Storage system, information recording method and three-dimensional storage device

Abstract

The present invention relates to a storage system, an information recording method and a three-dimensional storage device. A plurality of information recording points are arranged in a transparent material, and an information recording material is arranged at each information recording point. Different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics can map different colors under light irradiation. The data stored in the transparent material is represented by the color reflected by the information recording point.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of data storage, and more specifically, to a three-dimensional memory and storage system as well as an information recording method in the data storage link. Background Art

[0002] Early memories were planar two-dimensional memories, but the number of transistors that two-dimensional memories could accommodate was limited, resulting in low storage density. In order to overcome the limitations of two-dimensional memory devices, the industry has developed memory devices with three-dimensional structures, such as three-dimensional flash memory. Conventional measures are to increase the storage unit integration density and data capacity by arranging transistor-based memory cells three-dimensionally on a semiconductor substrate.

[0003] Regarding the concept of three-dimensional, for example, a three-dimensional flash memory is a flash memory device in which transistor storage units are constructed in a three-dimensional stacked manner. In the three-dimensional flash memory structure, it usually includes a stacking structure with a storage function and an external circuit structure with an interconnection function. The stacking structure can be formed by alternatingly stacking gate layers and dielectric layers, and the word line contact located in the step area of ​​the stacking structure leads out the word line to realize the electrical connection between the storage unit and the external circuit.

[0004] For the internal memory, such as the three-dimensional storage device of three-dimensional flash memory, the flash memory is usually physically divided into multiple blocks and each block includes a certain number of storage units. The erase operation of flash memory is generally based on the block as the basic unit, that is, the smallest unit. There is basically no difference in the block division in conventional three-dimensional memory. Three-dimensional memories from different manufacturers or three-dimensional memories from the same manufacturer but for different customer customization requirements may have different block division standards.

[0005] The biggest drawback of traditional 3D memory is that it is based on semiconductor transistors, and high heat will be generated whether reading, writing, erasing or other data operations. In addition, many types of memory need to be powered on to retain data, and data loss often occurs when the power is off. Although non-volatile memory can keep data from being lost when the power is off, most non-volatile memories use floating gates to store charges to express data, and charges will also leak. For example, NAND FLASH storage cells will lose most of their charges in about a few years or decades without operation. This is a problem to be solved. Summary of the invention

[0006] The present application relates to an information recording method, which is characterized in that it mainly includes: setting a plurality of information recording points in a transparent material, each of which is provided with information recording materials and different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics map different colors under light irradiation; and representing the data stored in the transparent material by the color reflected by the information recording points.

[0007] The above-mentioned information recording method extracts image data mapped by its many information recording points under light irradiation from a transparent material provided in advance as a sample, and provides a series of data sets of the image data in a manner of changing preset characteristics; uses the data sets to train a neural network, and uses the trained neural network to infer the data represented by the images to be identified mapped by the information recording points of other transparent materials under light irradiation.

[0008] In the above-mentioned information recording method, the transparent material at least includes a transparent semiconductor wafer or a transparent ceramic wafer or a glass or an acrylic plate.

[0009] In the above-mentioned information recording method, the information recording material includes quantum dot material, and different preset characteristics include setting the particle radius of the quantum dot material to be different; or the information recording material includes gem material, and different preset characteristics include adjusting the proportions of various chemical components in the gem material to be different; or the information recording material includes three-primary color material for color mixing, and different preset characteristics include setting the color mixing ratio relationship of the three-primary color material to be different.

[0010] In the above-mentioned information recording method, a transparent material is divided into multiple pages, and multiple information recording points arranged on the plane of any page serve as a pixel point set of the any page, and each information recording point serves as a pixel point.

[0011] In the above information recording method, two pixels belonging to any two different pages but with the same address are set to not overlap each other; or two pixels belonging to any two different pages but with the same address are overlapped up and down.

[0012] In the above-mentioned information recording method, each page of the image containing all the pixels is read in parallel as parallel data at one time, that is, the data of one page is read at a time with the page as the minimum reading unit; or the data of each pixel of each page is read after the data is read, and then the data of the next pixel is read, that is, a single pixel is used as the minimum reading unit and the data of one pixel is read at a time in a serial manner.

[0013] In the above information recording method, all pixels located in different pages but with the same address are arranged in a spiral in a direction perpendicular to the page, and there is a spiral angle value between any pixel of the latter page and a pixel of the same address of the adjacent previous page.

[0014] The present application relates to a storage system, characterized in that it includes:

[0015] A transparent material, in which a plurality of information recording points are arranged, each of which is provided with information recording materials and different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics reflect different colors under light irradiation;

[0016] A computer is provided with a camera, the camera captures image data mapped by a plurality of information recording points of a transparent material as a sample under light irradiation, and provides a series of data sets of the image data by changing a preset characteristic (e.g., changing the preset characteristic of the information recording material of the plurality of information recording points on the transparent material of the sample), and the camera interacts (or transmits) the data sets to the computer;

[0017] The computer uses the data set to train a neural network, and uses the trained neural network to infer the data represented by the image to be identified mapped by the information recording points of other transparent materials under light irradiation.

[0018] In the above-mentioned storage system, the computer includes a classical computer or a server or a quantum computer or a mobile terminal computer (such as a tablet or a mobile phone or an industrial computer or a laptop computer, etc.) or an optical quantum computer, etc.

[0019] The present application relates to a three-dimensional (3D) memory, characterized in that it comprises: a plurality of information recording points are arranged in a transparent material, each information recording point is arranged with an information recording material, and different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics map different colors under light irradiation;

[0020] The transparent material is divided into a plurality of pages, wherein a plurality of information recording points arranged on a plane of any page are used as a pixel point set of the any page, and each information recording point is used as a pixel point;

[0021] The data stored in the transparent material is represented by the color reflected by the information recording point.

[0022] In the above three-dimensional (3D) memory, two pixels belonging to any two different pages but with the same address are set to not overlap each other; or two pixels belonging to any two different pages but with the same address are set to overlap each other.

[0023] In the above-mentioned three-dimensional (3D) memory, each page of the image containing all the pixels is used as parallel data to be read in parallel at one time, that is, the data of one page is read at a time with the page as the minimum reading unit; or the data of each pixel of each page is read after the data is read, and then the data of the next pixel is read, that is, the data of one pixel is read at a time in a serial manner with a single pixel as the minimum reading unit.

[0024] In the above three-dimensional (3D) memory, all pixels located in different pages but with the same address are arranged in a spiral in a direction perpendicular to the page, and there is a spiral angle value between any pixel of the latter page and a pixel of the same address of the adjacent previous page.

[0025] The present application relates to a method for information recording, which is characterized by comprising: setting a plurality of information recording points in a transparent material, each of which is provided with information recording material and different data are represented by information recording materials with different colors, wherein different colors refer to: different information recording materials project different colors under light irradiation; and the data stored in the transparent material is represented by the color reflected by the information recording point.

[0026] The present application relates to a three-dimensional (3D) memory, characterized in that it comprises: a plurality of information recording points are arranged in a transparent material, each information recording point is arranged with an information recording material, and different data are represented by information recording materials with different colors, and different information recording materials are mapped to different colors under light irradiation;

[0027] The transparent material is divided into a plurality of pages, wherein a plurality of information recording points arranged on a plane of any page are used as a pixel point set of the any page, and each information recording point is used as a pixel point;

[0028] The data stored in the transparent material is represented by the color reflected by the information recording point.

[0029] The present application relates to a storage system, characterized in that it includes:

[0030] A transparent material, in which a plurality of information recording points are arranged, each of which is provided with information recording materials and different data are represented by information recording materials with different colors, wherein different colors refer to different information recording materials reflecting different colors under light irradiation;

[0031] A computer is provided with a camera, which captures image data mapped by a plurality of information recording points of a transparent material as a sample under light illumination, and provides a series of data sets of the image data by changing the light illumination color of the information recording material, and the camera interacts (or transmits) the data sets to the computer;

[0032] The computer uses the data set to train a neural network, and uses the trained neural network to infer the data represented by the image to be identified mapped by the information recording points of other transparent materials under light irradiation.

[0033] Traditional mechanical hard disks (HDDs) in electronic products contain mechanical parts. The magnetic head needs to move to the target position on the fast-spinning disk to write and read, and a large number of tasks are consumed in inefficient mechanical movements. Mechanical movement not only wastes energy but also generates a lot of heat. Traditional mechanical hard disks have major disadvantages.

