Cigarette packet paperboard positioning method based on image analysis

Abstract

The invention discloses a cigarette packet paperboard positioning method based on image analysis. The method comprises the following steps: projecting a structured light pattern to a target ink area; pulse energy is applied to a target printing ink area on the surface of the cigarette packet card paper to cause selective photo-thermal excitation, so that the cigarette packet card paper generates transient thermal elastic deformation; structured light stripe images are synchronously collected before and after the pulse energy is applied; based on the collected reference stripe image and the deformation stripe image, calculating to obtain a dynamic response characteristic field DRSF representing transient thermoelastic deformation generated by photo-thermal excitation of the target ink area; performing image segmentation on the DRSF to identify a deformation region, and extracting geometric features from the deformation region; and determining the position or posture of the target ink area according to the geometrical characteristics and the system calibration parameters. According to the invention, the limitation of the traditional visual positioning under specific conditions is overcome by utilizing the selective absorption of the ink on energy and the generated physical response, and the early warning of the microdefect can be realized at the same time.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a cigarette pack paper positioning method based on image analysis. Background Art

[0002] Cigarette pack paperboard is a crucial component of the modern packaging industry. Precise positioning of the paperboard during high-speed automated production processes, such as printing, cutting, coding, and folding, is crucial for ensuring product quality, improving production efficiency, and reducing material waste. Traditional methods for cigarette pack paperboard positioning rely primarily on visible light-based machine vision technologies, such as template matching and feature point extraction algorithms that identify printed marks, edge contours, or specific patterns. However, these methods often face numerous challenges in practical industrial applications: The printed patterns on cigarette pack paperboard are increasingly complex and may have low color contrast; lighting conditions in production environments are prone to fluctuations; and the paperboard surface may contain minor scratches, stains, or reflective coatings. These factors can interfere with the stable extraction of visual features, resulting in reduced positioning accuracy or even failure. Furthermore, the printing quality of cigarette pack paperboard directly impacts the appearance and functionality of the final product. Conventional machine vision defect detection mainly focuses on macroscopic defects visible on the surface. Existing technologies often find it difficult to effectively and quickly detect some microscopic, hidden defects or early defects involving the interaction between ink and paper substrate, such as tiny breakpoints, pinholes, and uneven thickness within the ink layer, or potential printing quality risks caused by microscopic unevenness of the paper substrate.

[0003] Although there are methods in the existing technology that use photothermal effects for material analysis or non-destructive testing, most of these methods target the steady-state or quasi-steady-state thermal properties of bulk materials. Their equipment, principles, detection speed, and accuracy requirements are significantly different from the specific needs of online high-speed positioning and micro-defect detection of thin, fast-moving printed materials such as cigarette pack paper. They also fail to reveal the technical potential of using the transient physical response induced by selective photothermal excitation for both high-precision positioning and early warning of micro-defects. Summary of the Invention

[0004] The present invention provides a cigarette pack cardboard positioning method based on image analysis to solve the technical problems in the prior art of insufficient positioning robustness and accuracy when facing complex industrial environments such as low contrast, complex background, illumination fluctuation or surface contamination.

[0005] In view of the above problems, the present invention provides a method for locating cigarette pack paper based on image analysis, the method comprising the following steps: S100: Projecting a structured light pattern onto a target ink area on the surface of the cigarette pack paper; S200: Acquire a first structured light fringe image of the target ink area under the illumination of the structured light pattern as a reference fringe image; S300: applying pulse energy to a target ink area on the surface of the cigarette pack cardboard to induce selective photothermal excitation, thereby generating transient thermoelastic deformation; S400: At a preset response moment of transient thermoelastic deformation of the target ink region due to the selective photothermal excitation, acquiring a second structured light fringe image of the target ink region under the illumination of the structured light pattern as a deformation fringe image; S500: Calculating a dynamic response characteristic field representing transient thermoelastic deformation of the target ink region due to photothermal excitation based on the reference fringe image and the deformed fringe image; S600: performing image segmentation on the dynamic response characteristic field to identify a deformation region corresponding to the target ink region; S700: Extracting at least one geometric feature from the identified deformation region; S800: Determine the position or posture of the target ink area on the cigarette pack paper based on the at least one geometric feature and in combination with preset system calibration parameters.

[0006] Preferably, the method further comprises the following steps: S900: Based on the dynamic response characteristic field calculated in step S500, analyzing abnormal features in the dynamic response characteristic field that are inconsistent with a preset response pattern of the target ink area in a non-defective state; S910: Based on the abnormal features analyzed in step S900, identify microscopic defects on the cigarette pack paper related to the printing quality of the target ink or the interface state between the target ink and the cigarette pack paper substrate or the cigarette pack paper substrate itself, and generate early warning information.

