A card image feature recognition method and system

Abstract

The application provides a card image feature recognition method and system, and relates to the technical field of card image feature recognition.The method comprises the following steps: acquiring macro image data and optical response image data of a card to be detected; extracting the boundary of a target geometric feature according to the macro image data, and generating a limited area based on the boundary of the target geometric feature; performing intensity evaluation on the optical response signal in the limited area according to the limited area and the optical response image data, and obtaining an intensity evaluation result; performing spatial distribution characteristic evaluation on the optical response signal in the limited area according to the limited area and the optical response image data, and obtaining a spatial distribution characteristic evaluation result; and identifying the authenticity of the card according to the intensity evaluation result and the spatial distribution characteristic evaluation result.The application can effectively identify the authenticity of the card by combining the macro image data and the optical response image data to perform intensity and spatial distribution characteristic evaluation on the optical response signal in a specific area.

Description

Technical Field

[0001] This application relates to the field of card image feature recognition technology, and in particular to a card image feature recognition method and system. Background Technology

[0002] In related technologies, especially for the authentication and condition grading of high-value collectible cards, there are increasingly severe challenges. Traditional authentication methods, such as analysis based on macroscopic visual characteristics (printing dots, edge wear, text sharpness) or simple material composition comparison, are no longer effective in dealing with increasingly sophisticated counterfeiting techniques.

[0003] While adhering to unified design standards, subtle differences in the selection of auxiliary materials can lead to the formation of invisible physicochemical imprints at specific microscopic interfaces after long-term natural circulation and aging. These imprints, due to their microscopic nature, invisibility, and correlation with specific production sources and aging processes, become crucial for distinguishing genuine products from counterfeits. However, accurately capturing and utilizing these subtle characteristics, which are detached from macroscopic vision, remains a pressing challenge for current authentication technologies. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a card image feature recognition method and system, aiming to achieve effective identification of card authenticity.

[0005] In a first aspect, embodiments of this application provide a card image feature recognition method, including:

[0006] Acquire macroscopic image data and optical response image data of the card to be tested;

[0007] The boundaries of the target geometric features are extracted from the macroscopic image data, and a defined region is generated based on the boundaries of the target geometric features.

[0008] The intensity of the optical response signal within the defined region is evaluated based on the defined region and the optical response image data to obtain the intensity evaluation result;

[0009] The spatial distribution characteristics of the optical response signal within the defined region are evaluated based on the defined region and the optical response image data, and the spatial distribution characteristic evaluation result is obtained.

[0010] The authenticity of the card is identified based on the strength assessment results and the spatial distribution characteristic assessment results.

[0011] According to some embodiments of this application, the step of evaluating the optical response signal within the defined region and its spatial distribution characteristics based on the defined region and the optical response image data to obtain the spatial distribution characteristic evaluation result includes:

[0012] Obtain the intensity sequence of the optical response signal within the defined region;

[0013] Based on the intensity sequence, a reference benchmark is obtained for other areas within the defined region;

[0014] Based on the intensity sequence and the reference benchmark, the target area with signal intensity lower than the reference benchmark within the defined area is analyzed to obtain the relative drop amplitude, continuous length, and signal change smoothness condition of the target area;

[0015] The spatial distribution characteristics evaluation result is obtained based on the relative drop amplitude of the target area, the continuous length, and the signal change smoothness condition.

[0016] According to some embodiments of this application, acquiring optical response image data includes:

[0017] Multiple frames of optical response images are acquired according to a preset frequency;

[0018] The multiple frames of optical response images are synchronously demodulated to obtain optical response image data.

[0019] According to some embodiments of this application, the step of extracting the boundary of the target geometric features from the macroscopic image data and generating a defined region based on the boundary of the target geometric features includes:

[0020] Obtain the surface morphology data of the card to be tested;

[0021] Geometric correction is performed on the macroscopic image data based on the surface morphology data to obtain geometrically corrected macroscopic image data;

[0022] The boundary of the target geometric feature is extracted based on the geometrically corrected macroscopic image data, and a defined region is generated based on the boundary of the target geometric feature.

[0023] According to some embodiments of this application, it also includes:

[0024] The optical response signal within the defined region is evaluated in segments to obtain multiple segment evaluation results;

[0025] Based on the segmentation evaluation results, segments with forgery characteristics within the defined region are identified;

[0026] A distribution pattern is obtained based on the segments with forgery characteristics, wherein the distribution pattern includes the total number of segments, the average length, and the maximum continuous length;

[0027] The authenticity of the card is identified based on the strength assessment results, the spatial distribution characteristic assessment results, and the distribution pattern.

[0028] According to some embodiments of this application, a distribution pattern is obtained based on the segments with forgery characteristics, wherein the distribution pattern includes the total number of segments, the average length, and the maximum continuous length, including:

[0029] The total number of segments with forgery characteristics is obtained based on the segments with forgery characteristics;

[0030] The average length of the segments with forgery characteristics is calculated based on the segments with forgery characteristics.

