Partition imaging system, determination method and storage medium
Through partition imaging system and optical sensor technology, the problem of unavailability to obtain image information of multiple parts in the prior art is solved, cost savings and imaging accuracy are achieved, and imaging efficiency and image algorithm accuracy are improved.
Patent Information
- Application Number
- CN202410985507.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-22
AI Technical Summary
In the prior art, it is impossible to obtain image information of multiple parts in a unified manner, resulting in increased cost of imaging systems and differences in imaging effects of multiple parts, affecting imaging efficiency and the accuracy of image algorithms.
Using a partition imaging system, modulated light corresponding to different parts of the target object is formed through two modulation regions, corresponding image data is generated using optical sensors, and image processing is performed in combination with processors and neural network models and the focal length of the metasurface structure is adjusted to Improve image accuracy.
It realizes the unified acquisition of image data from multiple parts of the target object, reduces hardware costs, avoids errors introduced by imaging perspective differences, and improves imaging efficiency and image algorithm accuracy.
Smart Images

Figure CN119011973A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of optical imaging technology, and in particular to a partition imaging system, a determination method and a storage medium. Background Art
[0002] In computer vision applications, in addition to obtaining image information of a single part, in some scenarios, it is also necessary to obtain image information of multiple parts for users to conduct targeted analysis. For example, in a driver monitoring system (DMS), in addition to observing information of other parts of the driver, at least the driver's eye image needs to be obtained to determine the driver's current fatigue state, etc.
[0003] In the prior art, multiple cameras are usually set up to obtain image information of multiple parts of the target object, and then processed and displayed. However, this technical solution greatly increases the hardware cost, and multiple cameras obtain image information from different shooting angles, which will introduce errors in the analysis of image information of multiple parts. Summary of the invention
[0004] One of the purposes of the present invention is to provide a partitioned imaging system to solve the technical problems in the prior art that image information of multiple parts cannot be uniformly acquired, resulting in increased cost of the imaging system, and differences in imaging effects of multiple parts lead to poor imaging efficiency and image algorithm accuracy.
[0005] One of the objects of the present invention is to provide a method for determining a partitioned imaging system.
[0006] One of the objects of the present invention is to provide a computer storage medium.
[0007] To achieve one of the above-mentioned purposes of the invention, one embodiment of the present invention provides a partitioned imaging system, comprising: a first modulation zone and a second modulation zone, for forming a metasurface structure; an optical sensor, for generating first image data according to the modulated light at the first modulation zone, and for generating second image data according to the modulated light at the second modulation zone; the first image data corresponds to a first part of the target object, and the second image data corresponds to a second part of the target object; the first part is different from the second part, or the first part is included in the second part.
[0008] As a further improvement of an embodiment of the present invention, the second part corresponds to the entire target object, and the first part corresponds to a part of the target object.
[0009] As a further improvement of an embodiment of the present invention, the accuracy of the first image is higher than the accuracy of the second image.
[0010] As a further improvement of one embodiment of the present invention, the optical sensor includes: a first processing area for generating the first image data; a second processing area for generating the second image data; or, the partitioned imaging system includes: a first optical sensor for generating the first image data; and a second optical sensor for generating the second image data.
[0011] As a further improvement of one embodiment of the present invention, it includes: a first supersurface light-guiding element, used to form a first modulation zone; a second supersurface light-guiding element, used to form a second modulation zone; or, includes: a third supersurface light-guiding element, used to form a first modulation zone and a second modulation zone.
[0012] As a further improvement of one embodiment of the present invention, the configuration of the supersurface structure at the first modulation zone is determined according to the size of the first part of the target object, and / or the configuration of the supersurface structure at the second modulation zone is determined according to the size of the second part of the target object.
[0013] As a further improvement of one embodiment of the present invention, the focal length of the metasurface structure at the first modulation zone is determined according to the size of the first portion; the focal length of the metasurface structure at the second modulation zone is determined according to the size of the second portion.
[0014] As a further improvement of an embodiment of the present invention, the focal length of the metasurface structure is adjusted by adjusting at least one of the following: the size of the nanostructure unit; the rotation angle of the nanostructure unit; the duty cycle of the nanostructure unit.
[0015] As a further improvement of one embodiment of the present invention, the configuration of the metasurface structure at the first modulation zone is determined according to the target accuracy of the first image, and / or the configuration of the metasurface structure at the second modulation zone is determined according to the target accuracy of the second image, and / or the configuration of the metasurface structure at the first modulation zone and the second modulation zone is determined according to the target accuracy relationship between the first image and the second image.
[0016] As a further improvement of an embodiment of the present invention, the partition imaging system further includes: a processor, configured to generate a target image including the first image and the second image according to the first image data and the second image data.
[0017] As a further improvement of one embodiment of the present invention, the partition imaging system also includes: a processor, used to generate a second image corresponding to the second part based on the second image data; the second image is an encrypted image and / or a posture image; the encrypted image is determined by fuzzy processing, and the posture image includes a mark indicating the current posture of the second part.
[0018] As a further improvement of one embodiment of the present invention, the processor generates the encrypted image and / or the posture image by implementing a neural network model; the neural network model is determined according to the target accuracy of the first image, and / or according to the target accuracy of the second image, and / or according to the target accuracy relationship between the first image and the second image, and / or according to the loss value training between the second image data and the sample image data.
[0019] To achieve one of the above-mentioned purposes of the invention, an embodiment of the present invention provides a method for determining a partitioned imaging system, wherein the partitioned imaging system is configured as any one of the above-mentioned technical solutions, and the method comprises: obtaining a first precision value of a first image corresponding to a first modulation zone, and a second precision value of a second image corresponding to a second modulation zone; determining the difference between the first precision value and a set target precision, and adjusting the focal length of the metasurface structure at the corresponding first modulation zone according to the difference; and synchronously or separately determining the difference between the second precision value and the set target precision, and adjusting the focal length of the metasurface structure at the corresponding second modulation zone according to the difference.
[0020] As a further improvement of one embodiment of the present invention, the partitioned imaging system also includes: a neural network model, used to generate a second image corresponding to the second part based on the second image data; the second image is an encrypted image and / or a posture image; the method includes: obtaining sample image data corresponding to the second part; determining a loss value based on the second image data and the sample image data; and adjusting the focal length and / or neural network model parameters of the supersurface structure at the corresponding second modulation area based on the loss value to obtain a trained neural network model.
[0021] To achieve one of the above-mentioned purposes of the invention, an embodiment of the present invention provides a computer storage medium having an application stored thereon, characterized in that when the application is executed, the steps of the method for determining a partitioned imaging system described in the above-mentioned technical solution are implemented.
