A carton recycling environment information verification method and system

Abstract

The application discloses a carton recycling environment information verification method and system, and relates to the technical field of carton recycling.The method comprises the following steps: scanning three-dimensional structure features on a non-corrugated environment-friendly packaging box body based on a three-dimensional scanning device, wherein the geometric arrangement mode of the three-dimensional structure features is encoded with environment authentication information; converting the three-dimensional structure features into three-dimensional image data; extracting geometric feature data corresponding to the three-dimensional structure features from the three-dimensional image data; analyzing the geometric feature data based on a preset coding rule to obtain the environment authentication information; and determining a carton recycling type in a recycling environment information library based on the environment authentication information.The embodiment of the application can stably and reliably obtain environment authentication information on a non-corrugated environment-friendly packaging box body, and has the advantages of improving the robustness of information identification and verification efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of carton recycling, and in particular to a carton recycling environmental information verification method and system. BACKGROUND

[0002] In modern industrial production and consumption cycles, the environmental attributes of packaging materials, particularly their recyclability and sustainability, are increasingly attracting attention. In order to ensure and prove that packaging materials meet specific environmental standards, they are usually subjected to environmental certification and attached with corresponding certification information on the packaging carrier. However, in actual supply chain circulation and recycling processing links, these attached information often face serious challenges, especially for some non-traditional structure or material packaging boxes, the reliability of their information identification and verification becomes a problem to be solved.

[0003] For example, the surface of a non-corrugated pulp molded box may not be as smooth as corrugated cardboard, and the printing quality can be easily affected. During transportation and storage, the box often suffers physical stress. For example, when the box is stacked, especially the bottom box, it will bear significant vertical load, causing compression deformation of the box structure. This deformation can cause the two-dimensional code pattern printed on the surface of the box to be distorted or stretched, beyond the tolerance range of conventional scanning equipment, resulting in identification failure. At the same time, friction and collision during handling can also leave wear marks on the surface of the box. If the wear area happens to be on the two-dimensional code or serial number, the information may be destroyed and cannot be read. Secondly, the packaging box will inevitably be exposed to various environmental factors during the process. For example, in a humid environment, the pulp material will absorb moisture, causing the box to swell and soften, and the printed ink may spread, making the edges of the two-dimensional code blurred and difficult to identify. In a dusty environment, dust and dirt can easily accumulate on the surface of the box, directly covering the printed information and hindering the normal operation of the scanning equipment. Long-term exposure to sunlight can cause the printed ink to fade, reducing the visibility and readability of the information. Furthermore, different products may use packaging boxes of different sizes and structural designs (although they are all non-corrugated), and may obtain environmental certifications of different types or from different agencies. This means that the information format, identification style, and corresponding verification platform on the box may vary. When handling a large number of different types of boxes, users or automated systems need to accurately identify the type and location of the information carrier and adapt the corresponding reading and verification process. If the information carrier is difficult to identify due to the above unfavorable factors, it cannot be determined which certification system it corresponds to, and subsequent verification cannot be performed. In addition, for recycling stations, it is necessary to quickly and batch process recyclables. If each non-corrugated environmentally friendly packaging box needs to be manually checked for the status of the information carrier, or the verification process is time-consuming and prone to failure, it will significantly reduce the recycling efficiency. The environment of the recycling station is also not conducive to accurate information identification, such as unstable lighting conditions and cluttered backgrounds. Considering the material properties (such as the hygroscopicity of pulp molding and surface roughness) and structural characteristics (deformation mode under stress) of non-corrugated environmentally friendly packaging boxes, as well as the various unfavorable conditions (physical damage, stains, humidity, light changes, batch processing requirements) they face in actual supply chains and recycling scenarios, traditional surface printed information carriers and their verification methods cannot guarantee stability and efficiency. SUMMARY

[0004] The purpose of the present application is to provide a carton recycling environmental information verification method and system, which can stably and reliably obtain environmental certification information on non-corrugated environmentally friendly packaging boxes, and improve the robustness of information identification and verification efficiency.

[0005] The present application provides a carton recycling environmental information verification method, comprising:

[0006] The three-dimensional structure features on the non-corrugated environment-friendly packaging box body are scanned based on a three-dimensional scanning device, and the geometric arrangement mode of the three-dimensional structure features is encoded with environment authentication information;

[0007] The three-dimensional structure features are converted into three-dimensional image data;

[0008] Geometric feature data corresponding to the three-dimensional structure features is extracted from the three-dimensional image data;

[0009] The geometric feature data is parsed based on a preset encoding rule to obtain the environment authentication information;

[0010] The carton recycling type is determined in a recycling environment information library based on the environment authentication information;

[0011] The geometric feature data is parsed based on a preset encoding rule to obtain the environment authentication information, including:

[0012] The numerical range and spatial position relationship in the geometric feature data are parsed;

[0013] The numerical range and spatial position relationship are mapped to a sequence of encoding symbols;

[0014] The sequence of encoding symbols is parsed based on the preset encoding rule to obtain the environment authentication information.

[0015] The above scheme can stably and reliably obtain the environment authentication information on the non-corrugated environment-friendly packaging box body, and improve the robustness and verification efficiency of information identification.

[0016] Optionally, extracting the geometric feature data corresponding to the three-dimensional structure features from the three-dimensional image data includes:

[0017] Image preprocessing is performed on the three-dimensional image data to obtain three-dimensional standardized image data;

[0018] The geometric feature data corresponding to the three-dimensional structure features is extracted from the three-dimensional standardized image data.

[0019] The above scheme improves the accuracy and reliability of extracting geometric features from three-dimensional image data.

[0020] Optionally, image preprocessing is performed on the three-dimensional image data to obtain three-dimensional standardized image data, including:

[0021] The spatial position and depth data under the three-dimensional array structure in the three-dimensional image data are identified;

[0022] Local reference data is constructed based on the spatial position and depth data under the three-dimensional array structure;

[0023] The relative depth of the three-dimensional structural features relative to the local reference data is calculated to obtain three-dimensional normalized image data.

[0024] The above approach, by constructing local benchmark data, further improves the accuracy of 3D image data preprocessing, laying the foundation for subsequent feature extraction.

[0025] Optionally, identifying the spatial location and depth data of a three-dimensional array structure in the three-dimensional image data includes:

[0026] Identifying the target region where a 3D array structure is located based on a geometric template matching algorithm;

[0027] Each coded feature is identified from the target region based on surface curvature analysis or local point cloud clustering methods. The coded features are pits or bumps.

[0028] Calculate the spatial coordinates and depth data for each encoded feature.

[0029] The above scheme employs a precise recognition algorithm, ensuring the accuracy of acquiring the spatial location and depth data of the encoded features.

[0030] Optionally, constructing local reference data based on spatial location and depth data under a three-dimensional array structure includes:

[0031] Identify reference features within the spatial neighborhood of each encoded feature;

[0032] Based on the spatial coordinates, depth data, and reference features of each encoded feature, the local reference height of the location of each information encoded feature is determined by a weighted interpolation method.

[0033] The local reference height set of all information encoding features corresponding to the three-dimensional array structure is used to form local reference data.

[0034] The above scheme improves the robustness and accuracy of local benchmark data construction by introducing reference features and weighted interpolation methods.

[0035] Optionally, the relative depth of the three-dimensional structural features relative to the local reference data is calculated to obtain three-dimensional normalized image data, including:

[0036] Identify 3D point cloud data in 3D image data;

[0037] Outlier filtering and point cloud registration are performed on the 3D point cloud data to obtain preprocessed point cloud data;

[0038] Calculate the vertical distance from each point in the preprocessed point cloud data to the local surface reference;

[0039] The relative depth of each coding feature in the three-dimensional structure feature relative to the local surface reference is obtained by statistical processing of the vertical distance.

[0040] Through the above scheme, the calculation accuracy and reliability of the relative depth of the three-dimensional structure feature are ensured by fine processing of the three-dimensional point cloud data.

[0041] Optionally, the geometric arrangement pattern of the three-dimensional structure feature is solidified and formed together with the box material in the box material forming process through the mold structure corresponding to the three-dimensional structure feature.

[0042] Through the above scheme, the permanence and tamper resistance of the three-dimensional structure feature are ensured, and the reliability of the information carrier is improved.