[0031] The maximum continuous length of the segment with forgery characteristics is determined based on the segment with forgery characteristics.

[0032] According to some embodiments of this application, the step of evaluating the intensity of the optical response signal within the defined region based on the defined region and the optical response image data to obtain an intensity evaluation result includes:

[0033] Obtain the local baseline intensity of the optical response signal within the defined region;

[0034] Background compensation is performed on the optical response signal within the defined region based on the local baseline intensity to obtain the background-compensated optical response signal;

[0035] The background-compensated optical response signal is subjected to noise suppression to obtain a noise-suppressed optical response signal.

[0036] The intensity of the noise-suppressed optical response signal is evaluated based on the defined region and the optical response image data to obtain the intensity evaluation result.

[0037] According to some embodiments of this application, the surface morphology data of the card to be tested is obtained through the following steps:

[0038] Multiple structured light patterns with different spatial codes are projected onto the surface of the card to be tested to obtain image data of the card under each structured light pattern projection.

[0039] The surface morphology data of the card to be detected is obtained by performing fusion processing on the image data.

[0040] According to some embodiments of this application, obtaining the macroscopic image data of the card to be detected includes:

[0041] Under the condition that the card to be tested is illuminated by a uniformly diffused light source, macroscopic image data of the card to be tested is acquired.

[0042] Secondly, embodiments of this application provide a card image feature recognition system, including:

[0043] The acquisition module is used to acquire macroscopic image data and optical response image data of the card to be detected;

[0044] The generation module is used to extract the boundaries of the target geometric features based on the macroscopic image data, and generate a defined region based on the boundaries of the target geometric features;

[0045] The first evaluation module is used to evaluate the intensity of the optical response signal within the defined region based on the defined region and the optical response image data, and obtain the intensity evaluation result;

[0046] The second evaluation module is used to evaluate the optical response signal and spatial distribution characteristics within the defined region based on the defined region and the optical response image data, and to obtain the spatial distribution characteristic evaluation result.

[0047] The identification module is used to identify the authenticity of the card based on the strength assessment results and the spatial distribution characteristic assessment results.

[0048] According to the technical solution of this application embodiment, at least the following beneficial effects are achieved: First, macroscopic image data and optical response image data of the card to be detected are acquired; the boundary of the target geometric features is extracted based on the macroscopic image data, and a defined region is generated based on the boundary of the target geometric features; the intensity of the optical response signal within the defined region is evaluated based on the defined region and the optical response image data to obtain an intensity evaluation result; the spatial distribution characteristics of the optical response signal within the defined region are evaluated based on the defined region and the optical response image data to obtain a spatial distribution characteristic evaluation result; the authenticity of the card is identified based on the intensity evaluation result and the spatial distribution characteristic evaluation result. This application embodiment, by combining macroscopic image data and optical response image data, and evaluating the intensity and spatial distribution characteristics of the optical response signal in a specific region, can effectively identify the authenticity of cards.

[0049] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0050] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0051] Figure 1 This is a flowchart illustrating a card image feature recognition method provided in one embodiment of this application;

[0052] Figure 2 This is a schematic diagram of the process for obtaining spatial distribution characteristic evaluation results according to one embodiment of this application;

[0053] Figure 3 This is a schematic diagram of a process for acquiring optical response image data according to one embodiment of this application;

[0054] Figure 4 This is a schematic diagram of the process for generating a defined region according to one embodiment of this application;

[0055] Figure 5 This is a schematic diagram illustrating the process of identifying the authenticity of a card according to one embodiment of this application;

[0056] Figure 6 A schematic diagram illustrating the process of obtaining a distribution pattern according to one embodiment of this application;

[0057] Figure 7 A schematic diagram illustrating the process of obtaining strength assessment results according to one embodiment of this application;

[0058] Figure 8 This is a schematic diagram of a process for obtaining surface morphology data of a card to be tested, provided in one embodiment of this application.

[0059] Figure 9 This is a schematic diagram illustrating the process of acquiring macroscopic image data of a card to be detected, provided in one embodiment of this application.

[0060] Figure 10 This is a schematic diagram of a card image feature recognition system provided in one embodiment of this application. Detailed Implementation

[0061] To make the objectives, technical methods, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] It should be noted that the meaning of "multiple" (or "more than") in the description of the embodiments of this application refers to two or more, and "greater than," "less than," "exceeding," etc. are understood to exclude the number itself, while "above," "below," "within," etc. are understood to include the number itself. If "first," "second," etc. are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0063] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: the existence of a alone, the existence of b alone, the existence of c alone, the simultaneous existence of a and b, the simultaneous existence of a and c, the simultaneous existence of b and c, or the simultaneous existence of a, b, and c, where a, b, and c can be single or multiple.

[0064] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0065] Based on the above, this application proposes a card image feature recognition method and system, aiming to achieve effective identification of card authenticity.

[0066] The card image feature recognition method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms; the software can be an application that implements the card image feature recognition method, but is not limited to the above forms.