[0007] Preferably, analyzing the abnormal characteristics of the dynamic response characteristic field in step S900 includes at least one of the following operations: aligning and comparing the currently acquired dynamic response characteristic field with a reference dynamic response characteristic field template of the response pattern of the target ink area in a defect-free state, and identifying an area where the difference between the two exceeds a preset tolerance threshold as the abnormal feature; In the dynamic response characteristic field, the signal amplitude representing the phase change amount or the converted physical deformation amount caused by the transient thermoelastic deformation is detected, and a local area where the signal amplitude is outside the statistical average value of its neighboring pixels or outside a predefined normal value range is defined as the difference feature; Calculating the spatial gradient distribution of the dynamic response characteristic field, and defining as the difference feature an area where the gradient amplitude exceeds a preset gradient threshold, an area where the gradient direction change rate exceeds a preset direction change rate threshold, or an area where a breakpoint or a jump point is detected in the gradient distribution; For the ink area in the dynamic response characteristic field that is expected to exhibit a spatially continuous or uniform response, the actual spatial continuity or uniformity index of its signal is evaluated, and the detected breakage or discontinuity points, or the area where the uniformity index is lower than the preset standard value is defined as the difference feature.

[0008] The technical solution provided by this application has at least the following technical effects or advantages: This method detects the transient thermoelastic deformation (DRSF) produced by the selective absorption of pulse energy by the target ink region. Because the DRSF primarily depends on the physical response characteristics of the ink and its interaction with the substrate, it demonstrates greater positioning robustness and potentially improved accuracy in situations difficult to address with traditional visual methods, such as low-contrast printing, complex background texture interference, fluctuating ambient lighting conditions, partially transparent stain coverage, and the reflective effects of specific ink materials. DRSF-based analysis is particularly promising for more precise edge definition and position determination of small or irregularly shaped ink features.

[0009] The present invention can, while performing positioning, detect microscopic defects with high sensitivity that are difficult to detect with conventional visible light vision detection methods or require complex post-processing algorithms to indirectly infer by deeply analyzing the local or global abnormal characteristics of DRSF.

[0010] Because the DRSF is a direct reflection of the physical response of the ink material after selectively absorbing energy, it is not only related to the printing quality of the ink itself, but may also be subtly affected by upstream processes such as ink formulation, printing parameters, and paper material characteristics. Therefore, through long-term, continuous monitoring and statistical analysis of the DRSF, "by-product" data on the stability of the production process can be indirectly obtained. For example, a systematic drift in the average response intensity of the DRSF may indicate a change in the ink batch or improper printing pressure; an increase in the frequency of specific defect modes in the DRSF may indicate that the printing equipment needs maintenance or adjustment. This information can provide valuable reference for the continuous optimization of the production process and preventive maintenance.

[0011] The present invention adopts optical methods for excitation and measurement, which is a non-contact detection method and will not cause physical damage to the surface of the cigarette package cardboard. It is suitable for products with high surface quality requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a flow chart of a method for positioning cigarette pack cardboard based on image analysis according to the present invention. DETAILED DESCRIPTION

[0013] The present invention discloses a method for positioning cigarette pack paper based on image analysis. The method selectively photothermally excites the ink area on the cigarette pack paper and combines structured light measurement technology to capture its transient thermoelastic deformation, thereby constructing a dynamic response characteristic field, and based on this, achieves precise positioning of the target ink area.

[0014] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention, rather than all, are shown in the drawings.

[0015] In a specific embodiment, the present invention provides a method for locating cigarette pack paper based on image analysis, which includes the following configurations: Pulse energy source unit: used for selective photothermal excitation of the target ink area. For example, it can be a near-infrared pulsed laser or a high-power pulsed light-emitting diode (LED) array. The unit has a tunable wavelength, adjustable pulse width (for example, in the range of 1 nanosecond to 100 nanoseconds), and adjustable pulse energy (for example, in the range of 0.01 joules per square centimeter to 1 joule per square centimeter, which must be sufficient to cause measurable thermoelastic deformation of the target ink but far below the threshold of material ablation). The unit is also equipped with corresponding optical components, such as lenses, reflectors, scanning galvanometers, or fiber couplers, to precisely focus or direct the pulse energy to the target ink area on the surface of the cigarette pack cardboard.

[0016] Structured light projection unit: Used to project a structured light pattern onto the target area. For example, this unit can be a digital light processing (DLP) projector or a laser interferometer-based fringe projection device. This unit can stably project a structured light pattern with a preset spatial frequency and modulation depth, typically a high-contrast sinusoidal fringe pattern. Projection parameters, such as fringe period and projection angle, can be configured based on measurement requirements.