[0043] Optionally, the three-dimensional structure feature on the non-corrugated environment-friendly packaging box is scanned based on a three-dimensional scanning device, including:

[0044] The structured light is projected onto the surface of the non-corrugated environment-friendly packaging box by the three-dimensional scanning device.

[0045] The deformed pattern of the structured light is captured by the three-dimensional scanning device, and the deformed pattern is used to form three-dimensional image data by the three-dimensional scanning device.

[0046] Through the above scheme, the specific implementation of three-dimensional scanning is clarified, and the accuracy and efficiency of initial data acquisition are improved.

[0047] Optionally, the present application also proposes a carton recycling environmental information verification system, the system comprising:

[0048] A three-dimensional scanning module is configured to scan a three-dimensional structure feature on a non-corrugated environment-friendly packaging box based on a three-dimensional scanning device, and the geometric arrangement pattern of the three-dimensional structure feature is encoded with environmental authentication information.

[0049] A three-dimensional imaging module is configured to convert the three-dimensional structure feature into three-dimensional image data.

[0050] A data processing module is configured to extract geometric feature data corresponding to the three-dimensional structure feature from the three-dimensional image data.

[0051] An analysis and verification module is configured to analyze the geometric feature data based on a preset encoding rule to obtain environmental authentication information, and the analysis of the geometric feature data based on the preset encoding rule to obtain the environmental authentication information includes: analyzing the numerical range and spatial position relationship in the geometric feature data; mapping the numerical range and spatial position relationship to a sequence of encoding symbols; and analyzing the sequence of encoding symbols based on the preset encoding rule to obtain the environmental authentication information.

[0052] A recycling matching module is configured to determine the carton recycling type in a recycling environmental information library based on the environmental authentication information.

[0053] Through the above scheme, a system architecture for implementing the above method is provided, which facilitates automatic and integrated operation and improves the efficiency of the overall verification process.

[0054] As can be seen, the paper box recycling environmental information verification method and system provided by the application effectively solves the problems of traditional surface printing information being easily damaged by physical damage, environmental factors, low recognition rate and information format diversity leading to verification difficulties by using three-dimensional structural features as the carrier of environmental certification information and combining three-dimensional scanning, image processing and data analysis technology, and has the advantages of being able to stably and reliably obtain environmental certification information on non-corrugated environmentally friendly packaging boxes, improving the robustness of information recognition and verification efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, a brief introduction will be given below to the drawings needed to be used in the embodiments or prior art description, and obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0056] Figure 1 is a paper box recycling environmental information verification method flowchart in the embodiment of the application;

[0057] Figure 2 is a method flowchart for calculating the relative depth of three-dimensional structural features relative to local reference data to obtain three-dimensional standardized image data in the embodiment of the application;

[0058] Figure 3 is a paper box recycling environmental information verification system structure schematic diagram in the embodiment of the application. DETAILED DESCRIPTION

[0059] The technical solutions in the application will be described in detail below with reference to the drawings in the application. Obviously, the described embodiments are only some of the embodiments of the application, not all. The components of the application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the application provided in the drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0060] It should be noted that similar reference numerals and letters refer to like items in the accompanying drawings, and once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0061] In the embodiments of the present application, the three-dimensional structural features of the box body are taken as information carriers, and the geometric arrangement mode thereof is obtained through three-dimensional scanning technology, so as to realize the identification and verification of the environmental authentication information. That is, the present application proposes a carton recycling environmental information verification method, comprising: scanning the three-dimensional structural features on the non-corrugated environmentally friendly packaging box body based on a three-dimensional scanning device, the geometric arrangement mode of the three-dimensional structural features being encoded with environmental authentication information; converting the three-dimensional structural features into three-dimensional image data; extracting the geometric feature data corresponding to the three-dimensional structural features from the three-dimensional image data; analyzing the geometric feature data based on a preset encoding rule to obtain the environmental authentication information; and determining the carton recycling type in the recycling environmental information library based on the environmental authentication information. The method aims to solve the problem of how to realize the stable and accurate identification of the environmental authentication information carriers attached to the surface or inside of the non-corrugated environmentally friendly packaging box body in a complex environment, such as physical damage, surface contamination, structural deformation or humid environment, and quickly complete the verification of the environmental authentication information, so as to ensure the recycling efficiency and information reliability.

[0062] Specifically, Figure 1 The flow chart of the carton recycling environmental information verification method in the embodiments of the present application is shown, and specifically as follows:

[0063] S101, scanning the three-dimensional structural features on the non-corrugated environmentally friendly packaging box body based on a three-dimensional scanning device;

[0064] It should be noted that the geometric arrangement mode of the three-dimensional structural features is encoded with environmental authentication information.

[0065] The three-dimensional scanning device herein refers to a device capable of acquiring the three-dimensional spatial information of an object surface, which can be implemented using technologies such as laser triangulation, structured light projection, time-of-flight (ToF), or stereo vision, for example, by projecting light with a known pattern onto the object surface and analyzing its deformation to obtain depth information, which is mainly for obtaining non-contact three-dimensional geometric data. The non-corrugated environmentally friendly packaging box refers to a packaging container made of other environmentally friendly materials and forming processes instead of traditional corrugated paperboard structures, which can be made of paper pulp molding, plant fiber pressing, or composite materials, for example, a paper pulp molding box, which is mainly to provide a packaging carrier with specific environmental properties and possibly uneven surface. The three-dimensional structural feature refers to a pre-set, specific geometric shape and spatial arrangement of concave-convex, texture, or pattern on the surface or inside the box, which can be formed by mold pressing, injection molding, or post-engraving, for example, a series of regularly arranged pits or protrusions, which is mainly to serve as a physical, non-wearable information carrier.

[0066] Encoding the geometric arrangement pattern with environmental authentication information refers to representing predetermined environmental authentication information through specific combinations or arrangements of geometric properties such as relative position, shape, size, or depth of three-dimensional structural features in space, which can be implemented using binary encoding, multi-encoding, or encoding based on the relative positions of feature points, for example, using the depth difference of pits or the arrangement order of protrusions to represent different information bits, which is mainly to solidify environmental authentication information on the box in a physical, non-visual readable manner, improving the robustness of information identification.

[0067] S102, converting the three-dimensional structural feature into three-dimensional image data;

[0068] Three-dimensional image data refers to a set of digital data obtained by a three-dimensional scanning device describing the three-dimensional spatial coordinates and / or depth information of an object surface, which can be represented in formats such as point cloud data, mesh models, or depth maps, for example, a data set composed of a series of points with X, Y, Z coordinates, which is mainly to convert three-dimensional geometric information of the physical world into a computer-processed digital form.

[0069] S103, extracting geometric feature data corresponding to the three-dimensional structural feature from the three-dimensional image data;

[0070] Geometric feature data refers to quantized data obtained by processing and extracting from three-dimensional image data to represent key geometric properties of three-dimensional structural features, which can include spatial coordinates, relative depth, curvature, normal vector, or topological relationship of feature points, for example, the center coordinates of a specific pit or protrusion and its depth value relative to the reference plane, which is mainly to refine the original three-dimensional image data into structured data that can be used for information analysis.

[0071] S104, based on the preset coding rule, the geometric feature data is parsed to obtain the environment authentication information;

[0072] The preset coding rule refers to the mapping relationship and analysis logic defined in advance when the environment authentication information is encoded into the three-dimensional structure feature. It can include numerical range definition, spatial position relationship analysis rule or symbol mapping table, etc. For example, a certain depth value range corresponds to a certain binary bit, or a certain arrangement mode corresponds to a certain authentication type. The main purpose is to ensure that the geometric feature data can be accurately inversely analyzed into the original environment authentication information.

[0073] S105, based on the environment authentication information, the carton recycling type is determined in the recycling environment information library.

[0074] The recycling environment information library refers to a database system that stores various types of carton environment authentication information, corresponding recycling standards and processing procedures. It can include authentication agency information, material composition data, recycling grade classification or processing suggestions, etc. For example, a database containing recycling guidelines for different environmentally friendly grade cartons. The main purpose is to provide an authoritative data source for verifying and matching the parsed environment authentication information to guide the subsequent recycling classification.

[0075] The method in the embodiment of the application solves the technical problems that the traditional surface printed information is easy to be damaged and unstable in recognition by taking the three-dimensional structure feature on the non-corrugated environmentally friendly packaging box as the physical carrier of the environment authentication information, combining with the three-dimensional scanning device to acquire data, and subsequently extracting the geometric feature data and parsing the preset coding rule. The effect of stably and accurately verifying the carton environment authentication information and efficiently determining the recycling type in a complex environment is achieved.