[0034] Solid-state drives (SSDs), which have a much faster reading speed than mechanical hard drives, no longer focus on mechanical movements. Moreover, solid-state drives are more shock-resistant and drop-resistant than mechanical hard drives in terms of physical structure. This is because solid-state drives are made of solid-state electronic storage chip arrays and use integrated circuits instead of traditional rotating disks. As mentioned earlier, solid-state drive memory mostly uses floating gates to store charges to express data, and charges will also leak. For example, NAND FLASH storage cells will lose most of their charges in about a few years or decades without operation. This is a problem to be solved.

[0035] The example introduced in the context of this application has the advantages that: it does not use electric charge to store data, so there is almost no charge leakage and the accompanying data loss problem; there is no need to move to the target position on a fast-rotating disk to write and read like a mechanical hard disk, which consumes a lot of tasks on mechanical actions. Another biggest difference from traditional transistor-based memory is that traditional transistor memory is arranged into minimum read and write units or minimum erase units of different sizes. This also means that each read and write action must be implemented with a minimum operation unit including tens of thousands or more transistors to implement data operations, and a large number of bit lines and word lines inside the minimum operation unit need to use a lot of control information to classify and manipulate, which not only makes the read and write operations complicated, but also requires the introduction of many auxiliary circuits such as instruction decoding and address decoding to complete these operations, and it is easy to make mistakes. The minimum operation unit of traditional transistor memory is often a partitioning mode such as block, page or sector.

[0036] The example introduced in the context of this application also has the advantage that it is no longer limited to the traditional transistor memory that must perform data operations on the minimum operation unit (block, page, sector), and allows a single minimum operation unit such as a single pixel to be used as the object to perform data operations, without the need to operate the data of other pixels at the same time. When the traditional transistor memory performs data operations on the minimum operation unit such as a block, page or sector, it often also operates the data of other transistors of the same operation unit. In this article, for example, each page of the image containing all pixels is used as parallel data to be read in parallel at one time, that is, each time the page is used as the minimum reading unit and the data of one page is read at a time; after the data of each pixel of each page is read, the data of the next pixel is read, that is, each time a single pixel is used as the minimum reading unit and the data of one pixel is read at a time in a serial manner. It has a high degree of data operation flexibility and convenience. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to make the above purposes, features and advantages more understandable, the following text will explain the specific implementation methods in detail in conjunction with the drawings. After reading the following description and referring to the following drawings, the features and advantages of the present application will become apparent.

[0038] Figure 1 As for the transparent material, options such as transparent semiconductor wafer, transparent ceramic wafer or glass can be selected.

[0039] Figure 2 The transparent material is divided into multiple pages and each information recording point is regarded as a pixel point.

[0040] Figure 3 It is an image mapped out by the camera capturing the many information recording points of transparent materials under light.

[0041] Figure 4 Any two pixels belonging to two different pages but with the same address are set to not overlap.

[0042] Figure 5 Two pixels belonging to any two different pages but with the same address are set to overlap vertically.

[0043] Figure 6 The data stored in the transparent material is represented by the color of the information recording point.

[0044] Figure 7 It uses a trained neural network to infer the data of information recording points of other transparent materials.

[0045] Figure 8 All pixels located on different pages but with the same address are arranged in a spiral shape in the vertical direction. DETAILED DESCRIPTION

[0046] The technical solutions disclosed in the present application will be clearly and completely described below in conjunction with specific embodiments. However, the embodiments described are only embodiments used for narrative explanation in the present application and are not all embodiments. Based on these embodiments, technicians in this field should realize that any solution obtained without creative work falls within the scope of protection of the present application.

[0047] See also Figure 1 In order to make the memory mentioned in this application compatible with traditional manufacturing processes, the wafer WFR0 shown in the figure is used as an example for explanation. The size of wafer WFR0 is 4 inches, 6 inches, 8 inches or 12 inches, and part of the manufacturing of the wafer is compatible with traditional wafer processes such as CMOS process, Bipolar process, BiCMOS process, typically crystal pulling process and wafer cutting process, as well as sputtering or plating or chemical vapor deposition or physical vapor deposition or coating of information recording materials on the wafer. Note that in addition to being compatible with traditional wafer manufacturing processes, information recording materials can also be replaced by new technologies that have emerged recently, such as printing a coating with information recording materials.

[0048] See also Figure 1 Regarding the material requirements of wafer WFR0: it should use transparent semiconductor wafers or transparent ceramic wafers or directly use glass materials, or even transparent acrylic materials. If it is purely from the perspective of material selection, acrylic materials or plastic materials can also be selected. However, compared with rigid materials such as semiconductors, ceramics or glass, the inherent flexibility or flexure of acrylic materials or plastic materials will cause local changes in the physical properties of information recording materials. For example, the color of the stretched information recording material is not completely consistent with the color of the original undeformed information recording material. Especially for nano-level information recording materials, superimposed with the radiation characteristics of light, there is a difference in color between the two cases.

[0049] See also Figure 1 , about the wafer WFR0 uses a transparent semiconductor wafer: one of the biggest features of the wafer used for SiC-based GaN materials is the transparency of the wafer. The third-generation compound semiconductor can be used. As a representative device of the third-generation compound semiconductor, GaN HEMT devices are widely used in the field of microwave power amplification with high electron mobility, high breakdown voltage, high current density and high reliability. The lattice constant of SiC material is close to that of GaN material, so high-quality GaN HEMT heterojunction structure can be epitaxially grown on SiC substrate. At the same time, the thermal conductivity of SiC material is high and guarantees the requirements of high-power heat dissipation. Therefore, SiC-based GaN HEMT material is one of the options.

[0050] See also Figure 1, about the wafer WFR0 uses a transparent ceramic wafer: magnesium aluminate MgAl 2 O 4 Ceramic wafer or crystal SiO 2 Ceramic wafers, lanthanum strontium aluminum tantalate LSAT ceramic wafers, strontium titanate SrTiO 3 Ceramic wafer, or titanium dioxide TiO 2 Ceramic wafer, YSZ ceramic wafer, MgO ceramic wafer, ZnO ceramic wafer or sapphire Al 2 O 3 Ceramic wafer, terbium scandate TbScO 3 Ceramic wafer, DyScO 3 Wafers, etc.

[0051] See also Figure 1 , Regarding the glass used for wafer WFR0: ZAO electric glass, FTO electric glass, ITO electric glass or traditional household ordinary glass, quartz glass, etc., can be used, and even acrylic plates can be used to replace glass. The main components of glass are silicon dioxide and other oxides or silicate complex salts, which are amorphous solids with irregular structures. When selecting glass materials, it should be noted that they have a certain compatibility with the information recording materials mentioned later.

[0052] See also Figure 1 If the wafer WFR0 is arranged with information recording materials, it can be subjected to the experience cutting (Saw) process according to the conventional wafer process, and the cut wafer WFR1 is shown in the figure. After the wafer WFR1 is cut, it will have many small-sized memory chips or memory particles DIE. The memory particles DIE are transparent. The preparation example of information recording materials is shown in the following Figure 8 .

[0053] See also Figure 2 The internal structure of the storage particle DIE is roughly as shown in the figure. The storage particle DIE is a transparent material and its preparation process is not limited to Figure 1 Use Figure 1 The reason for the process is that it can provide output quickly and efficiently based on the equipment and process conditions of traditional wafer factories. In fact, traditional glass manufacturers can also manufacture storage particles DIE by using diamond knives to cut glass. For example, after each layer of glass is manufactured, the glass manufacturer will coat the glass layer with information recording material, and then stack another layer of glass with information recording material on it. In this way, after stacking multiple layers of glass, it can form a Figure 2 The storage particles DIE shown.

[0054] See also Figure 2, the storage particle DIE is first explained by taking the preparation process of a semiconductor wafer factory as an example: a number of information recording points pix are arranged on a layer of transparent substrate, which is regarded as the first (carrying the first page Page1) transparent layer; the second transparent layer (carrying the second page Page2) continues to grow epitaxially on the first transparent layer, which also arranges information recording points pix; the third transparent layer (carrying the third page Page3) continues to grow epitaxially on the second transparent layer, which also arranges information recording points pix; the fourth transparent layer (carrying the fourth page Page4) continues to grow epitaxially on the third transparent layer, which also arranges information recording points pix. According to this rule, multiple layers of superimposed transparent material layers can be generated and each transparent layer carries a number of information recording points pix. The many information recording points pix on each transparent layer can be called a page. Based on the wafer process, multiple levels of pages Page1 to Page3 as data storage carriers, and even more layers of undisplayed pages, are finally prepared in the storage particle DIE.