[0017] High-speed image acquisition unit: This unit is used to capture the structured light pattern. For example, this can be a high-speed, high-resolution complementary metal oxide semiconductor (CMOS) camera or a scientific-grade CMOS (sCMOS) camera with a global shutter function. This camera is equipped with an imaging objective lens (such as a telecentric lens) with an appropriate working distance and magnification, capable of clearly capturing the fringe image projected by the structured light projection unit and modulated on the cigarette pack paper surface at a sufficiently high frame rate (e.g., several kilohertz or even higher).

[0018] Synchronization control unit: This coordinates the operational timing of each unit in the system. For example, this could be a high-precision programmable digital delay generator (DDG) or a field-programmable gate array (FPGA)-based control board. This unit is responsible for precisely coordinating the excitation timing of the pulse energy source unit, the pattern refresh of the structured light projection unit (if multi-frame phase shifting technology is used), and the exposure and acquisition timing of the high-speed image acquisition unit, ensuring sub-microsecond or even nanosecond synchronization between the operations of each unit.

[0019] Data processing unit: This unit processes the acquired image data and executes the positioning algorithm. For example, it can be an industrial computer or embedded processing system equipped with a high-performance central processing unit (CPU), a graphics processing unit (GPU), and sufficient memory and storage. This unit is pre-installed or configured with specialized algorithm modules.

[0020] like Figure 1 A method for locating cigarette pack paper based on image analysis is shown in the flowchart, comprising the following steps: S100: Projecting a structured light pattern onto the target ink area; Specifically, the structured light projection unit steadily projects a sinusoidal fringe pattern with a preset spatial frequency and high contrast onto the target ink area and surrounding areas on the cigarette pack's paperboard surface. Parameters such as the fringe period, direction, and projection angle are configured based on measurement accuracy and field of view requirements.

[0021] S200: Acquire a first structured light fringe image of the target ink area as a reference fringe image before applying the pulse energy; Specifically, under the instruction of the synchronization control unit, the high-speed image acquisition unit captures a frame of digital image at a reference moment before the pulse energy source applies the excitation pulse. The image records the original modulation state of the structured light pattern when the target ink area is not excited, that is, the reference fringe image.

[0022] S300: applying pulse energy to the target ink area on the surface of the cigarette pack cardboard to induce selective photothermal excitation; further, the applied pulse energy comes from a pulse laser in the near-infrared band or a high-power pulse light-emitting diode LED array.

[0023] Specifically, the pulse width of the pulse energy source unit is precisely controlled, for example, between 1 nanosecond and 100 nanoseconds, and the pulse energy is precisely adjusted, for example, between 0.01 joules and 1 joule per square centimeter. This ensures that the energy is sufficient to induce thermoelastic deformation in the target ink that can be clearly measured by the subsequent optical system, while remaining well below the ablation threshold of the ink material to avoid damage to the cigarette pack paper. This selective excitation exploits the differences in light absorption characteristics and thermophysical properties between the target ink and the surrounding paper substrate at specific wavelengths.

[0024] S400: After the pulse energy is applied, a second structured light fringe image of the target ink area is obtained as a deformation fringe image at a preset moment when the target ink area undergoes transient thermoelastic deformation; further, the preset moment for obtaining the deformation fringe image is a moment when the deformation reaches a peak or has significant characteristics, which is pre-estimated based on the thermophysical properties of the target ink material (such as thermal conductivity, heat capacity, thermal expansion coefficient) and the applied excitation pulse parameters (such as energy density, pulse width), or a specific moment within the optimal detection time window determined by experimental calibration.

[0025] Specifically, the synchronization control unit precisely coordinates to ensure that after the pulse energy source applies the excitation pulse, the high-speed image acquisition unit captures another digital image, namely the deformation fringe image, at the preset deformation moment. The time interval between the acquisition moments of the reference fringe image and the deformation fringe image (i.e., the time interval obtained by subtracting the "reference moment" from the "deformation moment") is controlled to be within an extremely short range (e.g., a few microseconds to hundreds of microseconds). This time interval needs to be long enough to allow the high-speed image acquisition unit and the subsequent DRSF calculation step to clearly identify and quantify the thermoelastic deformation, but it must be as short as possible to effectively capture this highly transient response and minimize the impact of low-frequency environmental vibrations (such as jitter of the mechanical platform) on the measurement results.

