RU2799240C2 - Determination of characteristics of crepe paper - Google Patents

Abstract

FIELD: imaging.

SUBSTANCE: invention relates to a method for obtaining characteristics of the structure of a creped plate. The method includes providing an imaging system comprising one or more television or photographic sensors or devices, generating one or more signals defining one or more images of a creped plate structure zone, and evaluating the generated image of the creped plate structure by two-dimensional spectral analysis using one or more indicators to determine crepe ridge characteristics, crepe groove characteristics, embossed three-dimensional structure characteristics, plate swelling, plate internal tears, and/or loose fibre ends.

EFFECT: increasing the accuracy of detecting the characteristics of the structure of the creped plate.

13 cl, 20 dwg

Description

Cross reference to related applications

This application seeks conventional priority from US 16/100,719, filed August 10, 2018, the contents of which are hereby incorporated by reference into this application in its entirety.

Technical field

The present invention relates to crepe paper, such as napkins or towels, which is made by pressing a wet fibrous material against the surface of a heated cylindrical dryer, referred to as a Yankee dryer or a Yankee cylinder (these terms will be used interchangeably), so that the wet fibrous material adheres to this surface and dried, after which the dried fibrous material is removed from the surface of the Yankee dryer with a stripping knife or creping scraper.

State of the art

Methods for producing creped materials are well known in the art. The fibrous material is pressed against the surface of a dryer, such as a Yankee dryer, and removed from that surface with a flexible creping scraper. The terms "creping scraper" and "clearing knife" are used interchangeably in this application. The creping scraper may be made of metal, ceramic, or other known materials. The degree of adhesion of the material to the surface of the dryer is a factor that determines the effectiveness of creping and, accordingly, the bulk, stretch, tear strength and softness of the creped material. Creping improves the softness of the paper by forming a creped structure, breaking a significant amount of bonds between the fibers and moving the ends of the fibers away from the surface of the web. The creped structure is characterized by ridges and grooves, which are usually oriented along the largest dimension, perpendicular to the machine direction (movement in the machine), that is, the direction of advance of the web. The size, shape and spatial frequency of these scallops affect the bulk of the web. The creped structure allows for a paper that is mechanically shortened in the machine direction, resulting in increased bulk, stretch and softness. Individual manufacturers define specifications for bulk, stretch, tear strength, and softness of creped material based on their products and the requirements of their customers.

To produce high quality crepe paper, the ability to adhere to and remove the fibrous material from the heated cylindrical drum of the dryer is important, and this greatly affects the structure of the creped web. The peelability is determined by the properties of the web, the composition of the material used to adhere the wet web to the Yankee cylinder, the setting of the creping knife, the temperature of the Yankee cylinder, and other factors known to those skilled in the art.

After the fibrous material on the surface of the Yankee cylinder has dried, the dried fibrous material or crepe paper is scraped off the surface of the Yankee cylinder using a stripping knife. The adhesion must be sufficient to form a satisfactory crepe structure that will provide the desired feel, bulk, stretch and tear strength to the final paper product (crepe paper). However, the grip should not be too strong so that it does not interfere with the removal of paper material from the Yankee cylinder with a stripping knife.

Creped structured web is a modification of creped web and is well known to those skilled in the art. To form a creped structured fabric, TAD technology (from the English. Through Air Drying - drying through air flow) was originally used. Other technologies are currently in use, including Valmet's NTT technology and Voith's ATMOS technology. Creped structured webs have a specific three-dimensional structure created on the web immediately after initial formation, prior to being fed to the Yankee dryer. At this stage of the manufacturing process, the web contains a significant amount of water, both in the fibers that form the web and in the spaces between the fibers, with the result that the web is particularly susceptible to mechanical deformation. Typically, the process uses vacuum and/or mechanical force to press the web against the fabric, the surface of which has a three-dimensional structure with a certain pattern, usually proprietary to the manufacturer and designed for a specific manufactured product. The web is pressed into the fabric to produce a mirror image of the three-dimensional structure present on the fabric. After the structure has been imprinted on the web, the web is partially dehydrated using thermal and/or mechanical methods. The web is then transferred to a Yankee cylinder where it is completely dried and then creped. The ridge and groove structure of the crepe paper resulting from the creping operation is superimposed on the previously extruded three-dimensional structure. In the case of a creped structured web, the embossed three-dimensional structure and the structure of the ridges and grooves determine the characteristics of the web, e.g. bulk, stretch, tear strength and softness. The proposed method is particularly well suited for monitoring both the structure of the ridges and grooves of the crepe, and the extruded three-dimensional structure. The extruded three-dimensional structure has the same dimensions as the ridges and grooves of the creping, or larger, and has a regular pattern that is much easier to assess.

The structure of a napkin or towel web, whether creped or embossed, is one of the main factors that determine the characteristics of the final product in the manufacture of creped web. If the texture of the creped web can be monitored and its characteristics can be determined in real time using the machine's sensor, then the tissue paper manufacturer can adjust or control process variables such as creping blade vibration, creping blade wear, Yankee adhesive composition. cylinder, amount of Yankee adhesive coating, Yankee temperature, direction and intensity of hot air flow, fiber bleaching and chemical additive before the Yankee cylinder (release agent, softener, enhancer) to ensure the required product. Although various real-time creping monitoring systems have been proposed, none of them consider the web as a source of multidimensional data. Also known systems do not monitor the edge of the web, although such monitoring can provide important information about the creped structure, including bulk of the web, internal tears in the web, and free ends of the fibers.

Current technologies include a number of imaging and data analysis techniques that use a variety of illumination sources, imaging technologies, and methods for analyzing them. Various technologies are used to evaluate crepe paper, which measure the angle of the ridges and the length of the ridges on the surface of the creped web in the transverse direction. Other techniques are used to characterize the scallops, such as crepe spacing, crepe distribution, orientation angles, and spatial frequency of linear crepes in the machine direction. However, in none of the modern methods, the ridges and grooves of the creping and any extruded structures were considered as three-dimensional structures having complex geometric characteristics, varying both in the machine direction (MD, from the English Machine-Direction) and in the transverse direction ( CD, from the English Cross-Direction), as well as in the direction perpendicular to the plane of the canvas, as well as in them the image of the canvas was not considered as a source of multidimensional information.

