A machine vision-based fabric dyeing uniformity online control method and system

Abstract

The application discloses a kind of based on machine vision's fabric dyeing uniformity online control method and system, it is related to machine vision technical field, its technical solution key points include the following steps: image acquisition position point is obtained by image acquisition position marking to the online production target fabric dyeing area, the interval parameter between two adjacent image acquisition position points is obtained to obtain the interval to be measured;The historical dyeing concentration distribution characteristics of historical period fabric are extracted from historical fabric dyeing monitoring data, and the color difference attenuation correlation factor between color deviation value and the interval to be measured in historical fabric dyeing monitoring data is counted according to historical dyeing concentration distribution characteristics;Effect is to make each batch fabric color consistent and texture uniform, so that product quality stability is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of machine vision technology, and more specifically, to a method and system for online control of fabric dyeing uniformity based on machine vision. Background Technology

[0002] Dyeing uniformity is one of the indicators for measuring product quality. With the advancement of machine vision technology, its application in industrial production inspection is becoming increasingly widespread. Machine vision uses cameras and image algorithms to quickly collect and analyze the surface features of objects. Its advantages in fabric dyeing are gradually becoming apparent; it can capture the differences in color and texture after dyeing, converting visual information into data, thus providing an objective basis for dyeing uniformity analysis. However, traditional machine vision-based online control methods for fabric dyeing uniformity are simple color comparisons, not deeply integrated with the entire production process, and cannot accurately output process adjustment parameters. This forces companies to repeatedly experiment with processes when facing quality issues, thus delaying production and increasing costs. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a method and system for online control of fabric dyeing uniformity based on machine vision.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A machine vision-based online control method for fabric dyeing uniformity, comprising the following steps: Image acquisition location points are obtained by marking the image acquisition location of the dyeing area of ​​the target fabric in online production, and the distance parameter between two adjacent image acquisition location points is obtained to obtain the distance to be measured. Historical dyeing concentration distribution characteristics of fabrics in historical periods were extracted from historical fabric dyeing monitoring data. Based on the historical dyeing concentration distribution characteristics, the color difference attenuation correlation factor between color deviation value and test distance in historical fabric dyeing monitoring data was statistically analyzed. Based on the current dyeing concentration distribution characteristics of the target fabric, a preprocessing correlation factor is extracted from the color difference attenuation correlation factor. Based on the preprocessing correlation factor, the color deviation status of the target monitoring area in the target fabric that is prone to uneven dyeing is determined, and the first color deviation value and the second color deviation value are obtained. The comprehensive color difference factor is obtained based on the first color deviation value and the second color deviation value. The uniformity of the dyeing in the target monitoring area affected by the dyeing concentration is judged based on the comprehensive color difference factor to obtain the uniformity deviation value. Based on the uniformity deviation value, an adjustment warning notification is output, and the optimal adjustment parameters are extracted from the dyeing process optimization parameters based on the adjustment warning notification.

[0005] Preferably, the color difference attenuation correlation factor between the color deviation value and the distance to be measured in historical fabric dyeing monitoring data is statistically analyzed based on the historical dyeing concentration distribution characteristics. This specifically includes the following steps: Based on the historical color concentration distribution characteristics, the historical color deviation values ​​of the image acquisition locations are extracted. The deviation difference is obtained by subtracting the historical color deviation values ​​of two adjacent image acquisition points based on the distance to be measured. The ratio of the deviation difference to the distance to be measured is used to obtain the deviation ratio. The color difference attenuation correlation factor is obtained by averaging all the measured deviation ratios.

[0006] Preferably, the preprocessing correlation factor is extracted from the color difference attenuation correlation factor based on the current dyeing concentration distribution characteristics of the target fabric, specifically including the following steps: Obtain the current dyeing concentration distribution characteristics of the target fabric; After matching the current staining concentration distribution characteristics with the historical staining concentration distribution characteristics, the preprocessed correlation factor is extracted from the color difference attenuation correlation factor.

[0007] Preferably, the color deviation status of the target monitoring area in the target fabric prone to uneven dyeing is determined based on the preprocessing correlation factor to obtain a first color deviation value and a second color deviation value, specifically including the following steps: The target monitoring point is obtained by monitoring the location of the target fabric in the target monitoring area. The image acquisition position points are marked on the left and right edges of the target fabric in the width direction to obtain the first acquisition point and the second acquisition point. The first preprocessing spacing is obtained by calculating the distance between the target monitoring point and the first collection point; The first color deviation value of the target monitoring point is obtained based on the first preprocessing interval, the ratio of the deviation to be measured, and the preprocessing correlation factor. The second preprocessing interval is obtained by calculating the distance between the target monitoring point and the second collection point. The second color deviation value of the target monitoring point is obtained based on the second preprocessing interval, the ratio of the deviation to be measured, and the preprocessing correlation factor.