[0076] The method in the embodiments of the present application overcomes the disadvantage that the traditional surface printed information is easily damaged by using the three-dimensional structure features inherent in the non-corrugated environmentally friendly packaging box as the carrier of environmental authentication information. First, the three-dimensional scanning device non-contact scans the three-dimensional structure features on the box to obtain accurate three-dimensional geometric information. The geometric arrangement pattern of these three-dimensional structure features is pre-encoded with environmental authentication information, which means that the information is physically solidified in the box structure, rather than just attached to the surface. The three-dimensional structure features obtained by scanning are then converted into three-dimensional image data, which digitizes the three-dimensional geometric information of the physical world and lays the foundation for computer processing. Then, the system accurately extracts the geometric feature data corresponding to the three-dimensional structure features from the three-dimensional image data, such as the spatial coordinates and relative depth of specific concave-convex points, which are quantitative representations of the original three-dimensional structure features. Subsequently, based on the preset encoding rules, the system parses the extracted geometric feature data to reverse decode the quantitative geometric information into the original environmental authentication information. This parsing process relies on the pre-established mapping relationship between the geometric features and the authentication information. Finally, the system uses the parsed environmental authentication information to compare and query in the environmental information library, thereby accurately determining the recycling type of the carton. The entire process forms a closed loop from the acquisition of physical features to the parsing and application of information, ensuring that the environmental authentication information of the box can still be stably and accurately identified and verified under adverse conditions such as physical damage, surface stains or deformation, thereby efficiently guiding the recycling classification of the carton.

[0077] In the embodiments of the present application, a high-precision structured light three-dimensional scanning device can be used to scan the non-corrugated environmentally friendly packaging box. The box is integrally formed by a pulp molding process, and a series of regularly arranged small pits are preformed on the surface as three-dimensional structural features. The depth difference or spatial arrangement sequence of these pits pre-encodes the environmental certification information of the box. For example, deeper pits represent binary "1", and shallower pits represent binary "0". The scanning device generates high-density point cloud data, i.e. three-dimensional image data, of the box surface by projecting structured light and capturing its deformed pattern. Subsequently, the data processing module processes the point cloud data, accurately identifies the position of each pit and its relative depth with respect to the box reference plane, and extracts the geometric feature data. For example, the center coordinates and depth values of each pit are extracted and stored. Then, the analysis and verification module parses the geometric feature data into environmental certification information according to the pre-set encoding rules. For example, by comparing the depth values of the pits with the pre-set depth threshold, the depth information of a series of pits is converted into a binary sequence, and the binary sequence is decoded into specific environmental certification codes, such as "biodegradable" or "high recovery rate paper pulp". Finally, the recycling matching module sends the parsed environmental certification information to a cloud-based environmental information database for query. The database stores the paper box recycling types and processing guidelines corresponding to different certification codes, so as to determine which recycling type the paper box should be classified into, such as "A-grade paper pulp recycling" or "special material recycling".

[0078] Through the above technical solutions, the present application can effectively solve the problem of unstable identification of traditional information carriers in non-corrugated environmentally friendly packaging boxes in complex environments, such as physical damage, surface contamination, structural deformation or in humid environments. Since the environmental certification information is encoded in the three-dimensional structural features of the box itself, rather than in the easily damaged surface printing, the information carrier has higher robustness and anti-interference ability. The combination of three-dimensional scanning technology, subsequent geometric feature extraction and pre-set encoding rule analysis ensures that the environmental certification information can be stably and accurately obtained and verified even in adverse conditions. This significantly improves the reliability and verification efficiency of information identification in the paper box recycling process, and helps to realize accurate classification and efficient recycling of environmentally friendly packaging boxes, thereby promoting sustainable development.

[0079] The step of extracting geometric feature data corresponding to three-dimensional structural features from three-dimensional image data in the embodiments of the present application includes: image pre-processing of three-dimensional image data to obtain three-dimensional standardized image data; extracting geometric feature data corresponding to three-dimensional structural features from three-dimensional standardized image data.

[0080] The image preprocessing here refers to a series of operations on the original three-dimensional image data to eliminate or reduce noise, inconsistency or bias in the data, thereby improving the data quality and the accuracy of subsequent processing. Various techniques can be used, such as statistical-based filtering methods to smooth the data and remove random noise, or model fitting-based methods to correct data bias, data interpolation or completion algorithms to handle data missing, or coordinate system conversion and alignment, the purpose of which is to optimize the quality of three-dimensional image data, making it more suitable for subsequent feature extraction, ensuring that the extracted geometric feature data has higher accuracy and reliability; the three-dimensional standardized image data refers to the three-dimensional image data after image preprocessing, the data quality is improved, the noise and bias are effectively suppressed, which can be specifically manifested as a data set with a unified coordinate system, consistent depth or size ratio, and clearer geometric contour and surface details, the purpose of which is to provide a clean, accurate and consistent data basis for subsequent geometric feature extraction, thereby improving the robustness and precision of feature extraction.

[0081] Here, by introducing an image preprocessing step, the process of extracting geometric feature data corresponding to three-dimensional structural features from three-dimensional image data is optimized. Specifically, after obtaining the original three-dimensional image data, first, the data is preprocessed. This preprocessing step aims to identify and eliminate possible noise, incompleteness or systematic bias in the original data, such as data distortion caused by scanning device accuracy, environmental light changes or irregularities on the box surface. Through filtering, smoothing, denoising or data completion operations, the original three-dimensional image data is converted into three-dimensional standardized image data, which can more accurately reflect the true geometric shape and spatial position of the three-dimensional structural features. On this basis, the geometric feature data corresponding to the three-dimensional structural features is extracted from the preprocessed three-dimensional standardized image data. Since the quality of the input data has been significantly improved, the subsequent feature extraction algorithm can more accurately identify and quantify the geometric properties of the three-dimensional structural features, such as the depth, size, shape of the concave or convex, and their relative positional relationship. This step-by-step processing method ensures that the extracted geometric feature data has higher accuracy and reliability. This processing flow is closely integrated with the overall framework of the carton recycling environment information verification method. After converting three-dimensional structural features into three-dimensional image data, preprocessing the image data provides high-quality input for subsequent geometric feature data extraction, making it possible to analyze geometric feature data based on pre-set coding rules, obtain environmental authentication information, and determine the carton recycling type in the recycling environment information library based on the environmental authentication information. This effectively avoids inaccurate feature extraction due to poor quality of the original data, which in turn affects the reliability of environmental authentication information analysis and verification results, thereby improving the robustness and accuracy of the entire verification method, especially in the complex and variable environment of recycling stations, which can stably identify and verify the environmental authentication information of the box.

[0082] The step of image preprocessing of three-dimensional image data in the present application to obtain three-dimensional standardized image data includes: identifying the spatial position and depth data under the three-dimensional array structure in the three-dimensional image data; constructing local reference data based on the spatial position and depth data under the three-dimensional array structure; calculating the relative depth of the three-dimensional structural features relative to the local reference data to obtain the three-dimensional standardized image data.

[0083] The three-dimensional array structure refers to a specific and regular arrangement pattern of three-dimensional structure features (such as pits or protrusions) for encoding environmental certification information on the surface of a non-corrugated environmentally friendly packaging box. This pattern can be matrix, linear arrangement or other predefined geometric layout, and the purpose is to provide an identifiable and analyzable framework for information coding. The spatial position and depth data refer to the precise coordinates of each independent coding feature in the three-dimensional coordinate system and the vertical distance of these features relative to the box surface or a reference plane. These data are the basis for quantifying the geometric shape of the three-dimensional structure features and are used for subsequent benchmark construction and relative depth calculation. The local benchmark data refer to the reference plane or reference surface obtained by calculation or fitting based on the spatial position and depth data of each coding feature in the three-dimensional array structure, which can reflect the local surface morphology of the box. This benchmark is "local", meaning it can adapt to the slight deformation or unevenness of the box surface. Its purpose is to provide a dynamic and adaptive reference zero point for subsequent depth standardization, thereby eliminating measurement errors caused by box posture, scanning angle or surface deformation. The relative depth refers to the actual depth value of each coding feature in the three-dimensional structure, minus the depth value represented by the local benchmark data at its corresponding position. In this way, the absolute depth is converted into a difference value relative to the local reference, thereby eliminating the influence of external environment (such as scanning distance, box inclination) and box deformation on depth measurement, making the depth data under different box conditions and different scanning conditions comparable and consistent.