[0067] This application can be applied to numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices. It should be noted that in various specific embodiments of this invention, when processing is required based on data related to the characteristics of an object (e.g., user attributes or sets of attribute information), permission or consent from the corresponding object is obtained first, and the collection, use, and processing of this data comply with relevant laws and standards. Furthermore, when the embodiments of the present invention need to obtain the attribute information of an object, they will obtain the separate permission or separate consent of the corresponding object through pop-up windows or redirection to a confirmation page. After obtaining the separate permission or separate consent of the corresponding object, they will then obtain the relevant data of the object necessary for the embodiments of the present invention to operate normally.

[0068] See Figure 1 , Figure 1 This is a flowchart illustrating a card image feature recognition method according to an embodiment of this application. The card image feature recognition method provided in this embodiment includes, but is not limited to, steps S110 to S150, which will be described in detail below.

[0069] Step S110: Acquire macroscopic image data and optical response image data of the card to be detected;

[0070] Step S120: Extract the boundary of the target's geometric features based on the macroscopic image data, and generate a defined region based on the boundary of the target's geometric features;

[0071] Step S130: Evaluate the intensity of the optical response signal within the defined area based on the defined area and the optical response image data to obtain the intensity evaluation result;

[0072] Step S140: Evaluate the optical response signal and spatial distribution characteristics within the defined region based on the defined region and optical response image data, and obtain the spatial distribution characteristic evaluation result;

[0073] Step S150: Identify the authenticity of the cards based on the strength assessment results and spatial distribution characteristic assessment results.

[0074] It should be noted that macroscopic image data refers to image information of the card under test under normal visible light conditions. This can be achieved by using a high-resolution camera to capture images under uniform diffused light sources, such as RGB images acquired by a CCD or CMOS sensor. This is mainly to obtain the overall visual appearance and geometric structure information of the card. Optical response image data refers to the response signal image generated by the microstructure or material properties of the card under test when excited in a specific non-visible light band. This can be achieved by irradiating the card with a light source of a specific band (near-infrared light source) and acquiring it with an image sensor (infrared camera) sensitive to the corresponding band. The boundary of the target geometric feature is the outline of the area on the card that has a specific shape or position and requires special attention. It can be extracted from macroscopic image data using image processing algorithms, such as identifying the boundary between the metal foil and the paper base using Canny edge detection or the Sobel operator. The defined region is a specific image area determined based on the boundary of the target geometric feature for optical response signal analysis. Intensity assessment is a quantitative analysis of the overall brightness or energy level of the optical response signal within the defined region, which can be achieved by calculating the average, sum, or statistical histogram of pixel gray values ​​within the region. Spatial distribution characteristic assessment is an analysis of the spatial variation pattern, structure, or texture of the optical response signal within the defined region. It can be achieved by analyzing the gradient, local extrema, continuity, or smoothness of the signal. For example, by obtaining the intensity sequence, identifying the reference benchmark, and analyzing the relative drop amplitude, continuous length, and smoothness of the signal change in the target area where the signal is lower than the reference benchmark, it is mainly to reveal subtle differences in microstructure or material distribution that may not be discovered through simple intensity assessment, thus providing a more comprehensive judgment on the authenticity of the card.

[0075] In one embodiment, when acquiring macroscopic image data of the card to be detected, an industrial camera equipped with a high-resolution CMOS sensor is used to photograph the front of the card under uniformly diffused LED light source illumination, obtaining a 24-bit true-color image. A near-infrared laser of a specific wavelength, such as a pulsed laser with a wavelength of 850nm, can be used as the excitation source, and an InGaAs camera sensitive to this wavelength is used for image acquisition. Then, multiple frames of images are continuously acquired, and synchronous demodulation processing is performed using lock-in amplification technology to filter out ambient light interference and obtain clean optical response image data. When extracting the boundary of the target's geometric features based on the macroscopic image data, the macroscopic image data is first preprocessed. A deep learning-based image segmentation model can be used to perform pixel-level segmentation of the metal foil region in the image, thereby accurately extracting the boundary line between the metal foil and the card paper base. Based on the boundary line, a band-shaped region can be generated by extending 5-10 pixels inward or outward to define the bounding area. Local baseline correction can be performed on the optical response image data within this bounding area first, followed by noise suppression of the compensated signal using methods such as non-local mean filtering or wavelet transform. Finally, the intensity assessment result is obtained by calculating the average or cumulative gray value of all effective pixels within the bounding area.

[0076] In one embodiment, when evaluating the spatial distribution characteristics of optical response signals within a defined region, a series of intensity sequences of optical response signals are extracted along the centerline of the defined region or a specific scanning path. A sliding window method or an adaptive threshold method is used to identify a reference benchmark representing a normal background signal from this intensity sequence. Continuous regions with signal intensities lower than the reference benchmark are marked as target regions. Next, the relative drop amplitude, continuous length, and signal change smoothness condition of each target region relative to the reference benchmark are calculated. Combining these parameters, the spatial distribution characteristic evaluation result can be obtained. The presence of multiple target regions with long continuous lengths, large drop amplitudes, and smooth signal changes indicates forgery features.