[0022] Compared with the prior art, the partitioned imaging system provided by the present invention provides two modulation zones to respectively form modulated lights corresponding to two parts of the target object, and the optical sensor generates two items of image data corresponding to the two modulated lights, so that the partitioned imaging system can uniformly obtain image data corresponding to at least two parts of the target object. Compared with configuring two imaging systems, it can not only save costs, but also avoid the introduction of errors due to the difference in imaging angles of the two imaging systems, thereby helping to improve imaging efficiency and the accuracy of image algorithms. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 It is a schematic structural diagram of a partition imaging system in the first embodiment of the present invention.
[0024] Figure 2is a schematic diagram of a target image in one embodiment of the present invention.
[0025] Figure 3 It is a schematic diagram of the structure of the first modulation area and the second modulation area in one embodiment of the present invention.
[0026] Figure 4 It is a schematic diagram of the structure of the first modulation area and the second modulation area in another embodiment of the present invention.
[0027] Figure 5 Schematic diagram of the structure of the partition imaging system in the second embodiment of the present invention.
[0028] Figure 6 Schematic diagram of the structure of the partition imaging system in the third embodiment of the present invention.
[0029] Figure 7 Schematic diagram of the structure of the partition imaging system in the fourth embodiment of the present invention.
[0030] Figure 8 is a schematic diagram of a target image in another embodiment of the present invention.
[0031] Fig. 9 is a schematic diagram of a second image in one embodiment of the present invention.
[0032] Fig.10 It is a schematic diagram of the steps of a method for determining a partitioned imaging system in one embodiment of the present invention.
[0033] Fig.11 It is a schematic diagram of some steps of a specific example of a method in an embodiment of the present invention. DETAILED DESCRIPTION
[0034] The present invention will be described in detail below in conjunction with the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional changes made by a person skilled in the art based on these embodiments are all within the scope of protection of the present invention.
[0035] It should be noted that the term "comprises" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In addition, the terms "first", "second", etc. are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.
[0036] like Figure 1 As shown, an embodiment of the present invention provides a partition imaging system.
[0037] The "zoning" may refer to imaging different parts or regions of the target object. The zoning imaging system may output a unified image containing image information of different parts, or may output images of different parts separately.
[0038] In one embodiment, “partitioning” may also refer to forming images corresponding to different parts in different regions.
[0039] The regions may be different regions of the light-guiding element. For example, the first part of the target object and the first image data correspond to the first modulation area of the metasurface light-guiding element; the second part of the target object and the second image data correspond to the second modulation area of the metasurface light-guiding element.
[0040] The first modulation area and the second modulation area may be arranged at different light guide elements, or may be arranged at the same light guide element.
[0041] The regions may be different regions of the optical sensor. For example, a first portion of the target object and first image data correspond to a first processing region of the optical sensor; a second portion of the target object and second image data correspond to a second processing region of the optical sensor.
[0042] The first processing area and the second processing area may be disposed at different optical sensors or at the same optical sensor.
[0043] The regions may be different regions of a display screen (e.g., rendered at a display module) or an output image. For example, a first portion of a target object and first image data correspond to a first image included in a target image, or correspond to a first display position of a display screen or a display module; a second portion of a target object and second image data correspond to a second image included in a target object, or correspond to a second display position of a display screen or a display module.
[0044] In one embodiment, the whole partition imaging system, or at least two parts of the partition imaging system, are integrated into a partition imaging module. For example, the first modulation area, the second modulation area, and the optical sensor of the partition imaging system are integrated into a partition imaging module, or the first modulation area, the second modulation area, the optical sensor, and the processor of the partition imaging system are integrated into a partition imaging module. In this way, the compactness of the system can be improved, and the imaging efficiency and the accuracy of the image algorithm can be improved.
[0045] The partitioned imaging system comprises a first modulation zone 11 .
[0046] The first modulation region 11 is used to form a supersurface structure.
[0047] When the metasurface structure is defined as a metasurface light-guiding element, the first modulation area 11 can be formed in the metasurface light-guiding element; the first modulation area 11 is a part of the metasurface light-guiding element. When the metasurface structure is defined as a nanostructure unit of the metasurface, the first modulation area 11 can be formed by a nanostructure unit; a plurality of nanostructure units are arranged in the first modulation area 11.
[0048] In one embodiment, the first modulation area 11 is used to modulate the light input thereto and output it to a first processing position of a subsequent device. In one embodiment, the input light of the first modulation area 11 corresponds to a first part of the target object. In one embodiment, the output light of the first modulation area 11 is used to generate first image data and a first image.
[0049] The zoned imaging system comprises a second modulation zone 12 .
[0050] The second modulation region 12 is used to form a super surface structure.
[0051] The second modulation area 12 can be formed in the metasurface light-guiding element; the second modulation area 12 is a part of the metasurface light-guiding element. When the metasurface structure is defined as a nanostructure unit of the metasurface, the second modulation area 12 can be formed by a nanostructure unit; a plurality of nanostructure units are arranged in the second modulation area 12.
[0052] In one embodiment, the second modulation area 12 is used to modulate the light input thereto and output it to a second processing position of a subsequent device. In one embodiment, the input light of the second modulation area 12 corresponds to a second part of the target object. In one embodiment, the output light of the second modulation area 12 is used to generate second image data and a second image.
[0053] The partitioned imaging system includes an optical sensor 200 .
[0054] The optical sensor 200 is used to generate first image data. The optical sensor 200 is used to receive modulated light at the first modulation area 11. The first image data is used to form a first image.
[0055] The optical sensor 200 is used to generate first image data according to the modulated light at the first modulation area 11 .
[0056] The optical sensor 200 is used to generate second image data. The optical sensor 200 is used to receive the modulated light at the second modulation area 12. The second image data is used to form a second image.
[0057] The optical sensor 200 is used to generate second image data according to the modulated light at the second modulation area 12 .
[0058] In one embodiment, the optical sensor 200 may include a CCD (Charge-coupled Device) sensor or a CMOS (Complementary Metal Oxide Semiconductor) sensor.
[0059] In one embodiment, the first modulation area 11 and the second modulation area 12 are arranged at the metasurface light-guiding element. The first modulation area 11 of the metasurface light-guiding element modulates the input light and outputs it to the optical sensor 200, and the optical sensor 200 generates first image data according to the output light of the first modulation area 11. The second modulation area 12 of the metasurface light-guiding element modulates the input light and outputs it to the optical sensor 200, and the optical sensor 200 generates second image data according to the output light of the second modulation area 12.
[0060] In one embodiment, the first modulation area and the second modulation area are arranged at the optical sensor 200. Light is input to the first modulation area for modulation, and the optical sensor 200 generates first image data according to the modulated light of the first modulation area. Light is input to the second modulation area for modulation, and the optical sensor 200 generates second image data according to the modulated light of the second modulation area.
[0061] The first image data corresponds to a first part P1 of the target object TO.
[0062] The first part P1 may be the whole or part of the target object TO that can be imaged. The first part P1 may be any area of the target object TO, and may also include the whole target object TO and the surrounding area of the target object TO.
[0063] The second image data corresponds to a second part P2 of the target object TO.