[0084] The scheme of the present application aims to overcome the noise, outliers and data inconsistency problems caused by box deformation in the original data through fine image preprocessing of the original three-dimensional image data. First, by identifying the three-dimensional array structure in the three-dimensional image data and the spatial position and depth data thereunder, the geometric features carrying environmental authentication information can be accurately located and quantified. This step is the basis for subsequent processing, which ensures accurate capture of information encoding features, even if there is a certain degree of interference in the data, the key area can be locked. On this basis, based on the identified three-dimensional array structure, spatial position and depth data, local reference data is constructed. The establishment of local reference is the key of the present scheme, which is different from the simple global plane fitting, but can dynamically adapt to the possible local unevenness or deformation of the surface of the non-corrugated environmentally friendly packaging box. By establishing a local reference plane or surface for each information encoding feature or its neighborhood, the depth measurement deviation introduced by the box deformation, scanning angle change or placement attitude difference can be effectively eliminated. This makes the subsequent depth calculation in a more stable and accurate reference system. Further, the relative depth of the three-dimensional structure feature with respect to the local reference data is calculated, thereby obtaining three-dimensional standardized image data. This calculation process converts the original absolute depth value into a difference value relative to the local reference, completely eliminating the influence of external environmental factors (such as the distance between the scanning device and the box, the overall inclination of the box) on the depth measurement. Thus, no matter what attitude the box is in or what slight deformation it experiences, the depth information of its three-dimensional structure feature can be standardized to a value with high consistency and comparability. Through the above preprocessing process, the inherent noise and inconsistency in the original three-dimensional image data are significantly reduced, and the data quality is greatly improved. This high-quality three-dimensional standardized image data provides a solid foundation for subsequent extraction of geometric feature data corresponding to three-dimensional structure feature data from three-dimensional standardized image data. This means that even if the original scanning data has defects, after preprocessing by the present scheme, stable and accurate geometric feature data can be obtained, thereby ensuring the accuracy and robustness of environmental authentication information based on the preset encoding rule, and ultimately improving the reliability of the paper box recycling environmental information verification. The implementation of the present scheme enables efficient and accurate acquisition and verification of environmental authentication information on non-corrugated environmentally friendly packaging boxes in complex and variable environments that are not conducive to traditional identification, effectively solving the shortcomings of traditional methods in data consistency and reliability.

[0085] In some preferred embodiments, the application is implemented as follows: when the three-dimensional scanning device obtains the three-dimensional image data of the non-corrugated environmentally friendly packaging box, firstly, the three-dimensional array structure can be identified by using an image processing algorithm. For example, a method based on template matching or feature point detection can be used to find the preset geometric pattern representing the information coding features (such as pits or protrusions) in the three-dimensional image data. Once these patterns are identified, the overall layout of the three-dimensional array structure can be determined. At the same time, for each identified coding feature in the array structure, the precise coordinates in the three-dimensional space and the depth data of its surface points can be extracted. These depth data can be extracted from the original depth map or point cloud data obtained directly from the scanning device. Further, based on these identified three-dimensional array structures, spatial positions and depth data, local reference data can be constructed. Specifically, for each coding feature in the three-dimensional array structure, the local area around it can be considered. For example, the coding feature and its immediately adjacent feature points or the plane area around it can be selected as a reference. By fitting the depth data of these reference points or areas, such as fitting a local plane using the least squares method, or using weighted average, interpolation, etc., the local reference depth of the position of the coding feature can be determined. Collecting the local reference depths of all coding features in the three-dimensional array structure forms the local reference data. This local reference data can reflect the ups and downs and tilts of the box surface in a local range. Finally, in order to obtain three-dimensional standardized image data, the relative depth of the three-dimensional structural features relative to the local reference data can be calculated. Specifically, for the original depth value of each coding feature in the three-dimensional image data, subtract the local reference depth value constructed at its corresponding position. For example, if the original depth of a pit is 10 mm and its local reference depth is 2 mm (indicating that the surface of this area itself is 2 mm lower than the overall reference surface), then its relative depth is 8 mm. Through this point-by-point or feature-by-feature subtraction operation, the effects of the overall posture of the box, the scanning distance and the local surface unevenness on the depth measurement can be eliminated, thereby obtaining a set of uniform and comparable relative depth values. These relative depth values constitute the three-dimensional standardized image data, which more accurately reflect the geometric shape of the three-dimensional structural features themselves, rather than their absolute position in space or depth affected by external factors.

[0086] The step of identifying the spatial position and depth data of the three-dimensional array structure in the three-dimensional image data in the embodiments of the application includes: identifying the target area where the three-dimensional array structure is located based on a geometric template matching algorithm; identifying each coding feature from the target area based on surface curvature analysis or local point cloud clustering method, the coding feature being a pit or a protrusion; calculating the spatial coordinates and depth data of each coding feature.

[0087] Among them, the geometric template matching algorithm refers to a method of searching and positioning similar geometric structures in the three-dimensional image data to be processed through a predefined geometric model (template), which can be implemented by adopting a feature point matching-based, shape context matching-based or iterative closest point (ICP) algorithm-based variant, and the purpose is to efficiently and accurately identify the specific area where the three-dimensional array structure may exist, thereby narrowing the scope of subsequent processing; the target area where the three-dimensional array structure is located refers to the local spatial range containing the three-dimensional array structure to be identified that is preliminarily positioned in the three-dimensional image data by the geometric template matching algorithm, which can be a bounding box, an irregular polygon area or a point cloud cluster, and the purpose is to limit the calculation amount of subsequent fine identification within the relevant area and avoid blind search on the entire image data; the surface curvature analysis refers to a method of describing the bending degree of the surface by calculating the curvature value of each point on the three-dimensional surface, which can be calculated by adopting mathematical methods such as Gaussian curvature, average curvature or principal curvature, and the purpose is to distinguish concave and convex coding features according to the surface geometric features; the local point cloud clustering method refers to a method of grouping points in the three-dimensional point cloud data that are close in space and have similar attributes (such as normal vector, color or density), which can be implemented by adopting algorithms such as K-means clustering, DBSCAN clustering or hierarchical clustering, and the purpose is to organize the discrete point cloud data into meaningful geometric entities, thereby identifying independent coding features; the coding feature refers to the smallest geometric unit in the three-dimensional array structure for bearing the environmental authentication information, which is specifically manifested as a pit or a protrusion on the surface of the box, and the purpose is to encode information through its specific geometric form and spatial arrangement; the spatial coordinates and depth data refer to data describing the accurate position (such as X, Y, Z coordinates) of each coding feature in the three-dimensional space and the vertical distance (depth value) thereof relative to the box surface reference, and the purpose is to provide accurate geometric quantitative basis for subsequent analysis of environmental authentication information.

[0088] The scheme of the present application realizes accurate identification of three-dimensional array structure and its encoding features in three-dimensional image data through a phased and refined processing flow. First, a geometric template matching algorithm is used to quickly locate the specific target area where the three-dimensional array structure may exist in the entire three-dimensional image data. This preprocessing step can effectively exclude the interference of irrelevant areas in the image, greatly reducing the search range of subsequent fine identification, thereby improving the processing efficiency and reducing the probability of misidentification. Once the target area is determined, the system focuses on the area inside. On this basis, in order to deal with the noise, dirt or local deformation that may exist on the surface of the box, the scheme further introduces surface curvature analysis or local point cloud clustering method. Surface curvature analysis can sensitively capture the small changes in surface geometry, thereby accurately distinguishing between pits or protrusions, these encoding features; while the local point cloud clustering method can aggregate point clouds with similar spatial properties, effectively identifying independent and discrete encoding features. Both of these methods can enhance the adaptability to complex surface conditions, ensuring that each encoding feature can be stably identified in adverse environments. Finally, for each identified encoding feature, the system will accurately calculate its coordinates in three-dimensional space and the depth data relative to the surface of the box. These accurate spatial coordinates and depth data are the key geometric quantitative basis for subsequent analysis of environmental authentication information, and they can accurately restore the geometric arrangement pattern of the encoding features. Through the synergistic effect of the above steps, the present scheme provides high-precision and high-robustness raw data input for subsequent construction of local reference data and calculation of the relative depth of three-dimensional structure features relative to the local reference data. In the preprocessing of three-dimensional image data, accurately identifying the spatial position and depth data of three-dimensional array structure and its encoding features is the premise of constructing reliable local reference, and also the basis for subsequent calculation of relative depth. It is precisely because the present scheme can overcome the challenges brought by noise, occlusion and surface irregularity, ensuring the accuracy of encoding feature identification, that the subsequent standardized image data generation, geometric feature data extraction and environmental authentication information analysis can be reliably carried out, thereby effectively solving the problem of difficult and stable and accurate identification of box surface environmental authentication information in complex environments, and significantly improving the reliability and efficiency of the entire carton recycling environmental information verification method.

