such as chairs and clothes enter the projection area, the anti-eye radiation function will not be triggered. Only when a human target enters the projection area can the anti-eye radiation function be triggered, and parts of the human body (such as fingers, half of the face, etc.) entering the projection area can also be detected, so as to trigger the anti-eye radiation function in time.

[0096] It should be noted that the encoder and decoder described in the embodiments of the present application can be implemented by image feature extraction calculation methods, such as SIFT (Scale-Invariant Feature Transform), HOG (Histogram of Oriented Gradients). It can also be implemented by matrix operations of SVM (Support Vector Machine). It can also be implemented by calculation methods of convolutional neural networks or Transformers. The present application does not make specific limitations on this.

[0097] In some embodiments, an encoder and a decoder implemented based on a convolutional neural network can be used to process a detection image to obtain a target mask. As Figure 11 shown, the mask model includes a plurality of convolutional operation modules for performing decoding operations and encoding operations on an image. Figure 11 Each square in it represents a convolutional operation module. Among them, the connection manner between the convolutional operation modules is represented by arrows, including two connection manners: one connection manner is the connection of adjacent modules horizontally ( Figure 11 the horizontal arrows in it), and this connection manner means that the result of the operation of the previous convolutional operation module is directly input into the next convolutional operation module. The other connection manner is the cross-module connection ( Figure 11 the curved arrows in it), and this connection manner means the process that the result of the operation of one convolutional operation module is added to another convolutional operation module through an addition operation. It should be noted that the above connection manner between the convolutional operation modules has better encoding and decoding effects in actual algorithm tests, but it does not mean that this module connection manner must be used.

[0098] As Figure 12 shown, it is a schematic structural diagram of the convolutional operation module provided by the embodiment of the present application. The convolutional operation module can be a superposition of calculation methods such as convolutional operation, depthwise separable convolutional operation, and linear integration unit. The convolutional operation module can sequentially execute convolutional operation, linear integration unit, depthwise separable convolutional operation, linear integration unit, and convolutional operation, and the calculation process is regarded as the process of encoding or decoding the input image into abstract features. Figure 12 There is also a skip connection that connects the input image to the output, which means adding the abstract features of the input and the output.

[0099] For the convolutional operation, assume that the output image output is an array matrix of I rows and J columns, the input image input is an array matrix of M rows and N columns and the convolutional kernel kernel, then the output image is:

[0100]

[0101] For the depthwise separable convolutional operation, it is a variant of the convolutional operation, and can be added or replaced with the original convolution in the module according to actual needs.

[0102] For the linear integration unit, it means that all pixels of the intermediate operation matrix undergo a non-linear calculation, and its non-linear function is:

[0103] f(X) = max(0, x)

[0104] It should be noted that any non - linear function can replace the linear integration unit, such as the sigmoid function, the tanh function, etc. However, in the implementation effect of this embodiment, the computational complexity of the linear integration unit is relatively small and the effect is better.

[0105] In some embodiments, the process of calculating the target mask from the distance dot - matrix image data collected by the distance sensor 600 through the encoder and decoder is similar to the process of calculating the target mask from the image data collected by the camera 700. However, since the resolution of the distance dot - matrix image data collected by the distance sensor 600 is smaller than the resolution of the image data collected by the camera 700, such as a resolution of 120 * 120. To reduce resource consumption, the structures of the encoder and decoder for the distance dot - matrix image data can be relatively simple. For example, as Figure 13 shown, the structures of the encoder and decoder for the distance dot - matrix image data can include seven convolutional operation modules, that is, the distance dot - matrix image data only needs to go through seven convolutional operation modules to achieve a better encoding and decoding effect.

[0106] In this regard, the mask model preset in the projection device can include a first mask model and a second mask model. Among them, the first mask model and the second mask model include convolutional operation modules, and the convolutional operation modules are used to perform encoding operations and decoding operations on the image. The number of convolutional operation modules in the first mask model is greater than the number of convolutional operation modules in the second mask model. As Figure 10 shown, in response to a projection instruction, the projection device detects the current projection scene. If the current projection scene is the first projection scene, it acquires the image data collected by the camera 700 when the light - emitting component projects the projection content, and inputs the image data into the first mask model to calculate the target mask. If the current projection scene is the second projection scene, it acquires the distance dot - matrix image data collected by the distance sensor 600 when the light - emitting component projects the projection content, and inputs the distance dot - matrix image data into the second mask model to calculate the target mask.