[0026] S500: Based on the reference fringe image and the deformed fringe image, a dynamic response characteristic field DRSF is calculated to characterize the transient thermoelastic deformation of the target ink area due to photothermal excitation; further, the process of calculating the dynamic response characteristic field DRSF can adopt the principle of single-frame or multi-frame Fourier transform profilometry FTP.

[0027] Specifically, taking single-frame Fourier transform profilometry as an example, the calculation steps include: S510: Perform a two-dimensional fast Fourier transform (2D-FFT) on the reference fringe image and the deformed fringe image after preprocessing (such as filtering and noise reduction, contrast enhancement, etc.), converting the images from the spatial domain to the frequency domain, thereby obtaining their respective complex frequency spectra, which are called the reference spectrum and the deformed spectrum.

[0028] S520: In the obtained reference spectrum and deformed spectrum, based on the fringe spatial frequency and direction information preset by the structured light projection unit, identify and locate the primary spectral component corresponding to the fringe fundamental frequency. This primary spectral component typically appears as a pair of conjugate symmetrical energy concentration regions deviating from the DC component (spectral center) in the spectrum graph. A two-dimensional bandpass filter (such as a Gaussian window, Hanning window, or ideal bandpass filter) is constructed with its passband center aligned with the center of the selected primary spectral component and its bandwidth optimized. This bandpass filter is then applied to the reference spectrum and deformed spectrum, respectively, to selectively extract the desired fundamental frequency information, i.e., the primary spectral component.

[0029] S530: The primary spectral components extracted after bandpass filtering in step S520 are frequency-shifted so that their centers coincide with the origin of the spectrogram (the location of the DC component). A two-dimensional inverse fast Fourier transform (2D-IFFT) is then performed on the centered primary spectra (corresponding to the reference state and the deformed state, respectively). This results in two complex images of the same size as the original image, referred to as the reference complex analytical signal and the deformed complex analytical signal.

[0030] S540: Extract a reference wrapped phase map and a deformed wrapped phase map from the reference complex analytical signal and the deformed complex analytical signal by calculating the argument of the complex value at each pixel (i.e., using its real and imaginary parts and applying the inverse tangent function atan2 to obtain the correct phase angle). The values ​​of these two wrapped phase maps are constrained to range from negative to positive pi (e.g., -π to +π radians), or from zero to twice pi (e.g., 0 to 2π radians), exhibiting a periodic sawtooth distribution.

[0031] S550: Calculate the difference between the phase value of each pixel in the deformation wrapped phase image and the phase value of the corresponding pixel in the reference wrapped phase image to obtain the original, wrapped differential phase image. This differential phase image intuitively reflects the phase changes caused by thermoelastic deformation and constitutes a preliminary form of the dynamic response signature field (DRSF) described in the present invention, or its basic data.

[0032] Furthermore, in order to optimize or further quantify the dynamic response characteristic field DRSF, the differential phase map in the wrapped form obtained in the aforementioned step S550 is subsequently processed.

[0033] Specifically, this subsequent processing may include at least one of the following optional operations: First, phase unwrapping is performed on the wrapped differential phase map. Due to the periodicity of the wrapped phase, the differential phase obtained by direct subtraction may contain jumps with a period of integer multiples of twice pi (i.e., phase ambiguity). Phase unwrapping algorithms (such as the branch-cut method based on path integrals, the quality map-guided algorithm based on reliability ranking, or the least squares method) aim to eliminate these jumps and restore a continuous "unwrapped differential phase map" that represents the total phase change. This unwrapped differential phase map can be used as a more accurate dynamic response signature field (DRSF).

[0034] Second, the wrapped differential phase image or the unwrapped differential phase image obtained through phase unwrapping is converted into a physical deformation map that directly represents the surface normal displacement, based on a phase-to-physical deformation conversion relationship (e.g., a system calibration coefficient representing the physical displacement corresponding to a unit phase change) determined in advance through system calibration (e.g., measurement using a step of known height or a standard deformation sample). This physical deformation map makes the deformation information more physically meaningful and constitutes a form of the dynamic response signature field (DRSF) described in the present invention. Its grayscale value or pseudo-color mapping can intuitively display the actual, quantified magnitude and spatial distribution of the transient thermoelastic deformation of the target ink area.

[0035] S600: performing image segmentation on the dynamic response feature field DRSF to identify the deformation area corresponding to the target ink area; further, the process of performing image segmentation on the dynamic response feature field (DRSF) (for example, the differential phase map, the unfolded differential phase map or the physical deformation map obtained in the above step) is intended to separate the target ink area that has undergone significant deformation from the background.

