Visual invisibility method, device and readable medium

Abstract

The present application provides a visual invisibility method, device and readable medium, the method comprising: generating at least one synthetic field of view image based on scene information within a set range in which the device is located; within the field of view of all detection entities, using the synthetic field of view image to cover the imaging of the device in the field of view of each detection entity. The present application achieves an effect of integration with the surrounding environment by utilizing the surrounding scene in which the device is located to synthesize the field of view image, and using the synthetic field of view image to cover the original device imaging in the field of view of the detection entity, thereby achieving visual invisibility, avoiding photoelectric detection and detection by the naked eye of the detection personnel, and realizing true visual invisibility.

Description

Technical Field

[0001] The present application relates to the field of visual recognition, and in particular to a visual invisibility method, device and readable medium. Background Art

[0002] Currently, there are few studies on the visual invisibility of equipment. Generally, most of them use physical methods to make the equipment invisible, which can avoid radar monitoring. However, in actual applications, there may be detection methods such as photoelectric equipment detection and on-site detection by detection personnel. Physical methods cannot achieve true invisibility. Therefore, there is an urgent need for a visual invisibility technology that can avoid detection by photoelectric equipment and detection personnel. Summary of the Invention

[0003] In view of the problem that in actual applications of the existing technology, there may be detection methods such as photoelectric equipment detection and on-site detection by detection personnel, and physical methods cannot achieve true invisibility, this application provides a visual invisibility method, device and readable medium.

[0004] A first embodiment of the present application provides a method for visually invisibility of a device, comprising:

[0005] generating at least one synthetic field of view image based on scene information within a set range of the device, wherein the scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original background points within the field of view of the corresponding detection entity after removing the device;

[0006] Within the field of view of all detection entities, the synthetic field of view image is used to cover the imaging of the device in the field of view of each detection entity.

[0007] In some embodiments, generating at least one synthetic field of view image based on scene information within a set range of the device includes:

[0008] Determine the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of the detection entity;

[0009] The synthetic field of view image is generated according to the virtual color; correspondingly, covering the imaging of the device in the field of view of each detection entity with the synthetic field of view image specifically includes: displaying the synthetic field of view image on the outer surface of the device to cover the imaging of the device in the field of view of the detection entity.

[0010] In some embodiments, the environmental information further includes a distance between the light source and the device, and an angle at which the light source illuminates the device, and the detecting entity includes a detecting person; and generating at least one synthetic field of view image based on scene information within a set range of the device includes:

[0011] Identify the detection personnel in each detection entity and determine the location information of the detection personnel;

[0012] Determine the virtual reflectivity of the background point originally covered by the device after the device is removed based on the position information of the detecting person, the distance between the light source and the device, and the angle at which the light source illuminates the device;

[0013] Determine the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of the detection entity;

[0014] The synthetic field of view image is generated according to the virtual reflectivity and the virtual color.

[0015] In some embodiments, the environmental information further includes a distance between the light source and the device, and an angle at which the light source illuminates the device; the detection entity includes a plurality of detection personnel, and the visual cloaking method further includes:

[0016] Identify the detection personnel in each detection entity, and determine the position information of each detection personnel and the pupil position of each detection personnel;

[0017] For each detected person, determine the virtual reflectivity of the background point originally covered by the device after the device is removed based on the position information of each detected person, the distance between the light source and the device, and the angle at which the light source illuminates the device;

[0018] Determine the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of each detection entity;

[0019] The synthetic field of view image corresponding to each detection person is generated according to the virtual color and the virtual reflectivity corresponding to each detection person; correspondingly, the imaging of the device in the field of view of each detection entity is covered with the synthetic field of view image, specifically including: projecting the corresponding synthetic field of view image onto the pupil of each detection person according to the pupil position of each detection person, so as to cover the imaging of the device in the field of view of each detection entity.

[0020] In some embodiments, determining the virtual reflectivity of the background point originally covered by the device after the device is removed based on the location information of the detection entity, the distance between the light source and the device, and the angle at which the light source illuminates the device includes:

[0021] determining illumination parameters according to the distance between the light source and the device and the angle at which the light source illuminates the device;

[0022] The virtual reflectivity is generated according to the lighting parameters.

[0023] In some embodiments, determining the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of the detection entity includes:

[0024] generating a three-dimensional model of the environment by combining multiple pre-stored background images;

[0025] Performing color marking on each coordinate point on the surface of the three-dimensional model to obtain a color mark for each color point in all coordinate points of the three-dimensional model; the coordinate points also include color default points;

[0026] According to the color mark of each color point of the three-dimensional model, for each coordinate point of the three-dimensional model on a straight line in a direction, a color simulation line is generated along multiple directions, and the coordinate point is located on each corresponding color simulation line;

[0027] Determine the color of the color-default point at each coordinate point according to the color of each color simulation line corresponding to each coordinate point on the three-dimensional model;

[0028] generating a background color image according to the color of the color point and the virtual color of the color default point;

[0029] The area occupied by the original covered background points after the device is removed is determined according to the position of the device and the corresponding occupied space, and the color of the coordinate points in the area in the background color image is used as the virtual color.