Current methods for evaluating a creped web do not take into account that the web contains three-dimensional structures, such as ridges and creping grooves, as well as three-dimensional structures embossed on the web during the formation of a given pattern on the web, and also that these combined structures can be used as a source of multidimensional data. In the proposed new method, the image is considered as a source of multidimensional information. In the plane of the canvas, the scallops and grooves form irregular two-dimensional features in the image, the characteristics (indicators) of which can be obtained using various geometric equations, and these indicators can vary both in space and in time. These indicators are also used to characterize the three-dimensional structures obtained on the web during the extrusion (embossing) process, as well as their changes in space and time. Changes in brightness in the image provide indirect information about the height of the creping structures, the depth of the grooves, and the out-of-plane changes created by the structuring of the web, these variations being characterized based on changes in intensity, spatial location, and time. Since the image is considered as a source of multidimensional information, two-dimensional spectral analysis methods are used to identify features of creping patterns based on the size and frequency of occurrence of these features. Time variations of these dimensions provide the ability to monitor and control the operation of the machine.

The proposed method can also be applied to the edge of the web, where the image of the edge of the web represents two-dimensional features in the MD direction and in thickness (in the direction perpendicular to the plane of the web). Scallops, grooves, crepe break zones, and fiber ends form irregular two-dimensional features, which are characterized by different geometric dimensions, as well as changes in these dimensions in space and time. In the case of scallop images, changes in brightness in the image indicate the difference between areas of higher and lower density, and the changes are characterized based on changes in intensity, spatial arrangement, and time. The edge image also provides information about the plumpness. To identify features of creping patterns based on the size and frequency of occurrence of these features, two-dimensional spectral analysis methods are applied and used. With respect to surface images, changes over time in these dimensions provide the ability to monitor and control the operation of the machine.

Disclosure of invention

The proposed method provides for the determination in the production process of the characteristics of paper or paper products, such napkins or other crepe paper products. More specifically, a sensor or image source is used to provide a signal so that an image of a topographical or surface zone or region can be obtained.

Television or photographic sensors provide a signal that determines one or more images, and one or more devices and indicators can be used to extract various information about the formation of the crepe web from the images.

The proposed method also includes obtaining an image of the zone or region of the edge of the creped web in the same way as it is done when evaluating the zone or region of the surface of the creped web. One or more signals from one or more sensors are converted into one or more images. Images are evaluated using one or more indicators.

Combining the 2D spectral analysis of the surface and edge zone provides multidimensional characterization of the structure of the creped web, which is an additional new and unique application of the method.

Brief description of the drawings

The present invention will now be described with reference to the accompanying drawings in which like reference numbers indicate like elements.

In FIG. 1 - the original image of the surface area with the structure of the creped web;

in fig. 2 - image of a real sharp transition between two levels of illumination intensity;

in fig. 3 is a digital representation of the image of FIG. 2 after applying one of the algorithms to the image;

in fig. 4 is a digital representation of the image of FIG. 3 after sharpening;

in fig. 5 is an enlarged view of FIG. 1 with increased sharpness;

in fig. 6 is a representation of FIG. 5 with closed contours around the areas of crepe scallops;

in fig. 7 is a histogram of the distances between the closed loops shown in FIG. 6;

in fig. 8 is a histogram of the illumination intensity between the closed contours shown in FIG. 6;

in fig. 9 is a view of the edge of the creped web structure;

in fig. 10 is a representation of FIG. 9 after sharpening;

in fig. 11 is a representation of FIG. 10 with increased sharpness, with closed contours around the areas of the scallops of the creped web;

in fig. 12 is a histogram of the areas of closed contours of FIG. 11;

in fig. 13 is a histogram of the roundness of the closed contours of FIG. 11;

in fig. 14 is a view of the edge of the creped web structure with a reference line;

in fig. 15 is a plot of illumination intensity along the reference line of FIG. 14;

in fig. 16 is a histogram of illumination intensity along the reference line of FIG. 14;

in fig. 17 is a fast Fourier transform (FFT) view of the web image of FIG. 5;

in fig. 18 is a fast Fourier transform (FFT) view of the web image of FIG. 5 after removing information outside the contours;

in fig. 19 is an FFT inversion of the image of FIG. 5 after filtering;

in fig. 20 is half of the FFT of the image of FIG. 17.

Detailed description of the invention

The following detailed description is given for illustrative purposes only and should not be construed as limiting the scope of the invention or the application and applications of the invention. In addition, the application does not intend to be limited by any of the theories presented in the Background section above or in the detailed description below.

A method is provided for characterizing creped web structures using an imaging system that includes one or more sensors capable of providing a signal representing an area (part) of a surface or relief, or an area (part) of an edge of a creped web structure. The signal is converted into one or more images, and these images are evaluated using one or more indicators.

In embodiments of the method, one or more sensors may be television or photographic sensors, wherein the signal that defines said area (part) is displayed in the form of an image.

In some embodiments of the method, various indicators may be used to evaluate images of a zone (part) of the structure of the creped web. These indicators are used for both the ridge structure and the crepe groove structure, as well as the extruded three-dimensional structure of the creped structured web. These indicators can provide important information, including crepe ridge characteristics, crepe groove characteristics, as well as embossed three-dimensional structure, web bulk, web internal tears, and loose fiber ends.

Under the displayed area (part) of the structure of the creped web, is understood to be the resulting one or more images of the top surface, bottom surface and/or edge zone of the structure of the creped web. Images of the structure of the creped web are typically obtained during the manufacturing process and after the web is creped as it exits the Yankee cylinder. Although imaging of the surface of the web may be performed after the manufacturing process has been completed, for real-time optimization of this process, it is most preferred that the images be acquired after the creped web has left the Yankee cylinder. The imaging system may include one or more sensors capable of providing a signal that can be converted to an image, or a photographic or television device that can provide a clear image of the area (part) of the formed crepe web being evaluated. In this application, the terms "creped web structure" and "formed creped web" are used interchangeably.