[0055] See also Figure 2 The information recording material of the information recording points pix of the storage particles DIE can be arranged on the wafer in a variety of optional ways, such as sputtering, plating, chemical vapor deposition, physical vapor deposition or coating on the wafer and other semiconductor processes, or directly printing the information recording material onto the wafer.

[0056] See also Figure 2 , the storage particle DIE is further explained by taking the improvement of the ordinary glass preparation process as an example: a plurality of information recording points pix are coated on a layer of transparent glass, which is regarded as the first (carrying the first page Page1) glass layer; the second glass layer (carrying the second page Page2) is continuously stacked and laminated on the first glass layer, which is also arranged with information recording points pix; the third glass layer (carrying the third page Page3) is continuously stacked and laminated on the second glass layer, which is also arranged with information recording points pix; the fourth glass layer (carrying the fourth page Page4) is continuously stacked and laminated on the third glass layer, which is also arranged with information recording points pix. According to this rule, multiple layers of stacked transparent glass layers can be generated and each glass layer carries a plurality of information recording points pix. The plurality of information recording points pix on each glass layer can be called a page. Based on the glass preparation, multiple levels of pages Page1 to Page3 as data storage carriers and even more layers of undisplayed pages are finally prepared in the storage particle DIE.

[0057] See also Figure 2There are many optional ways for the information recording material of the information recording point pix of the storage particle DIE to be arranged on the glass, such as sputtering, plating, chemical vapor deposition, physical vapor deposition or coating on the glass, or directly printing the information recording material on the glass. The dusty preparation environment of glass cannot be compared with the dust-free preparation environment of wafers. If each color has eight-bit grayscale data, then any monochrome reference color will have 256 levels of grayscale. If each color has ten-bit grayscale data, then any monochrome reference color will have 1024 levels of grayscale. Due to process limitations, a single information recording point pix may only have relatively low-level grayscale data on ordinary glass, but thanks to the process, a single information recording point pix can provide grayscale data with higher-level resolution on the wafer. For example, the former expresses eight bits and the latter is ten bits. In addition, the information recording material can also be prepared into small-sized grooves reserved on glass or wafers. If the size of the information recording material is extremely small, such as reaching the nanoscale, there is no need to prepare it in the groove because it does not occupy physical space.

[0058] See also Figure 2 , a transparent material such as storage particles DIE is divided into multiple pages, and multiple information recording points arranged on the plane of any page are used as a set of pixel points of any page, and each information recording point is used as a pixel point. For example, in the figure, the transparent material is divided into multiple pages Page1 to Page3, or even more layers of pages that are not displayed. Multiple information recording points pix are arranged on the plane of any page such as page Page1 as a set of pixel points of any page such as page Page1, and each information recording point at this time can be used as a pixel point. Pixel arrangement: The coordinates of the upper left corner are P1_address (x0, y0), which is the address of the 0th pixel, and the coordinates of other pixel points are incremented on this basis.

[0059] See also Figure 2 For example, the address of a pixel with an RGB mark is P1_address(x1,y0) and its address shows that it is a pixel in the first row and second column, and the addresses of other pixels are similar.

[0060] See also Figure 2 For example, the address of a pixel with an RGB mark is P1_address(x1,y1) and its address shows that it is a pixel in the second row and second column, and the addresses of other pixels are similar.

[0061] See also Figure 2, multiple information recording points are set in the transparent material, information recording materials are arranged at each information recording point and different data are represented by information recording materials with different preset characteristics. The information recording materials with different preset characteristics map different colors under light irradiation; the data stored in the transparent material is represented by the color reflected by the information recording point.

[0062] See also Figure 2 There are many options for information recording materials of information recording point pix. Quantum dot materials, gem materials, three primary color materials for color mixing, and even ordinary dyes and fluorescent dyes are all possible options.

[0063] See also Figure 2 A transparent material such as a storage particle DIE is provided with a plurality of information recording points. For example, different pages each have a plurality of information recording points pix. Information recording materials are arranged at each information recording point pix. Different data are represented by information recording materials with different preset characteristics. The information recording materials include quantum dot materials, and the different preset characteristics include setting the particle radius of the quantum dot materials to be different. The data stored in the transparent material such as the storage particle DIE is represented by the color reflected by the information recording point pix.

[0064] See also Figure 2 Quantum dots, also known as semiconductor nanocrystals, are crystal particles with a diameter of less than 20nm composed of hundreds or thousands of atoms. The crystal particles of quantum dots can be composed of elements from Group II-VII, Group III-VI or Group I-III-VII. One of the important characteristics of quantum dots is the quantum confinement effect, that is, when the size of the bulk material gradually decreases, its energy band structure changes from a continuous energy band to a discrete energy level, and the band gap gradually increases as the crystal size decreases. Since the size of quantum dots is smaller than the Bohr radius of the exciton, it ensures that the excitons generated by light excitation can achieve efficient recombination, giving quantum dots unique optical properties, one of which is that quantum dots of different sizes will emit light of different colors.

[0065] See also Figure 2 Based on quantum dot materials, information recording materials of information recording dots pix can display particularly pure colors such as red, rose red, grapefruit red, etc. The important principle of quantum dot material luminescence is the interaction of hundreds to thousands of atoms and the ability to control their size and surface properties, thereby adjusting more colors. In the world of traditional luminescent materials, red is red and green is green, but quantum dot materials can achieve colorful colors.

[0066] See also Figure 2, during the conversion process of quantum dot color, material 1 can exist in the form of a thin film. When the incident light hits the quantum dot film, it excites the quantum dot to emit light. If the light field does not hit the quantum dot material before leaving the film, the light field will directly transmit through the quantum dot film and the propagation direction will remain unchanged. When the light field hits the quantum dot material but is not absorbed by the quantum dot material, the incident light is scattered, and the light field changes direction and continues to propagate in the film; the absorbed incident light will participate in the color conversion process of the quantum dot. When the quantum dot absorbs photons and the corresponding excitation energy level exceeds the band gap, the material electrons transition from the ground state to the excited state and form excitons with holes. The electron-hole pairs release energy in the form of radiation recombination or non-radiation recombination and generate photons in the radiation recombination process. Heat is released in the process of non-radiation transition. The band gaps of different types of quantum dot materials are different, so the photon energy required to be absorbed during the excitation reaction will also be different, which is related to the material structure properties of the quantum dots. The information recording point pix of the storage particle DIE can absorb photons to produce color at this time.

[0067] See also Figure 2 , the information recording point pix quantum dot of the storage particle DIE is a material with a size of about 1-10nm composed of group II-IV or group II-VI elements. Quantum dots of different sizes have different fluorescence effects and can excite light fields of different colors. In general, as the particle size decreases, the main wavelength of the corresponding excitation spectrum becomes shorter. Compared with traditional fluorescent dyes, the size of quantum dots is adjustable, and the emission and excitation bandwidth of fluorescence covers the spectral region from ultraviolet to infrared. The shape of the emission spectrum is narrow and symmetrical, with excellent photostability and biocompatibility.

[0068] See also Figure 2 In an optional embodiment, the information recording material includes a quantum dot material, and the different preset characteristics include setting the particle radius of the quantum dot material to be different. For example, in the case of a 4-nanometer particle radius, the information recording material has a first preset characteristic, such as a tendency toward red color. For another example, in the case of a 3-nanometer particle radius, the information recording material has a second preset characteristic, such as a tendency toward green color. For another example, in the case of a 1-nanometer particle radius, the information recording material has a third preset characteristic, such as a tendency toward blue color. Thus, it is achieved that the information recording material with different preset characteristics maps different colors under light irradiation.

[0069] See also Figure 2The transparent material such as storage particle DIE is provided with multiple information recording points. For example, different pages each have a plurality of information recording points pix. Information recording materials are arranged at each information recording point pix. Different data are represented by information recording materials with different preset characteristics. The information recording materials may include gemstone materials, and different preset characteristics include adjusting the proportions of various chemical components of the gemstone materials to different values. The data stored in the transparent material such as storage particle DIE is represented by the color reflected by the information recording point pix.

[0070] See also Figure 2 The gem material is typically ruby, sapphire or emerald, of course, the gem material can also be a combination of ruby ​​and sapphire, a combination of ruby ​​and emerald, a combination of sapphire and emerald, or a combination of ruby, sapphire and emerald. They can be artificially synthesized.