[0089] In some preferred embodiments, the identification of spatial position and depth data of each coding feature in the three-dimensional array structure in the three-dimensional image data can be implemented as follows: first, the system can pre-store one or more geometric templates of the three-dimensional array structure, which can be derived from CAD models or high-precision scanning data of standard boxes. When receiving the three-dimensional image data to be processed, a geometric template matching algorithm based on feature point matching can be used, for example, by extracting the fast point feature histogram (FPFH) descriptor in the three-dimensional image data and the geometric template, and using the random sample consensus (RANSAC) algorithm for matching, so as to identify the target region where the three-dimensional array structure is located. This target region can be a minimum bounding box containing the entire array. Subsequently, in the identified target region, in order to finely identify each coding feature, surface curvature analysis can be performed on the point cloud data in the region. Specifically, the normal vector of each point cloud in the target region can be calculated, and the principal curvature thereof can be calculated based on the normal vector. By setting a suitable curvature threshold, the surface flat region, the concave region (pit) and the convex region (projection) can be distinguished. Alternatively, a local point cloud clustering method such as the DBSCAN algorithm can be used to cluster the point clouds in the target region, wherein the clustering parameters can be adjusted according to the typical size and density of the coding features, so as to identify each independent pit or projection as a point cloud cluster. Finally, for each identified coding feature point cloud cluster, the spatial coordinates and depth data thereof can be calculated. The spatial coordinates can take the geometric center point coordinates (for example, the average value of the X, Y and Z coordinates of all points) of the point cloud cluster, and the depth data can be calculated as the average vertical distance of the Z coordinate value of the point cloud cluster relative to the local surface reference. For example, a local plane can be fitted as a reference, and then the average vertical distance of the point cluster to the plane is calculated as the depth data.

[0090] The step of constructing the local reference data based on the spatial position and depth data of the three-dimensional array structure in the embodiments of the present application includes: identifying a reference feature in the spatial neighborhood of each coding feature; determining the local reference height of the position of each information coding feature by a weighted interpolation method according to the spatial coordinates, depth data and reference feature of each coding feature; and forming the local reference data by collecting the local reference heights of all information coding features corresponding to the three-dimensional array structure.

[0091] wherein a reference feature within a spatial neighborhood of each coded feature is identified, the "reference feature" herein refers to other coded features or non-coding feature points that have certain spatial correlation or geometric similarity around the target coded feature, which can be realized by point set screening based on distance threshold, K-neighbor searching or screening based on local geometric attributes (such as planarity), the purpose of which is to provide auxiliary information for subsequent local reference height determination, and to smooth potential noise by using adjacent data. The "spatial neighborhood" herein refers to a specific range defined in three-dimensional space with the target coded feature as the center, such as a spherical region or a cubic region. In addition, the "weighted interpolation method" refers to a method of assigning different weights to multiple data points and calculating the target point value based on these weights, which can be realized by inverse distance weighted interpolation, Kriging interpolation or radial basis function interpolation, the purpose of which is to more reasonably estimate the reference height of the target position, and by assigning different weights to different reference features, the local surface change trend can be more accurately reflected.

[0092] The scheme of the present application first identifies the reference features within the spatial neighborhood of each coded feature, thereby providing necessary context information for subsequent local reference height calculation. It is precisely because the neighborhood information around the coded feature is considered, rather than processing individual features in isolation, that the scheme can utilize spatial correlation to enhance the robustness of the data. On this basis, according to the spatial coordinates of each coded feature, the depth data and the identified reference features, the local reference height of the position of each information coded feature is determined by a weighted interpolation method. This weighted interpolation mechanism can effectively fuse the information of multiple reference features in the neighborhood, and assign different weights according to their relative relationship with the target feature (such as distance or similarity), so that in the presence of noise, occlusion or incomplete data, a more smooth and accurate local reference height estimate can still be obtained. Finally, the local reference height set of all information coded features corresponding to the three-dimensional array structure forms the local reference data. In this way, a complete and accurate local reference surface can be constructed for subsequent three-dimensional standardized image data processing. The local reference data serves as the basis for subsequent calculation of the relative depth of three-dimensional structural features, and its accuracy directly affects the quality of the standardized image data. Through this refined reference construction process, the defects in the original data can be effectively overcome, providing high-quality input for subsequent environment authentication information extraction, thereby ensuring the reliability of the entire verification process.

[0093] In some preferred embodiments, in particular, when identifying reference features within the spatial neighborhood of each encoding feature, a spherical neighborhood with a radius R centered at each encoding feature, e.g. a pit or a bump, in the three-dimensional array structure can be set. Then, all other identified encoding features and non-encoding feature points determined to belong to the flat surface portion of the box within the spherical neighborhood are searched and used as reference features for the encoding feature. In determining the local reference height at the location of each information encoding feature from the spatial coordinates of each encoding feature, the depth data, and the reference features, an inverse distance weighted interpolation (IDW) method can be used. In particular, for each target encoding feature, its local reference height can be calculated as a weighted average of the depth values of all reference features, where the weight of each reference feature is proportional to the inverse of its Euclidean distance to the target encoding feature, i.e. the closer the reference feature, the greater the weight. Finally, in forming the local reference data from the set of local reference heights of all information encoding features corresponding to the three-dimensional array structure, the calculated local reference height of each encoding feature can be stored in a two-dimensional array or a point cloud data structure, where the indices of the two-dimensional array can correspond to the logical positions of the encoding features in the three-dimensional array structure, thereby forming a data set representing the local reference surface of the entire three-dimensional array structure.

[0094] In particular, Figure 2 A method flowchart for calculating the relative depth of features in a three-dimensional structure with respect to local reference data to obtain three-dimensional normalized image data is shown in the embodiments of the present application, which specifically includes:

[0095] S201, identifying three-dimensional point cloud data in three-dimensional image data;

[0096] S202, filtering out abnormal points and registering point clouds for the three-dimensional point cloud data to obtain pre-processed point cloud data;

[0097] S203, calculating the vertical distance of each point in the pre-processed point cloud data to the local surface reference;

[0098] S204, statistically processing the vertical distance to obtain the relative depth of each encoding feature of the three-dimensional structure feature with respect to the local surface reference.

[0099] The local surface reference refers to a reference plane or a set of reference heights constructed based on spatial positions and depth data under a three-dimensional array structure, which is used to measure the relative depth of three-dimensional structural features. It can be achieved by determining the local reference height of each information encoding feature position through a weighted interpolation method, and forming a set of local reference heights of all information encoding features. The purpose is to provide a stable and adaptive depth reference to the local surface morphology. The statistical processing refers to the mathematical summary and analysis of the calculated vertical distance. It can be achieved by calculating the mean, median, mode or Gaussian filtering. The purpose is to eliminate measurement noise and local fluctuations, so as to obtain more stable and representative relative depth values. The encoding feature refers to the geometric unit identified in the three-dimensional array structure for encoding environmental authentication information. Specifically, it can be a pit or a protrusion. The purpose is to serve as a physical carrier for carrying environmental authentication information.