[0008] Preferably, the comprehensive color difference factor is obtained based on the first color deviation value and the second color deviation value, specifically including the following steps: The preprocessed change frequency is obtained by detecting the color change frequency at the first acquisition point; The first color difference factor is obtained by multiplying the first color deviation value and the preprocessing change frequency. The second color difference factor is obtained by multiplying the second color deviation value and the preprocessing change frequency. The first color difference factor and the second color difference factor are summed to obtain the comprehensive color difference factor.

[0009] Preferably, the uniformity deviation value is obtained by determining the dyeing uniformity of the target monitoring area affected by dyeing concentration based on the comprehensive color difference factor, specifically including the following steps: Extract a reference deviation database of the dyeing uniformity deviation of the corresponding target monitoring area under different color deviation values ​​and different color change frequencies; The uniformity deviation value of the target monitoring area is obtained by matching the comprehensive color difference factor with the reference deviation database.

[0010] Preferably, an adjustment warning notification is output based on the uniformity deviation value, specifically as follows: If the uniformity deviation value is greater than or equal to the preset dyeing uniformity warning threshold, an adjustment warning notification message will be output.

[0011] Preferably, the preferred adjustment parameters are extracted from the dyeing process optimization parameters based on the adjustment warning notification information, specifically including the following steps: After generating the dyeing process optimization parameters for the target fabric based on the adjusted early warning notification information, the candidate optimization parameters are output. The degree of improvement in staining effect corresponding to each candidate optimization parameter is detected to obtain parameter optimization feature data; The baseline process characteristic data is obtained by detecting the baseline value of the dyeing effect to which the current dyeing process parameters belong; The optimal adjustment parameters are obtained by extracting the parameter optimization feature data of the candidate optimization parameters from the baseline process feature data.

[0012] A machine vision-based online control system for fabric dyeing uniformity includes: Acquisition module: Marks the image acquisition location points in the dyeing area of ​​the target fabric in online production, and obtains the distance parameter between two adjacent image acquisition location points to obtain the distance to be measured; Extraction module: Extracts the historical dyeing concentration distribution characteristics of fabrics from historical fabric dyeing monitoring data, and calculates the color difference attenuation correlation factor between color deviation values ​​and the distance to be measured in historical fabric dyeing monitoring data based on the historical dyeing concentration distribution characteristics; First processing module: Based on the current dyeing concentration distribution characteristics of the target fabric, extract the preprocessing correlation factor from the color difference attenuation correlation factor, and based on the preprocessing correlation factor, determine the color deviation status of the target monitoring area in the target fabric that is prone to uneven dyeing to obtain the first color deviation value and the second color deviation value. The second processing module obtains a comprehensive color difference factor based on the first color deviation value and the second color deviation value, and judges the dyeing uniformity of the target monitoring area affected by the dyeing concentration based on the comprehensive color difference factor to obtain the uniformity deviation value. Control module: Outputs adjustment warning notification information based on the uniformity deviation value, and extracts the optimal adjustment parameters from the dyeing process optimization parameters based on the adjustment warning notification information.

[0013] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement an online control method for fabric dyeing uniformity based on machine vision.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This invention utilizes machine vision to mark image acquisition points and combines historical data to mine color difference attenuation correlation factors to construct a dyeing uniformity analysis system. This application can capture fabric color deviations and control dyeing uniformity within a fluctuating range, ensuring consistent color and uniform texture in each batch of fabric, thus significantly improving product quality stability. Based on comprehensive color difference factors, it judges uniformity deviations and outputs early warnings, and optimizes and adjusts process parameters. If dyeing uniformity deviates from the standard during production, a timely warning is issued. Through image acquisition, uniformity judgment, and process optimization during online fabric production, end-to-end online control is achieved. Attached Figure Description

[0015] Figure 1 A schematic diagram illustrating the steps of an online control method for fabric dyeing uniformity based on machine vision proposed in this invention; Figure 2 This invention presents a schematic diagram of a machine vision-based online control system for fabric dyeing uniformity. Detailed Implementation

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0018] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0019] Reference Figures 1-2 As shown.

[0020] The embodiments further illustrate the online control method and system for fabric dyeing uniformity based on machine vision proposed in this invention.