[0071] See also Figure 2 The main component of ruby ​​is aluminum oxide (Al 2 O 3 ), generally red in color, hence the name. The color comes from trace amounts of chromium, which makes it appear red. The higher the chromium content, the redder it is. The reddest is commonly known as pigeon blood red. Luminescence of ruby: It has obvious weak red fluorescence under long-wave and short-wave ultraviolet light. Different preset characteristics include adjusting the proportions of various chemical components of the gem material to different levels: for example, aluminum oxide (Al 2 O 3 ) or chromium (mainly Cr 2 O 3 ) to different proportions. For example, adjusting the chromium content from 0.12% to 0.20% is equivalent to changing the preset characteristics.

[0072] See also Figure 2 In an optional embodiment, the information recording material includes a ruby ​​material, and different preset characteristics include adjusting the proportions of various chemical components in the ruby ​​to different proportions. For example, when the chromium proportion is 0.15%, the information recording material has a first preset characteristic, such as a tendency toward weak red. For another example, when the chromium proportion is 0.25%, the information recording material has a second preset characteristic, such as a tendency toward dark red. For another example, when the chromium proportion is 0.35%, the information recording material has a third preset characteristic, such as a tendency toward blood red. Thereby achieving: information recording materials with different preset characteristics map different colors under light irradiation.

[0073] See also Figure 2 The main component of sapphire is aluminum oxide (Al 2 O 3), sapphire is the general name for corundum gemstones other than ruby, and sapphire has pink, yellow, green, white and other colors, and even the same stone has multiple colors. Blue sapphire is due to the presence of a small amount of titanium (Ti) and iron (Fe) impurities. Different preset characteristics include adjusting the proportions of various chemical components of the gem material to different proportions: the proportions of elements such as iron or titanium can be adjusted to different proportions. For example, the iron content can be adjusted from 0.1WT% to 0.15WT%, which is equivalent to a change in the preset characteristics.

[0074] See also Figure 2 In an optional embodiment, the information recording material includes sapphire material, and different preset characteristics include adjusting the proportions of various chemical components in the sapphire to different proportions. For example, when the iron proportion is 0.09%, the information recording material has a first preset characteristic, such as a tendency towards weak blue. For another example, when the iron proportion is 0.15%, the information recording material has a second preset characteristic, such as a tendency towards light blue. For another example, when the iron proportion is 0.21%, the information recording material has a third preset characteristic, such as a tendency towards dark blue. Thereby achieving: information recording materials with different preset characteristics map different colors under light irradiation.

[0075] See also Figure 2 The main component of emerald is beryllium aluminum silicate mineral. The color range caused by iron ions in emerald is very wide, such as yellow, green, yellow-green, blue, etc. Iron has two valence states and often exists in two forms inside emerald. Emeralds are mostly light green. When the composition is rich in cesium, it is pink and is called rose emerald. When it contains chromium, it is bright emerald green and is called emerald. When it contains divalent iron, it is light blue and is called aquamarine. When it contains trivalent iron, it is yellow and is called yellow emerald. Different preset characteristics include adjusting the proportions of each chemical component of the gemstone to different proportions: the proportions of elements such as cesium, chromium or iron can be adjusted to different proportions.

[0076] See also Figure 2 In an optional embodiment, the information recording material includes an emerald material, and the different preset characteristics include adjusting the proportions of various chemical components in the emerald to different values. For example, when the gem material is rich in cesium, the information recording material has a first preset characteristic, such as a tendency toward pink. For another example, when the gem material is rich in chromium, the information recording material has a second preset characteristic, such as a tendency toward emerald green. For another example, when the gem material is rich in iron, the information recording material has a third preset characteristic, such as a tendency toward light blue. Thus, the information recording materials with different preset characteristics reflect different colors under light irradiation.

[0077] See also Figure 2, a transparent material such as storage particle DIE is provided with a plurality of information recording points, for example, different pages each have a plurality of information recording points pix, and information recording materials are arranged at each information recording point pix, and different data are represented by information recording materials with different preset characteristics, and the information recording materials include three primary color materials for color mixing, and different preset characteristics include setting the color mixing ratio of the three primary color materials to different. The data stored in the transparent material such as storage particle DIE is represented by the color embodied by the information recording point pix. The pixel points in the display field use the three primary colors of red, green and blue to mix colors to obtain full color. For example, if each color in RGB has eight-bit grayscale data, then any monochrome reference color will have 256 grayscale levels, and the three primary colors can form about 16 trillion colors. For another example, if each color has ten-bit grayscale data, any monochrome reference color will have 1024 grayscale levels, and the three primary colors can form nearly one billion colors. The figure shows the three primary colors of red, green and blue RGB mixed at the information recording point pix.

[0078] See also Figure 2 In an optional embodiment, the information recording material includes three-primary color phosphors, and the different preset characteristics include setting the color mixing ratio of each primary color material in the phosphor to be different. When the three-primary color phosphors are rich in red and green phosphors, the information recording material has a first preset characteristic such as leaning towards yellow, when the three-primary color phosphors are rich in red and blue phosphors, the information recording material has a second preset characteristic such as leaning towards purple, and when the three-primary color phosphors are rich in blue and green phosphors, the information recording material has a third preset characteristic such as leaning towards cyan. Thus, it is achieved that the information recording materials with different preset characteristics map different colors under light irradiation.

[0079] See also Figure 2 In an optional embodiment, the information recording material includes polystyrene microspheres, and the different preset characteristics include setting the mixing ratio of each primary color microsphere in the polystyrene microspheres to be different. The three primary color materials are mixed using red, blue and green polystyrene microspheres. When the mixed color polystyrene microspheres are rich in red and green microspheres, the information recording material has a first preset characteristic, such as leaning towards yellow; when the mixed color polystyrene microspheres are rich in red and blue microspheres, the information recording material has a second preset characteristic, such as leaning towards purple; when the mixed color polystyrene microspheres are rich in blue and green microspheres, the information recording material has a third preset characteristic, such as leaning towards cyan. Thereby achieving: information recording materials with different preset characteristics map different colors under light irradiation.

[0080] See also Figure 2Regarding the preparation of red, blue and green tricolor polystyrene microspheres: ethanol was used as the dispersion medium, polyvinyl pyrrolidone as the dispersant, and azobisisobutyronitrile as the initiator, and the polystyrene microspheres were prepared by dispersion polymerization. Chloroform and isopropanol were used as swelling agents, acridine orange and rhodamine were used as fluorescein, and the microspheres were dyed by the gradual evaporation of the solvent. As a result, monodisperse microspheres with a particle size range of 1-5 microns were prepared. The dyed microspheres can be observed to be excited to emit red, green, blue and other colors of fluorescence under a fluorescence microscope.

[0081] See also Figure 3 , a transparent material such as storage particles DIE is used to divide multiple pages, and multiple information recording points pix arranged on the plane of page PageK are used as the pixel point set of the page PageK. The camera Foc-Ca can obtain the corresponding image ImageK by shooting each information recording point pix or pixel point on page PageK, and the camera Foc-Ca can obtain the corresponding image ImageK+1 by shooting each information recording point pix or pixel point on page PageK+1. The camera preferably uses an industrial high-definition and high-pixel resolution camera with a focusing function. In situations where the data resolution is not required (for example, each pixel only needs to express 1-2 bits of data information), handheld devices such as conventional tablets or mobile phones can be used to shoot and capture images of page information recording points pix. In situations where the data resolution is required to be high (for example, each pixel needs to express 6-8 bits or even more bits), a high-resolution camera is required.

[0082] See also Figure 4 In an optional embodiment, two pixels belonging to any two different pages but having the same address are set to overlap each other. An example of pixel arrangement of any two different pages is as follows.

[0083] See also Figure 4 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page1) but with exactly the same addresses are set to overlap up and down: the pixel with address P0_address(x0,y0) of page Page0 and the pixel with address P1_address(x0,y0) of another page Page1 are set to overlap up and down.

[0084] See also Figure 4 In an optional embodiment, two pixels belonging to any two different pages (Page1 and Page2) but with exactly the same addresses are set to overlap up and down: the pixel with address P1_address(x0,y0) of page Page1 and the pixel with address P2_address(x0,y0) of another page Page2 are set to overlap up and down.

[0085] See also Figure 4In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page2) but with exactly the same addresses are set to overlap up and down: the pixel with address P0_address(x0,y0) of page Page0 and the pixel with address P2_address(x0,y0) of another page Page2 are set to overlap up and down.

[0086] See also Figure 4 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page1) but with exactly the same addresses are set to overlap up and down: the pixel with address P0_address(xn,yn) of page Page0 and the pixel with address P1_address(xn,ym) of another page Page1 are set to overlap up and down.