[0036] Specifically, the segmentation process may include determining one or more appropriate thresholds based on prior knowledge of the expected DRSF signal intensity (i.e., the pixel amplitude in the DRSF image) of the target ink under specific excitation conditions, or through statistical analysis of the DRSF image (e.g., histogram analysis). Pixels in the DRSF image with signal amplitudes above (or below, depending on the deformation direction and DRSF signal definition) the preset threshold are identified as belonging to the target deformation region, while the remaining pixels are considered the background region. This generates a binary target region mask image, where pixel values ​​in the target region are 1 (or 255) and pixel values ​​in the background region are 0. To improve the accuracy and robustness of the segmentation results, morphological image processing techniques can be further applied to the binary target region mask image obtained from the initial segmentation. For example, an opening operation (i.e., erosion followed by dilation) can be used to remove small, isolated noise points and fine burrs that may exist in the image; a closing operation (i.e., dilation followed by erosion) can be used to fill small holes within the target region and connect small, broken parts of the target region. These morphological optimization operations help to obtain a more complete and regular outline of the target deformation region.

[0037] S700: extracting at least one geometric feature from the identified deformed area; further, extracting at least one geometric feature for positioning from the identified deformed area (eg, the binary mask image obtained in step S600 or its corresponding DRSF area).

[0038] Specifically, you can choose to extract at least one of the following geometric features: First, use edge detection algorithms (such as the Canny algorithm or the Sobel operator) or contour tracking algorithms (such as the Suzuki algorithm) to extract the precise boundary contour points of the target deformation area. Based on this contour point set, its length (i.e., perimeter) and the area enclosed by the contour can be further calculated.

[0039] Second, the image moments of the target deformed region are calculated. For example, the zero-order moment can be used to calculate the area of ​​the region (which can be verified with the area calculated by the contour mentioned above); the first-order moment is used to calculate the pixel coordinates of the region's geometric center of mass in the image; and the second-order central moment can be used to calculate the direction of the region's principal axis of inertia (usually expressed as an angle relative to the image coordinate axis, representing the overall orientation of the region) and shape description parameters such as the region's eccentricity.

[0040] Third, if the target ink feature itself contains clear and predictable geometric shapes (for example, straight line segments in barcodes, circular or rectangular elements in trademark logos), then in the area with significant deformation in DRSF, methods such as Hough transform (for detecting straight lines, circles, etc.) and least squares fitting can be used to directly detect and fit these geometric primitives, thereby obtaining their precise parameters (such as the position and angle of the straight line, the center coordinates and radius of the circle, the vertex coordinates of the rectangle, etc.).

[0041] Fourth, if the target deformation area in the DRSF presents clearly identifiable corners (such as the vertices of a rectangular pattern), endpoints (such as the endpoints of a barcode line) or specific inflection points, algorithms such as Harris corner detection, Shi-Tomasi corner detection or FAST corner detection can be applied to extract the precise pixel coordinates of these key feature points.

[0042] S800: Determine the position or posture of the target ink area on the cigarette pack paper based on the at least one geometric feature and in combination with preset system calibration parameters.

[0043] Specifically, preliminary work first requires system calibration. This includes camera calibration (obtaining the camera's intrinsic parameter matrix and distortion coefficients) and hand-eye calibration or world coordinate system calibration (establishing a precise transformation relationship between the image pixel coordinate system and the physical world coordinate system of the cigarette pack paper surface, for example, by calculating a homography matrix or a more complete projective transformation model). Then, using the one or more geometric features extracted in step S700 (e.g., the pixel coordinates of the center of mass and the image pose angles derived from principal axis analysis, or a set of pixel coordinates of multiple feature points) as input, the obtained system calibration parameters (i.e., camera parameters and coordinate transformation relationships) are used to transform them from the image coordinate system to the physical coordinate system of the cigarette pack paper through appropriate mathematical operations (such as direct application of coordinate system transformation formulas, or, for multiple feature points, PnP solution, affine / projective transformation fitting, etc.). Thus, the planar position of the target ink feature in the physical coordinate system of the cigarette pack cardboard (e.g., world coordinates, attitude angle (e.g., rotation angle in the world coordinate system)) can be calculated. If a positioning strategy based on template matching is adopted, and a DRSF of a standard defect-free sample is pre-made as a reference template, this step can also be performed by calculating the optimal alignment transformation (such as translation, rotation, and scaling parameters) between the currently acquired DRSF (or its extracted features) and the reference DRSF template. These transformation parameters indicate the deviation of the current target ink feature from the standard posture of the template, which can be used for precise positioning and possible subsequent production calibration.