[0077] It is worth noting that the embodiments of this application can accurately locate regions by combining macroscopic image data and capture microscopic physicochemical imprints invisible to the naked eye by using optical response image data of specific non-visible light bands. Furthermore, by conducting dual evaluation of the intensity and spatial distribution characteristics of the optical response signal within a defined area, not only is the overall optical response of the material quantified, but also the subtle differences in microstructure or material distribution are revealed in depth, thereby significantly improving the accuracy and reliability of card authenticity identification.

[0078] See Figure 2 , Figure 2This is a schematic flowchart of obtaining spatial distribution characteristic evaluation results according to an embodiment of this application; regarding the above step S140, the optical response signal within the defined area and the spatial distribution characteristics are evaluated based on the defined area and the optical response image data to obtain the spatial distribution characteristic evaluation results, including but not limited to steps S210 to S240, each step will be described in turn below.

[0079] Step S210: Obtain the intensity sequence of the optical response signal within the defined region;

[0080] Step S220: Identify reference benchmarks for other areas within the defined region based on the intensity sequence;

[0081] Step S230: Based on the intensity sequence and the reference benchmark, analyze the target area within the defined area where the signal intensity is lower than the reference benchmark to obtain the relative drop amplitude, continuous length and signal change smoothness conditions of the target area.

[0082] Step S240: Based on the relative drop amplitude, continuous length, and signal change smoothness conditions of the target area, obtain the spatial distribution characteristic evaluation results.

[0083] It should be noted that the intensity sequence refers to converting two-dimensional image data of optical response signals within a defined region into one-dimensional signal intensity values ​​arranged according to a predetermined scanning path. This can be achieved by extracting and arranging the grayscale or brightness values ​​of image pixels sequentially. The purpose is to transform spatial information into a data format suitable for sequence analysis, facilitating subsequent feature extraction. The reference benchmark is a reference value used to measure the normal intensity level of the optical response signal within the defined region. It can be achieved by calculating the average signal intensity, median intensity, or dynamically determining the intensity value through a local adaptive thresholding algorithm for most non-target areas within the defined region. The target region is the set of continuous or discontinuous pixels within the defined region whose optical response signal intensity is lower than the reference benchmark. This can be achieved by comparing the intensity value of each point in the intensity sequence with the reference benchmark and identifying continuous or discontinuous segments below the reference benchmark. The purpose is to focus on specific areas that may have abnormal optical response characteristics, improving the targeting of the analysis. The relative drop amplitude is the degree to which the signal intensity within the target region decreases relative to the reference benchmark. It is achieved by calculating the difference between the minimum signal intensity within the target region and the reference benchmark, and then performing a ratio operation with the reference benchmark. The continuity length refers to the number of consecutive pixels or spatial dimensions of a target region in the intensity sequence. It can be achieved by counting the number of consecutive pixels in the target region that are lower than the reference baseline. The signal change smoothness condition refers to the smoothness of signal intensity changes within the target region. It can be determined by evaluating the local variance of signal intensity within the target region, the gradient rate of change, or by the smoothness of the fitted curve.

[0084] In one embodiment, pixels within a defined region extracted from optical response image data are arranged sequentially from left to right and top to bottom, with their grayscale or brightness values ​​forming a one-dimensional numerical sequence. If the defined region is a 100x100 pixel rectangle, a sequence of 10,000 intensity values ​​can be obtained. A fixed-size sliding window (e.g., 5% of the total sequence length) is set and slid across the intensity sequence. The average or median signal intensity within each window is calculated, and these averages or medians are used as local reference benchmarks. Alternatively, statistical methods can be employed, such as calculating the global average or median of the intensity sequence and combining it with the standard deviation to set a dynamic upper or lower limit as a reference benchmark. Subsequently, each intensity value in the intensity sequence is iterated and compared with its corresponding reference benchmark. When an intensity value is lower than its corresponding reference benchmark, and this lower value persists for a certain length (e.g., 5 consecutive pixels), these consecutive pixels are marked as a target region. A percentage value is obtained by calculating the difference between the lowest signal intensity within the target region and the reference benchmark, and then dividing by the reference benchmark. Simultaneously, the continuous length of the target region can be statistically determined, i.e., the number of pixels contained within that region. The local variance of the signal intensity within the target region is calculated, or a linear fit is performed on the intensity sequence within the target region, evaluating the slope or residual of the fitted line to determine its smoothness. By weighting and combining the multiple parameters obtained above or inputting them into a pre-trained classification model, a comprehensive evaluation score or classification result is output. This result directly reflects whether the spatial distribution characteristics of the optical response signal within the defined region conform to the characteristics of a genuine product.

[0085] See Figure 3 , Figure 3 This is a schematic flowchart illustrating the process of acquiring optical response image data according to an embodiment of this application. The acquisition of optical response image data in step S110 includes, but is not limited to, steps S310 to S320, which will be described in detail below.

[0086] Step S310: Acquire multiple frames of optical response images according to a preset frequency;

[0087] Step S320: Perform synchronous demodulation processing on multiple frames of optical response images to obtain optical response image data.