[0087] See also Figure 4 In an optional embodiment, two pixels belonging to any two different pages (Page1 and Page2) but with exactly the same addresses are set to overlap up and down: the pixel with address P1_address(xn,ym) of page Page1 and the pixel with address P2_address(xn,ym) of another page Page2 are set to overlap up and down.

[0088] See also Figure 4 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page2) but with exactly the same addresses are set to overlap up and down: the pixel with address P0_address(xn,ym) of page Page0 and the pixel with address P2_address(xn,ym) of another page Page2 are set to overlap up and down.

[0089] See also Figure 5 In an optional embodiment, two pixels belonging to any two different pages but having the same address are set to be non-overlapping. An example of pixel arrangement of any two different pages is as follows.

[0090] See also Figure 5 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page1) but with exactly the same address are set to be non-overlapping: the pixel with address P0_address (x0, y0) of page Page0 and the pixel with address P1_address (x0, y0) of the adjacent page Page1 are set to be non-overlapping.

[0091] See also Figure 5In an optional embodiment, two pixels belonging to any two different pages (Page1 and Page2) but with exactly the same address are set to be non-overlapping: the pixel with address P1_address(x0, y0) of page Page1 and the pixel with address P2_address(x0, y0) of the adjacent page Page2 are set to be non-overlapping.

[0092] See also Figure 5 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page2) but with exactly the same address are set to be non-overlapping: the pixel with address P0_address (x0, y0) of page Page0 and the pixel with address P2_address (x0, y0) of the separated page Page2 are set to be non-overlapping.

[0093] See also Figure 5 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page1) but with exactly the same address are set to be non-overlapping: the pixel with address P0_address(xn,yn) of page Page0 and the pixel with address P1_address(xn,ym) of the adjacent page Page1 are set to be non-overlapping.

[0094] See also Figure 5 In an optional embodiment, two pixels belonging to any two different pages (Page1 and Page2) but with exactly the same address are set to be non-overlapping: the pixel with address P1_address(xn,ym) of page Page1 and the pixel with address P2_address(xn,ym) of the adjacent page Page2 are set to be non-overlapping.

[0095] See also Figure 5 In an optional embodiment, two pixels belonging to any two different pages (Page0 and Page2) but with exactly the same address are set to be non-overlapping: the pixel with address P0_address(xn,ym) of page Page0 and the pixel with address P2_address(xn,ym) of the separated page Page2 are set to be non-overlapping.

[0096] See also Figure 5In an optional embodiment, there is a certain pre-set offset between two pixels of different pages but with the same address (i.e., they do not overlap each other), so the focusing ability of the camera Foc-Ca is not so high, and the camera is usually not interfered by other non-shooting objects (e.g., information recording points pix on page PageK+1) when aiming at the shooting object (e.g., each information recording point pix on page PageK) to collect image data. A specific example of such interference is that the image acquisition process of the information recording point pix with address PK+1_address (xn, ym) on page PageK+1 is interfered by the information recording point pix with address PK_address (xn, ym) on another page PageK. There are many factors for such interference, such as the light shielding of the latter to the former, the camera lens shielding of the latter to the former, the light reflection of the latter to the former, and various optical noises introduced by the latter to the former. Optical noise is usually reflected in zero exposure noise, thermal noise, readout noise, and optical noise based on the Poisson distribution of photons.

[0097] See also Figure 5 In an optional embodiment, two pixels of different pages but with the same address have a preset zero offset (i.e. overlap each other), which places relatively high demands on the focusing capability of the camera Foc-Ca and the camera should not be interfered by other non-shooting objects (e.g., information recording points pix on page PageK+1) when aiming at the shooting object (e.g., information recording points pix on page PageK) to collect image data. At this time, it is necessary to focus the shooting angle of view on the level or page position of the shooting object (e.g., page PageK).

[0098] See also Figure 6 Different data are represented by information recording materials with different preset characteristics. Information recording materials with different preset characteristics map different colors under light irradiation; the color reflected by the information recording point is used to represent the data stored in a transparent material such as storage particles DIE.

[0099] See also Figure 6 In an optional embodiment, the data represented by the information recording point with address PK_address (x0, y0) on page PageK is 0x0f, and the preset characteristic of its information recording material (such as using quantum dot material) is that the radius of the quantum dot particles of the quantum dot material is about 4.6 nanometers, and the preset characteristic excites and maps the red color.

[0100] See also Figure 6In an optional embodiment, the data represented by the information recording point with address PK_address (x1, y0) on page PageK is 0x0e, and the preset characteristic of its information recording material (such as using quantum dot material) is that the radius of the quantum dot particles of the quantum dot material is about 3.1 nanometers, and the preset characteristic excites and maps the green color.

[0101] See also Figure 6 In an optional embodiment, the data represented by the information recording point with address PK_address (x2, y0) on page PageK is 0x08, and the preset characteristic of its information recording material (such as using quantum dot material) is that the radius of the quantum dot particles of the quantum dot material is about 1.2 nanometers, and the preset characteristic excites and maps a blue color.

[0102] See also Figure 6 , other data of page PageK such as 0x0d, 0x0c, 0x0b, etc., the remaining data of the first line and more undisplayed data are not described one by one. The data of the second line 0xff, 0xfe, 0xf6, 0xe6, 0xef, 0xcf and more undisplayed data are not described one by one. The data format here can also be binary or octal.

[0103] See also Figure 6 Different data, such as 0x0f, 0x0e, and 0x08, can be represented by information recording materials, such as quantum dot materials, with different preset characteristics, such as different quantum dot particle radii. Information recording materials with different preset characteristics map different colors, such as red, green, blue, or other colors, under light irradiation. The color reflected by the information recording point is used to represent the data stored in a transparent material, such as storage particles DIE. For example, if the camera captures the red color of the information recording point, it knows that it is data 0x0f, if it captures the green information recording point, it knows that it is data 0x0e, and if it captures the blue information recording point, it knows that it is data 0x08.

[0104] See also Figure 6 , several data of page PageK use red, green and blue to express some data. In essence, each primary color has eight-bit grayscale data, so any monochrome reference color will have 256 grayscale levels, and the three-primary color mixture can constitute about sixteen trillion colors. For another example, if each color is equipped with ten-bit grayscale data, then any monochrome reference color can reach 1024 grayscale levels, and the three-primary color mixture can constitute nearly one billion colors. In other words, under the condition of eight-bit grayscale data, the monochrome of each information recording point pix can select one of 256 grayscales, and further, under the condition of eight-bit grayscale mixing, the mixed color of each information recording point pix can select one of sixteen trillion colors. The calculation method under other data width conditions, such as ten-bit grayscale data or sixteen-bit grayscale data, is similar.

[0105] See also Figure 6 In the optional example, the data represented by the information recording point at the address PK+1_address (x0, y0) on page PageK+1 is 0xf3, and the preset characteristic of its information recording material (such as a three-primary color phosphor material) is that the color mixing ratio of the phosphor is high, and the ratio of red and green phosphors is high, and the preset characteristic maps a yellow color.

[0106] See also Figure 6 In the optional example, the data represented by the information recording point at the address PK+1_address (x1, y0) on page PageK+1 is 0xf0, and the preset characteristic of its information recording material (such as a three-primary color phosphor material) is that the mixed color ratio of the phosphor is high, and the purple color is mapped under this preset characteristic.

[0107] See also Figure 6 In the optional example, the data represented by the information recording point at the address PK+1_address(x2,y0) on page PageK+1 is 0xf8, and the preset characteristic of its information recording material (such as a three-primary color phosphor material) is that the color mixing ratio of the phosphor is high in the blue-green phosphor ratio, and the cyan color is mapped under this preset characteristic.

[0108] See also Figure 6 , the data of page PageK+1 such as 0xec, 0xee, 0xef, etc., the remaining data of the first line and more undisplayed data are not described one by one. The data of the second line 0xce, 0xc6, 0xc2, 0xc3, 0xcb, 0xca and more undisplayed data are not described one by one. The data format here can also be binary or octal.

[0109] See also Figure 6 Different data, such as 0xf3, 0xf0, and 0xf8, can be represented by information recording materials with different preset characteristics, such as different color mixing ratios, such as three-primary color phosphor materials. Information recording materials with different preset characteristics map different colors, such as yellow, purple, cyan, or other colors, under light irradiation. The color reflected by the information recording point is used to represent the data stored in a transparent material, such as storage particles DIE. For example, if the camera captures the yellow color of the information recording point, it knows that it is data 0xf3, if it captures the purple information recording point, it knows that it is data 0xf0, and if it captures the cyan information recording point, it knows that it is data 0xf8.