The surface texture of the creped web is defined by the creped ridges, crepe grooves, internal web tears, outwardly exposed free fiber ends, and/or an extruded three-dimensional structure in the formed creped web, and these elements determine the properties of the final creped web, including softness, tear strength, stretch, and bulk. . Since it is necessary to acquire images of the moving web, the sensors or the imaging device or system must provide a clear image of the area of the formed creped web.

In some embodiments of the proposed method, imaging with a television or photographic device that provides the necessary clarity requires the use of a lens with an appropriate depth of field to obtain a focused image, despite small movements of the web towards or away from the imaging device. The method of selecting an appropriate lens system is well known to those skilled in the art. The structure of the ridges and grooves of the creping and/or the embossed three-dimensional structure of the creped structured web provide the basis for the Cartesian coordinate system associated with the MD and CD directions. Therefore, it is not necessary for the imaging system to be oriented such that the vertical and horizontal imaging axes are aligned with the MD and CD directions.

In some embodiments of the proposed method, a geometric transformation is used between the MD/CD coordinate system of the surface structure of the creped web and the coordinate system associated with the vertical and horizontal axes of the resulting image. Coordinate system rotation is a well-known method of geometric transformation, and there are various software packages available to perform these operations automatically.

In some embodiments of the proposed method, a television or photographic device is used to obtain an image. Given that crepe paper can move at speeds in excess of 1500 meters per second (m/s), a shutter speed of 0.5-5 microseconds (µs) is required to achieve the required clarity. Such shutter speeds require particularly powerful light sources. In some applications, halogen light sources may be used to provide the required light intensity. However, halogen and similar light sources generate a lot of heat, which in the vicinity of paper products (napkins, towels) is a potential fire hazard. Alternative light sources are high intensity LEDs. This type of source is more efficient in providing light and does not generate much heat. One manufacturer of high-speed LED systems is Veritas, Tallahassee, PA. Florida.

The image of the texture of the creped web provides contrast between creping ridges, which are lighter in color/shade, and creping grooves, which are darker in color/shade, regardless of the orientation of the light source relative to the surface of the web. A similar effect is created by the extruded three-dimensional structure of the creped structured web. In any particular application, there may be orientations of the light source relative to the creped web in which this difference is enhanced. Methods and indicators used to determine such orientations and the required illumination intensity are known to those skilled in the art. There are digital camera systems on the market that can provide images at more than 10,000 fps with shutter speeds of 0.5-5 µs. Photron USA, Inc., San Diego, pc. California is one of the providers of such imaging systems. These types of systems require the use of high capacity storage media due to the number of images acquired. The higher the acquisition frequency and the larger the storage capacity, the higher the cost of the entire system and the more difficult it is to process the recorded data.

In some embodiments of the proposed method, the signals or images are received at a frequency that allows real-time monitoring of the creping process. For example, the frequency of obtaining images may be on the order of one image every 10 seconds, in which case it is still possible to control the operation of the machine in real time. Any control action taken based on the image information will take more than 10 seconds to make a change to the production line. This is due to the lag time that is inherent in the tissue or towel making machine. The minimum required frequency of imaging depends on the application, since no two machines that produce napkins or towels work in the same way.

In some embodiments of the proposed method, an image of the surface or relief of the structure of the creped web is obtained. The surface may be the top or bottom surface of the creped web structure, for which 2D spectral analysis can be used to determine the size and frequency of the features that make up the surface structure of the post-creped web and/or the embossed 3D structure of the creped structured web. In the manufacture of any paper product, a lot of effort is spent on making the structure of the web as uniform as possible. However, all paper products are made up of individual fibers whose physical and chemical characteristics vary from fiber to fiber. As a result, the characteristics of the fabric will vary in space, which in turn leads to spatial variability in the structures of the ridges and grooves of the creped fabric. This variability manifests itself in different scales: from individual fibers to sizes exceeding the sizes of combs and crepe grooves. Such variability may be random, or there may be patterns in it.

Surface images of the creped web show variation in the crepe and groove structures and/or the three-dimensional structure extruded on the web, in the MD and CD directions. These structures have features that vary in a certain range of sizes and frequency of their occurrence. Web edge images show variability in web density due to web disruptions and loose fiber ends, both in the machine and in the direction perpendicular to the web plane. As in the case of images of the surface of the canvas, the mentioned structures have features that change in a certain range of sizes and frequency of their repetition. Both types of images can be analyzed using 2D spectral analysis. For this type of transformation, an image from the spatial domain, that is, a standard digital image, is used and converted to the frequency domain. All the information contained in the original image is retained, however it is present in units of frequency and magnitude or scale. Methods of two-dimensional spectral analysis provide the possibility of simultaneously obtaining the dimensions and frequency of repetition of signs of the structure of the creped web in two dimensions. One method of spectral analysis is the Fast Fourier Transform (FFT). Other spectral analysis methods include elementary wave transforms, Hartley transforms, and Wigner-Will transforms. These spectral decomposition technologies and the implementation of these technologies are standard in the field of signal processing and image analysis. The choice of the most appropriate method is made on the basis of the information content of the images, the computing power available, and the ease of implementation. The FFT is included in many image analysis packages available. One example is ImageJ, a free Java image processing program developed by the National Institutes of Health. An example of a paid image processing package that includes FFT is the Aphelion™ Image Software Suite available from Amrinex Applied Imaging, Inc., r. Monroe New Jersey.