[0088] It should be noted that the preset frequency is the frequency used to excite the specific optical response of the card or the periodic signal used for image acquisition when acquiring optical response images. It can be the modulation frequency of the light source or a specific synchronization frequency of the image sensor's acquisition frames. The purpose of selecting this frequency is to ensure that the periodic signal related to the card's specific optical response can be captured, while also being distinguishable from the frequency characteristics of ambient noise. Synchronous demodulation processing is a signal processing technique that extracts the useful signal of a specific frequency from the mixed signal by coherently processing the acquired multi-frame image data with a reference signal (usually synchronized with the preset frequency). This can include operations such as Fourier transform, phase detection, or lock-in amplification of image pixel values. Its purpose is to effectively filter out ambient light variations, sensor inherent noise, and other interference from non-target frequencies, significantly improving the signal-to-noise ratio of the optical response signal, thereby obtaining clean and accurate optical response image data.

[0089] It is worth noting that, in acquiring optical response image data, the embodiments of this application can effectively suppress interference such as ambient light changes and sensor noise, significantly improving the signal-to-noise ratio of the optical response signal. This results in high-quality, high-accuracy optical response image data, providing a more reliable input for subsequent card authenticity verification, thereby improving the overall accuracy of the verification.

[0090] See Figure 4 , Figure 4 This is a schematic diagram of the process for generating a defined region according to an embodiment of this application. Regarding step S120, which involves extracting the boundary of the target geometric features from the macroscopic image data and generating a defined region based on the boundary of the target geometric features, including but not limited to steps S410 to S430, each step will be described in turn below.

[0091] Step S410: Obtain the surface morphology data of the card to be tested;

[0092] Step S420: Perform geometric correction on the macroscopic image data based on the surface morphology data to obtain the geometrically corrected macroscopic image data;

[0093] Step S430: Extract the boundary of the target geometric features based on the geometrically corrected macroscopic image data, and generate a defined region based on the boundary of the target geometric features.

[0094] It should be noted that surface morphology data refers to the information on the three-dimensional geometric shape of the card surface to be inspected. This information can be obtained through technologies such as 3D scanning, structured light projection, or laser ranging, providing a quantitative basis for the unevenness of the card surface. Geometric correction is an image transformation process performed on macroscopic image data based on surface morphology data to eliminate or reduce image distortion caused by unevenness of the card surface. This is achieved by establishing a mapping relationship between image pixels and actual three-dimensional spatial points, and then using interpolation algorithms to restore the distorted image to an orthographic projection image, so that the macroscopic image data more accurately reflects the true geometric shape of the card.

[0095] In one embodiment, multiple structured light patterns with different spatial codes are projected onto the surface of the card to be detected, and image data of the card under each pattern projection is simultaneously acquired. Using phase unwrapping or triangulation principles, three-dimensional point cloud data of the card surface is reconstructed, thus obtaining accurate surface topography data. This three-dimensional topography data is then used to construct a geometric transformation model. A perspective transformation matrix or polynomial surface fitting model is calculated, mapping each pixel in the macroscopic image to a corrected planar coordinate system based on its corresponding three-dimensional spatial position. Bilinear interpolation or bicubic interpolation methods are employed to ensure that the corrected macroscopic image data has good visual quality and geometric accuracy. Next, for specific patterns or text on the card, grayscale processing is first performed, and then the Canny edge detection algorithm is used to identify its clear outline. If the target feature is a regular geometric shape, such as a rectangle or circle, methods such as Hough transform can be combined to further optimize the boundary extraction accuracy. Finally, appropriate dilation operations can be performed on the boundary lines to generate a closed region containing the boundary line and its surrounding area. This defined region is a polygonal area, and the pixels within it will be marked as the area to be analyzed, thus providing an accurate and distortion-free analysis range for subsequent evaluation of the optical response signal.

[0096] See Figure 5 , Figure 5 This is a schematic diagram of a process for identifying the authenticity of a card according to an embodiment of this application. It includes, but is not limited to, steps S510 to S540, which will be described in detail below.

[0097] Step S510: Perform segmented evaluation on the optical response signal within the defined area to obtain multiple segmented evaluation results;

[0098] Step S520: Based on the segmentation evaluation results, identify segments with forgery characteristics within the defined area;

[0099] Step S530: Obtain the distribution pattern based on the segments with forgery characteristics, wherein the distribution pattern includes the total number of segments, the average length, and the maximum continuous length.

[0100] Step S540: Identify the authenticity of the cards based on the strength assessment results, spatial distribution characteristic assessment results, and distribution patterns.