[0064] The second part P2 may be the whole or part of the target object TO that can be imaged. The second part P2 may be any area of the target object TO, or may include the whole target object TO and the surrounding area of the target object TO.
[0065] In one embodiment, the first portion P1 is different from the second portion P2.
[0066] The first part P1 and the second part P2 point to different parts of the target object TO. The first part P1 and the second part P2 may overlap or not overlap. The coverage of the first part P1 may be smaller than the second part P2, or greater than or equal to the second part P2.
[0067] In one embodiment, the first portion P1 is included in the second portion P2.
[0068] The first part P1 points to a part in the second part P2. The coverage of the first part P1 is smaller than that of the second part P2.
[0069] In this way, different independent parts of the target object can be imaged separately, or a certain part of the target object and its local part can be imaged separately, to assist in subsequent zoning comparison and analysis; since the two modulation areas forming the metasurface structure are used to form modulated light for the two parts respectively, the system can obtain image data in a unified manner, which not only saves costs, but also avoids the viewing angle error caused by configuring two imaging systems due to the setting of the two modulation areas, further improving the imaging efficiency and the accuracy of the image algorithm.
[0070] In one embodiment, the second portion P2 corresponds to the entirety of the target object TO.
[0071] The second image data corresponding to the second part P2 contains information of imaging the entire target object TO. The partition imaging system can at least generate image data corresponding to the entire target object TO for subsequent comparison and reference.
[0072] In one embodiment, the first portion P1 corresponds to a part of the target object TO.
[0073] The first image data corresponding to the first part P1 contains information of local imaging of the target object TO. The subarea imaging system can at least generate image data corresponding to a part of the target object TO for subsequent comparison and reference.
[0074] For different target objects TO, the first part P1 may point to the same position.
[0075] For example, the first position P1 may point to the upper half of the target object TO in the vertical direction.
[0076] For example, when the target object TO is the area above the shoulders of a person, the first part P1 may point to the head or eyes of different people.
[0077] Figure 2 The first image IMG1 corresponding to the first image data and the second image IMG2 corresponding to the second image data in this specific embodiment are shown. It can be seen that the first image IMG1 shows a part of the eye of the character, the corresponding first image data contains data information of the part of the eye, and the first part P1 is the part of the eye of the character. The second image IMG2 shows the entire area above the shoulders of the character, the corresponding second image data contains data information of the entire area above the shoulders of the character, and the second part P2 is the entire area above the shoulders of the character.
[0078] For different target objects TO, the first part P1 may point to different positions.
[0079] The first part P1 is determined according to the target object TO. The first part P1 is a region of interest of the imaging system.
[0080] For example, when the target object TO is the area above the shoulders of the person, the first part P1 may point to the head or eyes. When the target object TO is the area below the waist of the person, the first part P1 may point to the feet.
[0081] In other examples, the target object TO may be the whole or part of a person, an animal, or a device structure, and the first part P1 may be a predetermined part of interest of the target object TO.
[0082] In one embodiment, the range of the first portion P1 is smaller than the range of the second portion P2, and the field of view of the corresponding first modulation zone 11 is smaller than the field of view of the second modulation zone 12. That is, the field of view angle of the first modulation zone 11 is smaller than the field of view angle of the second modulation zone 12. In optical engineering, the field of view angle is also called the field of view, and the size of the field of view angle determines the field of view range of the optical instrument (here, the first modulation zone 11 and the second modulation zone 12). In this way, the first image corresponding to the first portion P1 displays a small range of the target object TO, and the second image corresponding to the second portion P2 displays a large range of the target object TO.
[0083] In a specific embodiment, the field of view of the first modulation area 11 with respect to the target object TO is configured to cover at least the first part P1 of the target object TO; in a specific embodiment, the field of view of the second modulation area 12 with respect to the target object TO is configured to cover at least the second part P2 of the target object TO. Specifically, the second part P2 of the target object TO can be the entire target object TO.
[0084] The field of view range / field of view angle of the modulation zone (first modulation zone 11 or second adjustment zone 12) can be adjusted by adjusting the focal length of the corresponding metasurface structure or metasurface light-guiding element or the shape, posture, and position of the nanostructure unit; the focal length of the metasurface structure or metasurface light-guiding element can be adjusted by adjusting the shape, posture, and position of the nanostructure unit, or the positional relationship between the metasurface structure and the target object TO, or the positional relationship between the metasurface structure and the optical sensor.
[0085] In one embodiment, the field of view is adjusted by the focal length of the metasurface structure. The focal length of the metasurface structure at the first modulation area 11 is configured to be greater than the focal length of the metasurface structure at the second modulation area 12.
[0086] In one embodiment, the accuracy of the first image is higher than that of the second image, which is helpful for distinguishing and analyzing the two images, for example, using the first image with high accuracy for detail analysis and using the second image with low accuracy for contour and trend analysis.
[0087] The accuracy can be defined as the accuracy and clarity of the image in describing and expressing the details of the real world. The accuracy can be at least one of resolution, color depth, signal-to-noise ratio, dynamic range, spatial frequency response, or is related to these parameters, or a combination of at least two of these parameters.
[0088] The focal length of the metasurface structure at the first modulation area 11 and / or the second modulation area 12 or the shape, posture, and position of the nanostructure unit may affect the accuracy of the corresponding image.
[0089] In one embodiment, the accuracy may be a training threshold of a prediction system including a partition imaging system and a neural network model. The prediction system outputs a prediction result for a specific service by using the partition imaging system in conjunction with a preset neural network model.
[0090] In a specific embodiment, the neural network model includes a first neural network model and a second neural network model.
[0091] The first neural network model is used to output a first prediction result of a first service based on the first image.
[0092] The second neural network model is used to output a second prediction result of the second business based on the second image.
[0093] The accuracy includes a first accuracy corresponding to the first neural network model and a second accuracy corresponding to the second neural network model. In one example, the first accuracy is higher than the second accuracy.
[0094] In a specific example, the first neural network model is a line of sight prediction neural network model.
[0095] In a specific example, the second neural network model is a posture prediction neural network model.
[0096] In a specific example, a first neural network model predicts the line of sight of a target object based on a first image generated by a partitioned imaging system, and a second neural network model predicts the posture of the target object based on a second image generated by the partitioned imaging system, so as to ultimately determine the focus of the target object based on the line of sight and posture of the target object.
[0097] In the first specific example, when training the prediction system, the first accuracy is used as the training threshold, and the first loss value is determined by the predicted line of sight and line of sight label of the target object. When the first loss value is less than the first accuracy, the configuration of the supersurface structure corresponding to the first modulation area 11 of the partitioned imaging system is determined by training.
[0098] In the second specific example, the second precision is used as the training threshold, and the second loss value is determined by the predicted posture and posture label of the target object. When the second loss value is less than the second precision, the configuration of the supersurface structure corresponding to the second modulation area 12 of the partitioned imaging system is determined by training.