In some embodiments of the proposed method, spectral analysis methods are used to quantify the magnitude and frequency of non-random changes in the structure of the creped web or the structure of the creped structured web. Non-random changes can be used to obtain web characteristics and to detect changes in web structure over space and time. Significant changes in the creped structure in the CD direction indicate a non-uniformly shaped and/or non-uniformly dried web. If the scale of variability is comparable to a significant part of the width of the machine, then in this case the conversion efficiency is reduced. The edge from one part of the width will have different characteristics compared to the edge from the other part of the width, which requires appropriate adjustment of the conversion hardware. If the dimensions of the changes are comparable to a fraction of the width of the final product, then the resulting product will not feel soft and/or the product will not perform its intended function. Significant changes in the structure in the direction of MD indicate poor control of the operation of the machine, which occurs due to a change over time in the operation of one or more components of the machine. Web quality monitoring requires characterization of crepe and crepe grooves in the MD and CD directions, as well as repeat rates in the MD and CD directions, both in space and time. Analyzing the web only in the MD or CD direction gives an incomplete picture of the web. Thus, the displayed area (part) of the formed creped web is evaluated using a two-dimensional spectral analysis.

In other embodiments of the proposed method, the image of the region of the edge of the structure of the creped web is evaluated. Edge zone images show changes in the structure of the creped web through its thickness, such as changes in bulk, changes in density, and changes in the occurrence of loose fiber ends. In the production process, significant efforts are spent on maintaining the stable characteristics of the web outside of its plane. However, the characteristics of the fibers that make up the web and the change in each manufacturing process result in spatial variability in the characteristics of the web, which in turn results in spatial variability in the ridge and crepe patterns, resulting in changes in the direction perpendicular to the plane of the web. characteristics such as bulk, density and the occurrence of free ends of the fibers. Such changes may be random, or there may be patterns in them. Spectral analysis makes it possible to obtain quantitative characteristics of the scale and frequency of non-random changes. Non-random variations can be used to characterize the web and to determine spatial and temporal changes in web structure.

In other embodiments of the proposed method, the 2D spectral analysis method can be used as an image filter. This is done by first transforming the image into the frequency domain using a tool such as the Fast Fourier Transform (FFT) for example. Then features are removed from the frequency domain in a certain frequency range or size range. At the final stage, an inverse transformation is carried out to return the image to the original spatial region. The result is an image in the time domain that is similar to the original image, but lacks features removed in the frequency domain. The two images can then be compared using any of the methods described herein and the differences between them can be quantified.

The inventive method can be used to quantify small features such as loose fiber ends, fine web disturbances, sometimes referred to as "microfolds", present in crepe and/or groove structures, and small changes in the embossed three-dimensional structure of a creped structured web. Since known methods are applied to a single dimension, the MD direction, the novelty of the proposed solution lies in the recognition of the two-dimensional nature of the structure of the creped web and the use of two-dimensional or three-dimensional analysis to fully describe the structure of the creped web. The idea of two-dimensional analysis in the proposed method is equally applicable to the evaluation of images of zones (parts) of the edges of the creped web structure. The proposed method provides the possibility of combining a two-dimensional spectral analysis of the surface and the edge zone, resulting in multidimensional characteristics of the structure of the creped fabric, which is an additional novelty and uniqueness of the application of the method.

If the creped web is considered as a source of multidimensional data, then a more complete description of the structures of this web can be obtained. The data may be used in one or more of the indicators discussed herein to characterize the structures of the formed creped web. In the resulting image of the area (part) of the surface of the creped web, there are areas that look lighter and areas that look darker. The lighter areas are the creping ridges, which appear lighter because more incident light is reflected from them towards the imaging device, so that the light intensity is greater in this case. The darker areas are the creping grooves, which appear darker because less incident light is reflected from them towards the imaging device, so that the light intensity is less in this case. In the case of a creped structured web, a similar effect is achieved, elements that are closest to the light source appear lighter, while elements further from the light source, i.e., recessed into the surface, appear darker. The extruded three-dimensional structure of the formed creped web can be considered as a geometrically uniform version of the structure of the ridges and grooves of the crepe, and can be evaluated and characterized using the same methods.

In a digital image, relative brightness and/or darkness is represented by pixel values for each of the pixels that make up the image. Therefore, the pixel value provides an indirect indication of the elevation of the elements that form the structure of the creped web. Light and dark zones form geometrically defined figures. A quantitative description of these figures based on indicators related to the area, perimeter, length, width, deviation from a symmetrical or regular shape, the spatial separation of these figures and the frequency of repetition of figures and signs of figures in space makes it possible to characterize the structure of the creped fabric shown in the image.

Images taken at different time points at the same location in the CD direction of the web can be used to determine the temporal change in crepe and groove zones and/or embossed three-dimensional structure zones of a creped structured web in terms of relative changes, frequency of change, and rate of change. . In addition to the shapes of the ridges and crepes and/or the embossed three-dimensional structure of the creped structured web, the light intensity in each zone, the light intensity between zones, and the variations in light intensity between similar zones can be used to characterize the vertical distribution of the ridges and crepes and/or the embossed three-dimensional structure of the creped structured canvas. Images taken at different times in the same location in the cross direction of the web can be used to determine the change over time in the vertical distribution of the ridges and grooves and/or zones of the extruded three-dimensional structure of the structured creped web in terms of relative changes, frequency of change and rate of change . An additional way to characterize is provided through the use of two-dimensional spectral analysis methods. Spectral analysis simultaneously takes into account and quantifies the scale and frequency of structural changes in the CD and MD directions. The information obtained is used to adjust the papermaking process in real time and produce higher quality paper products, especially tissue paper, towels and other crepe paper products.