[0044] Further S900: pre-processing the dynamic response characteristic field (DRSF) and establishing a reference benchmark for defect determination; using the dynamic response characteristic field DRSF calculated in the aforementioned step S500, by analyzing its abnormal characteristics, identifying microscopic defects related to the target ink or the substrate below it, and generating early warning information.

[0045] Specifically, first, the dynamic response characteristic field DRSF generated in the aforementioned step S500 is obtained, which may be, for example, a differential phase map, an unfolded differential phase map, or a physical deformation map.

[0046] Secondly, to ensure the accuracy and consistency of subsequent defect analysis, the currently acquired DRSF can be preprocessed as necessary. For example, if the subsequent analysis relies on comparison with a pre-stored template, it may be necessary to first perform precise spatial image registration of the current DRSF with the template to eliminate artifacts caused by slight positioning deviations or slight overall deformation of the DRSF itself (not caused by defects). For example, if there is a drift in the overall signal intensity of the DRSF due to slight systematic fluctuations in the energy of the excitation light source or the gain of the image acquisition unit, the signal intensity of the current DRSF can be normalized based on the signal intensity of certain known stable reference areas in the DRSF (for example, large areas of unprinted blank paper, whose DRSF response should be close to zero or a stable background value), or based on the statistical mean of the overall DRSF signal.

[0047] Next, establish or load a reference benchmark for defect determination. The reference benchmark can be one of the following forms or a combination thereof: Healthy sample DRSF template: This template is obtained by collecting and averaging (or selecting typical) DRSF images from a large number of confirmed defect-free, conforming cigarette packs of the same type as the currently inspected paper, under standard excitation and measurement conditions consistent with the current inspection. This template represents the DRSF response pattern under healthy conditions.

[0048] DRSF statistical property model: A model describing the normal parameter range of the DRSF is learned by statistically analyzing the DRSF of a large number of healthy samples. For example, for each pixel point or each pre-divided subregion in the DRSF, the statistical mean and standard deviation of its signal amplitude (such as phase difference or physical deformation) can be obtained; or, for the entire DRSF or a specific region thereof, the normal distribution range of its gradient and the expected value range of texture features can be obtained.

[0049] Design expectations based on internal consistency: For target ink areas on cigarette pack paper that are supposed to be uniformly printed (e.g., large solid color blocks) or regularly repeated (e.g., barcode lines), the resulting defect-free DRSF should also exhibit spatial uniformity, smoothness, or specific periodic patterns. This internal consistency based on design expectations can itself serve as a dynamic reference benchmark.

[0050] S910: Analyze abnormal features in the dynamic response feature field (DRSF) that are inconsistent with the reference datum, identify microscopic defects, and generate warning information; Specifically, by comparing the pre-processed current DRSF with the reference established or loaded in step S900, or by analyzing whether the signal characteristics within the current DRSF itself deviate from expectations, abnormal features that may indicate the presence of microscopic defects are extracted and quantified. At least one of the following analysis operations can be used: i) Abnormal Feature Extraction Based on Differences from a Reference Template: If a healthy sample DRSF template was loaded in step S900, a pixel-by-pixel difference map is calculated between the current DRSF (after registration and normalization, if necessary) and the reference DRSF. For example, the absolute difference between the corresponding pixel values ​​in the two images can be directly calculated, or the normalized correlation coefficient can be calculated and low-correlation regions identified. A difference tolerance threshold is then set, and pixels or connected regions in the difference map with differences exceeding this threshold are identified as abnormal features indicating potential defects.

[0051] ii) Feature extraction based on local anomalies in DRSF signal amplitude: Within the current DRSF (e.g., differential phase map or physical deformation map), for each pixel point or predefined local neighborhood window, analyze whether its signal amplitude significantly deviates from the normal level. For example, the difference between the signal amplitude at that point and the statistical mean of the signal amplitudes of the pixels in its surrounding M×M neighborhood can be calculated. If this difference exceeds a preset fluctuation threshold, it is determined to be abnormal. Alternatively, the signal amplitude at that point is compared with the normal value range for that location (or area of ​​that type) obtained based on statistics from a large number of healthy samples (e.g., the mean plus or minus N times the standard deviation). If it exceeds this range, it is determined to be abnormal. Such amplitude anomalies may indicate that the ink layer is too thin or too thick, the printing is not accurate, or the thermal and physical properties of the underlying substrate have changed.

[0052] iii) Feature extraction based on DRSF spatial gradient anomalies: First, the spatial gradient distribution map of the current DRSF is calculated. The value of each pixel in this gradient map represents the severity and direction of the DRSF change near that point. This gradient map is then analyzed for anomalies. For example, the gradient amplitude is detected to see if it increases abnormally in certain local areas (possibly indicating burrs on the ink edge, the tip of a small crack, or a discontinuity in the deformation field), or if the gradient direction undergoes irregular, unexpected, and drastic changes, or if the gradient distribution map itself exhibits breaks or jumps. These can all serve as abnormal features indicating defects.