[0030] In some embodiments, the multiple directions include a first direction, a direction opposite to the first direction, a second direction, and a direction opposite to the second direction, wherein the first direction and the second direction are perpendicular to each other; and determining the color of the color default point at each coordinate point on the three-dimensional model based on the color of each color simulation line corresponding to each coordinate point includes:

[0031] Determine the number of colors on each color simulation line corresponding to each color default point that are the same as each other, and perform one of the following steps based on the number:

[0032] If the number is four, confirm that the color of the default point is the same color;

[0033] If the number is three and the difference between the different colors and the three identical colors is greater than a first set threshold, determining the color of the color default point based on image integrity and whether a target exists;

[0034] If the number is two, and the colors on the other two color simulation lines are the same, and the difference between the two colors is greater than a second set threshold, then determining the color of the color default point based on image integrity and whether a target exists;

[0035] If the number is two, and the colors on the other two color simulation lines are different from each other, and the difference values ​​of the three colors are greater than a third set threshold, then determining the color of the color default point based on image integrity and whether there is a target;

[0036] If the colors on all color simulation lines are different and the difference values ​​of the four colors are greater than a fourth set threshold, the color of the color default point is determined based on image integrity and whether a target exists.

[0037] In some embodiments, identifying a detected person in each detected entity and determining the pupil position of each detected person includes:

[0038] Using the OpenPose algorithm to identify the image of the detection entity to obtain at least one eye image of each detected person;

[0039] All eye images are segmented and grayscaled to obtain at least one pupil grayscale image;

[0040] The pupil position of each detected person is determined according to the shapes of different grayscale parts in the pupil grayscale image.

[0041] A second embodiment of the present application provides a visual cloaking device, comprising:

[0042] A flexible housing for covering the device to be invisible or the internal components of a visually invisible device;

[0043] A plurality of display units are covered on the outer surface of the flexible housing, each display unit being capable of switching to emit one of a plurality of colors of light;

[0044] The detector emits a detection medium to detect whether there is a detection entity within a set range; wherein,

[0045] The visual cloaking device further comprises:

[0046] a communication module, configured to transmit the detection signal of the detector to an external processor, so that the external processor generates at least one synthetic field of view image based on the detection signal and scene information within a set range where the device to be cloaked is located, and transmits a display unit control instruction to the visual cloaking device based on the synthetic field of view image, so as to cover the imaging of the device to be cloaked in the field of view of each detecting entity with the synthetic field of view image within the field of view of all detecting entities; or

[0047] The visual cloaking device further comprises:

[0048] a processor configured to generate at least one synthetic field of view image based on the detection signal and scene information within a set range of the device to be cloaked, and to generate a display unit control instruction based on the synthetic field of view image to overlay an image of the device to be cloaked in the field of view of each detecting entity with the synthetic field of view image within the field of view of all detecting entities;

[0049] The scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original covered background points within the field of view of the corresponding detection entity after removing the device.

[0050] The third aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the visual invisibility method when executed by a processor.

[0051] It can be seen from the above technical solution that the present application provides a visual invisibility method, device and readable medium, which synthesizes the field of view image by utilizing the surrounding scene of the device, and covers the original device imaging with the synthesized field of view image in the field of view of the detection entity, thereby achieving the effect of integration with the surrounding environment, thereby realizing visual invisibility, avoiding photoelectric detection and detection by the naked eye of the detection personnel, and realizing true visual invisibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0053] Figure 1 It is a flowchart of the visual invisibility method in an embodiment of the present application.

[0054] Figure 2 In the embodiment of this application Figure 1 One of the specific flow diagrams of step S1 in FIG.

[0055] Figure 3 In the embodiment of this application Figure 2 One of the specific flow diagrams of step S111 in FIG.

[0056] Figure 4 It is a schematic diagram of the process of obtaining virtual color by using a color model in an embodiment of the present application.

[0057] Figure 5 In the embodiment of this application Figure 1 The second specific flow diagram of step S1.

[0058] Figure 6 Schematic diagram of the joints of a human object recognized by the OpenPose algorithm in an embodiment of the present application.

[0059] Figure 7 In the embodiment of this application Figure 1 The third specific flow diagram of step S1.

[0060] Figure 8 In the embodiment of this application Figure 7 Specific flow chart of step S131 in FIG.

[0061] Figure 9 It is a flow chart of how the smart sentinel achieves visual invisibility in an embodiment of the present application.

[0062] Figure 10 This is one of the structural diagrams of the visual invisibility device in the embodiment of the present application.

[0063] Figure 11 This is the second structural diagram of the visual invisibility device in the embodiment of the present application.

[0064] Figure 12 This is a schematic diagram of the specific structure of the diamond-shaped piece in the smart sentinel in the embodiment of the present application.

[0065] Figure 13 This is one of the specific structural diagrams of the smart sentinel in the embodiment of this application.

[0066] Figure 14 This is the second specific structural diagram of the smart sentinel in the embodiment of this application. DETAILED DESCRIPTION

[0067] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0068] It should be noted that the visual invisibility method, system, electronic device and computer-readable storage medium disclosed in this application can be used in the field of visual invisibility technology, and can also be used in any field other than the field of visual invisibility technology. The application field of the visual invisibility method, system, electronic device and computer-readable storage medium disclosed in this application is not limited.

[0069] This application provides an embodiment of a visual invisibility method, such as Figure 1 Shown, including:

[0070] S1: Generate at least one synthetic field of view image based on scene information within a set range where the device is located, wherein the scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original covered background points within the field of view of the corresponding detection entity after the device is removed.

[0071] S2: within the field of view of all detection entities, covering the imaging of the device in the field of view of each detection entity with the synthetic field of view image.

[0072] From the above description, it can be seen that the visual invisibility method provided in the embodiment of the present application utilizes the surrounding scene of the device to synthesize the field of view image, and uses the synthesized field of view image to cover the original device imaging in the field of view of the detection entity, thereby achieving the effect of integration with the surrounding environment, thereby achieving visual invisibility, avoiding photoelectric detection and detection by the naked eye of the detection personnel, and realizing true visual invisibility.