[0099] The first specific example and the second specific example may be implemented one by one, separately, or simultaneously.
[0100] In a specific embodiment, the first image corresponds to a part of the target object, and the second image corresponds to the entire target object, which is helpful for observing and analyzing the details of the designated part.
[0101] In one embodiment, the first modulation area 11 and the second modulation area 12 are formed at the same metasurface light-guiding element (eg Figure 1 or Figure 5 , a third metasurface light-guiding element 100).
[0102] In one embodiment, the partitioned imaging system includes a third metasurface light-guiding element 100 ; the third metasurface light-guiding element 100 is used to form a first modulation zone 11 and a second modulation zone 12 .
[0103] In a specific embodiment, the focal length of the metasurface structure at the first modulation area 11 and the metasurface structure at the second modulation area 12 is adjusted by adjusting the shape, posture, and position of the nanostructure unit. The first modulation area 11 and the second modulation area 12 are arranged in the same plane between the target object TO and the optical sensor.
[0104] like Figure 3 As shown, the third metasurface light-guiding element 100 can be configured to have a rectangular extension surface; the first modulation area 11 occupies a part of the extension surface, and the second modulation area 12 occupies another part of the extension surface. A plurality of nanostructure units 10 are arranged in the first modulation area 11 and the second modulation area 12. The nanostructure units located in the first modulation area 11 form a metasurface structure, and the nanostructure units located in the second modulation area 12 form another metasurface structure.
[0105] In one embodiment, the first modulation area 11 and the second modulation area 12 are set in two independent areas, and the ranges of the first modulation area 11 and the second modulation area 12 do not overlap.
[0106] like Figure 4As shown, the third metasurface light-guiding element 100 can be configured to have a circular or elliptical extension surface; the second modulation zone 12 occupies a partial area of the extension surface, and the first modulation zone 11 occupies a partial area within the second modulation zone 12.
[0107] In one embodiment, the first modulation area 11 and the second modulation area 12 are arranged in two areas that intersect each other, and the ranges of the first modulation area 11 and the second modulation area 12 overlap.
[0108] In one embodiment, the first modulation area 11 and the second modulation area 12 are formed at different metasurface light-guiding elements (eg Figure 6 or Figure 7 , a first supersurface light-guiding element 101, a second supersurface light-guiding element 102).
[0109] In one embodiment, the partitioned imaging system includes a first supersurface light-guiding element 101; the first supersurface light-guiding element 101 is used to form a first modulation zone 11.
[0110] In one embodiment, the partitioned imaging system includes a second metasurface light-guiding element 102 ; the second metasurface light-guiding element 102 is used to form a second modulation zone 12 .
[0111] The first metasurface light-guiding element 101 can be Figure 3 , Figure 4 The partial configuration corresponding to the first modulation area 11, or at least including Figure 3 , Figure 4 The part corresponding to the first modulation area 11.
[0112] The second metasurface light-guiding element 102 may be Figure 3 , Figure 4 The partial configuration corresponding to the second modulation area 12, or at least including Figure 3 , Figure 4 The portion corresponding to the second modulation zone 12.
[0113] The first metasurface light-guiding element 101 and the second metasurface light-guiding element 102 are arranged in the partitioned imaging system, and can be specifically arranged close to each other to improve the compactness of the structure, reduce the complexity of the optical path design, and avoid errors.
[0114] In one embodiment, the first metasurface light-guiding element 101 receives input light from the first portion P1 of the target object TO, modulates it, and outputs it to the optical sensor ( Figure 6 The first processing zone 21 shown in FIG. Figure 7 When the first part P1 is a part of the target object TO, local optical imaging can be achieved through the first metasurface light-guiding element 101.
[0115] In one embodiment, the second metasurface light-guiding element 102 receives input light from the first portion P2 of the target object TO, modulates it, and then outputs it to the optical sensor ( Figure 6 The second processing zone 22 shown in FIG. Figure 7 When the second part P2 is the entire target object TO, the second metasurface light-guiding element 102 can realize overall optical imaging.
[0116] In one embodiment, the range of the first part P1 is smaller than the range of the second part P2; the field of view of the first metasurface light-guiding element 101 with respect to the target object TO is smaller than the field of view of the second metasurface light-guiding element 102 with respect to the target object TO; the focal length of the first metasurface light-guiding element 101 is greater than the focal length of the second metasurface light-guiding element 102.
[0117] In one embodiment, the modulated light at the first modulation area 11 and the second modulation area 12 is detected by the same optical sensor (eg Figure 1 or Figure 6 , received by the optical sensor 200).
[0118] In one embodiment, the first image data corresponding to the first part P1 of the target object TO and the second image data corresponding to the second part P2 are obtained by the same optical sensor (eg, Figure 1 or Figure 6 , optical sensor 200) generates.
[0119] In one embodiment, the optical sensor 200 includes a first processing area 21; the first processing area 21 is used to generate first image data.
[0120] After the light from the first part P1 of the target object TO is modulated by the first modulation zone 11, the modulated light is at least input into the first processing zone 21, and the first processing zone 21 generates first image data according to the modulated light. The first image corresponds to the first part P1 of the target object TO, and the first image or target image formed thereby can show the state of the first part P1. When the first part P1 is a part of the target object TO, local optical imaging can be achieved through the first processing zone 21.
[0121] In one embodiment, the optical sensor 200 includes a second processing area 22; the second processing area 22 is used to generate second image data.
[0122] After the light from the second part P2 of the target object TO is modulated by the second modulation area 12, the modulated light is at least input into the second processing area 22, and the second processing area 22 generates second image data according to the modulated light. The second image corresponds to the second part P2 of the target object TO, and the second image or target image formed thereby can show the state of the second part P2. When the second part P2 is the entirety of the target object TO, the overall optical imaging can be achieved through the second processing area 22.
[0123] In a specific embodiment, the first processing area 21 and the second processing area 22 are arranged in the same plane behind the target object TO. In an embodiment, the first processing area 21 and the second processing area 22 are arranged in the same plane behind the first modulation area 11 and the second modulation area 12.
[0124] In a specific embodiment, the first processing area 21 and the second processing area 22 are arranged in two independent areas, and the ranges of the first processing area 21 and the second processing area 22 do not overlap.
[0125] In a specific embodiment, the first processing area 21 and the second processing area 22 are arranged in two areas that intersect each other, and the ranges of the first processing area 21 and the second processing area 22 overlap.
[0126] In one embodiment, the first processing area 21 and the second processing area 22 are formed on different optical sensors (eg Figure 5 or Figure 7 , a first optical sensor 201 and a second optical sensor 202).
[0127] In one embodiment, the first image data corresponding to the first part P1 of the target object TO and the second image data corresponding to the second part P2 are obtained by different optical sensors (eg, Figure 5 or Figure 7 , the first optical sensor 201 and the second optical sensor 202) generate.