[0107] S300: When there is an intersection between the target mask and the projection area in the detection image, add a first parameter value to the detection queue, and when there is no intersection between the target mask and the projection area in the detection image, add a second parameter value to the detection queue.

[0108] After obtaining the target mask, the projection device can detect whether there is an intersection between the target mask and the projection area in the detection image. If there is an intersection, it indicates that there is a target intrusion between the projection device and the projection surface 400, and the anti - eye - irradiation function can be enabled. If there is no intersection, it indicates that there is no target intrusion between the projection device and the projection surface 400, and the anti - eye - irradiation function can be disabled.

[0109] Since the user may appear briefly between the projection device and the projection surface 400, to avoid the anti-glaring function from being repeatedly turned on and off, the projection device can preset a detection queue. The detection queue refers to a container that can store or delete data and is used to store whether there is data of target intrusion in each frame of the detected image. It can be in the form of an array, a vector, etc. During the actual operation of the anti-glaring function of the projection device, the entire process continuously obtains the image data collected by the camera 700 or the distance dot matrix image data collected by the distance sensor 600, and determines whether there is a target intrusion between the projection device and the projection surface 400. If the current frame has a target intrusion, a first parameter value is added to the detection queue. If the current frame does not have a target intrusion, a second parameter value is added to the detection queue. For example, if the current frame has a human target intrusion, the value "1" is pushed into the detection queue. If the current frame does not have a human target intrusion, the value "0" is pushed into the detection queue.

[0110] That is to say, during the operation process, the intrusion results calculated based on each frame of the obtained image will be recorded by the detection queue, and the data recorded in the detection queue can be used for the subsequent determination of turning on or off the anti-glaring function.

[0111] In some embodiments, for the detection of whether there is a target intrusion between the projection device and the projection surface 400, as Figure 10 shown, the projection device can calculate the edge position of the target mask in the mask image. Among them, the edge position is the pixel position of the edge of the target mask. If the edge position is within the projection area, the step of adding a first parameter value to the detection queue is executed. If the edge position is not within the projection area, the step of adding a second parameter value to the detection queue is executed.

[0112] For the calculation of the edge position, the edge position of the target mask in the mask image can be calculated by means of image calculation, image feature extraction, or a trained neural network detection method, etc.

[0113] In some embodiments, the edge position of the target mask is calculated by means of image calculation. First, an opening operation and a closing operation are performed on the mask image to remove noise and holes within the target area. Then, the seed filling method is used to find the closed areas in the mask image. Among them, the seed filling method is to set a target pixel as a seed and spread it to the surrounding pixels to find. As long as it is also a target pixel, it is also set as a seed until no target pixel can be found among the pixels around the seed. For example, the mask image obtained through the encoder and the decoder is a grayscale image that marks the human target. The image matrix of this grayscale image includes two types of data, 0 / 1 or 0 / 255. 0 represents a non-human target (background), and 1 or 255 represents a human target. Based on the seed filling method, the pixels of 1 or 255 can be set as seeds, and the pixels around the seeds are checked. If the pixel is 1 or 255, it is also set as a seed. The above steps are continuously repeated until there are no pixels of 1 or 255 around the seed.

[0114] After using the seed filling method to find the closed areas in the mask image, all pixel positions within the closed areas are iterated to find the maximum and minimum pixel coordinates in the x and y directions, which are the edge position pixels of the target mask.

[0115] After calculating the edge position of the target mask, the projection area coordinates in the image data collected by the camera 700 or the projection area coordinates in the distance dot matrix image data collected by the distance sensor 600 can be calculated according to the parameters calibrated between the projection device and the camera 700 or the parameters calibrated between the projection device and the distance sensor 600. Among them, this coordinate position is a value that has been calculated during the production calibration of the projection device and is stored in the calibration information of the projection device and can be directly obtained. By judging whether the edge position of the target mask enters this projection area coordinate, it is judged whether the edge position invades the projection area in the image data or the projection area in the distance dot matrix image data.

[0116] S400: Control the projection brightness of the light-emitting component according to the detection queue.