[0021] A machine vision-based online control method for fabric dyeing uniformity, comprising the following steps: Image acquisition location points are obtained by marking the image acquisition location of the dyeing area of ​​the target fabric in online production, and the distance parameter between two adjacent image acquisition location points is obtained to obtain the distance to be measured. Historical dyeing concentration distribution characteristics of fabrics in historical periods were extracted from historical fabric dyeing monitoring data. Based on the historical dyeing concentration distribution characteristics, the color difference attenuation correlation factor between color deviation value and test distance in historical fabric dyeing monitoring data was statistically analyzed. Based on the current dyeing concentration distribution characteristics of the target fabric, a preprocessing correlation factor is extracted from the color difference attenuation correlation factor. Based on the preprocessing correlation factor, the color deviation status of the target monitoring area in the target fabric that is prone to uneven dyeing is determined, and the first color deviation value and the second color deviation value are obtained. The comprehensive color difference factor is obtained based on the first color deviation value and the second color deviation value. The uniformity of the dyeing in the target monitoring area affected by the dyeing concentration is judged based on the comprehensive color difference factor to obtain the uniformity deviation value. Based on the uniformity deviation value, an adjustment warning notification is output, and the optimal adjustment parameters are extracted from the dyeing process optimization parameters based on the adjustment warning notification.

[0022] First, the image acquisition location points are marked on the dyeing area of ​​the target fabric in online production. Then, the distance parameter between two adjacent location points is obtained after marking the image acquisition location points to obtain the distance to be measured.

[0023] Historical fabric dyeing concentration distribution characteristics were extracted from historical fabric dyeing monitoring data. This is because historical fabric dyeing monitoring data includes the correlation between concentration distribution and color deviation during the fabric dyeing process. Based on the historical dyeing concentration distribution characteristics, a color difference attenuation correlation factor between color deviation values ​​and the measured distance in the historical fabric dyeing monitoring data was statistically analyzed.

[0024] Based on the current dyeing concentration distribution characteristics of the target fabric, preprocessing correlation factors are extracted from the color difference attenuation correlation factors. This step is to adapt historically summarized patterns to the actual situation of current fabric dyeing. Then, based on the preprocessing correlation factors, the color deviation of the target monitoring area prone to uneven dyeing in the target fabric is judged, resulting in the first color deviation value and the second color deviation value. By marking the sampling points on the left and right edges of the fabric width, the first and second sampling points are obtained. The distances between the target monitoring point and the first and second sampling points are calculated to evaluate the color deviation of the monitoring area from different directions.

[0025] A comprehensive color difference factor is obtained based on the first and second color deviation values. The frequency of color changes at the first sampling point is detected, and the comprehensive color difference factor is obtained by summing the frequency of color changes with both the first and second color deviation values. This comprehensive color difference factor is then used to match a reference deviation database containing dyeing uniformity deviations under different color deviations and frequency of change conditions, thereby determining the influence of dyeing concentration on the dyeing uniformity of the target monitoring area and obtaining a uniformity deviation value.

[0026] If the uniformity deviation value is greater than or equal to the preset dyeing uniformity warning threshold, an adjustment warning notification is output, serving as a warning signal when dyeing uniformity issues arise. Based on the adjustment warning notification, optimal adjustment parameters are extracted from the dyeing process optimization parameters. Dyeing process optimization parameters for the target fabric are generated as candidate optimization parameters. Then, the degree of improvement in dyeing effect corresponding to each candidate parameter and the baseline value of the dyeing effect for the current dyeing process parameters are detected. The optimal adjustment parameters are obtained by extracting the parameter optimization feature data of the candidate optimization parameters that are superior to the baseline process feature data, thereby optimizing the dyeing process and solving the dyeing uniformity problem.

[0027] Based on the historical dyeing concentration distribution characteristics, the correlation factor between color difference attenuation and the measured distance in historical fabric dyeing monitoring data is statistically analyzed. This includes the following steps: Based on the historical color concentration distribution characteristics, the historical color deviation values ​​of the image acquisition locations are extracted. The deviation difference is obtained by subtracting the historical color deviation values ​​of two adjacent image acquisition points based on the distance to be measured. The ratio of the deviation difference to the distance to be measured is used to obtain the deviation ratio. The color difference attenuation correlation factor is obtained by averaging all the measured deviation ratios.

[0028] First, historical color deviation values ​​for the image acquisition locations are extracted based on the historical dyeing concentration distribution characteristics. A large amount of historical data is accumulated during fabric dyeing production, including the dyeing concentration distribution at various locations on the fabric under different production periods and process parameters. For example, during the production of a batch of fabric, images of multiple locations are acquired using machine vision, and the color deviation at each location is obtained by combining this with dyeing concentration detection, such as the color difference value compared to a standard color chart. In the current production process, the historical records corresponding to the current image acquisition location are first located in the historical data, and the historical color deviation values ​​for these locations are extracted.