[0073] In this application, the device can be a physical device that needs to implement the detection avoidance function, such as a sentry device used for coastal defense, or a stealth device set up to improve the user experience. For example, in a specific scenario, this device can be a sprinkler installed in a garden. In order to improve the visual experience of tourists during their visit, the sprinkler is made invisible so that tourists cannot detect the existence of the sprinkler, thereby improving the viewing experience.

[0074] Of course, the above examples are merely illustrative. This device can generally cope with any scenario that requires invisibility, and they are not enumerated here.

[0075] In this application, the setting range can be configured as needed. For example, the setting range can be a circular area with the device as the center and a diameter of 100 meters or 200 meters, or it can be the detection range of the scene information detector of the device. For example, the photoelectric detector generally has a visual range of 500 meters. In this case, the setting range is 500 meters. This application does not impose any restrictions.

[0076] Exemplarily, the device may include at least one detector for detecting scene information, and then obtain scene information within a set range through the detection of the detector on the device.

[0077] Exemplarily, the device may also separate the detector from the device body, that is, the device does not need to include the detector itself. Instead, the device may achieve communication connection with the detector through a communication device.

[0078] Furthermore, as a different example, the present device may not include any processing devices, that is, the acquisition and processing of scene information are all achieved through a remote processing device. The present device only includes the device parts that need to be invisible and a communication device. The communication device can be Bluetooth, a wireless transceiver, etc. Since the communication device transmits through wireless signals, as long as there is a signal-free shielding area on the outer shell of the device, signal transmission and reception can be achieved. Furthermore, the present device can be electrically connected to the processing device through a wire to perform signal transmission and reception. This application does not impose any restrictions.

[0079] In addition, illustratively, the detectors of the present application may include photoelectric detectors, such as cameras, infrared detectors, etc., and may also include radio wave detectors, such as ultrasonic detectors, pulse detectors, etc., and the present application does not impose any restrictions.

[0080] Furthermore, the scene information of the present application is environmental information of the surrounding environment, such as the color of the surrounding background points and the position of the detection entity.

[0081] It can be understood that in this application, the background environment in which the device is located can be defined as an environmental background point, and the device itself can be defined as a target point.

[0082] Exemplarily, the detection entity of the present application may be a photoelectric detector or a detection person, etc.

[0083] In addition, in this application, cooling materials and transparent nanomaterials can be applied to the equipment to avoid radar, which will not be elaborated in this application.

[0084] In one embodiment of the present application, the synthetic visual field map can be obtained based on a color model, specifically, Figure 2 As shown, step S1 of this application specifically includes:

[0085] S111: Determine the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of the detection entity;

[0086] S112: Generate the synthetic field of view image according to the virtual color; correspondingly, covering the imaging of the device in the field of view of each detection entity with the synthetic field of view image specifically includes: displaying the synthetic field of view image on the outer surface of the device to cover the imaging of the device in the field of view of the detection entity.

[0087] Specifically, in step S111, Figure 3 As shown, the color model can be used to determine the virtual color. Specifically, the virtual color is determined in the following ways:

[0088] S1111: Generate a three-dimensional model of the environment by combining multiple pre-stored background images.

[0089] For example, a photoelectric device can capture a 3D environment model composed of several background frames, and replace the pixels of the frame corresponding to the device with the background pixels, thereby creating a cloaking effect. First, it is necessary to store several background frames and stitch them together to form a 3D model.

[0090] S1112: Perform color marking on each coordinate point on the surface of the three-dimensional model to obtain a color mark for each color point in all coordinate points of the three-dimensional model; the coordinate points also include color default points.

[0091] Specifically, since the device covers the original scene color in the three-dimensional model, the color corresponding to the original covered background point after removing the device is the default color, that is, no color, and the original covered background point after removing the device is the default color point.

[0092] S1113: Generate color simulation lines along multiple directions for each coordinate point of the three-dimensional model on a straight line in a direction according to the color mark of each color point of the three-dimensional model, and the coordinate point is located on each corresponding color simulation line.

[0093] For example, the three-dimensional model is punctuated with colors horizontally and vertically. The denser the number of punctuations, the better the effect. Then, the color change rules are calculated from both sides of the color horizontally according to the color change rules. Then, the color change rules are calculated from both sides of the color vertically according to the color change rules. Finally, four simulated colors are formed.

[0094] S1114: Determine the color of the color default point at each coordinate point according to the color of each color simulation line corresponding to each coordinate point on the three-dimensional model.

[0095] Exemplarily, in some embodiments, the multiple directions include a first direction, a direction opposite to the first direction, a second direction, and a direction opposite to the second direction, and the first direction and the second direction are perpendicular to each other, such as the horizontal and vertical directions mentioned above.

[0096] In this embodiment, step S1114 performs one of the following steps based on the quantity:

[0097] S1114a: If the number is four, confirm that the color of the default point is the same color;

[0098] S1114b: If the number is three, and the difference between the different colors and the three identical colors is greater than a first set threshold, determining the color of the color default point based on image integrity and whether a target exists;

[0099] S1114c: If the number is two, and the colors on the other two color simulation lines are the same, and the difference between the two colors is greater than a second set threshold, then determining the color of the color default point based on image integrity and whether there is a target;

[0100] S1114d: If the number is two, and the colors on the other two color simulation lines are different from each other, and the difference values ​​of the three colors are greater than a third set threshold, then determining the color of the color default point based on image integrity and whether there is a target;

[0101] S1114e: If the colors on all color simulation lines are different and the difference values ​​of the four colors are greater than a fourth set threshold, the color of the color default point is determined based on image integrity and whether there is a target.