[0053] iv) Feature extraction based on abnormalities in DRSF spatial continuity or uniformity indicators: Analyze areas in the DRSF that, based on the cigarette package design, should exhibit spatially continuous (e.g., barcode lines) or uniform (e.g., large color blocks) responses. For example, image connectivity analysis algorithms (e.g., region growing or contour tracking) can be used to check whether the deformation bands corresponding to the barcode lines in the DRSF remain topologically intact. Detection of breaks or discontinuities in the deformation bands may indicate ink breakage. For color block regions that are expected to be uniform, the local standard deviation, entropy, or specific texture descriptors (e.g., contrast, correlation, energy, etc., calculated from the gray-level co-occurrence matrix) of the DRSF can be calculated. If these uniformity indicators fall below a preset quality standard or differ from the statistical values ​​of healthy samples by a value greater than a preset comparison threshold, this may indicate a non-uniform response pattern caused by uneven SF ink application, the presence of microscopic impurities or pigment agglomerates, or uneven properties of the underlying paper substrate.

[0054] Finally, based on the abnormal features extracted and quantified by one or more of the above operations, the data processing unit will make a comprehensive judgment on whether there are microscopic defects and possible defect types (such as broken wires, pinholes, uneven ink layers, poor interfaces, substrate abnormalities, etc.) according to pre-set defect judgment rules (for example, comprehensively considering the size, shape, number, position, degree of deviation from the normal state, etc. of the abnormal features, and may assign different weights or judgment logic to different types of abnormal features). If it is determined that there are microscopic defects that require attention, corresponding early warning information is generated. The early warning information can be output in the form of a visual method (such as marking the defect location on the image of the operation interface and displaying relevant parameters), an audio alarm, or a digital signal to guide subsequent quality control operations (such as product rejection, process parameter adjustment) or to generate a detailed quality inspection report for archiving and analysis.

[0055] Those skilled in the art will appreciate that the order of the above steps is not absolutely fixed. The order of certain steps may be adjusted, or certain steps may be combined or further subdivided, without affecting the core technical effect. Furthermore, the methods and systems of the present invention may also incorporate other auxiliary processing steps, such as confidence assessment of the final positioning result, historical data tracking and analysis, etc., depending on the needs of actual application scenarios.

[0056] The foregoing description is merely a specific embodiment of the present invention, and the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be readily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A cigarette pack paper positioning method based on image analysis, characterized in that: The method comprises the following steps: S100: Projecting a structured light pattern onto a target ink area on the surface of the cigarette pack paper; S200: Acquire a first structured light fringe image of the target ink area under the illumination of the structured light pattern as a reference fringe image; S300: applying pulse energy to a target ink area on the surface of the cigarette pack cardboard to induce selective photothermal excitation, thereby generating transient thermoelastic deformation; S400: At a preset response moment of transient thermoelastic deformation of the target ink region due to the selective photothermal excitation, acquiring a second structured light fringe image of the target ink region under the illumination of the structured light pattern as a deformation fringe image; S500: Calculating a dynamic response characteristic field representing transient thermoelastic deformation of the target ink region due to photothermal excitation based on the reference fringe image and the deformed fringe image; S600: performing image segmentation on the dynamic response characteristic field to identify a deformation region corresponding to the target ink region; S700: Extracting at least one geometric feature from the identified deformation region; S800: Determine the position or posture of the target ink area on the cigarette pack paper based on the at least one geometric feature and in combination with preset system calibration parameters.

2. The method for locating cigarette pack paper based on image analysis according to claim 1, characterized in that: The dynamic response characteristic field calculated in step S500 includes: S510: performing a two-dimensional Fourier transform on the reference fringe image and the deformed fringe image respectively to obtain a reference spectrum and a deformation spectrum thereof; S520: Identifying and extracting, from the reference spectrum and the deformation spectrum, primary spectrum components corresponding to the fundamental frequency of the structured light fringes according to the preset spatial frequency and direction of the structured light fringes; S530: Performing a two-dimensional inverse Fourier transform on the extracted primary spectral components to obtain a reference complex analytical signal and a deformed complex analytical signal; S540: Obtain a reference wrapped phase image and a deformed wrapped phase image by calculating the arguments of the reference complex analytical signal and the deformed complex analytical signal at each pixel point, respectively; S550: Calculate the pixel-by-pixel difference between the deformation wrapped phase map and the reference wrapped phase map to obtain a differential phase map in a wrapped form, wherein the differential phase map constitutes the dynamic response characteristic field or a preliminary form thereof.