[0128] In one embodiment, the partition imaging system includes a first optical sensor 201; the first optical sensor 201 is used to generate first image data.
[0129] In a specific embodiment, the first optical sensor 201 may be configured as the aforementioned optical sensor 200 .
[0130] After the light from the first part P1 of the target object TO is modulated by the first modulation area 11, the modulated light is at least input into the first optical sensor 201, and the first optical sensor 201 generates first image data according to the modulated light. The first image corresponds to the first part P1 of the target object TO, and the first image or target image formed thereby can show the state of the first part P1. When the first part P1 is a part of the target object TO, local optical imaging can be achieved through the first optical sensor 201.
[0131] In this embodiment, the first optical sensor 201 may include the aforementioned first processing area 21 .
[0132] In one embodiment, the optical sensor 200 includes a second optical sensor 202; the second optical sensor 202 is used to generate second image data.
[0133] After the light from the second part P2 of the target object TO is modulated by the second modulation area 12, the modulated light is at least input into the second optical sensor 202, and the second optical sensor 202 generates second image data according to the modulated light. The second image corresponds to the second part P2 of the target object TO, and the second image or target image formed thereby can show the state of the second part P2. When the second part P2 is the entirety of the target object TO, the entirety optical imaging can be achieved through the second optical sensor 202.
[0134] In this embodiment, the second optical sensor 202 may include the aforementioned second processing area 22 .
[0135] In a specific embodiment, the first optical sensor 201 and the second optical sensor 202 can be defined as being included in the aforementioned optical sensor 200 ; the optical sensor 200 includes the first optical sensor 201 and the second optical sensor 202 .
[0136] In one embodiment, the range of the first part P1 is smaller than the range of the second part P2; the field of view of the first metasurface light-guiding element 101 with respect to the target object TO is smaller than the field of view of the second metasurface light-guiding element 102 with respect to the target object TO; the focal length of the first metasurface light-guiding element 101 is greater than the focal length of the second metasurface light-guiding element 102.
[0137] In one embodiment, the configuration of the metasurface structure at the first modulation zone 11 and the second modulation zone 12 is designed based on computer optical simulation. The focal length of the metasurface structure at the modulation zone can be adjusted based on the optical simulation. The optical simulation can specifically adjust the configuration parameters of the nanostructure unit at the modulation zone to achieve the setting of the focal length at the modulation zone. In this way, the modulation intensity of the nanostructure unit to different input lights containing image information can be accurately predicted, thereby obtaining the first image data corresponding to the first modulation zone 11 and the second image data corresponding to the second modulation zone 12.
[0138] like Figure 1 , Figures 5 to 7 As shown, the configuration of the metasurface structure at the first modulation area 11 is determined according to the first portion P1 of the target object TO.
[0139] The metasurface structure at the first modulation zone 11 includes the properties of the metasurface structure as a whole (for example, focal length), the configuration of the nanostructure units therein (for example, the shape, posture, and position of the nanostructure units), the positional relationship between the metasurface structure and the target object TO (for example, the distance between the first modulation zone 11 and the first part P1), and the positional relationship between the metasurface structure and the optical sensor (for example, the distance between the first modulation zone 11 and the first processing zone 21 or the first optical sensor 201).
[0140] The configuration of the metasurface structure can be determined according to the position, size, and optical properties (eg, brightness, color) of the first portion P1.
[0141] In one embodiment, the configuration of the metasurface structure at the first modulation area 11 is determined according to the size of the first part P1 of the target object TO. In this way, the first modulation area 11 can image the target object TO within a preset range according to the first part P1 and form a first image with a preset accuracy.
[0142] In a specific embodiment, the focal length of the metasurface structure corresponding to the first modulation area 11 is determined according to the size of the first portion P1.
[0143] In a specific example, the focal length is negatively correlated with the size. When the first portion P1 is small, the first modulation area 11 can be set to have a larger focal length; otherwise, the first modulation area 11 can be set to have a smaller focal length.
[0144] The configuration of the metasurface structure at the second modulation region 12 is determined according to the second portion P2 of the target object TO.
[0145] The metasurface structure at the second modulation zone 12 includes the properties of the metasurface structure as a whole (e.g., focal length), the configuration of the nanostructure units therein (e.g., the shape, posture, and position of the nanostructure units), and the positional relationship between the metasurface structure and the target object TO (e.g., the distance between the second modulation zone 12 and the second part P2).
[0146] The configuration of the metasurface structure can be determined based on the position, size, and optical properties (eg, brightness, color) of the second portion P2.
[0147] In one embodiment, the configuration of the metasurface structure at the second modulation area 12 is determined according to the size of the second portion P2 of the target object TO. In this way, the second modulation area 12 can image the target object TO within a preset range according to the second portion P2, and form a second image with a preset accuracy.
[0148] In a specific embodiment, the focal length of the metasurface structure corresponding to the second modulation area 12 is determined according to the size of the second portion P2.
[0149] In a specific example, the focal length is negatively correlated with the size. When the second portion P2 is larger, the second modulation area 12 can be set to have a smaller focal length; otherwise, the second modulation area 12 can be set to have a larger focal length.
[0150] The focal length of the metasurface structure at the first modulation area 11 or the second modulation area 12 can be adjusted in the following manner.
[0151] In one embodiment, the focal length of the metasurface structure is adjusted by adjusting the size of the nanostructure units.
[0152] The size includes at least one of length, width and height.
[0153] By adjusting the size parameters, the response of the nanostructure units to different wavelengths of light can be precisely controlled to achieve the desired optical effects, such as a specific focal length or phase distribution.
[0154] In one embodiment, the focal length of the metasurface structure is adjusted by adjusting the rotation angle of the nanostructure unit.
[0155] The rotation angle refers to the rotation angle of the nanostructure unit relative to a reference direction (usually the light propagation direction or a symmetry axis of the lens). Adjusting the rotation angle can change the response of the nanostructure unit to the polarization state of light.
[0156] The rotation angle can change the response of the nanostructure unit to light of different polarization states, such as changing the phase delay or amplitude attenuation of the transmitted light. By rotating the nanostructure unit, fine control of the light wavefront can be achieved to generate a specific phase distribution and focal length.
[0157] In one embodiment, the focal length of the metasurface structure is adjusted by adjusting the duty cycle of the nanostructure units.
[0158] Duty cycle refers to the ratio of the "filled" part to the "empty" part of a nanostructure unit within a period. It is often used to describe nanopatterns in periodic structures.
[0159] Variations in the duty cycle change the overall optical properties of the periodic nanostructure units, such as reflectivity, transmittance, and diffraction efficiency.
[0160] In one embodiment, the configuration of the metasurface structure at the first modulation area 11 is determined according to the target accuracy of the first image. In this way, the image accuracy can be adjusted by adjusting the metasurface structure, or the metasurface structure can be reversely adjusted according to the set image accuracy.