[0102] Combine Figure 4 From the color model framework shown, we can know that the four colors are the same, which is the first confirmation item. If the three colors are the same and there is a large difference with the fourth threshold, then refer to the fourth framework. The entire module is a complete specific target object. Image integrity is the first consideration. The color complement method is used to simulate the remaining color of the specific object and match it with two colors. If it matches the fourth color successfully, it is the fourth color. If the match is successful, it is three identical colors, which is the second confirmation item; three identical colors, no specific target, if there is not much difference with the fourth threshold, then the three identical colors are the second confirmation item; two identical colors, no specific target, if there is not much difference with the other two color thresholds, then the same color is the third confirmation item; two identical colors, if there is a large difference with the other two color thresholds, then whether there is a complete specific target object at this position, image integrity is the first consideration, and color complementation is used. The color complement method is used to simulate the remaining color of the specific object and match it with two colors. If any one of the two colors is the same, it is the third confirmation item; if two colors are the same and the other two are different, and the threshold difference with the other two colors is large, whether there is a complete specific target object at this position, image integrity is the first consideration, and the color complement method is used to simulate the remaining color of the specific object and match it with three colors. If any one of the three colors is the same, it is the third confirmation item; if all four colors are different, whether there is a complete specific target object at this position, image integrity is the first consideration, and the color complement method is used to simulate the remaining color of the specific object and match it with four colors. If any one of the four colors is the same, it is the fourth confirmation item; if there is no specific target: the color threshold difference is small, then the neutral color of the color is selected.

[0103] S1115: Generate a background color image according to the color of the color point and the virtual color of the color default point.

[0104] Specifically, the entire three-dimensional model is obtained according to the virtual colors in the above steps, and the surface color of the three-dimensional model is the background color image.

[0105] S1116: Determine the area occupied by the original covered background points after removing the device according to the position of the device and the corresponding occupied space, and use the color of the coordinate points in the area in the background color image as the virtual color.

[0106] In this embodiment, the visual invisibility of the color fusion method implemented by the device can be achieved by arranging tiny lamps on the outer surface of the device. For example, LED color-changing lamp beads can be evenly arranged on the outer surface of the device, and the above-mentioned environmental color fusion can be performed by controlling the color displayed by the LED color-changing lamp beads, thereby avoiding detection by photoelectric detectors or the naked eye of the detection personnel.

[0107] For example, this embodiment may also be combined with arranging transparent cooling materials and transparent wave-absorbing materials on the device to form double stealth, thereby further avoiding radar search.

[0108] In addition, in the embodiment of the present application, in order to further improve the invisibility effect when detecting with the naked eye of the detector, the embodiment of the present application further considers the influence of reflectivity. In the embodiment of the present application, the environmental information also includes the distance between the light source and the device, and the angle at which the light source illuminates the device. The detection entity includes a detection person. Figure 5 As shown, the steps of synthesizing the field of view image specifically include:

[0109] S121: Identify the detecting personnel in each detecting entity and determine the location information of the detecting personnel.

[0110] The input of the OpenPose algorithm is an image, and the output is an image of the positions of all the human skeleton joints in the image detected by the algorithm. Figure 8 Some embodiments of this application provide Figure 7 Specific flow chart of step S131 in FIG.

[0111] For example, Figure 6 As shown in the figure, after OpenPose output, each human body has 18 joints, including the head, two eyes, two ears, nose, neck, two shoulders, two elbows, two wrists, two hips, two knees, and two ankles. Each joint position can be represented by a two-dimensional coordinate in OpenPose, and each human bone has a total of 36 values.

[0112] In a preferred embodiment, the present application may adopt the Small-OpenPose algorithm, which is a variant of the OpenPose algorithm. Through the Small-OpenPose algorithm, some identification points and actions in the human skeleton joints in the image that are not related to the detection person's identification position can be deleted, and only the part position information needs to be identified.

[0113] For example, the number of human joint positions that need to be identified becomes 2, and each joint position is represented by a two-dimensional coordinate in the image, so each human bone has a total of 4 values.

[0114] Compared with the OpenPose algorithm, the Small-OpenPose algorithm deletes unnecessary recognition parts, thereby reducing the number of human joint positions that need to be identified, thereby improving the efficiency of identifying the positions of human joints.

[0115] In some embodiments, the present application can train an image recognition model based on the OpenPose algorithm. First, a data set is constructed, and a two-point model is trained in the COCO data set. Then, images for training are obtained by extracting video frames, that is, real-time video access, video processing, and video frames (w*h size color images). Then, feature values ​​F are extracted through a convolutional network (VGG-19 10-layer network). Preferably, the network forms a loop branch for predicting confidence maps S: key points (human body joints). The first stage S 1 =ρ 1 (F) ρ 1 It is the inference of the convolutional network CNNsd.

[0116] It can be understood that for the OpenPose algorithm, the network is divided into two loop branches when used. One branch is used to predict the confidence map S: key points (human joints). In each subsequent stage, the predicted value of the previous stage is used as the output S of the previous stage. t- The feature map F is used as input and the formula is as follows:

[0117]

[0118] The network then outputs S, and uses the loss function to calculate the L2 norm between the predicted value of S and the groundtruth (S*), as follows:

[0119]

[0120] Among them S * j is the event library confidence map, W is the binary mask W(P) = 0 image location fixation missing.