In other embodiments of the proposed method, the image of the region of the edge of the structure of the creped web is evaluated. The images of the edge zones show changes in the structure of the creped web through its thickness, such as changes in bulk, changes in density, and changes in the repeatability of the free ends of the fibers. The surface of the web that has been pressed against the Yankee cylinder generally has a greater number of elongated darker zones. They are fibers or groups of fibers that have been drawn from the surface of the web during the creping process. As in the case of the surface image, the light and dark zones are geometrically defined figures. A quantitative description of these figures based on indicators related to the area, perimeter, length, width, deviation from a symmetrical or regular shape, the spatial separation of these figures and the frequency of repetition of figures and signs of figures in space makes it possible to characterize the structure of the creped fabric shown in the image. Images acquired at different points in time can be used to characterize the changes over time of denser and less dense zones in terms of relative changes, frequency of change, and rate of change. In addition to the shapes of the ridges and creping grooves, the light intensity in each zone, the light intensity between zones, and light changes between similar zones can be used to characterize bulk, areas of damage, and free ends of the fibers. Images taken at different points in time can be used to determine changes in these characteristics over time in terms of relative change, rate of change and rate of change, and spectral analysis simultaneously takes into account and quantifies the scale and frequency of structure changes, both in the direction of MD and in a direction perpendicular to the plane of the canvas.

In some embodiments of the proposed method, image processing technologies such as contouring can be used to highlight the structure of the creped surface and / or the embossed three-dimensional structure of the creped structured web, as well as to create closed contours around the zones of the crests of the crepe or the zones of the embossed three-dimensional structure of the creped structured web. The concept of a closed contour is similar to the concept of lines of constant heights (horizontals) on a topographic map. The contours are built on the image based on the pixel values of the image, with contour lines connecting pixels with the same numerical pixel values. Because the pixel values of the creped web image indirectly correspond to the local height, the resulting contour lines delineate the ridges of the creped web. The contours define the shape and dimensions of the combs, as well as the shape and dimensions of the associated grooves. By changing the value of the pixels for which the contours are drawn, it is possible to characterize the slope of the scallops. The scallop of the crepe web is equal to the change in pixel value divided by the number of pixels between the contours in question. If the distance between the imaging device and the web is known, or if a calibration image is available, then the physical distance can be related to the number of pixels between the contour lines. Such technologies are standard and well known in the field of image analysis.

In other embodiments of the proposed method, images of the surface or relief zone of the formed creped web in the machine (MD) and transverse (CD) directions are obtained, and these images are evaluated so that closed contours are built around the crests of the crepe, as a result of which they are designated as combs, and creping grooves. Closed contours, as outlines of the structure of the crepes, will generally have an elongated shape, and their larger dimension will be oriented parallel to the transverse direction, and the smaller dimension will be parallel to the machine direction. Due to variations in both web structure and the manufacturing process, closed loops will vary in a range of shapes and sizes. In addition, changes in the manufacturing process will lead to changes in the surface structure of the creped web, which in turn entail changes in both the shape and size of the closed contours, as well as the relative distribution of these shapes and sizes. Depending on the product being manufactured, there may be trends in both the shape and size of the ridges and crepes, and these trends may correlate with characteristics of the creped web, such as softness, stretch, tear strength, and bulk. There are a number of geometries that can be used to characterize the shape of the crepe combs. In the case of creped structured webs, the same technique is used to construct closed contours around the lighter areas of the extruded three-dimensional structure of the creped structured web, so that the characteristics of the structure can be obtained. Although the embossed structure must be uniform, both in time and space, the above manufacturing process variations result in changes in structure that can be characterized and correlated with web characteristics such as softness, stretch, tear strength and bulk.

In one variant of the proposed method, the indicators used to evaluate the resulting images are associated with the idea of a characteristic ratio. The "characteristic ratio" is usually defined in relation to regular shapes such as rectangles and ellipses as the ratio of the larger dimension to the perpendicular smaller dimension. This is a simple indicator that is used to establish a relationship between the dimensions of a figure in two perpendicular directions. This indicator is particularly suitable for evaluating the surface or relief zone (part) of the formed creped web to determine the change in the dimensions of the ridges and crepe grooves in the MD and CD directions. A long thin figure will have a higher aspect ratio compared to a short thick figure. In the case of low and high aspect ratio figures, the creped web generally does not feel soft, and depending on the relative separation of the ridges, the web may have unsatisfactory stretch and strength characteristics. Combs with a low aspect ratio will be more separated from each other in the machine direction, even in a close-packed configuration. As a result, the crests of the crepe will feel more like separate structures, which detracts from the soft feel. If the figures have a large characteristic ratio, then this may indicate scallops with pointed tops, which also impair the feeling of softness. This is exacerbated when the crepe combs are more separated from each other. Both the elongation of the creping ridges and the number of ridges per unit area and/or per unit length, in the MD and CD directions, are factors influencing the characteristics of the final web.

In some embodiments of the proposed method, if the creping ridges are not simple ellipses, then a modified form of the characteristic ratio is used instead of the standard form of the characteristic ratio to obtain characteristics of the structure of the creping ridges. As already mentioned, the standard form of the aspect ratio is the ratio of the largest dimension to the perpendicular minor dimension. If the sensor or imaging device remains at a relatively constant distance from the surface being imaged, then the number of pixels in either MD or CD direction can be used as an identifier for physical dimensions in the MD or CD direction. The standard aspect ratio definition can be modified for asymmetric figures by using the average length/average width ratio definition. In this case, the average length is the average of the pixel line lengths in the CD direction, and the average width is the average of the pixel lengths in the MD direction. Another modification of the characteristic ratio is the aspect ratio of the figure,

F _extension = (I _CD /I _MD ) ^1/2

where I _CD and I _MD are the second moment of inertia in the directions CD and MD, respectively, and where the second moment of inertia is defined as

I _x =ƒƒx ² dx dy.

This indicator requires additional calculations, but it potentially better quantifies the relationship between the sizes of the crepe combs in the MD and CD directions.

In other embodiments of the proposed method, several indicators are used to determine the relationship between the length and width of the scallop patterns and crepe grooves. In these indicators, one or both sizes are compared with the area or perimeter of the figure. The simplest of these measures refer to the MD and CD directions separately and include: i) closed loop perimeter/maximum dimension in the CD direction; ii) closed loop perimeter/maximum dimension in MD direction; iii) closed contour area/maximum dimension in direction CD; and iv) closed loop area/maximum dimension in MD direction. It is worth noting that i) divided by ii) or iii) divided by iv) results in a characteristic ratio. Divide i) by iv) or ii) by iii) to obtain an index in which the ratio of perimeter to area is multiplied by the ratio of the maximum dimension in the CD direction to the maximum dimension in the MD direction. More difficult to calculate, the turning radius for the MD or CD direction is defined as:

R _turn-CD = [I _CD /area] ^1/2

And

R _turn-MD = [I _MD /Area] ^1/2 .