[0161] In one embodiment, the configuration of the metasurface structure at the second modulation area 12 is determined according to the target accuracy of the second image. In this way, the image accuracy can be adjusted by adjusting the metasurface structure, or the metasurface structure can be reversely adjusted according to the set image accuracy.
[0162] It can be seen that the target accuracy can be set for the image, and the configuration of the hypersurface structure corresponding to the image can be adjusted according to the target accuracy.
[0163] The target accuracy of the first image and the target accuracy of the second image may be the same or different. For the convenience of description, the target accuracy of the first image is defined as the first target accuracy, and the target accuracy of the second image is defined as the second target accuracy.
[0164] In a specific embodiment, the configuration process of the above-mentioned metasurface structure can be implemented by optical simulation. Specifically, the metasurface structure at the first modulation area 11 (or the focal length of the first modulation area 11) and the metasurface structure at the second modulation area 12 (or the focal length of the second modulation area 12) can be adjusted by optical simulation until the following conditions are met:
[0165] The first image meets a preset first target accuracy; and / or,
[0166] The second image meets a preset second target accuracy.
[0167] The focal length of the metasurface structure at the first modulation area 11 or the first modulation area 11 and the focal length of the metasurface structure at the second modulation area 12 or the second modulation area 12 can be adjusted simultaneously or separately.
[0168] In one embodiment, the configuration of the metasurface structure at the first modulation area 11 and the configuration of the metasurface structure at the second modulation area 12 are determined according to the target accuracy relationship between the first image and the second image. In this way, the image accuracy relationship can be adjusted by adjusting the metasurface structure, or the metasurface structure can be reversely adjusted according to the set image accuracy relationship.
[0169] It can be seen that a target accuracy relationship can be set for the first image and the second image, and the configuration of the hypersurface structure corresponding to the first image and the second image can be adjusted according to the target accuracy relationship.
[0170] In a specific embodiment, the metasurface structure at the first modulation area 11 (or the focal length of the first modulation area 11) and the metasurface structure at the second modulation area 12 (or the focal length of the second modulation area 12) can be continuously adjusted by optical simulation until the following conditions are met:
[0171] The first image and the second image satisfy a preset target accuracy relationship.
[0172] The focal length of the metasurface structure at the first modulation area 11 or the first modulation area 11 and the focal length of the metasurface structure at the second modulation area 12 or the second modulation area 12 can be adjusted simultaneously.
[0173] In one embodiment, the partition imaging system further includes a processor 300 .
[0174] The processor 300 is coupled to the optical sensor. Figure 1 and Figure 6 As shown, the processor 300 is coupled to the optical sensor 200; Figure 5 and Figure 7 As shown, the processor 300 is coupled to the first optical sensor 201 and the second optical sensor 202 .
[0175] The processor 300 may be an integrated circuit chip with signal processing capability. In the implementation process, each step of the method for processing image data or the method for determining the partition imaging system may be completed by the hardware integrated logic circuit or software instructions in the processor 300. The above-mentioned processor 300 may be a general-purpose processor, a DSP (Digital Signal Processor), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 300 may implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention may be directly embodied as being executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in the decoding processor.
[0176] The processor 300 is configured to generate a target image according to the first image data and the second image data.
[0177] The target image includes information of the first image data and the second image data, and a comprehensive analysis of multiple parts of the target object can be performed based on the target image.
[0178] The target image includes the first image and the second image. Thus, the first image and the second image can be compared and analyzed based on the target image to provide a comprehensive reference for subsequent operations.
[0179] The target image may be determined according to the image data by generating a first image and a second image according to the target image, and then processing the two images to determine the target image.
[0180] For example, Figure 2 As shown, the first image and the second image are spliced, for example, the first image and the second image are mixed or superimposed in proportion.
[0181] Before performing the above operations, the first image and the second image may be enlarged or reduced.
[0182] Another method of determining the target image based on the image data is to directly process the data and then determine the target image based on the processed target image data.
[0183] For example, the image data is input into a neural network model or other computing model, and after data processing or result prediction, the target image data is determined and the target image is determined accordingly. The neural network model can be used for image encryption, morphological analysis, image quality improvement, etc.
[0184] In one embodiment, the partition imaging system further includes a processor 300 .
[0185] The processor 300 is configured to generate a second image corresponding to the second part P2 according to the second image data.
[0186] In a specific embodiment, the second image is an encrypted image. The encryption operation can be performed by the processor 300. When the second part P2 is the entire target object TO, the processor 300 can achieve the effect of encrypting the entire target object TO.
[0187] When the first part P1 is a part of the target object TO, combined with the above-mentioned encrypted imaging, the processor 300 can realize overall encrypted imaging and local clear imaging, provide users with reference through differentiated imaging, and keep non-interested areas confidential.
[0188] The encrypted image is determined by blurring. Figure 8As shown, the second image data may be blurred by optical encryption or electrical encryption to determine the second image IMG2.
[0189] In a preferred embodiment, the accuracy of the first image IMG1 can be kept higher than the accuracy of the second image IMG2, so that local clear imaging can be achieved.
[0190] In one scenario, the first image IMG1 images the eyes, and the second image IMG2 images the entire upper body of the person. When applied to DMS, the overall image of the person can be kept confidential, and only the details of the eyes are retained for identity recognition, driver off-duty inspection, fatigue detection, distraction detection, smoking detection, phone call detection, remote monitoring and warning, driver emotion analysis and assisted gesture control, etc. It can also assist in seat and rearview mirror adjustment, etc.
[0191] Fuzzy processing can be achieved through neural network models.
[0192] The neural network model is pre-trained and determined. For example, the target blur level or target blurred image of the sample image of the first part P1 is first determined, the loss value is determined by the difference between the target blur level or the target blurred image and the sample image, and the neural network model is determined by training according to the loss value.
[0193] In a specific embodiment, the second image is a posture image. The posture analysis operation can be performed by the processor 300. When the second part P2 is the entire target object TO, the processor 300 can achieve the effect of posture analysis on the entire target object TO.
[0194] When the first part P1 is a part of the target object TO, in combination with the above-mentioned posture analysis, the processor 300 can realize overall posture analysis and local original imaging, and provide the user with reference through differential imaging.
[0195] The posture image includes a mark indicating the current posture of the second part P2. Fig. 9 As shown, the second image IMG2 can be determined by posture analysis. The second image IMG2 can include the entire person, nodes indicating the joints of the person, and connection lines connecting the nodes and indicating the limbs of the person.
[0196] The posture image can be applied to the above-mentioned DMS to assist in realizing functions such as driver departure inspection, smoking detection, phone call detection and assisted gesture control.
[0197] Posture analysis can be achieved through neural network models.
[0198] The neural network model is pre-trained and determined. For example, the posture indication mark of the sample image of the first part P1 is first determined, the loss value is determined by the difference between the posture indication mark and the posture reasoning result of the sample image, and the neural network model is determined by training according to the loss value.