[0029] The deviation difference is obtained by subtracting the historical color deviation values ​​of two adjacent image acquisition points based on the distance to be measured. Assume that two adjacent image acquisition points A and B in the historical data have historical color deviation values ​​of ΔE1 and ΔE2, respectively, and the distance to be measured between these two points is d (e.g., the actual physical distance calculated from the image coordinates). That is, the deviation difference ΔE = |ΔE1 - ΔE2|. This difference reflects the variation in color deviation between adjacent positions within the distance to be measured d. This captures the difference in dyeing deviation between adjacent areas of the fabric; a large difference indicates a significant problem with the uniformity of dyeing between these two adjacent positions.

[0030] The ratio of the deviation difference to the distance to be measured is used to obtain the deviation ratio. The deviation ratio K = ΔE / d, thus quantifying the correlation between color deviation changes and spatial distance. If the distance between two adjacent positions is small, but the deviation difference is large, the deviation ratio K will be large, indicating that the color deviation changes drastically within a small space, reflecting a significant local uniformity problem during the dyeing process.

[0031] The color difference attenuation correlation factor is obtained by averaging all the measured deviation ratios. Historical data contains combinations of multiple adjacent image acquisition points, such as points AB, BC, CD, etc., and each combination yields measured deviation ratios K1, K2, K3, etc. The color difference attenuation correlation factor is obtained by averaging these measured deviation ratios.

[0032] Based on the current dyeing concentration distribution characteristics of the target fabric, preprocessing correlation factors are extracted from the color difference attenuation correlation factors, specifically including the following steps: To obtain the current dyeing concentration distribution characteristics of the target fabric, machine vision detection methods are used to collect dyeing concentration data of the target fabric being produced in the online production scenario of fabric dyeing. For example, image recognition technology is used to detect the color depth and color distribution of different areas of the fabric, and features that can characterize how the current dyeing concentration of the fabric is spatially distributed are extracted. For example, some areas have high dyeing concentration and present a darker color, while some areas have low concentration and present a lighter color, as well as the area size and distribution density of these different concentration areas. These are combined to obtain the current dyeing concentration distribution characteristics.

[0033] After matching the current dyeing concentration distribution characteristics with historical dyeing concentration distribution characteristics, a preprocessing correlation factor is extracted from the color difference attenuation correlation factor. The historical dyeing concentration distribution characteristics are extracted from previously stored fabric dyeing monitoring data, including the distribution patterns of fabric dyeing concentration under different production batches and process conditions. The current fabric dyeing concentration distribution characteristics are compared one by one with numerous historical dyeing concentration distribution characteristics to obtain the most similar and highest matching historical features. Since the color difference attenuation correlation factor is calculated based on the historical dyeing concentration distribution characteristics, and the color difference attenuation correlation factor corresponding to the historical features that match the current features is the most valuable reference for the analysis of the current fabric dyeing uniformity, it is extracted as a preprocessing correlation factor. Assuming that a batch of fabrics in the historical data was dyed using a low-temperature dyeing process, its dyeing concentration distribution shows a slightly lower concentration in the edge area and a relatively uniform concentration in the middle area. If the target fabric currently being produced is found to have a similar concentration distribution pattern through testing, then the relevant part of the color difference attenuation correlation factor corresponding to this batch of historical fabrics is extracted to improve the accuracy of judging the dyeing uniformity of the current fabric.