[0117] During the operation, the intrusion results calculated based on each frame of image obtained will be recorded by the detection queue. The projection device can turn on or off the anti-eye irradiation function based on the data recorded in the detection queue. In order to reduce the false detection phenomenon of the anti-eye irradiation function, the projection device can control the opening or closing of the anti-eye irradiation function in the way of state machine transition. Among them, the state machine consists of two parts: state and state transition judgment, and is used to make a judgment on the final detection queue data to execute the anti-eye irradiation function in line with the user's expectations.

[0118] In this regard, the projection device can create a state machine. The states of the state machine include a no-target state, a target state, a first intermediate state, and a second intermediate state. The continuous data recorded in the detection queue is used to make a final decision for the state machine. The projection device is also preset with a state transition policy, which is a state transition judgment method set based on the state maintenance time, the number of consecutive occurrences of the first parameter value, and the number of consecutive occurrences of the second parameter value in the detection queue.

[0119] After the projection device pushes a new parameter value into the detection queue, it can transfer the state of the state machine based on the preset state transition policy and the continuous data recorded in the detection queue.

[0120] The anti-eye radiation function refers to reducing the projection brightness of the light-emitting component and adjusting the projection brightness of the light-emitting component to at least one of the preset projection brightnesses. That is, the state of the state machine can be used to indicate controlling the projection brightness of the light-emitting component. The projection device can detect the current state of the state machine. If the current state is the no-target state, set the projection brightness of the light-emitting component to the first projection brightness. If the current state is the target state, set the projection brightness of the light-emitting component to the second projection brightness. Among them, the second projection brightness is less than the first projection brightness. If the current state is the first intermediate state or the second intermediate state, control the light-emitting component to maintain the current projection brightness.

[0121] For example, as Figure 14 shown, the states of the state machine include four states: a no-target state, a target state, and two intermediate states (a first intermediate state and a second intermediate state). Among them, the no-target state indicates performing an operation to turn off the anti-eye radiation function, the target state indicates performing an operation to turn on the anti-eye radiation function, and the intermediate state indicates not performing any operation. State transitions occur by judging the data recorded in the detection queue, and any state transition will clear the detection queue and record it again.

[0122] Specifically, the detection queue records the values "1" and "0", and the continuous data of 0 or 1 is used to make a final decision for the state machine. In a detection queue record of the anti-eye radiation function, for the transition from the no-target state to the target state, if the current state is the no-target state and wants to make a state transition to the first intermediate state, it needs to maintain the no-target state for more than 1 second, and the continuous data of 1 in the detection queue exceeds 10 times, that is, a human target appears between the projection device and the projection surface 400. If the above conditions are met, the state machine enters the first intermediate state and the detection queue is cleared.

[0123] If this human body target only appears between the projection device and the projection surface 400 for a short time, it will trigger the maintenance of the first intermediate state for 1 second, and if the consecutive 0 data in the detection queue exceeds 3 times, the state machine enters the no-target state. If this human body target appears between the projection device and the projection surface 400 for a long time, it will trigger the maintenance of the first intermediate state for 3 seconds, and if the consecutive 1 data in the detection queue exceeds 10 times, the state machine enters the target state, and the light-emitting component is controlled to adjust the projection brightness to the second projection brightness.

[0124] Similarly, for the transition from the target state to the no-target state. If the current state is the target state, to transfer to the second intermediate state, it is necessary to maintain the target state for 1 second, and if the consecutive 0 data in the detection queue exceeds 10 times, that is, no human body target appears between the projection device and the projection surface 400. If the above conditions are met, the state machine enters the second intermediate state and the detection queue is cleared. At this time, the projection brightness of the light-emitting component remains unchanged.

[0125] After entering the second intermediate state, if a human body target appears between the projection device and the projection surface 400, it triggers the maintenance of the second intermediate state for 1 second, and if the consecutive 1 data in the detection queue exceeds 3 times, the state machine enters the target state and the detection queue is cleared. At this time, the projection brightness of the light-emitting component continues to remain unchanged. After entering the second intermediate state, if no human body target appears between the projection device and the projection surface 400 for a long time, it will trigger the maintenance of the second intermediate state for 3 seconds, and if the consecutive 0 data in the detection queue exceeds 10 times, the state machine enters the no-target state, and the light-emitting component is controlled to adjust the projection brightness to the first projection brightness.

[0126] It should be noted that the 1 second, 10 times, and 3 times described in the above embodiments are adjustment parameters and are not limited to the above values. For actual functional applications, in order to improve the sensitivity of detection, the adjustment parameters can also be 0.2s, 5 times, and 2 times.