[0110] See also Figure 7, extracting the image data Image0_training of many information recording points pix mapped under light irradiation from the transparent material such as wafer WFR0_training provided in advance as a sample. Wafer WFR0_training is not much different from the wafer WFR0 mentioned above. Wafer WFR0_training needs to provide training data for the neural network NET and can also tag these training data.

[0111] See also Figure 7 , assuming that the image captured by the camera at the information recording points of the unillustrated page PageK of the wafer WFR0_training is Image0_training, and the images Image0_training and Figure 3 The images ImageK and so on are basically similar and their acquisition methods are basically the same. Therefore, the image Image0_training will not be described separately. In the figure, it is assumed that the image is Image0_training similar to Figure 3 of image ImageK, but Figure 3 Different from the above, the pixel data of the image Image0_training shown in the figure is the training data. Figure 3 The pixel data of the image ImageK is unknown data to be identified or inferred.

[0112] See also Figure 7 , the color value or color value or color-based gray value of each pixel on the image Image0_training is recorded in the color value table Color0_ext. For example, the color value of the information recording point or pixel at the address PK_address(x0,y0) is recorded as GS(x0,y0), and this color value corresponds to the tag 0xef (just an example and not a limitation), and the tag value of the same address PK_address(x0,y0) in the tag value table is recorded as 0xef. Therefore, in addition to the color value table itself, the tag value table Data0_tag corresponding to the color value table of the image Image0_training is also shown in the figure.

[0113] See also Figure 7 In an optional embodiment, for example, the color value of the information recording point or pixel with the address PK_address (x1, y0) is recorded as GS (x1, y0), and the color value corresponds to the label 0xde (just an example and not a limitation), and the label value of the same address PK_address (x1, y0) in the label value table is recorded as 0xde.

[0114] See also Figure 7In an optional embodiment, for example, the color value of the information recording point or pixel at the address PK_address (x2, y0) is recorded as GS (x2, y0), and the color value corresponds to the label 0xdc (just an example and not a limitation), and the label value of the same address PK_address (x2, y0) in the label value table is recorded as 0xdc.

[0115] See also Figure 7 In an optional embodiment, for example, the color value of the information recording point or pixel at the address PK_address (x0, y1) is recorded as GS (x0, y1), and the color value corresponds to the label 0xd7 (just an example and not a limitation), and the label value of the same address PK_address (x0, y1) in the label value table is recorded as 0xd7.

[0116] See also Figure 7 In an optional embodiment, for example, the color value of the information recording point or pixel at the address PK_address (x1, y1) is recorded as GS (x1, y1), and the color value corresponds to the label 0xd6 (just an example and not a limitation), and the label value of the same address PK_address (x1, y1) in the label value table is recorded as 0xd6.

[0117] See also Figure 7 In an optional embodiment, for example, the color value of the information recording point or pixel at the address PK_address (x2, y1) is recorded as GS (x2, y1), and the color value corresponds to the label 0xd5 (just an example and not a limitation), and the label value of the same address PK_address (x2, y1) in the label value table is recorded as 0xd5. The color values ​​and corresponding label values ​​of other addresses can be matched according to the above rules, so they are not described one by one.

[0118] See also Figure 7 It has been informed that the image data Image0_training mapped by many information recording points pix under light irradiation can be extracted from a transparent material such as wafer WFR0_training as a sample, and a series of image data sets can be provided by changing its preset characteristics. By changing the preset characteristics of wafer WFR0_training (such as the particle radius of quantum dots or the chemical composition ratio of gemstone materials or the mixed color ratio relationship of three primary color materials, etc.), many series of image data Image0_training data sets are provided. The preset characteristics of the information recording material of any information recording point pix on any page are changed once to obtain a corresponding color value. The amount of data in the label value form Data0_tag can be enriched and varied, and therefore it is regarded as a training data set.

[0119] See also Figure 7 In an optional embodiment, the color value GS(x0,y0) of the information recording point or pixel at the address PK_address(x0,y0) is taken as an example to explain the process, and a series of image data sets are provided by changing its preset characteristics such as the particle radius of the quantum dot. When the radius of the quantum dot particle is 4.5 nanometers, this information recording point has a corresponding color value GS(x0,y0)_4.5. When the radius changes, when the radius of the quantum dot particle is 3.2 nanometers, this information recording point has a corresponding color value GS(x0,y0)_3.2. When the radius changes again, when the radius of the quantum dot particle is 1.3 nanometers, this information recording point has a corresponding color value GS(x0,y0)_1.3. Suppose that when GS(x0,y0)_4.5, the tag is Tag_4.5, when GS(x0,y0)_3.2, the corresponding tag is Tag_3.5, and when GS(x0,y0)_1.3, the corresponding tag is Tag_1.3. By analogy, any change in the radius of the quantum dot particle will be assigned a corresponding label value, thereby providing a series of image data such as a data set of pixel values ​​GS(x0,y0)_4.5 or GS(x0,y0)_3.2 or GS(x0,y0)_1.3 by changing the preset characteristics of the information recording point or the information recording material, such as the radius of the quantum dot particle (for example, 4.5 or 3.2 or 1.3 nanometers, etc.). This data set may also include the pixel values ​​of the information recording points at other addresses PK_address(xn,ym).

[0120] See also Figure 7In an optional embodiment, the color value GS(x2,y1) of the information recording point or pixel at the address PK_address(x2,y1) is used as an example to explain the process, and a series of image data sets are provided by changing its preset characteristics such as the particle radius of the quantum dot. When the radius of the quantum dot particle is 4.0 nanometers, this information recording point has a corresponding color value GS(x2,y1)_4.0. When the radius is reduced, when the radius of the quantum dot particle is 3.0 nanometers, this information recording point has a corresponding color value GS(x2,y1)_3.0. When the radius is further reduced, when the radius of the quantum dot particle is 1.2 nanometers, this information recording point has a corresponding color value GS(x2,y1)_1.2. Suppose that when GS(x2,y1)_4.0, the tag is Tag_4.0, when GS(x2,y1)_3.0, the corresponding tag is Tag_3.0, and when GS(x2,y1)_1.2, the corresponding tag is Tag_1.2. By analogy, any change in the radius of the quantum dot particle will be configured with a corresponding label value, thereby providing a series of image data such as a data set of pixel values ​​GS(x2,y1)_4.0 or GS(x2,y1)_3.0 or GS(x2,y1)_1.2 by changing the preset characteristics of the information recording point or the information recording material, such as the radius of the quantum dot particle (for example, 4.0 or 3.0 or 1.2 nanometers, etc.).

[0121] See also Figure 7 In summary, n and m are set to 0, 1, 2, 3, etc. Therefore, the image data not only includes the adjustable pixel value of the information recording point at the address PK_address (x0, y0), but also includes the adjustable pixel value of the information recording point at other addresses PK_address (xn, ym). The adjustable pixel value is reflected in the preset characteristics of the information recording material, such as the adjustable radius of the quantum dot particle. Or the adjustable pixel value is reflected in the preset characteristics of the information recording material, such as the adjustable proportion of the three primary color phosphors in the mixed color phosphor. Or the adjustable pixel value is reflected in the preset characteristics of the information recording material, such as the adjustable proportion of various chemical components in the gem material.

[0122] See also Figure 7 , the image data Image0_training based on the wafer WFR0_training or the deformed form of the image data is used as the data set for neural network NET training, and the trained neural network NET is used to infer other transparent materials such as Figure 1 The information recording point pix of wafer WFR0 (see Figure 3 The image to be identified is mapped under illumination as shown in Figure 3The data represented by the images ImageK and ImageK+1 to be identified should include the data at a pixel or information recording point pix corresponding to each address on the image. This example provides a data set of image data by changing the preset characteristics of the information recording material of the sample transparent material.

[0123] See also Figure 3 , use the neural network NET to infer the data represented by the image ImageK to be identified by the information recording points of other transparent materials under illumination: the data of the information recording points at each address PK_address(xj,yq) are inferred by the neural network NET. 0≤j≤n, 0≤q≤m. j=0, 1…n and q=0, 1…m. The other transparent materials inferred no longer belong to the category of sample transparent materials, and it is necessary to distinguish the properties of transparent materials.

[0124] See also Figure 3 , use the neural network NET to infer the data represented by the image ImageK+1 to be identified mapped by the information recording points of other transparent materials under illumination: the data of the information recording points at each address PK+1_address(xj,yq) are inferred by the neural network NET. 0≤j≤n, 0≤q≤m. j=0, 1…n and q=0, 1…m. The other transparent materials inferred no longer belong to the category of sample transparent materials, and it is necessary to distinguish the properties of transparent materials.