[0121] The loss function is the sum of the loss functions of each layer of the recurrent network, that is:

[0122]

[0123] S122: Determine the virtual reflectivity of the background point originally covered after the device is removed based on the position information of the detecting person, the distance between the light source and the device, and the angle at which the light source illuminates the device.

[0124] Specifically, the device's photoelectric system detects the angle and distance between the sun or light source and the device; then confirms the degree of reflectivity (glossiness) and its relationship to color changes. Glossiness is expressed as the reflectivity of the sample relative to the standard surface in the mirror (regular reflection) direction multiplied by 100, that is, G = 100R / R; finally, the entire simulated three-dimensional model is given a change in color and reflectivity (glossiness).

[0125] For example, glossiness can be measured using a gloss meter, an instrument used to measure the glossiness of an object's surface. Gloss meters are widely used in the automotive, ceramics, paint, coatings, and building materials industries. Glossiness can be categorized into low gloss, medium gloss, and high gloss depending on the measurement angle, which is not detailed here.

[0126] S123: Determine the virtual color of the original covered background point after the device is removed according to the color of each environmental background point and the position information of the detection entity.

[0127] The specific process in this step is the same as that of the above step S111 and will not be repeated here. The specific process is shown in S1111 to S1116.

[0128] S124: Generate the synthetic field of view image according to the virtual reflectivity and the virtual color.

[0129] In this embodiment, since virtual reflectivity is taken into consideration, that is, the reflectivity is superimposed on the three-dimensional model for each background point, a more realistic effect can be achieved when the detector detects with the naked eye at close range, making the detector believe that there is no equipment here, and the visual invisibility effect is better through the feedback of reflectivity.

[0130] Furthermore, in a preferred embodiment, in order to cope with more complex detection situations, such as multiple detection personnel performing detection at the same time, when the implicit formation success rate achieved through reflectivity reaches its limit, the present application further replaces the method of using the synthetic field of view image to cover the imaging of the device in the field of view of each detection entity.

[0131] Specifically, in the above embodiment, "using the synthetic field of view image to cover the imaging of the device in the field of view of each detection entity" in this application means using the synthetic field of view image to cover the device housing, thereby covering the imaging of the device in the field of view of each detection entity. In order to cope with the above complex detection situations, such as Figure 7 As shown, step S1 in the method of the present application specifically includes:

[0132] S131: Identify the detection personnel in each detection entity, and determine the position information of each detection personnel and the pupil position of each detection personnel;

[0133] S132: For each detected person, determine the virtual reflectivity of the background point originally covered by the device after the device is removed based on the position information of each detected person, the distance between the light source and the device, and the angle at which the light source illuminates the device;

[0134] S133: Determine the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of each detection entity;

[0135] S134: Generate the synthetic field of view image corresponding to each detection person according to the virtual color and the virtual reflectivity corresponding to each detection person.

[0136] Specifically, the pupil position is Figure 6 The position of the left eye and the position of the right eye are not described in detail in this application.

[0137] Specifically, step S133 in this embodiment is the same as the above-mentioned step S111, which will not be described in detail here. The specific process is shown in S1111 to S1116.

[0138] The reflectivity of this device can be detected by the photoelectric system. The angle and distance between the sun or light source and the device; then the degree of reflectivity (glossiness) and its relationship with color change are confirmed. The glossiness is expressed by the reflectivity of the sample relative to the standard surface in the mirror (regular reflection) direction multiplied by 100, that is, G = 100R / R; finally, the color and reflectivity (glossiness) changes are given to the entire simulated 3D model.

[0139] For example, the reflectivity can be measured by a gloss meter, but the present application is not limited thereto.

[0140] Correspondingly, the use of the synthetic field of view image to cover the imaging of the device in the field of view of each detection entity specifically includes: projecting the corresponding synthetic field of view image onto the pupil of each detection personnel according to the pupil position of each detection personnel to cover the imaging of the device in the field of view of each detection entity.

[0141] In this embodiment, the synthetic field of view image is directly projected into the pupils of each detector by means of a projection image, and each detector obtains a different synthetic field of view map due to its different position, and then the corresponding synthetic field of view map is projected onto the detector. Therefore, when multiple detectors exist at the same time, the different reflectivities corresponding to the detectors in different positions are comprehensively considered, so that multiple detectors cannot perceive this device, achieving the effect of common invisibility, and further improving the visual invisibility effect and success rate of multiple detectors during reconnaissance.

[0142] In an optional embodiment, if Figure 8 As shown, the determination of the pupil position of each detected person in the above step S131 in this application specifically includes:

[0143] S1311: Using the OpenPose algorithm to identify the image of the detection entity to obtain at least one eye image of each detected person;

[0144] S1312: Slice all eye images and perform grayscale processing to obtain at least one pupil grayscale image;

[0145] S1313: Determine the pupil position of each detected person according to the shapes of different grayscale parts in the pupil grayscale image.

[0146] For example, the steps of using OpenPose and OpenCV algorithms to identify the pupils of a target object are as follows:

[0147] ①Use the minimalist OpenPose algorithm to identify the left and right eyes;

[0148] ②On this basis, the classifier in Opencv performs classification confirmation;

[0149] ③ Cut the entire left and right eye images;

[0150] ④ Grayscale the cut image;

[0151] ⑤ After grayscale processing, judge the shape of different grayscale parts;

[0152] ⑥The circular area is the location of the pupil.