[0101] It should be noted that segmented evaluation divides the optical response signal within a defined area into multiple smaller, continuous sub-regions or segments, and independently analyzes and quantifies the signal characteristics of each sub-region. The segmented evaluation result is the signal characteristic data of each sub-region obtained through segmented evaluation, which may include the average intensity, signal fluctuation range, or specific frequency response of each segment. Counterfeit feature segments are continuous or discontinuous sub-regions within the defined area whose optical response signals deviate from the expected signal of the genuine card. They can be identified using methods such as threshold comparison, pattern matching, or machine learning classification, and can pinpoint the specific location of counterfeit information. The distribution pattern describes the spatial arrangement and quantity characteristics of the identified counterfeit feature segments within the defined area. The total number of segments is the total number of independent segments with counterfeit features identified within the defined area. The average length is the average of the sum of the lengths of all segments with counterfeit features divided by the total number of segments, representing the size of the counterfeit feature. The maximum continuous length refers to the length of the longest continuous counterfeit region among all segments with counterfeit features.

[0102] In one embodiment, the optical response signal can be divided into a series of continuous segments along the length of the defined region, with a fixed step size (e.g., every 50 pixels) or an adaptive step size based on the signal change rate. For each segment, the average intensity value and standard deviation of the optical response signal within it are calculated; these calculations constitute the evaluation results for multiple segments. Subsequently, a threshold is set: if the average intensity value of a segment is lower than a preset percentage (e.g., lower than 90%) of the average intensity of the corresponding area of ​​the genuine card, or if its standard deviation is higher than the fluctuation range of the genuine card, then the segment is marked as having counterfeit characteristics. The system then counts the total number of all segments marked as counterfeit. Simultaneously, the length of these segments is calculated, and their average length is obtained. Furthermore, the system traverses these counterfeit characteristic segments to find the longest continuous counterfeit region, thereby determining the maximum continuous length. Finally, the system comprehensively analyzes the previously obtained intensity assessment results (e.g., the overall average intensity of the optical response signal within the defined area) and spatial distribution characteristic assessment results (e.g., the relative drop amplitude, continuous length, and signal change smoothness conditions of the signal drop area within the defined area) with the newly obtained distribution pattern (including the total number of segments, average length, and maximum continuous length). Using a support vector machine or neural network, these three types of assessment results are used as input features, and a pre-trained model outputs the card's authenticity judgment. This comprehensive judgment mechanism can more comprehensively consider the various characteristics of the card, thereby improving the accuracy of recognition.

[0103] See Figure 6 , Figure 6 This is a schematic flowchart illustrating the process of obtaining a distribution pattern according to an embodiment of this application. Regarding step S530, which obtains the distribution pattern based on segments with forgery characteristics, the distribution pattern includes the total number of segments, the average length, and the maximum continuous length, including but not limited to steps S610 to S630. Each step will be described in turn below.

[0104] Step S610: Obtain the total number of segments with forgery features based on the segments with forgery features;

[0105] Step S620: Calculate the average length of the segments with forgery features based on the segments with forgery features;

[0106] Step S630: Determine the maximum continuous length of the segment with forgery characteristics based on the segment with forgery characteristics.

[0107] In one embodiment, after identifying segments with forgery features within a defined region based on segmentation evaluation results, these segments can be further processed to obtain a distribution pattern. The identified segments with forgery features are stored as a list, where each element contains the start and end positions of the segment. To obtain the total number of segments, the elements in the list can be simply counted. When the list contains three segments, the total number is 3. To calculate the average length of the segments, the list can be traversed first, calculating the length of each segment (end position minus start position plus 1), then the lengths of all segments are summed, and finally the total sum is divided by the total number of segments. When the lengths of the three segments are 10, 15, and 5 pixels respectively, the total length is 30, and the average length is 10 pixels. To determine the maximum consecutive length of a segment, the list can be traversed, tracking the length of the longest segment currently encountered. During traversal, each time a segment is encountered, its length is compared with the currently recorded maximum consecutive length; if the current segment length is greater, the maximum consecutive length is updated. It can extract statistically significant distribution patterns from discrete, forged segmented information.

[0108] It is worth noting that the embodiments of this application extract the total number, average length and maximum continuous length of segments from segments with forgery features as distribution patterns, which can comprehensively and multidimensionally describe the distribution of forgery features. This allows the identification system to no longer rely solely on the existence of a single forgery point, but to comprehensively consider the density of forgery areas, average size and the most concentrated forgery range, thereby more effectively reflecting the forger's tampering strategies and methods.

[0109] See Figure 7 , Figure 7 This is a schematic flowchart illustrating the process of obtaining an intensity assessment result according to an embodiment of this application. The above step S130, which involves assessing the intensity of the optical response signal within a defined region based on the defined region and optical response image data to obtain an intensity assessment result, includes, but is not limited to, steps S710 to S740. Each step will be described in turn below.

[0110] Step S710: Obtain the local baseline intensity of the optical response signal within the defined area;

[0111] Step S720: Perform background compensation on the optical response signal within the defined area based on the local baseline intensity to obtain the background-compensated optical response signal;

[0112] Step S730: Perform noise suppression on the background-compensated optical response signal to obtain the noise-suppressed optical response signal;

[0113] Step S740: Evaluate the intensity of the noise-suppressed optical response signal based on the defined region and optical response image data to obtain the intensity evaluation result.