These figures will approach unity when the length or width of the ridges and crepe grooves reaches the numerical value of the perimeter or area. The closer the index is to one, the more elliptical the figure is, which is generally undesirable for creping structures for the same reason that a low aspect ratio is undesirable. An indicator that refers to the same shape parameter, but takes into account both the MD direction and the CD direction, is the figure compactness factor:

F _compactness = Area/[2 π (I _MD ² +I _CD ² ) ^1/2 ].

Isoperimetric coefficient

Q = 4 π zone area/zone perimeter ²

is an alternative indicator of compactness. As a rule, increased compactness characterizes more rounded figures. This is undesirable for the structure of the creping ridges, since in this case the surface quality will be similar to the quality typical for figures with a low aspect ratio.

It is generally assumed that the crepe comb shapes do not have concave portions. The concave areas are indicative of the heterogeneity of the structure of the creping combs, which, if extensively distributed among the combs, leads to a decrease in softness, bulk, tensile strength and stretch.

In some embodiments of the proposed method, image analysis technologies are used to determine how much of the figure's perimeter is convex. In this case, the waviness index is used, which can be determined in accordance with the equation:

F _waviness = P _bulge / P,

where P _bulge is the length of the perimeter with a convex shape, and P is the total length of the perimeter. The value of the waviness index close to one is preferable.

In other embodiments of the proposed method, the indicator of deviation from the round or rounded shape is used. This indicator is determined using the equation

R=(1/N) Σ R _i ,

where R is the distance from the center of the figure to the perimeter, N is the total number of measurements taken with equal angular increments along the perimeter line of the figure, and the summation is carried out from 1 to N. Then the parameters a and b are calculated:

a = (2/N) _{ΣRicos θi}

b = (2/N) Σ R _i sin θ _i ,

where θ is the angular increment for radial measurements, and are used in the expression

Δ=R _i - R - a cos θ _i - sin θ _i ,

where Δ is the roundness deviation that should be maximized in the case of crepe ridges. The crepe scallops, containing rounded shapes, feel like raised braille marks.

All of the above metrics can also be used to evaluate and characterize the embossed three-dimensional structure of a creped structured web.

In other embodiments of the proposed method, an alternative to characterizing the shapes of crepes and grooves is to directly quantify the light intensity in the ridges and grooves, the light intensity between the ridges and grooves, and the difference in light intensity between areas with similar shapes.

In other embodiments of the proposed method, after applying geometric indicators to the areas of ridges and grooves of the crepe and/or quantifying the intensity of illumination for these areas, statistical processing techniques can be applied to obtain additional characteristics of the structure of the creped web. The simplest means for characterizing ridges and creping grooves are, for example, average, median and standard deviation. In the case of a creped web, in which the ridges and crepe grooves have a range of shapes and sizes, these types of means do not fully describe the structure. A more complete description is to quantify the distribution of the indicator used. A simple example is a characteristic ratio histogram. It shows the measurement range of characteristic ratios as well as the relative frequency of repetition.

All of the methods described above relating to the quantitative description of light intensity can also be applied to evaluate and characterize the embossed three-dimensional structure of a creped structured web.

The same geometric parameters and quantitative description of the intensity of illumination are applied to images of the edge of the canvas. In the case of the edge of the web, an elongated shape with the largest dimension significantly larger than the smallest dimension is not possible due to the limited distance in the direction perpendicular to the plane of the web, and usually results in undesirable characteristics. Long elongated shapes result from large crepe combs with sharp edges. Shapes with smaller aspect ratios, but not round, are more desirable for a web with the desired bulk, stretch, tear strength, and softness characteristics.

Examples

The following examples refer to the characterization of ridge and crepe structures and it should be apparent that in the case of the more uniform structures created in the structured web process, the examples given are equally applicable to creped structured webs. In FIG. 1 shows the surface of a creped web moving in a conventional tissue paper machine at a typical speed of about 1200 m/min. The image shows the bottom surface of the web, the side of the web that was glued to the Yankee cylinder. As for the top surface of the web, this surface has more rounded (smoothed) creping ridges, less sharp transitions between ridges and grooves, and, as a rule, contains more open ends of the fibers. This image was taken using a high speed digital camera capable of 1000 fps. The lens system had a depth of field that maintained image focus despite normal changes in distance between the web and the imaging system. The shutter speed was 2 µs. An LED lighting system was used to illuminate the surface of the canvas. In the image, crepe ridges appear as brighter areas and crepe grooves appear as darker areas. The CD direction is generally parallel to the largest dimension of the crepes and the MD direction is generally perpendicular to this largest dimension. These directions are indicated by arrows superimposed on the image. The MD/CD coordinate system is not aligned with the vertical and horizontal image axes. Most modern image analysis and processing software packages include technologies for rotating the MD/CD coordinate system or vertical and horizontal image coordinate systems so that the two systems are aligned with each other.

When viewing the image, it can be seen that the structures of the ridges and crepe grooves have blurred borders and could be clearer. This is typical for digital imaging, which uses discrete pixels to form an image. The actual location of the change in light intensity may take place throughout the entire pixel instead of a pixel boundary. As a result, the pixel is assigned a value that represents a light intensity level that falls between the two. This can be corrected by applying an image sharpening algorithm (increasing the observed contrast). In FIG. 2, 3 and 4 illustrate an example. In FIG. 2 shows an actual sharp transition between two levels of illumination intensity. In FIG. 3 shows a digital representation of this image. The transition line between two levels of illumination intensity cannot pass through the boundaries of pixels, as a result of which the pixels located along the transition line have intermediate values. In FIG. 4 shows the result of applying the image sharpening algorithm. Such algorithms are well known in the field of image analysis and processing, and they are included in almost all software packages used for digital image processing. In FIG. 5 shows an image of FIG. 1 after applying the image sharpening and magnification algorithm.