[0125] See also Figure 7 One of the biggest differences between traditional transistor-based memory and traditional memory is the read and write format. Traditional memory usually uses pages, sectors, or blocks as the minimum basic unit for reading and writing data. It is impossible to read or write the data of each transistor separately. All transistors in the minimum basic unit can only be operated simultaneously in a piece. This is determined by the inherent characteristics of traditional memory, where word lines and bit lines must be connected to many transistors at the same time. This is obviously unreasonable, but this design is a compromise made to reduce the number of interconnects, reduce the difficulty of preparation, reduce the volume, and increase storage density.

[0126] See also Figure 6, if the page is taken as the smallest basic unit, the image containing all pixels of each page is used as the parallel data read in parallel at one time, that is, the data of one page is read at a time with the page as the smallest reading unit. For example, the camera can take a complete image ImageK of a page such as page PageK at a time, and the image ImageK contains all the information recording points or all the pixels of the page. For example, the information recording point data at each address PK_address (xj, yq) of the image is inferred by the neural network NET. 0≤j≤n, 0≤q≤m. j=0, 1…n and q=0, 1…m. Then the image ImageK containing all the pixels corresponding to the page is used as the parallel data read in parallel at one time (one shot), that is, the data of one page PageK is read at a time with the page as the smallest reading unit. Obviously, the data reading efficiency is extremely high, which can cope with scenes with massive processing data and high data bandwidth. In this scene, there are certain performance requirements for the camera resolution of the shooting and the computer equipment running the neural network, especially for a single clear shot of all the pixels of a single page.

[0127] See also Figure 6 In an optional embodiment, if a single pixel is used as the minimum basic unit, the data of each pixel of each page is read after the data of the next pixel is read, that is, a single pixel is used as the minimum reading unit each time and the data of one pixel is read in serial. In this embodiment, for example, the camera can only shoot the segmented image Image_address(xj,yq) corresponding to the single information recording point pix(xj,yq) at the address PK_address(xj,yq) at each time, 0≤j≤n, 0≤q≤m. j=0, 1…n and q=0, 1…m. The segmented image obviously only contains the color of one pixel rather than the image of all pixels. The data of the information recording point at each address PK_address(xj,yq) is inferred one by one by the neural network NET.

[0128] See also Figure 6 In an optional embodiment, for example, after the data of the pixel at the address PK_address (x1, y1) is read, the data of the next pixel at the address PK_address (x2, y1) is read, that is, a single pixel is used as the minimum reading unit each time and the data of one pixel is read in serial mode. The biggest advantage of serial reading is that it allows each information recording point or pixel to extract color by focusing and shooting separately, so the clarity of each pixel is higher and can express more data bit width, for example, from 4 bits to 8 bits.

[0129] See also Figure 8, an optional example of the internal situation of a transparent material such as storage particles DIE is shown in the figure, which contains different layers or different pages. The information recording points pix are arranged on the plane of the first page Page1 as a set of pixels of the page and the information recording material a in the groove is used as a mark. Each information recording point pix is ​​a pixel point and the color mapped by the information recording point pix can be changed by changing the preset characteristics of the information recording material a. In this embodiment, it is assumed that the information recording material a is a quantum dot material and the adjustable color or pixel value at the information recording material a is reflected in the preset characteristics of the information recording material a, such as the adjustable radius of the quantum dot particles. The information recording material a is attached to the groove of the first transparent layer or the first glass layer. If the radius of the information recording material a is small, the groove can be cancelled and the information recording material a can be directly attached to the layer of the first transparent layer or the first glass layer.

[0130] See also Figure 8 In an optional example, information recording points pix are arranged on the plane of the second page Page2 as a set of pixel points of the page and the information recording material b in the groove is used as a mark. Each information recording point pix is ​​a pixel point and the color mapped by the information recording point pix can be changed by changing the preset characteristics of the information recording material b. In this embodiment, it is assumed that the information recording material b is a quantum dot material and the adjustable color or pixel value at the information recording material b is reflected in the preset characteristics of the information recording material b, such as the adjustable radius of the quantum dot particles. The information recording material b is attached to the groove of the second transparent layer or the second glass layer. If the radius of the information recording material b is small, the groove can be cancelled and the information recording material b can be directly attached to the second transparent layer or the second glass layer.

[0131] See also Figure 8 In an optional example, information recording points pix are arranged on the plane of the third page Page3 as a set of pixel points of the page and the information recording material c in the groove is used as a mark. Each information recording point pix is ​​a pixel point and the color mapped by the information recording point pix can be changed by changing the preset characteristics of the information recording material c. In this embodiment, it is assumed that the information recording material c is a quantum dot material and the adjustable color or pixel value at the information recording material c is reflected in the preset characteristics of the information recording material c, such as the adjustable radius of the quantum dot particles. The information recording material c is attached to the groove of the third transparent layer or the third glass layer. If the radius of the information recording material c is small, the groove can be cancelled and the information recording material c can be directly attached to the layer of the third transparent layer or the third glass layer.

[0132] See also Figure 8In an optional example, information recording points pix are arranged on the plane of the fourth page Page3 as a set of pixels of the page and the information recording material d in the groove is used as a mark. Each information recording point pix is ​​a pixel point and the color mapped by the information recording point pix can be changed by changing the preset characteristics of the information recording material d. In this embodiment, it is assumed that the information recording material d is a quantum dot material and the adjustable color or pixel value at the information recording material d is reflected in the preset characteristics of the information recording material d, such as the adjustable radius of the quantum dot particles. The information recording material d is attached to the groove on the level of the fourth transparent layer or the fourth glass layer. If the radius of the information recording material d is small, the groove can be cancelled and the information recording material d can be directly attached to the level of the fourth transparent layer or the fourth glass layer. Three-primary color phosphor materials or artificial gemstone materials can also be used as raw materials for information recording materials a, b, c, and d.

[0133] See also Figure 8 , the information recording materials a, b, c, d seem to be on the same level when viewed from the direction perpendicular to the transparent material or from the vertical direction perpendicular to the page. The vertical projection images pro_img of the information recording materials a, b, c, d can better reflect this feature, but in fact they are not on the same level.

[0134] See also Figure 8 , the image acquisition process of the information recording point pix with address PK+1_address(xn,ym) on page PageK+1 is interfered by the information recording point pix with address PK_address(xn,ym) on another page PageK. The factors of this interference are various, and the specific possible situations of the interference results have been informed in the previous text. In the optional implementation example, it can be set that all information recording materials or all pixels located on different pages but with the same address are arranged in a spiral in the direction perpendicular to the page, and there is a spiral angle value between any pixel of the latter page and a pixel with the same address of the adjacent previous page. This is one of the optional ways to eliminate interference.

[0135] See also Figure 8 , all information recording materials a, b, c, d or all corresponding pixels located on different pages Page1-Page4 but with the same address (e.g. PQ_address(xj, yq) is the same address of each page and Q=1 to 4) are arranged in a spiral structure in a direction perpendicular to these pages Page1-Page4. A single information recording point pix(xj, yq) at the address PQ_address(xj, yq) satisfies 0≤j≤n, 0≤q≤m. j=0, 1…n and q=0, 1…m.

[0136] See also Figure 8There is a pre-designed spiral angle value between the information recording material b or the corresponding pixel point of the next page Page2 with the address P2_pix(x0,y0) and the information recording material a or the corresponding pixel point of the adjacent previous page Page1 with the address P1_pix(x0,y0), for example, a spiral angle value of 90° or other angles.

[0137] See also Figure 8 There is a pre-designed spiral angle value between the information recording material c or the corresponding pixel point of the next page Page3 with the address P3_pix(x0,y0) and the information recording material b or the corresponding pixel point of the adjacent previous page Page2 with the address P2_pix(x0,y0), for example, a spiral angle value of 90° or other angles.

[0138] See also Figure 8 There is a pre-designed spiral angle value between the information recording material d or the corresponding pixel point of the next page Page4 with the address P4_pix(x0,y0) and the information recording material c or the corresponding pixel point of the adjacent previous page Page3 with the address P3_pix(x0,y0), for example, a spiral angle value of 90° or other angles.