[0199] The aforementioned neural network model can be determined independently or in unison with the supersurface structure in the first modulation zone 11 and / or the second modulation zone 12.
[0200] The aforementioned neural network can be mounted in the processor 300, or in a storage module independent of the processor 300, and when the processor 300 executes the neural network model stored in the storage module, at least one of the above-mentioned encryption operations or posture analysis is implemented.
[0201] In one embodiment, the neural network model is determined based on a target accuracy of the first image.
[0202] In one embodiment, the neural network model is determined based on a target accuracy of the second image.
[0203] It can be seen that the target accuracy can be set for the image and the neural network model can be adjusted according to the target accuracy.
[0204] In one embodiment, the neural network model is determined based on a target accuracy relationship between the first image and the second image.
[0205] It can be seen that a target accuracy relationship can be set for the first image and the second image, and the neural network model can be adjusted according to the target accuracy relationship.
[0206] In one embodiment, the neural network model is determined by training based on the loss value between the second image data and the sample image data.
[0207] In a specific embodiment, the second image is an encrypted image, and the loss value is determined according to the difference between the blur degree determined by the second image data and the target blur degree determined by the sample image data, or the loss value is determined according to the difference between the second image data and the sample image data.
[0208] In a specific embodiment, the second image is a posture image, and the loss value is determined based on the difference between the posture indication mark determined by the second image data and the posture indication mark determined by the sample image, or the loss value is determined based on the difference between the second image data and the sample image data.
[0209] The target accuracy of the first image and the target accuracy of the second image may be the same or different.
[0210] In a specific embodiment, the supersurface structure at the first modulation area 11 (or the focal length of the first modulation area 11), the supersurface structure at the second modulation area 12 (or the focal length of the second modulation area 12), and the neural network model can be adjusted until the following conditions are met:
[0211] The first image meets a preset first target accuracy; and / or the second image meets a preset second target accuracy; and,
[0212] The loss value between the second image data and the sample image data satisfies a set condition; the set condition is such as: convergence.
[0213] Continue as Figure 1 , Figures 5 to 7 In one embodiment, the partition imaging system further includes a display module 400 .
[0214] The display module 400 is used to display the first image; the display module 400 is used to display the second image; and the display module 400 is used to display the target image.
[0215] The display module 400 may be a display screen integrated in the partition imaging system. When the partition imaging system is integrated in the partition imaging module, it is arranged on one side of the outer surface of the partition imaging module to realize user interaction.
[0216] The display module 400 can also be defined as a display device outside the partition imaging system. When the partition imaging system is integrated in the partition imaging module, it is arranged outside the partition imaging module. The user can monitor the image output or control the partition imaging module by operating the display device.
[0217] Figure 1 , Figures 5 to 7 Four embodiments of a zoned imaging system are shown, and the differences between the embodiments are listed below.
[0218] Figure 1 In the illustrated embodiment, the first modulation zone 11 and the second modulation zone 12 are arranged at the same third metasurface light-guiding element 100; the modulated light output by the two modulation zones is received by the same optical sensor 200; the optical sensor 200 includes a first processing zone 21 corresponding to the first modulation zone 11; the optical sensor 200 includes a second processing zone 22 corresponding to the second modulation zone 12.
[0219] Figure 5 In the illustrated embodiment, the first modulation zone 11 and the second modulation zone 12 are arranged at the same third metasurface light-guiding element 100; the modulated light output by the two modulation zones is received by different optical sensors; the modulated light output by the first modulation zone 11 is received by the first optical sensor 201; and the modulated light output by the second modulation zone 12 is received by the second optical sensor 202.
[0220] Figure 6 In the illustrated embodiment, the first modulation zone 11 and the second modulation zone 12 are arranged at different metasurface light-guiding elements; the first modulation zone 11 is arranged at the first metasurface light-guiding element 101; the second modulation zone is arranged at the second metasurface light-guiding element 102; the modulated light output by the two modulation zones is received by the same optical sensor 200; the optical sensor 200 includes a first processing zone 21 corresponding to the first modulation zone 11; the optical sensor 200 includes a second processing zone 22 corresponding to the second modulation zone 12.
[0221] Figure 7 In the illustrated embodiment, the first modulation zone 11 and the second modulation zone 12 are arranged at different metasurface light-guiding elements; the first modulation zone 11 is arranged at the first metasurface light-guiding element 101; the second modulation zone is arranged at the second metasurface light-guiding element 102; the modulated light output by the two modulation zones is received by different optical sensors; the modulated light output by the first modulation zone 11 is received by the first optical sensor 201; the modulated light output by the second modulation zone 12 is received by the second optical sensor 202.
[0222] like Fig.10 , an embodiment of the present invention provides a method for determining a partitioned imaging system.
[0223] The method can be used to fix the structural features of components in a partition imaging system, such as the structural features of the components themselves, or the positional relationship between components.
[0224] This method can be used to configure operation parameters in a fixed partition imaging system, such as image processing methods, neural network model parameters, etc.
[0225] The partition imaging system can be configured as the technical solution provided in any of the above embodiments.
[0226] The method for determining a partitioned imaging system specifically includes at least one of the following steps.
[0227] Step S1, obtaining a first precision value of a first image corresponding to a first modulation area, and a second precision value of a second image corresponding to a second modulation area.
[0228] The first image corresponds to the first modulation area, and a first precision value of the first image can be determined, and the first modulation area can be adjusted based on the first precision value.
[0229] The second image corresponds to the second modulation area, and the second precision value of the second image can be determined, and the second modulation area can be adjusted based on the second precision value.
[0230] Step S2, determining the difference between the first accuracy value and the set target accuracy, and adjusting the focal length of the metasurface structure corresponding to the first modulation area according to the difference.
[0231] Step S3, determining the difference between the second precision value and the set target precision, and adjusting the focal length of the metasurface structure corresponding to the second modulation area according to the difference
[0232] In one embodiment, step S2 and step S3 are performed synchronously.
[0233] In one embodiment, step S2 and step S3 are performed separately.
[0234] The configuration scheme and explanation of the target accuracy, as well as the adjustment method and explanation of the focal length of the metasurface structure, can be implemented with reference to the previous text.
[0235] In one embodiment, the partition imaging system further includes: a neural network model for generating a second image corresponding to the second part based on the second image data.
[0236] In a specific embodiment, the second image is an encrypted image. The encrypted image can be explained with reference to the technical solution provided above.
[0237] In a specific embodiment, the second image is a posture image. The posture image can be explained with reference to the technical solution provided above.
[0238] In one embodiment, Fig.11 As shown, the method for determining a partitioned imaging system includes at least one of the following steps.
[0239] Step S11, obtaining sample image data corresponding to the second part.
[0240] Step S12, determining a loss value according to the second image data and the sample image data.
[0241] Step S13, according to the loss value, adjusting the focal length and / or neural network model parameters of the metasurface structure corresponding to the second modulation area to obtain a trained neural network model.