Web surface area with closed contours

In FIG. 6 shows an image of FIG. 5 with closed contours indicating areas of crepe scallops. The techniques used to define the contour lines are well known in the field of image analysis and processing. These technologies involve the application of numerical algorithms to digital images in a manner similar to the application of algorithms that have been used to sharpen an image. It should be noted that the zones are generally oriented with their largest size in the CD direction, but they are not uniform in shape and size. This is a typical case. The shape, size and relative spatial frequency of these zones determines the quality of the canvas. This relationship varies with fiber composition, product type and other factors. The above-described geometric dimensions are used to characterize the structures of the ridges and crepe grooves with uniform and quantifiable dimensions. It should be noted that with the help of the described indicators, the multidimensional nature of the structure of the creped web is naturally quantified and characterized. Changes in performance, both spatially and temporally, are then used to identify and quantify changes in machine performance and web quality. After that, appropriate changes can be made to the operation of the machine to ensure the required web quality.

Statistical estimates

One way to monitor changes in the indicators used is through the use of statistical tools that provide the ability to quantify changes, both in space and time, and the frequency of these changes. In FIG. 7, 8 are examples of the use of statistical tools such as histograms. In FIG. 7 shows the distribution of distances between crepe ridge zones, which are defined by closed contours. In this case, the distribution is compact, which indicates a uniform distribution of scallops in space. In FIG. 8 shows the distribution of illumination intensity for zones that are defined by closed contours. As can be seen, the distribution has two vertices (the second one is not clearly expressed), which may be an indication of some heterogeneity in the shapes of the crepe ridges. In addition, histograms from different locations on the web or at different times can be compared, and relative differences, rates of change, and regularities can be identified and used to characterize changes in web structure. Any of the above metrics or sizes can be handled in a similar way. The performance will vary depending on the composition of the compositions used in the creping, the moisture content of the web or coating, the contamination of the coating with fibers/fine fibers, the wear of the creping knife, and the like. In this way, relationships can be established between specific operating conditions and/or specific web quality characteristics, as well as the information provided by the indicators used and the application of statistical tools to these indicators. These relationships can then be used to optimize machine performance and web quality. The most useful relationships will depend on the application.

Web edge area with closed contours

In FIG. 9 shows an image of the web edge area. This image was acquired using the same illumination and imaging system used to acquire the image of FIG. 1. As in the case of the image of the surface area, a sharpening algorithm was applied to the original image to compensate for blurring caused by pixels, resulting in an image with increased clarity and contrast. In FIG. 10 shows the resulting image. In FIG. 11 shows the same image with lines of closed contours delimiting denser and less dense areas of the web. It should be noted that longer zones, having larger dimensions, are usually oriented vertically. Relative repeatability, spacing between zones, and total ridge height are related to web bulkiness. Contours can be described using mean values, median values and standard deviation values, however, as in the case of a surface image, this does not give a complete description of the structure.

In FIG. 12, 13 are additional examples of applied geometries, with the results presented in the form of histograms. In FIG. 12 shows a histogram for areas covered by closed contours. It can be seen that most of the zones have a relatively small area, but there are zones with very large areas, but their frequency is low. They are a possible indication of the heterogeneity of the formation, but this requires further study. In FIG. 13 shows a histogram of the roundness of closed contours. As you can see, the histogram shows a more or less smooth distribution despite some gaps. This is the case in which time variation can provide additional information about the characteristics of the distribution. A possible additional measure would be the distribution over time of the mean.

In the case of an edge image, a line may be superimposed on the image and measurements may be taken relative to that line. In FIG. 14 is an example. The line may pass through various points outside the plane of the web, and characteristics may be determined along this line. This can be done at many points outside the plane of the web so that changes in web thickness can be quantified. In FIG. 15 shows the illumination intensity along the line, and FIG. 16 shows a histogram for this measurement. Since illumination intensity is a measure of density, performing this evaluation at many points out of plane of the web allows a density map to be obtained across the thickness of the web, effectively providing a high resolution bulk measurement.

Example - Applying Spectrum Decomposition

The web surface image provides spatial information in the MD and CD directions, and the web edge image provides spatial information in the MD direction and in the direction perpendicular to the web plane. Thus, any type of image can be analyzed using 2D spectral analysis. Two-dimensional spectral analysis methods quantify the spectral content of an image, the repetitive features that make up the image. The most commonly used method is the Fourier transform. In the case of digital images, an FFT is used to implement the Fourier transform. This provides the transformation of the image from the spatial domain to the frequency domain, and all the information contained in the original image is stored in the converted image.

This example uses the image shown in Fig. 5. This image is converted from the spatial domain to the frequency domain and the result is shown in FIG. 17. The right half of the image is the inversion of the left half of the image, which is typical for this transformation. The lighter zones contain information for the largest, dominant spatial features, while the darker zones contain information for the smaller, non-dominant features. In FIG. 17, a contour is drawn around the brightest area of the image. The contour is constructed using the same methods used to construct the contours around the crepe combs in FIG. 6. In FIG. 18 shows an image of FIG. 17, from which all information that was contained outside the outlines has been removed. In other words, information describing smaller, non-dominant features, such as, for example, the free ends of the fibers, has been removed. The image was then inverted to return to the spatial domain, but without the deleted information, and the result is shown in FIG. 19. Comparing FIG. 5 and 19, it can be seen that both images contain the main ridges and crepe grooves. The contribution of smaller features, such as free ends of fibers, can be assessed by applying the same scores to both images and comparing the results. For example, histograms similar to the histograms of FIG. 7, 8 (distance between zones and illumination intensity), followed by their comparison. The process can then be repeated many times, contouring around areas of progressively increasing or decreasing brightness, so that the contribution of sizes over a range to the overall structure of the creped web can be quantified.