[0139] See also Figure 8, two pixels belonging to two different pages but with the same address are set to be non-overlapping. In the example shown in the figure, two pixels belonging to two different pages Page2 and Page3 but with the same address: the information recording material b or the corresponding pixel with the address P2_pix(x0,y0) in the example of page Page2, and the information recording material c or the corresponding pixel with the address P3_pix(x0,y0) in the example of page Page3, do not overlap. Moreover, by adjusting the value of the spiral angle between the two, the size of their overlapping area can be changed. For example, by adjusting the spiral angle between the two to less than 90°, the two can be changed from non-overlapping to overlapping. Therefore, two pixels belonging to any two different pages but with the same address can also overlap up and down, which is another option. It is worth noting that transparent materials have a concern that needs to be resolved. The light reflected back from the transparent material will naturally increase the noise in the read incident light beam, thereby causing the signal-to-noise ratio of the read light to drop significantly. The important advantage or use of the above example is to suppress the noise in the reading light used to read the data of the information recording point and improve the signal-to-noise ratio of the reading light. All the pixels located in different pages but with the same address are arranged in a spiral shape in the direction perpendicular to the page, and the two pixels belonging to any two different pages but with the same address are set not to overlap each other. The purpose is: during the period of extracting the data of the read page, the light reflected from the non-read page in the transparent material is mixed into the reading light, which induces a decrease in the signal-to-noise ratio of the reading light. In particular, the light reflected from the information recording point in the non-read page in the transparent material with the same reading address as the reading page but located in different pages is suppressed from mixing into the reading light, which further induces a decrease in the signal-to-noise ratio of the reading light. Specifically, the reflected light will oscillate back and forth at the information recording points of each layer in the non-read page with the same reading address as the reading page but located in different pages to form a series of parasitic light pulse sequences. The light pulse sequences are mixed into the reading light and will destroy the real data of the information recording point at the reading address of the reading page. The intensity or amplitude of the light pulse sequence formed by the oscillation between two information recording points of any two pages tends to become stronger as the distance between the two pages decreases. The oscillating light pulse sequence of any two pages will also partially transmit to the information recording points with the same address on the adjacent page. The light pulse sequence often carries local data information of other information recording points with the same address but not on the read page. The incidental achievement of this example in achieving the relevant advantages is that it can still ensure the high storage density of information recording points in the transparent material.

[0140] See also Figure 8, and another example is that two pixels on different pages but with the same address do not overlap. In the example shown in the figure, there are two pixels belonging to two different pages Page3 and Page4 but with the same address: the information recording material c or the corresponding pixel with the address P3_pix(x2,y1) in the example of page Page3, and the information recording material d or the corresponding pixel with the address P4_pix(x2,y1) in the example of page Page4, they do not overlap. And by adjusting the value of the spiral angle between the two, the size of their overlapping area can be changed, for example, by adjusting the spiral angle between the two to less than 90°, the two can be changed from non-overlapping to overlapping.

[0141] See also Figure 8 There are many ways of illumination. A typical method is to use an LED light source to illuminate the page object to be read or to illuminate a single or several information recording points to be read. The camera focuses on the object to be read and takes an image. The light source can illuminate from the front side of the transparent material or from the back side of the transparent material. If a handheld device such as a mobile phone or a tablet is used for shooting, the built-in light source of the handheld device can be used to illuminate the transparent material. If a focusing camera is used, a focusable light source can be separately equipped for it to illuminate the transparent material. There are many optional light sources such as laser light sources or ordinary lamp light sources. Although natural light is not very friendly to reading, it is also one of the alternatives. In addition, a waveguide that can guide light can be directly integrated into the transparent material, and some light sources can be integrated with the transparent material, such as light-emitting diodes or laser tubes, etc., and transparent materials such as storage particles DIE are encapsulated in the same package. The light source integrated in the package guides the light to illuminate the reading object, such as the information recording material of the information recording point, through the waveguide. The power and reachable depth or focus point of the light source allow for adjustment, such as the illumination depth or focus position changes with the reading object. It can be seen that the memory based on transparent materials in this article is a cold storage medium, which generates almost no heat and has a very high storage density. Other advantages include the flexibility to choose parallel data reading or serial data reading, and information is almost never lost.

[0142] The above description and drawings provide typical embodiments of specific structures of specific embodiments. The above invention provides existing preferred embodiments, but these contents are not intended to be limiting. For those skilled in the art, various changes and modifications will undoubtedly be obvious after reading the above description. Therefore, the attached claims should be regarded as covering all changes and modifications of the true intent and scope of the present invention. Any and all equivalent scopes and contents within the scope of the claims should be considered to still be within the intent and scope of the present invention.

Claims

1. A method for recording information, characterized in that: include: A plurality of information recording points are arranged in the transparent material, each of which is provided with information recording materials, and different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics reflect different colors under light irradiation; The data stored in the transparent material is represented by the color reflected by the information recording point; The information recording material includes quantum dot material, and the different preset characteristics include setting the particle radius of the quantum dot material to be different; or The information recording material includes gemstone material, and the different preset characteristics include adjusting the proportions of various chemical components in the gemstone material to different proportions; or The information recording material includes three primary color materials for color mixing, and the different preset characteristics include setting the color mixing ratio relationship of the three primary color materials to be different.

2. The information recording method according to claim 1, characterized in that: Extracting image data mapped by a plurality of information recording points of a transparent material provided in advance as a sample under light irradiation, and providing a series of data sets of the image data in a manner of changing preset characteristics; A neural network is trained using the data set, and the trained neural network is used to infer the data represented by the image to be identified mapped by the information recording points of other transparent materials under light irradiation.

3. The information recording method according to claim 1, characterized in that: The transparent material at least includes a transparent semiconductor wafer or a transparent ceramic wafer or glass.

4. The information recording method according to claim 1, characterized in that: The transparent material is divided into a plurality of pages, and a plurality of information recording points arranged on the plane of any page serve as a pixel point set of the any page, and each information recording point serves as a pixel point.

5. The information recording method according to claim 4, characterized in that: Two pixels belonging to any two different pages but with the same address are set to be non-overlapping; or Two pixels belonging to any two different pages but with the same address overlap each other.

6. The information recording method according to claim 4, characterized in that: The image of all pixels on each page is used as parallel data to be read in parallel at one time, that is, the data of one page is read at a time with the page as the minimum reading unit; or After the data of each pixel of each page is read, the data of the next pixel is read, that is, a single pixel is used as the minimum reading unit each time and the data of one pixel is read at a time in a serial manner.

7. The information recording method according to claim 4, characterized in that: All pixels located in different pages but with the same address are arranged in a spiral in a direction perpendicular to the page, and there is a spiral angle value between any pixel of the latter page and a pixel of the same address of the adjacent previous page.

8. A storage system, characterized in that: include: A transparent material, in which a plurality of information recording points are arranged, each of which is provided with information recording materials and different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics reflect different colors under light irradiation; A computer is provided with a camera, the camera captures image data mapped by a plurality of information recording points of a transparent material as a sample under light irradiation, and provides a series of data sets of the image data in a manner of changing preset characteristics, and the camera exchanges the data sets to the computer; The computer trains a neural network using the data set, and uses the trained neural network to infer the data represented by the image to be identified mapped by the information recording points of other transparent materials under light illumination; The information recording material includes quantum dot material, and the different preset characteristics include setting the particle radius of the quantum dot material to be different; or The information recording material includes gemstone material, and the different preset characteristics include adjusting the proportions of various chemical components in the gemstone material to different proportions; or The information recording material includes three primary color materials for color mixing, and the different preset characteristics include setting the color mixing ratio relationship of the three primary color materials to be different.

9. A three-dimensional memory, characterized in that: include: A plurality of information recording points are arranged in the transparent material, each of which is provided with information recording materials, and different data are represented by information recording materials with different preset characteristics, and the information recording materials with different preset characteristics reflect different colors under light irradiation; The transparent material is divided into a plurality of pages, wherein a plurality of information recording points arranged on a plane of any page are used as a pixel point set of the any page, and each information recording point is used as a pixel point; The data stored in the transparent material is represented by the color reflected by the information recording point; The information recording material includes quantum dot material, and the different preset characteristics include setting the particle radius of the quantum dot material to be different; or The information recording material includes gemstone material, and the different preset characteristics include adjusting the proportions of various chemical components in the gemstone material to different proportions; or The information recording material includes three primary color materials for color mixing, and the different preset characteristics include setting the color mixing ratio relationship of the three primary color materials to be different.

10. The three-dimensional memory according to claim 9, characterized in that: The image of all pixels on each page is used as parallel data to be read in parallel at one time, that is, the data of one page is read at a time with the page as the minimum reading unit; or After the data of each pixel of each page is read, the data of the next pixel is read, that is, a single pixel is used as the minimum reading unit each time and the data of one pixel is read at a time in a serial manner.

Images