[0034] Based on the preprocessing correlation factors, the color deviation of the target monitoring area in the target fabric that is prone to uneven dyeing is determined, resulting in a first color deviation value and a second color deviation value. The specific steps include: The target monitoring point is obtained by monitoring the location of the target monitoring area on the target fabric. Image acquisition points are marked on the left and right edges of the target fabric width to obtain the first and second acquisition points. In the fabric dyeing uniformity monitoring scenario, the target monitoring area of ​​interest is first identified, and the specific location of this area on the target fabric is determined using machine vision monitoring methods. This location is the target monitoring point. Simultaneously, considering that the edge areas of the fabric width are of significant reference value for judging dyeing uniformity, a point is selected on the left edge of the width as the first acquisition point, and a point on the right edge as the second acquisition point. For example, on a wide fabric, if we want to monitor a region in the middle that is prone to uneven dyeing, we first determine the center point or feature point of this region as the target monitoring point. Then, we mark the first acquisition point on the leftmost edge of the fabric and the second acquisition point on the rightmost edge. These three points form the basis for subsequent analysis of color deviation. The first preprocessing distance is obtained by statistically analyzing the distance between the target monitoring point and the first acquisition point. The actual distance between the target monitoring point and the first acquisition point on the fabric is calculated by measurement or based on image coordinate conversion, thus obtaining the first preprocessing distance. This reflects the spatial distance relationship between the target monitoring point and one edge of the fabric. Assuming the coordinates of the target monitoring point on the fabric are (x1, y1) and the coordinates of the first acquisition point on the left edge are (x2, y2), the straight-line distance between them is calculated using the distance calculation formula to obtain the first preprocessing distance. The first color deviation value of the target monitoring point is obtained based on the first preprocessing interval, the measured deviation ratio, and the preprocessing correlation factor. The first color deviation value of the target monitoring point relative to the first acquisition point is then calculated by substituting these factors into the formula. For example, the first color deviation value = first preprocessing interval × measured deviation ratio × preprocessing correlation factor. This calculation combines spatial spacing, historical deviation patterns, and the current fabric's correlation characteristics to obtain a result that reflects the color deviation of the target monitoring point relative to the left edge acquisition point, thus assessing the dyeing uniformity from one side edge of the fabric. The second preprocessing distance is obtained by calculating the distance between the target monitoring point and the second acquisition point. Similarly, the distance from the target monitoring point to the second acquisition point on the right edge is calculated by means of measurement or coordinate conversion, thus obtaining the second preprocessing distance.

[0035] The second color deviation value of the target monitoring point is obtained based on the second preprocessing interval, the measured deviation ratio, and the preprocessing correlation factor. The second color deviation value of the target monitoring point relative to the second acquisition point is then calculated by substituting these values ​​into the formula. For example, the second color deviation value = second preprocessing interval × measured deviation ratio × preprocessing correlation factor.

[0036] The comprehensive color difference factor is obtained based on the first color deviation value and the second color deviation value, specifically including the following steps: The preprocessed change frequency is obtained by detecting the color change frequency at the first sampling point. During the fabric dyeing process, the color at the first sampling point is not constant; machine vision continuously monitors this point. For example, with continuous fabric production, the rotation of the dyeing rollers, and dynamic changes in dye concentration, the color at the first sampling point may undergo periodic or non-periodic changes. The preprocessed change frequency is obtained by counting the number of color changes at the first sampling point within a certain period. This reflects the frequency of dynamic color changes in the edge areas during fabric dyeing, because more frequent color changes indicate a more unstable dyeing process and a greater impact on the final dyeing uniformity. The first color difference factor is obtained by multiplying the first color deviation value and the preprocessing change frequency. The first color deviation value reflects the degree of color deviation of the target monitoring point relative to the first acquisition point, while the preprocessing change frequency reflects the frequency of color changes at the first acquisition point. Assuming the first color deviation value is ΔE1 and the preprocessing change frequency is n, the first color difference factor F1 = ΔE1 × n. For example, if the color changes frequently at the first acquisition point, and the color deviation of the target monitoring point relative to the first acquisition point is also large, then the first color difference factor obtained after multiplication will be large, indicating a higher risk and impact of uneven dyeing from the direction of the left edge acquisition point. The second color difference factor is obtained by multiplying the second color deviation value and the preprocessing change frequency. The second color deviation value represents the degree of color deviation of the target monitoring point relative to the second acquisition point. Let the second color deviation value be ΔE2, and the preprocessing change frequency be n (since it is based on the color change frequency of the first acquisition point, it is assumed here that the color change frequencies of the two edges of the fabric width are correlated in the same production process, and the same value can be used; of course, in practice, the two sides may be detected separately). Then, the second color difference factor F2 = ΔE2 × n. Taking into account the influence of dyeing deviation and color change frequency on dyeing uniformity from the right edge of the fabric width, an evaluation dimension for the dyeing status of the target monitoring area is constructed from the left and right edges of the fabric. The comprehensive color difference factor is obtained by summing the first and second color difference factors, i.e., F = F1 + F2 = ΔE1×n + ΔE2×n. A larger comprehensive color difference factor indicates a greater combined influence of the dynamic changes and static deviations in dyeing at the left and right edges of the target monitoring area, leading to a higher probability and severity of uneven dyeing. Subsequently, this comprehensive color difference factor is used to match a reference database to determine the deviation in dyeing uniformity, providing a basis for adjusting the dyeing process.