In yet another embodiment of the proposed method, a 2D transform is used to estimate the position and shape of a contour used to delineate the brightest area in an image. Because the image resulting from the FFT is symmetrical, only half of the FFT is needed for evaluation, as can be seen in FIG. 20. If the creping pattern is a sine wave, then the FFT result will consist of a single dot on a black background. The degree of deviation from this result can be used as an indication of the uniformity and periodicity of the structure of the creped web. The geometry and illumination intensity that have been applied to the contours at the surface and at the edge can be applied to the contour in the FFT image, resulting in the ability to accurately measure the selected area in the 2D FFT image of FIG. 20.

Although in FIG. 20 shows a single contour, however, multiple contours can be built with progressively lower illumination intensity thresholds to define the contour. The location, shape and dimensions of the contours will change if there are changes in the structure of the creped web. A series of contours can be obtained which represents the optimum structure of the creped web. During web production, the optimal contours are compared with the contour measured in real time and the deviation from the optimum is determined. The degree of deviation from this result can be used as an indication of the uniformity and periodicity of the creping pattern. The advantage of this process is that there is no need to measure specific image features, which simplifies the analysis process.

While embodiments of the invention are illustrated in the description using various examples, and these options are described in sufficient detail, the applicant does not intend to limit in any way the scope of the attached form to the described options. Additional advantages and modifications may be apparent to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific elements represented by the methods and illustrative examples described above. Accordingly, other elements can be used without going beyond the essence and scope of the main idea of the invention proposed by the applicant.

While at least one illustrative embodiment of the invention is presented in the above detailed description, it should be understood that there are a large number of modifications. It should also be understood that the illustrative example or illustrative examples are merely examples and do not limit the scope, applicability or structure of the invention in any way. On the contrary, the foregoing detailed description will provide a convenient guide for those skilled in the art in implementing the illustrative embodiment, it being understood that various changes may be made to the function and arrangement of the elements described in the illustrative embodiment without departing from the scope of the invention as defined by the appended claims and their legally binding claims. equivalents.

Claims (16)

1. A method for obtaining the characteristics of the structure of a creped web, including: providing an imaging system comprising one or more television or photographic sensors or devices; generating one or more signals defining one or more images of a creped web structure zone; And evaluating the formed image of the structure of the creped web by 2D spectral analysis using one or more indices to determine crepe ridge characteristics, crepe groove characteristics, embossed 3D structure characteristics, web bulk, web internal tears and/or loose fiber ends. 2. The method of claim 1 wherein the structure of the creped web is an extruded three-dimensional structure. 3. The method of claim. 1, in which two-dimensional spectral analysis is used to determine the frequency of repetition of the pattern on the depicted area of the structure of the creped web. 4. The method of claim. 1, which uses image processing technologies to highlight the structure of the creped fabric and to create closed contours around the areas of crepe ridges. 5. The method of claim. 1, in which the depicted area of the structure of the creped web is evaluated in the machine direction and in the transverse direction. 6. The method according to p. 5, in which the structure of the creped fabric in the machine and transverse directions is evaluated using one or more indicators selected from the group consisting of the following indicators: zone area; zone perimeter; characteristic ratio, defined as the largest dimension/smallest dimension, perpendicular to the largest dimension, average length/average width; closed loop perimeter/length; zone border/width; zone area/length; zone area/width; figure elongation factor, defined as

Where

And

is the second moment of inertia in the directions MD and CD, and the second moment of inertia is defined as

figure compactness factor, defined as

turning radius, defined as

convexity factor, defined as P _convexity = convex part of the perimeter of the zone/full perimeter of the zone; isoperimetric coefficient, defined as Q=4 π zone area/zone perimeter ² ; roundness of the zones, defined as a deviation from the circle, the intensity of illumination in the zones, the intensity of illumination between the zones, the number of zones in the image, which are then converted into a frequency number, minimum and maximum heights, the vertical distribution of zones in size and shape, the frequency of repetition of long vertical zones, frequency of repetition of short zones. 7. The method according to any one of paragraphs. 1-6, wherein the creping ridges and/or surrounding areas in the embossed three-dimensional structure of the creped structured web are evaluated within the depicted creped structure. 8. The method according to any one of paragraphs. 1-6, in which the image is used to determine the bulk of the web, the internal violations of the web, the free ends of the fibers, the structure of the creping and/or the embossed three-dimensional structure of the creped structured web. 9. The method of claim. 1, in which the depicted zone of the structure of the creped fabric is the edge of the structure of the creped fabric in the machine direction. 10. The method of claim. 9, which uses two-dimensional spectral analysis to determine the frequency of repetition of the pattern on the depicted area of the structure of the creped web. 11. The method of claim. 9, which uses image processing technologies to highlight the structure of the creped web and to create closed contours around the areas of crepe ridges. 12. The method according to p. 9, in which the zone of the edge of the structure of the creped web is evaluated using one or more indicators selected from the group consisting of the following indicators: zone area; zone perimeter; aspect ratio, defined as the largest dimension/smallest dimension perpendicular to the largest dimension, and ii) average length/average width; zone boundary/length; zone border/width; zone area/length; zone area/width; figure elongation factor, defined as

Where

And

is the second moment of inertia in the directions MD and CD, and the second moment of inertia is defined as

figure compactness factor, defined as

turning radius, defined as

And

convexity factor, defined as P _convexity = convex part of the perimeter of the zone / full perimeter of the zone; isoperimetric coefficient, defined as Q=4 π zone area/zone perimeter ² ; roundness of the zones, defined as a deviation from the circle, the intensity of illumination in the zones, the intensity of illumination between the zones, the number of zones in the image, which are then converted into a frequency number, minimum and maximum heights, the vertical distribution of zones in size and shape, the frequency of repetition of long vertical zones, frequency of repetition of short zones. 13. The method of claim 10, wherein the creping ridges are evaluated within the area of the depicted creped web structure.

Images