[0114] It should be noted that the local baseline intensity is the intensity level of the background illumination or non-target response of the optical response signal within a defined region. This can be achieved by statistically analyzing the pixel values ​​of non-target feature areas within the defined region. Background compensation involves subtracting or adjusting the background component in the optical response signal to eliminate the influence of ambient illumination or non-target responses. This can be achieved by subtracting the local baseline intensity pixel-by-pixel from the original optical response signal or through adaptive thresholding. Noise suppression uses algorithms to reduce the interference of random or systematic noise in the optical response signal, thereby improving the signal-to-noise ratio and sharpness. This can be achieved using spatial domain filtering methods such as mean filtering, Gaussian filtering, and median filtering, or transform domain processing methods such as wavelet transform and Fourier transform.

[0115] See Figure 8 , Figure 8 This is a schematic flowchart illustrating the process of obtaining surface morphology data of a card to be tested according to an embodiment of this application. The above-described step S410, obtaining the surface morphology data of the card to be tested, includes, but is not limited to, steps S810 to S840, which will be described in detail below.

[0116] Step S810: Project multiple structured light patterns with different spatial codes onto the surface of the card to be detected, and obtain image data of the card under each structured light pattern projection.

[0117] Step S820: Perform fusion processing based on image data to obtain the surface morphology data of the card to be detected.

[0118] In one embodiment, firstly, a digital projector sequentially projects a series of Gray code patterns and phase-shift fringe patterns onto the surface of the card to be inspected. Gray code patterns can be used to quickly determine a rough correspondence between pixels, while phase-shift fringe patterns provide high-precision phase information. During each pattern projection, a high-resolution industrial camera simultaneously captures image data of the card under the current structured light pattern projection. A set of binary-coded Gray code patterns can be projected; for each pattern projected, the camera captures an image, forming a series of coded images. Subsequently, a set of three-step or multi-step phase-shift fringe patterns is projected; again, for each pattern projected, the camera captures an image, forming a series of phase-shift images. Next, fusion processing is performed based on these captured image data. First, the Gray code image data is decoded to establish a preliminary correspondence between projector pixels and camera pixels, thereby eliminating ambiguity. Then, using the phase-shift fringe image data, the absolute phase value corresponding to each camera pixel is calculated using a phase unwinding algorithm. Combining the calibration parameters of the projector and camera, and the principle of triangulation, the phase value is converted into the three-dimensional coordinates of each point on the surface of the card to be inspected. Ultimately, these 3D coordinate points can be organized into point cloud data or depth maps, thus obtaining the surface topography data of the card to be inspected. This method ensures that high-density, high-precision 3D topography information can be obtained even on complex surfaces.

[0119] See Figure 9 , Figure 9 This is a schematic flowchart illustrating the process of acquiring macroscopic image data of a card to be detected according to an embodiment of this application. The above-described step S110, acquiring macroscopic image data of the card to be detected, includes, but is not limited to, steps S910 to S920, which will be described in detail below.

[0120] Step S910: Under the condition of illuminating the card to be tested with a uniform diffuse light source, acquire macroscopic image data of the card to be tested.

[0121] It should be noted that a uniform diffuse light source is a device that can provide uniform, non-directional illumination to the surface of the card to be tested. Specifically, it can be an integrating sphere light source, a ring diffuser, or an illumination system composed of multiple LED arrays and a diffuser. Its purpose is to eliminate or significantly reduce shadows, highlights, and reflections on the card surface caused by uneven illumination, and to ensure the consistency of image data in terms of brightness.

[0122] In one embodiment, the card to be inspected is placed within a closed imaging cavity. A uniformly diffused light source consisting of a high-brightness LED array is positioned at the top or around the cavity. A frosted glass or milky-white acrylic sheet is placed in front of the LED array as a diffuser to further ensure uniform diffuse illumination on the card surface. The image sensor can be a high-resolution industrial camera, positioned perpendicular to the card surface, to capture a macroscopic image of the card under uniformly diffused illumination. During image acquisition, the brightness of the light source is adjusted to ensure that the image sensor obtains image data with a good signal-to-noise ratio without saturation or overexposure. This results in a macroscopic image of the card under inspection with uniform illumination and clear details, providing reliable input for subsequent image processing and feature recognition.

[0123] See Figure 10 , Figure 10 This is a schematic diagram of a card image feature recognition system provided in one embodiment of this application. The card image feature recognition system 1000 includes:

[0124] The acquisition module 1010 is used to acquire macroscopic image data and optical response image data of the card to be detected;

[0125] The generation module 1020 is used to extract the boundary of the target geometric features based on the macroscopic image data, and generate a defined region based on the boundary of the target geometric features;

[0126] The first evaluation module 1030 is used to evaluate the intensity of the optical response signal within the defined area based on the defined area and the optical response image data, and obtain the intensity evaluation result.

[0127] The second evaluation module 1040 is used to evaluate the optical response signal and spatial distribution characteristics within the defined area based on the defined area and optical response image data, and to obtain the spatial distribution characteristic evaluation result.