[0037] The uniformity deviation value is obtained by determining the dyeing uniformity of the target monitoring area affected by dyeing concentration based on the comprehensive color difference factor, specifically including the following steps: A reference deviation database is extracted to record the dyeing uniformity deviation of the target monitoring area under different color deviation values ​​and color change frequencies. In long-term fabric dyeing production practice, different color deviation values ​​and color change frequencies are recorded. Different color deviation values ​​range from slight to severe deviations, and different color change frequencies represent the frequency of color changes, such as 1 or 3 changes per minute. Then, for each combination of color deviation value and color change frequency, high-precision machine vision inspection is used to determine the corresponding dyeing uniformity deviation of the target monitoring area. This data is then compiled to form the reference deviation database. For example, when the color deviation value is ΔE=5 and the color change frequency is n=3 times / meter, what is the corresponding dyeing uniformity deviation? When the color deviation value changes to ΔE=8 and the color change frequency changes to n=5 times / meter, what is the uniformity deviation? The results of these different combinations are stored in the database, providing a standard reference for subsequently judging the uniformity deviation of the fabric being produced. The uniformity deviation of the target monitoring area is obtained by matching the comprehensive color difference factor with a reference deviation database. The comprehensive color difference factor is then compared with the reference deviation database to find the uniformity deviation data that best matches the combination of color deviation value and color change frequency. For example, the color deviation value combination and frequency combination corresponding to the current comprehensive color difference factor are calculated, and the closest or perfect match is found in the reference deviation database. The corresponding dyeing uniformity deviation data is the uniformity deviation value of the target monitoring area. By comparing with a large amount of historical data, the dyeing uniformity deviation of the current production fabric in the target monitoring area can be determined, thus determining whether the dyeing uniformity deviation is within the normal range and providing clear data basis for whether subsequent adjustments to the dyeing process are needed.

[0038] Based on the uniformity deviation value, an adjustment warning notification message is output, specifically: If the uniformity deviation value is greater than or equal to the preset dyeing uniformity warning threshold, an adjustment warning notification will be output. The preset dyeing uniformity warning threshold is set based on industry standards, historical production experience, or customer quality requirements. If the uniformity deviation value is less than the preset dyeing uniformity warning threshold, it means that the dyeing uniformity is within an acceptable range and therefore no adjustment is needed. If the uniformity deviation value is greater than or equal to the preset dyeing uniformity warning threshold, it is determined that the dyeing uniformity is abnormal, and an adjustment warning notification will be output, such as sending an audible and visual alarm to the central control room and pushing a process adjustment instruction to the production system.

[0039] Based on the adjustment warning notification information, the optimal adjustment parameters are extracted from the dyeing process optimization parameters, specifically including the following steps: After generating optimized dyeing process parameters for the target fabric based on the adjustment warning notification information, candidate optimization parameters are output. The adjustment warning notification information includes the specific details of the current fabric dyeing uniformity deviation, and a pre-defined process adjustment strategy library is invoked based on this. For example, if it is known that the uniformity deviation is caused by uneven dye concentration distribution, different combinations of process parameters such as adjusting the dye liquor concentration replenishment frequency and adjusting the dyeing roller speed are generated. These combinations are the candidate optimization parameters, which form the basis for subsequent screening and selection of optimal adjustment parameters, providing multiple possible process adjustment directions for solving the dyeing uniformity problem. The degree of improvement in dyeing effect corresponding to each candidate optimization parameter is detected to obtain parameter optimization feature data. The candidate optimization parameters are then applied to the dyeing process to detect changes in the dyeing effect index of fabric dyeing uniformity. For example, after applying a candidate parameter that adjusts the dye liquor concentration replenishment frequency to a small section of fabric in actual dyeing production, machine vision is used to detect the dyeing uniformity of the newly produced fabric. The uniformity data is compared with the uniformity data before adjustment to calculate the improvement in uniformity and the data on the uniformity of dye concentration distribution. These data constitute the parameter optimization feature data for that candidate parameter, reflecting the actual improvement in dyeing effect that each candidate optimization parameter can bring. The baseline process characteristic data is obtained by detecting the baseline values ​​of the dyeing effect corresponding to the current dyeing process parameters. These parameters are those currently used in production. Detecting the baseline values ​​of the dyeing effect of the current dyeing process parameters clarifies the level of dyeing effect without parameter adjustments. For example, data on the dyeing uniformity and dyeing concentration distribution characteristics of the fabric produced by the current process are collected. This data constitutes the baseline process characteristic data and is compared with the effects of candidate optimization parameters to determine whether the candidate optimization parameters can truly improve the dyeing effect. The optimal adjustment parameters are obtained by extracting the parameter optimization feature data of candidate optimization parameters that are superior to the baseline process feature data. The parameter optimization feature data of each candidate optimization parameter is compared with the baseline process feature data. If the optimization feature data of a certain candidate optimization parameter (such as a higher improvement in dyeing uniformity or a more uniform dyeing concentration distribution) is better than the baseline process feature data, it indicates that this candidate parameter can effectively improve the dyeing effect, and this candidate parameter is selected as the optimal adjustment parameter. For example, if the parameter optimization feature data after adjusting the dye liquor concentration replenishment frequency shows a 10% improvement in dyeing uniformity, while the baseline process feature data shows a 0% improvement in uniformity, then this parameter is used as the optimal adjustment parameter for actual production adjustments, thereby solving the dyeing uniformity problem and improving the fabric dyeing quality.