[0153] From the above technical solution, it can be seen that the present application utilizes the surrounding scene of the device to synthesize the field of view image, and uses the synthesized field of view image to cover the original device imaging in the field of view of the detection entity, thereby achieving the effect of integration with the surrounding environment, thereby achieving visual invisibility, avoiding photoelectric detection and detection by the naked eye of the detection personnel, and realizing true visual invisibility.

[0154] Furthermore, the above-mentioned stealth embodiments of the present application can be combined with each other. For example, in some embodiments, a four-step stealth mechanism can be set. Figure 9 It shows that this device is a combination of the above-mentioned visual stealth methods of the smart sentinel device. It can be seen that this application is combined with the four-time stealth mechanism to achieve the best visual stealth effect.

[0155] Furthermore, the present application provides a visual cloaking device, such as Figure 10 Shown, including:

[0156] The flexible housing 101 is used to be mounted on the device to be invisible or the internal components of the visual invisible device;

[0157] Multiple display units (not shown in the figure) are covered on the outer surface of the flexible shell, each display unit can switch to emit one of multiple colors of light;

[0158] The detector 102 emits a detection medium to detect whether a detection entity exists within a set range; and

[0159] The communication module 103 is used to send the detection signal of the detector to an external processor, so that the external processor generates at least one synthetic field of view image based on the detection signal and scene information within a set range of the device to be cloaked, and sends a display unit control instruction to the visual cloaking device based on the synthetic field of view image, so as to cover the imaging of the device to be cloaked in the field of view of each detecting entity with the synthetic field of view image within the field of view of all detecting entities.

[0160] In this embodiment, the detection medium may be an optical medium, an acoustic wave medium, or an ultrasonic medium. It is understandable that the detection medium of the present application may also be micro detection particles, etc., and the present application does not impose any limitation thereto.

[0161] In this embodiment, a flexible shell is mounted on the device to be invisible, and the flexible shell is covered with a display unit. For example, the display unit in the embodiment of the present application can be the LED lamp beads described in the above embodiment. Of course, the display unit can also be other display devices, such as a material layer composed of a luminescent material, etc. This application does not limit this.

[0162] In this embodiment, the processing device is placed in the cloud or a remote location, and signal interaction is achieved through a communication module. The communication module can be a Bluetooth module, a wireless transceiver, etc., and this application does not impose any restrictions.

[0163] It can be understood that the visual stealth device of the present application is mounted on equipment that needs to be invisible, such as the current smart sentinel device used for coastal defense. The visual stealth device can be mounted on the smart sentinel device to achieve visual invisibility at sea.

[0164] In other embodiments, the visual cloaking device of the present application may be a device that itself needs to be invisible. Taking the smart sentinel as an example, the visual cloaking device in this embodiment may be the smart sentinel device itself, that is, the outer shell of the smart sentinel device is a flexible shell, and the internal device is the internal device of the visual cloaking device.

[0165] For example, the present application provides a specific application of a visual stealth device, such as a smart sentinel device, such as Figure 12 As shown, the smart sentinel device includes a shell composed of flexible diamond-shaped pieces. The flexible diamond-shaped pieces are the smallest components of the invisible coat. They are miniature in design and are magnetically attracted. They can be attached to objects of different shapes at will. The components are magnetically attracted to each other to prevent disconnection.

[0166] Illustratively, the internal components of the diamond-shaped piece provided in the present application are mainly composed of LED color-changing lamp beads, transparent cooling materials and transparent absorbing materials. For example, the LED color-changing lamp beads can change colors such as red, yellow, and blue.

[0167] Furthermore, this application does not limit the shape of the smart sentinel device. For example, Figure 14 As shown, the smart sentinel device is cylindrical, or flat (not shown in the figure).

[0168] In addition, in an embodiment of the present application, the diamond-shaped sheet can be combined with the smart sentinel by magnetic attraction. The magnetic adsorption device is located at the bottom of the flexible invisible diamond-shaped sheet structure. The adsorption device adopts the principle of magnetic adsorption and is adsorbed on the smart sentinel with components.

[0169] For example, Figure 2 As shown, the upper side 2 of the interface in the figure is located on one side of the flexible invisible diamond sheet structure, which is a part of the flexible invisible diamond sheet structure, and is composed of LED color-changing lamp beads, transparent cooling material and transparent absorbing material, and half of a magnetic adsorption device, which is used to bond with the lower end of the interface.

[0170] Further, Figure 12 The lower side 3 of the interface is located opposite to the upper side of the interface of the flexible invisible diamond-shaped sheet structure, is installed at the bottom end of the flexible invisible diamond-shaped sheet, and is composed of half a magnetic adsorption device for bonding with the upper side of the interface.

[0171] For example, the present application can use nano-stitching materials attached to diamond-shaped sheets. In some embodiments, the nano-stitching materials are composed of transparent cooling materials and transparent absorbing materials, which are arranged in a regular nano-scale arrangement. The location of the lamp beads and the content of the cooling materials and absorbing materials can be configured based on actual conditions.

[0172] In a preferred embodiment, a transparent cooling material and a transparent wave-absorbing material may be stacked in sequence, but the present application is not limited thereto.

[0173] It can be understood that the smart sentinel of the present application may include a projector 6, located at the top of the smart sentinel, for projecting images onto the pupils of a person.

[0174] For example, in the embodiment shown in the figure, the smart sentinel device includes a rod body, which can be located at the bottom of the underwater smart sentinel in the coastal defense field to support various sensor front ends of the entire smart sentinel.