[0127] Based on the projection control method provided in the above embodiments, some embodiments of the present application further provide a projection device, which includes a light-emitting component, a target detection component, and a controller 500. Among them, the light-emitting component is configured to project projection content onto the projection surface, and the projection surface can be a wall or a screen. The target detection component includes a distance sensor 600 and a camera 700. The camera 700 is configured to collect image data of the projection surface 400 according to the first acquisition period; the distance sensor 600 is configured to collect distance dot matrix image data of the projection surface 400 according to the second acquisition period. The controller 500 is then configured to execute the following program steps:

[0128] In response to a projection instruction, obtain a detection image.

[0129] Among them, the detection image is image data collected by controlling the camera 700 when the light-emitting component projects the projection content in the first projection scenario, or the detection image is distance dot matrix image data collected by controlling the distance sensor 600 when the light-emitting component projects the projection content in the second projection scenario.

[0130] Based on the mask model, identify the target mask in the detection image. The mask model includes an encoder and a decoder for identifying the target contour based on pixel points.

[0131] When there is an intersection between the target mask and the projection area in the detection image, add the first parameter value to the detection queue, and set the projection brightness of the light-emitting component according to the detection queue.

[0132] When there is no intersection between the target mask and the projection area in the detection image, add the second parameter value to the detection queue, and set the projection brightness of the light-emitting component according to the detection queue.

[0133] As can be seen from the above technical solutions, the projection device and projection control method provided in the above embodiments can obtain a detection image in response to a projection instruction. Among them, the detection image is image data collected by controlling the camera when the light-emitting component projects the projection content in the first projection scenario or distance dot matrix image data collected by controlling the distance sensor when the light-emitting component projects the projection content in the second projection scenario. Based on the mask model, identify the target mask in the detection image. The mask model includes an encoder and a decoder for identifying the target contour based on pixel points. When there is an intersection between the target mask and the projection area in the detection image, add the first parameter value to the detection queue. When there is no intersection between the target mask and the projection area in the detection image, add the second parameter value to the detection queue. Finally, set the projection brightness of the light-emitting component according to the detection queue. The method uses the camera in cooperation with the distance sensor, and the two sensors cooperate to solve the problem that the anti-eye projection function does not work in special scenarios. And uses encoding and decoding methods to perform pixel segmentation of the target on the image data, improving the sensitivity of human recognition. At the same time, it also uses the state machine transition method to alleviate the problem of false detection.

[0134] For the same and similar parts between the various embodiments in this specification, they can be referred to each other and will not be repeated here.

[0135] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments or some parts of the embodiments of the present invention.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.

[0137] For the sake of explanation, the above description has been made in conjunction with specific embodiments. However, the above exemplary discussion is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed above. According to the above teachings, various modifications and variations can be obtained. The selection and description of the above embodiments are for the purpose of better explaining the principles and practical applications, so that those skilled in the art can better use the embodiments and various different variations of the embodiments suitable for specific use considerations.

Claims

1. A projection device, characterized in that, it includes: a light-emitting component configured to project projection content onto a projection surface; a target detection component, the target detection component includes a distance sensor and a camera, the camera is configured to collect image data of the projection surface according to a first acquisition period; the distance sensor is configured to collect distance dot matrix image data of the projection surface according to a second acquisition period; a controller configured to: in response to a projection instruction, obtain a detection image, the detection image is image data collected by controlling the camera when the light-emitting component projects projection content in a first projection scenario, or, the detection image is distance dot matrix image data collected by controlling the distance sensor when the light-emitting component projects projection content in a second projection scenario; identify a target mask in the detection image based on a mask model, the mask model includes an encoder and a decoder for identifying a target contour based on pixel points; when there is an intersection between the target mask and the projection area in the detection image, add a first parameter value to a detection queue, and set the projection brightness of the light-emitting component according to the detection queue; when there is no intersection between the target mask and the projection area in the detection image, add a second parameter value to the detection queue, and set the projection brightness of the light-emitting component according to the detection queue.

2. The projection device according to claim 1, characterized in that, the controller is further configured to: obtain a sampling image, the sampling image is an image taken by the camera when the light-emitting component projects projection content; extract a non-projection area that does not contain projection content in the sampling image; detect a color attribute value of the non-projection area, and mark the current projection scenario as a first projection scenario or a second projection scenario based on the color attribute value.