[0040] A machine vision-based online control system for fabric dyeing uniformity includes: Acquisition module: Marks the image acquisition location points in the dyeing area of ​​the target fabric in online production, and obtains the distance parameter between two adjacent image acquisition location points to obtain the distance to be measured; Extraction module: Extracts the historical dyeing concentration distribution characteristics of fabrics from historical fabric dyeing monitoring data, and calculates the color difference attenuation correlation factor between color deviation values ​​and the distance to be measured in historical fabric dyeing monitoring data based on the historical dyeing concentration distribution characteristics; First processing module: Based on the current dyeing concentration distribution characteristics of the target fabric, extract the preprocessing correlation factor from the color difference attenuation correlation factor, and based on the preprocessing correlation factor, determine the color deviation status of the target monitoring area in the target fabric that is prone to uneven dyeing to obtain the first color deviation value and the second color deviation value. The second processing module obtains a comprehensive color difference factor based on the first color deviation value and the second color deviation value, and judges the dyeing uniformity of the target monitoring area affected by the dyeing concentration based on the comprehensive color difference factor to obtain the uniformity deviation value. Control module: Outputs adjustment warning notification information based on the uniformity deviation value, and extracts the optimal adjustment parameters from the dyeing process optimization parameters based on the adjustment warning notification information.

[0041] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements an online control method for fabric dyeing uniformity based on machine vision.

[0042] like Figure 2 As shown, the electronic device may include a processor, a communication interface, a memory, and a communication bus, wherein the processor, communication interface, and memory communicate with each other via the communication bus. The processor can call logical instructions in the memory to execute a machine vision-based online control method for fabric dyeing uniformity.

[0043] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0044] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute a machine vision-based online control method for fabric dyeing uniformity.

[0045] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform a machine vision-based online control method for fabric dyeing uniformity.

[0046] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0047] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A machine vision based on-line control method for fabric dyeing uniformity, characterized in that, The method includes the following steps: The image acquisition location points are obtained by visually inspecting the dyeing area of ​​the target fabric in online production. The distance parameter between two adjacent image acquisition location points is obtained to obtain the distance to be measured. The historical dyeing concentration distribution characteristics of fabrics from historical periods are extracted from historical fabric dyeing monitoring data. Based on the historical dyeing concentration distribution characteristics, the color difference attenuation correlation factor between the color deviation value and the distance to be measured in the historical fabric dyeing monitoring data is statistically analyzed. The specific steps include: Based on the historical color concentration distribution characteristics, the historical color deviation values ​​of the image acquisition locations are extracted. The deviation difference is obtained by subtracting the historical color deviation values ​​of two adjacent image acquisition points based on the distance to be measured. The ratio of the deviation difference to the distance to be measured is used to obtain the deviation ratio. The color difference attenuation correlation factor is obtained by averaging all the measured deviation ratios. Based on the current dyeing concentration distribution characteristics of the target fabric, a preprocessing correlation factor is extracted from the color difference attenuation correlation factor. The color deviation status of the target fabric's monitoring area prone to uneven dyeing is then determined based on the preprocessing correlation factor to obtain the first color deviation value and the second color deviation value. The specific steps include: The target monitoring point is obtained by monitoring the location of the target fabric in the target monitoring area. The image acquisition position points are marked on the left and right edges of the target fabric in the width direction to obtain the first acquisition point and the second acquisition point. The first preprocessing spacing is obtained by calculating the distance between the target monitoring point and the first collection point; The first color deviation value of the target monitoring point is obtained based on the first preprocessing interval, the ratio of the deviation to be measured, and the preprocessing correlation factor. The second preprocessing interval is obtained by calculating the distance between the target monitoring point and the second collection point. The second color deviation value of the target monitoring point is obtained based on the second preprocessing interval, the ratio of the deviation to be measured, and the preprocessing correlation factor. The comprehensive color difference factor is obtained based on the first color deviation value and the second color deviation value. The uniformity of the dyeing in the target monitoring area affected by the dyeing concentration is judged based on the comprehensive color difference factor to obtain the uniformity deviation value. Based on the uniformity deviation value, an adjustment warning notification is output. Based on the adjustment warning notification, the optimal adjustment parameters are extracted from the dyeing process optimization parameters. The specific steps include: After generating the dyeing process optimization parameters for the target fabric based on the adjustment warning notification information, the candidate optimization parameters are output. The degree of improvement in staining effect corresponding to each candidate optimization parameter is detected to obtain parameter optimization feature data; The baseline process characteristic data is obtained by detecting the baseline value of the dyeing effect to which the current dyeing process parameters belong; The optimal adjustment parameters are obtained by extracting the parameter optimization feature data of the candidate optimization parameters from the baseline process feature data.