[0175] For example, Figure 13 and Figure 14 As shown, the detector of the present application can be a photoelectric detector, such as a camera. Specifically, the detector can be located on one side of the underwater smart sentinel and used to determine the person's orientation and pupil position. For example, this system consists of two three-variable cameras, which can achieve 6x optical zoom. The image sensor uses an industrial CMOS sensor. The entire gimbal is controlled by five servos, which can achieve 5 degrees of freedom, including horizontal rotation of the two cameras, vertical pitch, and horizontal rotation of the neck.

[0176] In other embodiments, the photodetector may be located inside the device, and the detection light of the photodetector may be guided out through a transparent material.

[0177] Furthermore, in other optional embodiments, the present application also provides a visual cloaking device, as shown in FIG11 , comprising:

[0178] The flexible housing 101 is used to be mounted on the device to be invisible or the internal components of the visual invisible device;

[0179] Multiple display units (not shown in the figure) are covered on the outer surface of the flexible shell, each display unit can switch to emit one of multiple colors of light;

[0180] The detector 102 emits a detection medium to detect whether there is a detection entity within a set range.

[0181] Unlike the above-mentioned embodiment, the visual cloaking device of this embodiment includes a processor 104, which generates at least one synthetic field of view image based on the detection signal and scene information within a set range of the device to be cloaked, and generates a display unit control instruction based on the synthetic field of view image to cover the image of the device to be cloaked in the field of view of each detecting entity with the synthetic field of view image within the field of view of all detecting entities;

[0182] The scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original covered background points within the field of view of the corresponding detection entity after removing the device.

[0183] In this embodiment, a flexible shell is mounted on the device to be invisible, and the flexible shell is covered with a display unit. For example, the display unit in the embodiment of the present application can be the LED lamp beads described in the above embodiment. Of course, the display unit can also be other display devices, such as a material layer composed of a luminescent material, etc. This application does not limit this.

[0184] In this embodiment, the processing device is placed inside the device, and the specific calculation process can be performed in the processor.

[0185] It can be understood that the visual invisibility device provided by the present application utilizes the surrounding scene of the device to synthesize the field of view image, and uses the synthesized field of view image to cover the original device imaging in the field of view of the detection entity, thereby achieving the effect of integration with the surrounding environment, thereby achieving visual invisibility, avoiding photoelectric detection and detection by the naked eye of the detection personnel, and realizing true visual invisibility.

[0186] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the visual cloaking method in the above embodiments. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the computer program implements all steps of the visual cloaking method in the above embodiments in which the execution subject is a server or a client. For example, when the processor executes the computer program, the following steps are implemented:

[0187] S1: Generate at least one synthetic field of view image based on scene information within a set range where the device is located, wherein the scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original covered background points within the field of view of the corresponding detection entity after the device is removed.

[0188] S2: within the field of view of all detection entities, covering the imaging of the device in the field of view of each detection entity with the synthetic field of view image.

[0189] From the above description, it can be seen that the computer-readable medium provided in the embodiment of the present application achieves the effect of integration with the surrounding environment by utilizing the surrounding scene of the device to synthesize the field of view image, and using the synthesized field of view image to cover the original device imaging in the field of view of the detection entity, thereby achieving visual invisibility, avoiding photoelectric detection and detection by the naked eye of the detection personnel, and realizing true visual invisibility.

[0190] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0191] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (apparatus), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0192] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0193] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0194] Specific embodiments are used in this application to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for visually invisible equipment, characterized in that: include: generating at least one synthetic field of view image based on scene information within a set range of the device, wherein the scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original background points within the field of view of the corresponding detection entity after removing the device; Within the field of view of all detection entities, covering the imaging of the device in the field of view of each detection entity with the synthetic field of view image; The generating of at least one synthetic field of view image according to scene information within a set range of the device includes: Determine the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of each detection entity; Generating the synthetic field of view image according to the virtual color; correspondingly, covering the imaging of the device in the field of view of each detection entity with the synthetic field of view image specifically includes: displaying the synthetic field of view image on the outer surface of the device to cover the imaging of the device in the field of view of the detection entity; The step of determining the virtual color of the original covered background point after removing the device according to the color of each environmental background point and the position information of the detection entity includes: generating a three-dimensional model of the environment by combining multiple pre-stored background images; Performing color marking on each coordinate point on the surface of the three-dimensional model to obtain a color mark for each color point in all coordinate points of the three-dimensional model; the coordinate points also include color default points; According to the color mark of each color point of the three-dimensional model, for each coordinate point of the three-dimensional model on a straight line in a direction, a color simulation line is generated along multiple directions, and the coordinate point is located on each corresponding color simulation line; Determine the color of the color-default point at each coordinate point according to the color of each color simulation line corresponding to each coordinate point on the three-dimensional model; generating a background color image according to the color of the color point and the virtual color of the color default point; The area occupied by the original covered background points after the device is removed is determined according to the position of the device and the corresponding occupied space, and the color of the coordinate points in the area in the background color image is used as the virtual color.

2. The visual invisibility method according to claim 1, characterized in that: The scene information further includes a distance between the light source and the device, and an angle at which the light source illuminates the device. The detection entity includes a detection person. Generating at least one synthetic field of view image based on the scene information within a set range of the device includes: Identify the detection personnel in each detection entity and determine the location information of the detection personnel; Determine the virtual reflectivity of the background point originally covered by the device after the device is removed based on the location information of the detecting person, the distance between the light source and the device, and the angle at which the light source illuminates the device; Generating the synthetic visual field image according to the virtual color includes: The synthetic field of view image is generated according to the virtual reflectivity and the virtual color.