[0100] The scheme of the present application realizes the accurate calculation of the relative depth of three-dimensional structural features relative to the local reference data through a series of refined point cloud processing and distance calculation, thereby obtaining standardized three-dimensional image data. Specifically, first, the three-dimensional point cloud data in the three-dimensional image data is identified. These point cloud data are the direct embodiment of the three-dimensional scanning result, containing the spatial information of the object surface, providing a basis for subsequent processing. Then, the original three-dimensional point cloud data is filtered and registered to eliminate noise data and integrate data from different angles or multiple scans, ensuring the cleanliness, accuracy and consistency of the data, thereby obtaining preprocessed point cloud data. On this basis, the vertical distance of each point in the preprocessed point cloud data to the local surface reference is calculated. This step is critical, as it utilizes the previously constructed local reference data to effectively eliminate the effects of object surface inclination or curvature, thereby more accurately describing the depth of three-dimensional structural features. Finally, the vertical distances are statistically processed to further eliminate noise interference and obtain stable and reliable relative depth values of each encoding feature relative to the local surface reference. It is due to the synergistic effect of these steps that the present scheme can convert the original three-dimensional image data into standardized three-dimensional image data. This standardization process, especially the calculation of the relative depth relative to the local reference, effectively overcomes the problem of inaccurate depth information caused by box deformation, uneven surface or changes in scanning angle in traditional methods. By providing accurate and standardized three-dimensional image data, the present scheme provides a reliable basis for subsequent extraction of geometric feature data corresponding to three-dimensional structural features from three-dimensional standardized image data, thereby making the process of analyzing environmental authentication information based on a predetermined encoding rule more stable and accurate, and ultimately improving the overall reliability and efficiency of carton recycling environmental information verification.

[0101] In some preferred embodiments, the calculation of the relative depth of the three-dimensional structural features with respect to the local reference data to obtain the three-dimensional normalized image data can be implemented as follows: first, identify the three-dimensional point cloud data in the three-dimensional image data. For example, after the three-dimensional scanning device completes the scanning of the non-corrugated environmentally friendly packaging box, the raw data stream output by it can be parsed into a series of points with X, Y, Z coordinates, which together constitute the three-dimensional point cloud data of the surface and three-dimensional structural features of the box. Then, the three-dimensional point cloud data is subjected to outlier removal and point cloud registration to obtain preprocessed point cloud data. The outlier removal can use the statistical outlier removal (SOR) algorithm to identify and remove data points far from most points by calculating the neighborhood average distance of each point. The point cloud registration can use the iterative closest point (ICP) algorithm to align the point cloud data of multiple scans or different perspectives to a unified coordinate system, ensuring the integrity and consistency of the data. Then, the vertical distance of each point in the preprocessed point cloud data to the local surface reference is calculated. For example, for each preprocessed point cloud data point, its spatial coordinates can be used to calculate the nearest vertical distance from the point to the local surface reference (e.g., a plane or curved surface composed of multiple local reference height points) constructed previously. This distance reflects the degree of protrusion or depression of the point with respect to the local reference. Finally, the vertical distances are statistically processed to obtain the relative depth of each coded feature in the three-dimensional structural features with respect to the local surface reference. For example, for an identified coded feature (such as a pit), the vertical distances of all points in the pit region to the local surface reference can be collected, and then the average or median of these distances is calculated as the relative depth of the coded feature. This can effectively smooth local noise and obtain a more representative depth value.

[0102] The step of obtaining the environment authentication information based on the preset encoding rule by analyzing the geometric feature data in the embodiments of the present application comprises: analyzing the numerical range and spatial position relationship in the geometric feature data; mapping the numerical range and spatial position relationship into a sequence of encoding symbols; and analyzing the sequence of encoding symbols based on the preset encoding rule to obtain the environment authentication information.

[0103] The numerical range and spatial position relationship in the geometric feature data refers to the specific numerical interval of the three-dimensional structural features in the dimensions of depth, height, width, etc., as well as the relative position, arrangement pattern or mutual distance of the feature points or regions in the three-dimensional space, which can be realized by statistical analysis, geometric measurement or topological relationship identification technology, and the purpose is to preliminarily identify and quantify the information elements carried by the three-dimensional structural features. The coding symbol sequence refers to the numerical range and spatial position relationship extracted from the geometric feature data, which is converted into a series of discrete, standardized symbol or character combinations according to a specific mapping rule, which can be realized by using a lookup table, a hash function or a self-defined coding algorithm, and the purpose is to convert the complex geometric feature information into a standardized format that is easy to process and transmit, and to improve the robustness of information identification. The preset coding rule refers to a set of rules defined in advance for one-to-one or many-to-one correspondence between the coding symbol sequence and the specific environmental authentication information, which can be realized by using a decision tree, a state machine or a database query table, and the purpose is to ensure that the coding symbol sequence can be accurately restored to the corresponding environmental authentication information, such as environmental standards, certification agencies or recycling types.

[0104] The scheme of the present application realizes accurate acquisition of environment authentication information by fine analysis of geometric feature data extracted from three-dimensional image data. Specifically, after obtaining geometric feature data corresponding to three-dimensional structural features, the data is first analyzed to identify the numerical range and spatial position relationship contained therein. This is because the geometric arrangement pattern of three-dimensional structural features itself encodes environment authentication information, and the core carrier of this information is the numerical value of the geometric features and their spatial position relative to each other. Through this analysis step, information elements related to environment authentication can be preliminarily extracted. On this basis, the numerical range and spatial position relationship obtained by analysis are mapped to a sequence of code symbols. This conversion process converts complex, possibly different geometric feature data into a standardized, discrete symbolic representation. This intermediate representation not only simplifies the subsequent processing process, but also effectively eliminates the geometric feature differences caused by factors such as box size, scanning angle or slight deformation, thereby significantly improving the robustness of information recognition. Further, based on a preset encoding rule, the sequence of code symbols is analyzed to obtain the final environment authentication information. The preset encoding rule is a pre-established knowledge base that defines the correspondence between geometric feature encoding and specific environment authentication information. By applying these rules, the system can accurately restore the standardized code symbol sequence to specific environment authentication information, such as a specific environmental standard, certification agency or recycling type. It is precisely because the system has obtained more regular and reliable geometric feature data through image preprocessing and standardization (such as identifying three-dimensional array structures, constructing local reference data, calculating relative depths, etc.) before analyzing the geometric feature data that the subsequent numerical range and spatial position relationship analysis is more accurate, the mapping of the code symbol sequence is more stable, and the final environment authentication information obtained based on the preset encoding rule is more accurate. The combination of data preprocessing and fine analysis enables the present scheme to reliably restore environment authentication information even in the face of noise, irregularities or slight deformation that may exist in complex recycling environments, effectively solving the deviation problem that may occur when directly analyzing.

[0105] In some preferred embodiments, the present application is implemented as follows: after obtaining the geometric feature data corresponding to the three-dimensional structural features on the non-corrugated environmentally friendly packaging box, for example, these data may contain a series of pit or protrusion depth values, their two-dimensional coordinates on the box surface, and their relative distances. The system first analyzes these geometric feature data, identifies the numerical range in which each pit or protrusion depth value is located, and analyzes their spatial arrangement pattern. Then, the numerical range and spatial position relationship obtained by the analysis are mapped to a sequence of encoding symbols. For example, if a pit with a depth value in a certain range is identified, and its spatial position relationship conforms to the preset "A" pattern, it is mapped to symbol "A"; if a protrusion with a depth value in another range is identified, and its spatial position relationship conforms to "B" pattern, it is mapped to symbol "B". In this way, a series of geometric features are converted into a sequence composed of symbols such as "A", "B", "C", etc., for example, "AABBC". Finally, the environmental authentication information is obtained by analyzing the encoding symbol sequence based on the preset encoding rules. The preset encoding rules can be stored in a database or lookup table, which defines the specific environmental authentication information corresponding to different encoding symbol sequences. For example, the encoding symbol sequence "AABBC" may correspond to "EU CE certification, recyclable paper pulp material", while "ABCCA" may correspond to "US FDA certification, biodegradable material". The system determines the environmental properties and recycling type of the box by querying these preset rules and analyzing "AABBC" to specific environmental authentication information.

[0106] It should be noted that the geometric arrangement pattern of the three-dimensional structural features is formed by the mold structure corresponding to the three-dimensional structural features together with the box material during the box material forming process.

[0107] Among them, the mold structure refers to a tool for giving the box material a specific three-dimensional shape, which can be made of metal, ceramic or high-strength composite material, etc., and has corresponding concave-convex or texture corresponding to the geometric arrangement pattern of the required three-dimensional structural features, and its purpose is to directly form the preset three-dimensional features when the box material is formed. The box material forming process refers to the manufacturing process of converting raw materials into a box with a specific shape, which can be realized by paper pulp molding, injection molding, compression molding or extrusion process, etc., and its purpose is to provide a basic structure for the box and a carrier for the integration of three-dimensional features. Solidification is the process of changing the material from a liquid, semi-solid or powder state to a stable solid shape under certain conditions, which can be realized by drying, cooling, chemical reaction or physical compaction, etc., and its purpose is to ensure that the three-dimensional structural features and the box material form an integrated stable structure.