[0128] The identification module 1050 is used to identify the authenticity of cards based on the strength assessment results and spatial distribution characteristic assessment results.

[0129] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0130] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0131] The foregoing has provided a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined in this application.

Claims

1. A method for recognizing card image features, characterized in that, include: Acquire macroscopic image data and optical response image data of the card to be tested; The boundaries of the target geometric features are extracted from the macroscopic image data, and a defined region is generated based on the boundaries of the target geometric features. The intensity of the optical response signal within the defined region is evaluated based on the defined region and the optical response image data to obtain the intensity evaluation result; The spatial distribution characteristics of the optical response signal within the defined region are evaluated based on the defined region and the optical response image data to obtain the spatial distribution characteristic evaluation result; Based on the strength assessment results and the spatial distribution characteristic assessment results, the authenticity of the card is identified; The step of evaluating the spatial distribution characteristics of the optical response signal within the defined region based on the defined region and the optical response image data, to obtain the spatial distribution characteristic evaluation result, includes: Obtain the intensity sequence of the optical response signal within the defined region; Based on the intensity sequence, a reference benchmark for non-target regions within the defined area is obtained; Based on the intensity sequence and the reference benchmark, the target area with signal intensity lower than the reference benchmark within the defined area is analyzed to obtain the relative drop amplitude, continuous length, and signal change smoothness condition of the target area; The spatial distribution characteristics evaluation result is obtained based on the relative drop amplitude of the target area, the continuous length, and the signal change smoothness condition.

2. The method according to claim 1, characterized in that, Acquire optical response image data, including: Multiple frames of optical response images are acquired according to a preset frequency; The multiple frames of optical response images are synchronously demodulated to obtain optical response image data.

3. The method according to claim 1, characterized in that, The step of extracting the boundary of the target geometric features based on the macroscopic image data and generating a defined region based on the boundary of the target geometric features includes: Obtain the surface morphology data of the card to be tested; Geometric correction is performed on the macroscopic image data based on the surface morphology data to obtain geometrically corrected macroscopic image data; The boundary of the target geometric feature is extracted based on the geometrically corrected macroscopic image data, and a defined region is generated based on the boundary of the target geometric feature.

4. The method according to claim 1, characterized in that, Also includes: The optical response signal within the defined region is evaluated in segments to obtain multiple segment evaluation results; Based on the segmentation evaluation results, segments with forgery characteristics within the defined region are identified; A distribution pattern is obtained based on the segments with forgery characteristics, wherein the distribution pattern includes the total number of segments, the average length, and the maximum continuous length; The authenticity of the card is identified based on the strength assessment results, the spatial distribution characteristic assessment results, and the distribution pattern.

5. The method according to claim 1, characterized in that, The step of evaluating the intensity of the optical response signal within the defined region based on the defined region and the optical response image data to obtain an intensity evaluation result includes: Obtain the local baseline intensity of the optical response signal within the defined region; Background compensation is performed on the optical response signal within the defined region based on the local baseline intensity to obtain the background-compensated optical response signal; The background-compensated optical response signal is subjected to noise suppression to obtain a noise-suppressed optical response signal. Based on the defined region and the optical response image data, the noise-suppressed optical response signal is processed... Strength assessment is conducted to obtain the strength assessment results.

6. The method according to claim 3, characterized in that, The surface morphology data of the card to be tested is obtained through the following steps: Multiple structured light patterns with different spatial codes are projected onto the surface of the card to be tested to obtain image data of the card under each structured light pattern projection. The surface morphology data of the card to be detected is obtained by performing fusion processing on the image data.

7. The method according to claim 1, characterized in that, The acquisition of macroscopic image data of the card to be detected includes: Under the condition that the card to be tested is illuminated by a uniformly diffused light source, macroscopic image data of the card to be tested is acquired.

8. A card image feature recognition system, characterized in that, include: The acquisition module is used to acquire macroscopic image data and optical response image data of the card to be detected; The generation module is used to extract the boundaries of the target geometric features based on the macroscopic image data, and generate a defined region based on the boundaries of the target geometric features; The first evaluation module is used to evaluate the intensity of the optical response signal within the defined region based on the defined region and the optical response image data, and obtain the intensity evaluation result; The second evaluation module is used to evaluate the spatial distribution characteristics of the optical response signal within the defined region based on the defined region and the optical response image data, and to obtain the spatial distribution characteristic evaluation result. The identification module is used to identify the authenticity of the card based on the strength assessment result and the spatial distribution characteristic assessment result; The second evaluation module is also used for: Obtain the intensity sequence of the optical response signal within the defined region; Based on the intensity sequence, a reference benchmark for non-target regions within the defined area is obtained; Based on the intensity sequence and the reference benchmark, the target area with signal intensity lower than the reference benchmark within the defined area is analyzed to obtain the relative drop amplitude, continuous length, and signal change smoothness condition of the target area; The spatial distribution characteristics evaluation result is obtained based on the relative drop amplitude of the target area, the continuous length, and the signal change smoothness condition.

Images