3. The visual invisibility method according to claim 1, characterized in that: The scene information further includes the distance between the light source and the device, and the angle at which the light source illuminates the device; the detection entity includes a plurality of detection personnel, and the visual cloaking method further includes: Identify the detection personnel in each detection entity, and determine the position information of each detection personnel and the pupil position of each detection personnel; For each detected person, the virtual reflectivity of the background point originally covered by the device after the device is removed is determined based on the position information of the detected person, the distance between the light source and the device, and the angle at which the light source illuminates the device; Generating the synthetic visual field image according to the virtual color includes: The synthetic field of view image corresponding to each detection person is generated according to the virtual color and the virtual reflectivity corresponding to each detection person; correspondingly, the imaging of the device in the field of view of each detection entity is covered with the synthetic field of view image, specifically including: projecting the corresponding synthetic field of view image onto the pupil of each detection person according to the pupil position of each detection person, so as to cover the imaging of the device in the field of view of each detection entity.

4. The visual invisibility method according to claim 2 or 3, characterized in that: The determining, based on the position information of the detecting person, the distance between the light source and the device, and the angle at which the light source illuminates the device, of the virtual reflectivity of the background point originally covered by the device after the device is removed includes: determining illumination parameters according to the distance between the light source and the device and the angle at which the light source illuminates the device; The virtual reflectivity is generated according to the lighting parameters.

5. The visual invisibility method according to any one of claims 1 to 3, characterized in that: The multiple directions include a first direction, a direction opposite to the first direction, a second direction, and a direction opposite to the second direction, wherein the first direction and the second direction are perpendicular to each other; and determining the color of the color default point at each coordinate point on the three-dimensional model according to the color of each color simulation line corresponding to each coordinate point includes: Determine the number of colors on each color simulation line corresponding to each color default point that are the same as each other, and perform one of the following steps based on the number: If the number is four, confirm that the color of the default point is the same color; If the number is three and the difference between the different colors and the three identical colors is greater than a first set threshold, determining the color of the color default point based on image integrity and whether a target exists; If the number is two, and the colors on the other two color simulation lines are the same, and the difference between the two colors is greater than a second set threshold, then determining the color of the color default point based on image integrity and whether a target exists; If the number is two, and the colors on the other two color simulation lines are different from each other, and the difference values ​​of the three colors are greater than a third set threshold, then determining the color of the color default point based on image integrity and whether there is a target; If the colors on all color simulation lines are different and the difference values ​​of the four colors are greater than a fourth set threshold, the color of the color default point is determined based on image integrity and whether a target exists.

6. The visual invisibility method according to claim 3, characterized in that: Identify the detecting personnel in each detecting entity and determine the pupil position of each detecting personnel, including: Using the OpenPose algorithm to identify the image of the detection entity to obtain at least one eye image of each detected person; All eye images are segmented and grayscaled to obtain at least one pupil grayscale image; The pupil position of each detected person is determined according to the shapes of different grayscale parts in the pupil grayscale image.

7. A visual cloaking device, characterized in that: include: A flexible housing for covering the device to be invisible or the internal components of a visually invisible device; A plurality of display units are covered on the outer surface of the flexible housing, each display unit being capable of switching to emit one of a plurality of colors of light; The detector emits a detection medium to detect whether there is a detection entity within a set range; wherein, The visual cloaking device further comprises: a communication module, configured to transmit the detection signal of the detector to an external processor, so that the external processor generates at least one synthetic field of view image based on the detection signal and scene information within a set range where the device to be cloaked is located, and transmits a display unit control instruction to the visual cloaking device based on the synthetic field of view image, so as to cover the imaging of the device to be cloaked in the field of view of each detecting entity with the synthetic field of view image within the field of view of all detecting entities; or The visual cloaking device further comprises: a processor configured to generate at least one synthetic field of view image based on the detection signal and scene information within a set range of the device to be cloaked, and to generate a display unit control instruction based on the synthetic field of view image to overlay an image of the device to be cloaked in the field of view of each detecting entity with the synthetic field of view image within the field of view of all detecting entities; The scene information includes the color of each environmental background point and the position information of all detection entities, and each synthetic field of view image includes an image of the original covered background points within the field of view of the corresponding detection entity after removing the device; The communication module and the processor are further configured to: Determining, based on the colors of the environmental background points and the position information of the detection entities, a virtual color of the original background points after the device is removed; generating the synthetic field of view image based on the virtual color; and generating a display unit control instruction based on the synthetic field of view image to display the synthetic field of view image on the outer surface of the device to cover the imaging of the device in the field of view of the detection entity; The communication module and the processor are further configured to: A three-dimensional model of an environment is generated by combining multiple pre-stored background images; each coordinate point on the surface of the three-dimensional model is color-marked to obtain a color mark for each color point in all coordinate points of the three-dimensional model; the coordinate points also include color default points; based on the color mark of each color point of the three-dimensional model, color simulation lines are generated along multiple directions for each coordinate point on a straight line in a direction of the three-dimensional model, and the coordinate point is located on the corresponding color simulation lines; based on the color of each coordinate point on the three-dimensional model on the corresponding color simulation lines, the color of the color default point in each coordinate point is determined; a background color image is generated based on the color of the color point and the virtual color of the color default point; based on the position of the device and the corresponding occupied space, the area occupied by the original covered background point after removing the device is determined, and the color of the coordinate point in the area in the background color image is used as the virtual color.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images