[0242] The sample image is used to provide the second image data for training reference. When the second image is an encrypted image, an image that meets the target blur level and corresponds to the second part can be determined as the sample image. When the second image is a posture image, an image that includes a posture indicator and corresponds to the second part can be determined as the sample image.
[0243] The sample image data may be data information of the sample image. In this embodiment, the second image data used to determine the loss value may be data information of an image inferred based on the current system and model.
[0244] The sample image data may also be a target blur level or an expected posture indicator. In this embodiment, the second image data used to determine the loss value may be a blur level or a posture indicator corresponding to an image inferred by the current system or model.
[0245] In one embodiment, the metasurface structure at the second modulation region is adjusted according to the loss value. The adjustment content of the metasurface structure can be configured according to the technical solution provided in any of the above embodiments.
[0246] In one embodiment, the focal length of the metasurface structure corresponding to the second modulation area is adjusted according to the loss value. The focal length adjustment method can be configured according to the technical solution provided in any of the above embodiments.
[0247] In one embodiment, the neural network model parameters are adjusted based on the loss value.
[0248] An embodiment of the present invention provides a computer storage medium.
[0249] The computer storage medium stores the computer program executed by the processor mentioned above, or the instructions of the method for determining the partition imaging system in any of the technical solutions mentioned above.
[0250] When the processor executes the computer program, it can execute the description of the method for determining the partition imaging system in any of the technical solutions below, so it will not be repeated here. In addition, the description of the beneficial effects of using the same method will not be repeated. For technical details not disclosed in the computer storage medium embodiment involved in the present invention, please refer to the description of the method embodiment of the present invention.
[0251] The computer storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.
[0252] An embodiment of the present invention provides a computer program product.
[0253] The computer program product may include a computer program, which may be stored in a computer storage medium. A processor of an electronic device reads the computer program from the computer storage medium, and the processor may execute the computer program, so that the electronic device executes the description of the method for determining a partition imaging system in any of the technical solutions below.
[0254] In summary, the partitioned imaging system provided by the present invention provides two modulation zones to respectively form modulated lights corresponding to two parts of the target object, and the optical sensor generates two items of image data corresponding to the two modulated lights, so that the partitioned imaging system can uniformly obtain image data corresponding to at least two parts of the target object. Compared with configuring two imaging systems, it can not only save costs, but also avoid the introduction of errors due to the difference in imaging angles of the two imaging systems, thereby helping to improve imaging efficiency and the accuracy of image algorithms.
[0255] It should be understood that although this specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each implementation mode may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.
[0256] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent implementation methods or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.
Claims
1. A partition imaging system, characterized in that: include: A first modulation region and a second modulation region are used to form a metasurface structure; an optical sensor for generating first image data based on the modulated light at the first modulation area, and for generating second image data based on the modulated light at the second modulation area; The first image data corresponds to a first part of the target object, and the second image data corresponds to a second part of the target object; the first part is different from the second part, or the first part is included in the second part.
2. The partition imaging system according to claim 1, characterized in that: The second part corresponds to the entire target object, and the first part corresponds to a part of the target object.
3. The partition imaging system according to claim 1, characterized in that: The accuracy of the first image is higher than the accuracy of the second image.
4. The partition imaging system according to claim 1, characterized in that: The optical sensor comprises: A first processing area, used to generate the first image data; A second processing area, used for generating the second image data; Alternatively, the partition imaging system comprises: a first optical sensor, configured to generate the first image data; The second optical sensor is used to generate the second image data.
5. The partition imaging system according to claim 1, characterized in that: include: A first metasurface light-guiding element, used to form a first modulation area; A second metasurface light-guiding element, used to form a second modulation area; Or, include: The third metasurface light-guiding element is used to form the first modulation area and the second modulation area.
6. The partition imaging system according to claim 1, characterized in that: The configuration of the metasurface structure at the first modulation zone is determined according to the size of the first part of the target object, and / or the configuration of the metasurface structure at the second modulation zone is determined according to the size of the second part of the target object.
7. The partition imaging system according to claim 6, characterized in that: The focal length of the metasurface structure corresponding to the first modulation area is determined according to the size of the first part; the focal length of the metasurface structure corresponding to the second modulation area is determined according to the size of the second part.
8. The partition imaging system according to claim 7, characterized in that: The focal length of the metasurface structure is adjusted by adjusting at least one of the following: The size of the nanostructured unit; The rotation angle of the nanostructure unit; Duty cycle of the nanostructure unit.
9. The partition imaging system according to claim 1, characterized in that: The configuration of the metasurface structure at the first modulation zone is determined according to the target accuracy of the first image, and / or, the configuration of the metasurface structure at the second modulation zone is determined according to the target accuracy of the second image, and / or, the configuration of the metasurface structure at the first modulation zone and the second modulation zone is determined according to the target accuracy relationship between the first image and the second image.
10. The partition imaging system according to claim 1, characterized in that: The partition imaging system further comprises: A processor is used to generate a target image including a first image and a second image according to the first image data and the second image data.
11. The partition imaging system according to claim 1, characterized in that: The partition imaging system further comprises: A processor, configured to generate a second image corresponding to the second part according to the second image data; the second image is an encrypted image and / or a posture image; The encrypted image is determined by blurring, and the posture image includes a mark indicating a current posture of the second part.
12. The partition imaging system according to claim 11, characterized in that: The processor generates the encrypted image and / or the posture image by implementing a neural network model; The neural network model is determined according to the target accuracy of the first image, and / or according to the target accuracy of the second image, and / or according to the target accuracy relationship between the first image and the second image, and / or according to the loss value training between the second image data and the sample image data.
13. A method for determining a partitioned imaging system, characterized in that: The partition imaging system is configured as claimed in any one of claims 1 to 12, and the method comprises: Obtaining a first precision value corresponding to a first image of a first modulation area, and a second precision value corresponding to a second image of a second modulation area; Determine the difference between the first precision value and the set target precision, adjust the focal length of the metasurface structure at the corresponding first modulation area according to the difference, and synchronously or separately The difference between the second precision value and the set target precision is determined, and the focal length of the metasurface structure at the corresponding second modulation area is adjusted according to the difference.
14. The method according to claim 13, characterized in that The partition imaging system further includes: a neural network model for generating a second image corresponding to the second part according to the second image data; the second image is an encrypted image and / or a posture image; the method includes: obtaining sample image data corresponding to the second part; determining a loss value based on the second image data and the sample image data; According to the loss value, the focal length and / or neural network model parameters of the metasurface structure corresponding to the second modulation area are adjusted to obtain a trained neural network model.
15. A computer storage medium having an application program stored thereon, characterized in that: When the application is executed, the steps of the method for determining a partitioned imaging system according to any one of claims 13 to 14 are implemented.
Citation Information
Cited By
Method and device for generating training and task processing results of photoelectric hybrid computing system
CN121351909A