[0108] The scheme of the present application realizes the economic and reliable integration of three-dimensional structural features on non-corrugated environmentally friendly packaging boxes by closely combining the formation of three-dimensional structural features with the manufacturing process of box materials. Specifically, in the box material forming process, such as in paper pulp molding or injection molding, the pre-designed mold structure directly contacts the box material. The mold structure has concave-convex or texture corresponding to the geometric arrangement pattern of the required three-dimensional structural features. When the box material is cured and formed under the action of the mold, the material will fill or imprint into the corresponding structure of the mold, so that the three-dimensional structural features are formed together with the box material. This means that the three-dimensional structural features are not added extra after the completion of box manufacturing, but are integrated as part of the box itself.

[0109] The step of scanning the three-dimensional structural features on the non-corrugated environmentally friendly packaging box based on the three-dimensional scanning device in the embodiment of the present application includes: projecting structured light onto the surface of the non-corrugated environmentally friendly packaging box through the three-dimensional scanning device; and capturing the deformation pattern of the structured light through the three-dimensional scanning device, the deformation pattern being used for the three-dimensional scanning device to form three-dimensional image data.

[0110] Among them, the structured light refers to a light with a specific spatial coding pattern, which can be realized in the form of stripes, dot matrix or coding pattern, and its purpose is to form an identifiable geometric pattern on the surface of the object to assist in the acquisition of three-dimensional information; the deformation pattern refers to the pattern that changes in shape, position, density, etc. after the structured light is projected onto the surface of the object due to the ups and downs of the object surface, which can be captured by an image sensor, and its purpose is to contain the three-dimensional geometric information of the object surface, and by analyzing the deformation degree, the height and shape of the object surface can be deduced; the three-dimensional image data refers to a data set describing the three-dimensional spatial geometric information of the object, which can be expressed as point cloud data, depth map or mesh model, and its purpose is to provide basic data for subsequent feature extraction and information analysis.

[0111] The scheme of the present application realizes the accurate scanning of the three-dimensional structural features on the surface of the non-corrugated environmentally friendly packaging box by actively projecting structured light and capturing its deformed pattern. Specifically, the three-dimensional scanning device first projects a predetermined structured light pattern, such as parallel stripes or coded dot array, onto the surface of the box. When these structured lights encounter the uneven three-dimensional structural features on the surface of the box, their projection on the surface of the box will be deformed, and this deformation directly reflects the geometric undulation of the surface of the box. Subsequently, the image sensor equipped with the three-dimensional scanning device will capture these deformed patterns. Since the pattern of the structured light is known, by analyzing the differences between the captured deformed pattern and the original pattern, such as the degree of bending of the stripes and the displacement of the dot array, the three-dimensional coordinates and depth information of each point on the surface of the box can be accurately calculated. It is precisely because of this principle based on active illumination and geometric coding that the scheme can effectively overcome the challenges encountered by traditional direct scanning methods in the face of non-corrugated environmentally friendly packaging box surface material unevenness, existence of reflection or unstable lighting conditions and other complex environments. Through structured light, even on low-contrast or reflective surfaces, clear and identifiable features can be formed on the surface of the object, thereby ensuring the integrity and accuracy of the three-dimensional image data. This high-quality three-dimensional image data, as the input of the subsequent steps (such as converting the three-dimensional structural features into three-dimensional image data and extracting the geometric feature data from it), significantly improves the reliability and robustness of the entire paper box recycling environmental information verification method, making it possible to extract and analyze environmental certification information under various adverse conditions, thereby ensuring the recycling efficiency and accuracy of information verification.

[0112] In some preferred embodiments, the application is implemented as follows: a structured light three-dimensional scanning device based on fringe projection can be used to scan the non-corrugated environmentally friendly packaging box. The device includes a projector and one or more cameras. During the scanning process, the projector projects a series of fringe patterns with specific spatial frequency and phase onto the surface of the non-corrugated environmentally friendly packaging box. When these fringe patterns are projected onto the surface of the box, due to the existence of three-dimensional structural features on the surface of the box, the fringes will be bent and deformed. For example, if there is a protrusion on the surface of the box, the fringes projected onto the protrusion will bend outward; if there is a pit, the fringes will bend inward. Subsequently, the camera captures the deformed fringe patterns from different angles. The captured image data is transmitted to the built-in processor of the scanning device or the connected computer. The processor analyzes the phase information of the captured deformed fringe patterns using a preset algorithm, such as Fourier transform profilometry or phase shift method. By calculating the phase value corresponding to each pixel point and combining the calibration parameters of the camera and the projector, the three-dimensional coordinates of each point on the surface of the box can be accurately inferred, and a high-precision three-dimensional image data, such as point cloud data or depth map, can be formed. This way can effectively avoid the influence of environmental light changes and box surface material properties on the scanning results, ensuring the quality of the three-dimensional image data and providing a reliable basis for subsequent environmental authentication information extraction and verification.

[0113] Specifically, Figure 3 A structure diagram of a carton recycling environment information verification system in an embodiment of the application is shown, and the system includes:

[0114] A three-dimensional scanning module is configured to scan three-dimensional structural features on the non-corrugated environmentally friendly packaging box based on a three-dimensional scanning device, and the geometric arrangement mode of the three-dimensional structural features is encoded with environmental authentication information.

[0115] A three-dimensional imaging module is configured to convert the three-dimensional structural features into three-dimensional image data.

[0116] A data processing module is configured to extract geometric feature data corresponding to the three-dimensional structural features from the three-dimensional image data.

[0117] An analysis and verification module is configured to analyze the geometric feature data based on a preset encoding rule to obtain the environmental authentication information.

[0118] A recycling matching module is configured to determine the carton recycling type in the recycling environment information library based on the environmental authentication information.

[0119] The three-dimensional scanning module refers to a hardware and software collection for obtaining the three-dimensional geometric information of the surface of the non-corrugated environmentally friendly packaging box, which can be realized by using technologies such as structured light scanning, laser triangulation or time of flight (ToF), and the purpose is to accurately capture the microstructure features of the box surface; the three-dimensional imaging module refers to a functional unit for converting the original three-dimensional data obtained by the three-dimensional scanning module into a three-dimensional image format that can be processed by a computer, which can be realized by using methods such as point cloud to depth map conversion, mesh reconstruction or voxelization, and the purpose is to provide standardized input for subsequent data processing; the data processing module refers to a functional unit for analyzing three-dimensional image data and identifying and extracting geometric features related to environmental authentication information therefrom, which can be realized by using algorithms such as image filtering, feature point detection, region segmentation or pattern recognition, and the purpose is to accurately locate and quantify the information from complex three-dimensional data; the analysis and verification module refers to a functional unit for decoding the extracted geometric feature data according to the preset coding rules, thereby restoring the original environmental authentication information, which can be realized by using methods such as rule-based analysis, lookup table mapping or machine learning classification, and the purpose is to convert the geometric features into understandable authentication content; the recycling matching module refers to a functional unit for comparing the environmental authentication information obtained by analysis with the pre-established recycling environment information library to determine the specific recycling type of the carton, which can be realized by using methods such as database query, information retrieval or decision tree matching, and the purpose is to realize the automatic classification and recycling guidance of the carton.