3. The method for locating cigarette pack paper based on image analysis according to claim 2, characterized in that: The step S500 also includes at least one of the following operations to process the wrapped differential phase map: performing phase unwrapping processing on the wrapped differential phase map to eliminate integer multiple jumps of twice the value of pi, obtaining an unwrapped differential phase map, and using this unwrapped differential phase map as the dynamic response characteristic field; or converting the wrapped differential phase map or the unwrapped differential phase map into a physical deformation variable map representing the surface normal displacement according to a preset phase-physical deformation variable conversion relationship, and using this physical deformation variable map as the dynamic response characteristic field.

4. The method for locating cigarette pack paper based on image analysis according to claim 1, characterized in that: The image segmentation of the dynamic response feature field in step S600 includes: setting one or more thresholds according to the signal amplitude in the dynamic response feature field, identifying the pixel points whose signal amplitude meets the preset conditions as belonging to the target deformation area, and the remaining pixel points as the background, thereby generating a binary target area mask image.

5. The method for locating cigarette pack paper based on image analysis according to claim 1, characterized in that: Extracting at least one geometric feature from the identified deformed region in step S700 includes at least one of the following operations: extracting a boundary contour of the deformed region using an edge detection algorithm or a contour tracking algorithm, and calculating its perimeter or the area enclosed; calculating an image moment of the deformed region to obtain pixel coordinates of its geometric centroid or an inertial principal axis angle representing its direction; if the target ink region contains an expected geometric shape, detecting and fitting geometric primitives of the geometric shape within the deformed region using a Hough transform or a least squares fitting method to obtain parameters thereof; A corner detection algorithm is applied to extract pixel coordinates of corner points, endpoints or specific inflection points in the deformation area.

6. The method for locating cigarette pack paper based on image analysis according to claim 1, characterized in that: The pulse energy applied in step S300 is derived from a near-infrared pulse laser or a high-power pulsed light-emitting diode array, and its pulse width and pulse energy are adjusted to be sufficient to cause a measurable thermoelastic deformation of the target ink but well below its ablation threshold; Furthermore, the time interval between the reference fringe image obtained in step S200 and the deformation fringe image obtained in step S400 is controlled within a range that is sufficient to cause the target ink to undergo quantified thermoelastic deformation but is as short as possible, so as to effectively capture the transient thermoelastic deformation and suppress the influence of low-frequency environmental vibrations. The preset moment for obtaining the deformation fringe image is the deformation peak moment estimated based on the thermophysical properties and excitation parameters of the target ink material or a moment within the optimal detection time window determined experimentally.

7. The method for locating cigarette pack paper based on image analysis according to claim 1, characterized in that: The method further comprises the following steps: S900: Based on the dynamic response characteristic field calculated in step S500, analyzing abnormal features in the dynamic response characteristic field that are inconsistent with a preset response pattern of the target ink area in a non-defective state; S910: Based on the abnormal features analyzed in step S900, identify microscopic defects on the cigarette pack paper related to the printing quality of the target ink or the interface state between the target ink and the cigarette pack paper substrate or the cigarette pack paper substrate itself, and generate early warning information.

8. The method for locating cigarette pack paper based on image analysis according to claim 7, characterized in that: Analyzing the abnormal characteristics of the dynamic response characteristic field in step S900 includes at least one of the following operations: aligning and comparing the currently acquired dynamic response characteristic field with a reference dynamic response characteristic field template of the response pattern of the target ink area in a defect-free state, and identifying an area where the difference between the two exceeds a preset tolerance threshold as the abnormal feature; In the dynamic response characteristic field, the signal amplitude representing the phase change amount or the converted physical deformation amount caused by the transient thermoelastic deformation is detected, and a local area where the signal amplitude is outside the statistical average value of its neighboring pixels or outside a predefined normal value range is defined as the difference feature; Calculating the spatial gradient distribution of the dynamic response characteristic field, and defining as the difference feature an area where the gradient amplitude exceeds a preset gradient threshold, an area where the gradient direction change rate exceeds a preset direction change rate threshold, or an area where a breakpoint or a jump point is detected in the gradient distribution; For the ink area in the dynamic response characteristic field that is expected to exhibit a spatially continuous or uniform response, the actual spatial continuity or uniformity index of its signal is evaluated, and the detected breakage or discontinuity points, or the area where the uniformity index is lower than the preset standard value is defined as the difference feature.

Images