3. The projection device according to claim 2, characterized in that, when the controller executes marking the current projection scenario as a first projection scenario or a second projection scenario based on the color attribute value, it is further configured to: convert the image format of the sampling image to a target image format, and the color attribute values in the target image format include hue, saturation, and brightness; obtain the saturation and brightness of the non-projection area in the sampling image; if the saturation is within a first threshold range and the brightness is within a second threshold range, mark the current projection scenario as a second projection scenario; if the saturation is not within the first threshold range, or the brightness is not within the second threshold range, mark the current projection scenario as a first projection scenario.

4. The projection device according to claim 1, characterized in that, the controller is further configured to: detect the current projection scenario; if the current projection scenario is a first projection scenario, control the camera to collect image data of the projection surface according to a first acquisition period; if the current projection scenario is a second projection scenario, control the distance sensor to periodically collect distance dot matrix image data of the projection surface according to a second acquisition period.

5. The projection device according to claim 1, characterized in that, The mask model includes a first mask model and a second mask model. The first mask model and the second mask model include a convolution operation module, which is used to perform encoding and decoding operations on an image. The number of convolution operation modules in the first mask model is greater than that in the second mask model. The controller is further configured to: If the current projection scenario is the first projection scenario, acquire the image data collected by the camera when the light emitting component projects the projection content, and input the image data into the first mask model; If the current projection scenario is the second projection scenario, acquire the distance dot matrix image data collected by the distance sensor when the light emitting component projects the projection content, and input the distance dot matrix image data into the second mask model.

6. The projection device according to claim 1, wherein, The controller is further configured to identify the target mask in the detected image based on the mask model, and is further configured to: Input the detected image into the mask model to identify the target contour in the detected image based on the mask model; Acquire the mask image output by the mask model. The mask image includes a first pixel value and a second pixel value. The first pixel value is used to represent the target, and the second pixel value is used to represent non-target; Extract the area of the first pixel value in the mask image to obtain the target mask.

7. The projection device according to claim 6, wherein, The controller is further configured to: Calculate the edge position of the target mask in the mask image. The edge position is the pixel position of the edge of the target mask; If the edge position is within the projection area, perform the step of adding a first parameter value to the detection queue; If the edge position is not within the projection area, perform the step of adding a second parameter value to the detection queue.

8. The projection device according to claim 1, wherein, The controller is further configured to: Create a state machine. The states of the state machine include a no-target state, a target state, a first intermediate state, and a second intermediate state. The states of the state machine are used to indicate controlling the projection brightness of the light emitting component; Acquire a preset state transition strategy. The state transition strategy is a state transition judgment method set based on the state maintenance time, the consecutive number of first parameter values, and the consecutive number of second parameter values in the detection queue; Transfer the state of the state machine according to the state transition strategy and the detection queue.

9. The projection device according to claim 8, wherein, The controller is further configured to: Detect the current state of the state machine; If the current state is the no-target state, set the projection brightness of the light emitting component to a first projection brightness; If the current state is the target state, set the projection brightness of the light emitting component to a second projection brightness, and the second projection brightness is less than the first projection brightness; If the current state is the first intermediate state or the second intermediate state, control the light emitting component to maintain the current projection brightness.

10. A projection control method, It is characterized in that applied to a projection device, the projection device includes a light-emitting component, a target detection component and a controller; the light-emitting component is configured to project projection content onto a projection surface; the target detection component includes a distance sensor and a camera; the camera is configured to collect image data of the projection surface according to a first acquisition period; the distance sensor is configured to collect distance dot matrix image data of the projection surface according to a second acquisition period; the projection control method includes: in response to a projection instruction, obtaining a detection image, where the detection image is image data collected by controlling the camera when the light-emitting component projects projection content in a first projection scenario, or the detection image is distance dot matrix image data collected by controlling the distance sensor when the light-emitting component projects projection content in a second projection scenario; identifying a target mask in the detection image based on a mask model, where the mask model includes an encoder and a decoder for identifying a target contour based on pixel points; when there is an intersection between the target mask and the projection area in the detection image, adding a first parameter value to a detection queue, and setting the projection brightness of the light-emitting component according to the detection queue; when there is no intersection between the target mask and the projection area in the detection image, adding a second parameter value to the detection queue, and setting the projection brightness of the light-emitting component according to the detection queue.

Images