[0120] The scheme of the present application forms a complete system by materializing each step in the carton recycling environment information verification method into a functional module, thereby realizing automatic and efficient verification of the non-corrugated environmentally friendly packaging box environment certification information. Specifically, the three-dimensional scanning module first scans the three-dimensional structural features on the non-corrugated environmentally friendly packaging box based on a three-dimensional scanning device. The geometric arrangement mode of these features is encoded with environment certification information, ensuring accurate acquisition of the original data. Subsequently, the three-dimensional imaging module converts these original three-dimensional structural feature data into standardized three-dimensional image data, laying the foundation for subsequent digital processing. Next, the data processing module accurately extracts the geometric feature data corresponding to the three-dimensional structural features from the three-dimensional image data, effectively filtering out noise and focusing on key information. Further, the analysis and verification module analyzes the extracted geometric feature data according to the preset encoding rules, thereby accurately restoring the environment certification information. Finally, the recycling matching module determines the recycling type of the carton in the recycling environment information library based on the analyzed environment certification information. This systematic design integrates the complex process that may require manual intervention or step-by-step processing by general software into automated and continuous operations. Through the coordinated work of special modules, the system can overcome the problems of traditional printed information being easily damaged and poor environmental adaptability, ensuring that the environment certification information of the non-corrugated environmentally friendly packaging box can still be stably and accurately identified and verified in complex environments such as recycling stations. Compared to merely limiting the method steps, the present system provides specific execution carriers and functional division, making the verification process more efficient and reliable, significantly improving the automation level of carton recycling and the robustness of information verification, thereby effectively solving the practical application problems of carton recycling environment information verification.

[0121] In some preferred embodiments, the application is implemented as follows: a three-dimensional scanning module can use an industrial-grade structured light three-dimensional scanner that can capture the small concave-convex structure on the surface of the non-corrugated environmentally friendly packaging box with sub-millimeter accuracy. The scanner can be integrated on an automated assembly line and automatically trigger scanning when the box passes by. A three-dimensional imaging module can be an image processing software module built into the system's main control unit, which receives the point cloud data output by the scanner and converts it into a standardized depth map or three-dimensional mesh model. The data processing module can be a software module based on computer vision and pattern recognition algorithms running on a powerful processor. This module can perform image segmentation and feature point detection on the depth map, identify geometric features such as specific shape and depth pit arrays that are encoded with environmental authentication information. The module can further calculate the relative depth, spatial position and other geometric feature data of these pits. The parsing verification module can be a software module preloaded with a coding rule database and parsing algorithm. This module can map the pit depth values and their spatial positions in the array extracted by the data processing module into a binary sequence according to the preset binary coding rules, and then parse the binary sequence into specific environmental authentication information through table lookup or decoding algorithm. The recycling matching module can be a software module connected to the database of the back-end recycling management system. This module receives the parsed environmental authentication information and automatically queries and matches it in the recycling environmental information library. The entire system can be coordinated and managed by a central controller to ensure smooth data transmission and collaborative work between modules, thereby achieving fast, accurate and automated verification of environmental authentication information on non-corrugated environmentally friendly packaging boxes.

[0122] Through the above technical solutions, the application provides a carton recycling environmental information verification system. The system integrates three-dimensional scanning, three-dimensional imaging, data processing, parsing verification and recycling matching functions through modular design to form a complete automated verification process. This enables the system to overcome physical damage, surface stains, structural deformation or wet environment and other unfavorable conditions that non-corrugated environmentally friendly packaging boxes may encounter in actual recycling scenarios, and to stably and accurately obtain and verify the environmental authentication information on the boxes. The system significantly improves the robustness of information identification and the efficiency of the verification process, effectively solving the practical application problem of quickly and reliably verifying carton recycling environmental information in complex environments, thereby ensuring the effective recycling and classification of environmentally friendly packaging.

[0123] The above only describes embodiments of the application and is not intended to limit the protection scope of the application. For those skilled in the art, the application can have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the application shall be included in the protection scope of the application.

Claims

1. A method for verifying environmental information related to cardboard box recycling, characterized in that, include: Based on the scanning of three-dimensional structural features on non-corrugated environmentally friendly packaging boxes using a three-dimensional scanning device, the geometric arrangement pattern of the three-dimensional structural features is encoded with environmental certification information; The three-dimensional structural features are converted into three-dimensional image data; Extract the geometric feature data corresponding to the three-dimensional structural features from the three-dimensional image data; The geometric feature data is parsed based on preset encoding rules to obtain the environmental authentication information; Based on the environmental certification information, determine the type of cardboard box to be recycled in the environmental information database; The process of parsing the geometric feature data based on preset encoding rules to obtain the environmental authentication information includes: Analyze the numerical range and spatial relationships in the geometric feature data; The numerical range and spatial relationship are mapped into a sequence of coded symbols; The environmental authentication information is obtained by parsing the encoded symbol sequence based on the preset encoding rules. The extraction of geometric feature data corresponding to the three-dimensional structural features from the three-dimensional image data includes: The three-dimensional image data is preprocessed to obtain three-dimensional standardized image data; Extract the geometric feature data corresponding to the three-dimensional structural features from the three-dimensional standardized image data; The step of preprocessing the three-dimensional image data to obtain three-dimensional standardized image data includes: Identify spatial location and depth data of a 3D array structure in 3D image data; Local reference data is constructed based on the spatial position and depth data of the aforementioned three-dimensional array structure; The relative depth of the three-dimensional structural features with respect to the local reference data is calculated to obtain three-dimensional normalized image data.

2. The method for verifying environmental information related to cardboard box recycling according to claim 1, characterized in that, The spatial location and depth data of the three-dimensional array structure in the identified three-dimensional image data include: Identifying the target region where a 3D array structure is located based on a geometric template matching algorithm; Each coded feature, which is a pit or a protrusion, is identified from the target region based on surface curvature analysis or local point cloud clustering methods. Calculate the spatial coordinates and depth data for each encoded feature.

3. The method for verifying environmental information related to cardboard box recycling according to claim 2, characterized in that, The construction of local reference data based on the spatial location and depth data under the three-dimensional array structure includes: Identify reference features within the spatial neighborhood of each encoded feature; Based on the spatial coordinates, depth data, and reference features of each encoded feature, the local reference height of the location of each information encoded feature is determined by a weighted interpolation method. The local reference data is formed by collecting the local reference heights of all information encoding features corresponding to the three-dimensional array structure.

4. The method for verifying environmental information related to cardboard box recycling according to claim 1, characterized in that, The calculation of the relative depth of the three-dimensional structural features with respect to the local reference data to obtain three-dimensional normalized image data includes: Identify 3D point cloud data in 3D image data; The three-dimensional point cloud data is subjected to outlier filtering and point cloud registration to obtain preprocessed point cloud data; Calculate the vertical distance from each point in the preprocessed point cloud data to the local surface reference; The vertical distance is statistically processed to obtain the relative depth of each coded feature in the three-dimensional structural features relative to the local surface reference.

5. The method for verifying environmental information related to cardboard box recycling according to claim 1, characterized in that, The geometric arrangement of the three-dimensional structural features is solidified and formed together with the box material during the molding process through a mold structure corresponding to the three-dimensional structural features.

6. The method for verifying environmental information related to cardboard box recycling according to claim 1, characterized in that, The scanning of three-dimensional structural features on non-corrugated environmentally friendly packaging boxes using a three-dimensional scanning device includes: Structured light is projected onto the surface of the non-corrugated environmentally friendly packaging box using the 3D scanning device; The deformation pattern of the structured light is captured by the 3D scanning device, and the deformation pattern is used by the 3D scanning device to form the 3D image data.

7. A cardboard box recycling environmental information verification system, characterized in that, The system includes: A 3D scanning module is used to scan the 3D structural features on non-corrugated environmentally friendly packaging boxes using a 3D scanning device. The geometric arrangement pattern of the 3D structural features is encoded with environmental certification information. A three-dimensional imaging module is used to convert the three-dimensional structural features into three-dimensional image data; The data processing module is used to extract the geometric feature data corresponding to the three-dimensional structural features from the three-dimensional image data; It is also used to perform image preprocessing on the three-dimensional image data to obtain three-dimensional standardized image data; Extract the geometric feature data corresponding to the three-dimensional structural features from the three-dimensional standardized image data; It is also used to identify spatial location and depth data under a three-dimensional array structure in three-dimensional image data; Local reference data is constructed based on the spatial position and depth data of the aforementioned three-dimensional array structure; Calculate the relative depth of the three-dimensional structural features with respect to the local reference data to obtain three-dimensional normalized image data; The parsing and verification module is used to parse the geometric feature data based on preset encoding rules to obtain the environmental authentication information; the step of parsing the geometric feature data based on preset encoding rules to obtain the environmental authentication information includes: parsing the numerical range and spatial position relationship in the geometric feature data; mapping the numerical range and spatial position relationship into an encoded symbol sequence; and parsing the encoded symbol sequence based on the preset encoding rules to obtain the environmental authentication information. The recycling matching module is used to determine the recycling type of cardboard boxes in the recycling environmental information database based on the environmental certification information.

Images