2. The method according to claim 1, wherein, Based on the current dyeing concentration distribution characteristics of the target fabric, preprocessing correlation factors are extracted from the color difference attenuation correlation factors, specifically including the following steps: Obtain the current dyeing concentration distribution characteristics of the target fabric; After matching the current staining concentration distribution characteristics with the historical staining concentration distribution characteristics, the preprocessed correlation factor is extracted from the color difference attenuation correlation factor.

3. A machine vision based on-line control method of fabric dyeing uniformity according to claim 2, characterized in that, The comprehensive color difference factor is obtained based on the first color deviation value and the second color deviation value, specifically including the following steps: The preprocessed change frequency is obtained by detecting the color change frequency at the first acquisition point; The first color difference factor is obtained by multiplying the first color deviation value and the preprocessing change frequency. The second color difference factor is obtained by multiplying the second color deviation value and the preprocessing change frequency. The first color difference factor and the second color difference factor are summed to obtain the comprehensive color difference factor.

4. The method according to claim 3, wherein, The uniformity deviation value is obtained by determining the dyeing uniformity of the target monitoring area affected by dyeing concentration based on the comprehensive color difference factor, specifically including the following steps: Extract a reference deviation database of the dyeing uniformity deviation of the corresponding target monitoring area under different color deviation values ​​and different color change frequencies; The uniformity deviation value of the target monitoring area is obtained by matching the comprehensive color difference factor with the reference deviation database.

5. The online control method for fabric dyeing uniformity based on machine vision according to claim 1, characterized in that, Based on the uniformity deviation value, an adjustment warning notification message is output, specifically: If the uniformity deviation value is greater than or equal to the preset dyeing uniformity warning threshold, an adjustment warning notification message will be output.

6. A machine vision-based online control system for fabric dyeing uniformity, applied to the machine vision-based online control method for fabric dyeing uniformity as described in any one of claims 1 to 5, characterized in that, include: Acquisition module: The image acquisition location points are obtained by visually inspecting the dyeing area of ​​the target fabric in online production and marking the image acquisition location points. The distance parameter between two adjacent image acquisition location points is obtained to obtain the distance to be measured. Extraction Module: Extracts historical dyeing concentration distribution characteristics of fabrics from historical fabric dyeing monitoring data. Based on these characteristics, it calculates the color difference attenuation correlation factor between color deviation values ​​and the measured distance in the historical fabric dyeing monitoring data. This includes the following steps: Based on the historical color concentration distribution characteristics, the historical color deviation values ​​of the image acquisition locations are extracted. The deviation difference is obtained by subtracting the historical color deviation values ​​of two adjacent image acquisition points based on the distance to be measured. The ratio of the deviation difference to the distance to be measured is used to obtain the deviation ratio. The color difference attenuation correlation factor is obtained by averaging all the measured deviation ratios. The first processing module extracts preprocessing correlation factors from the color difference attenuation correlation factors based on the current dyeing concentration distribution characteristics of the target fabric. Based on these preprocessing correlation factors, it determines the color deviation status of the target fabric's monitoring areas prone to uneven dyeing, obtaining the first color deviation value and the second color deviation value. Specifically, this includes the following steps: The target monitoring point is obtained by monitoring the location of the target fabric in the target monitoring area. The image acquisition position points are marked on the left and right edges of the target fabric in the width direction to obtain the first acquisition point and the second acquisition point. The first preprocessing spacing is obtained by calculating the distance between the target monitoring point and the first collection point; The first color deviation value of the target monitoring point is obtained based on the first preprocessing interval, the ratio of the deviation to be measured, and the preprocessing correlation factor. The second preprocessing interval is obtained by calculating the distance between the target monitoring point and the second collection point. The second color deviation value of the target monitoring point is obtained based on the second preprocessing interval, the ratio of the deviation to be measured, and the preprocessing correlation factor. The second processing module obtains a comprehensive color difference factor based on the first color deviation value and the second color deviation value, and judges the dyeing uniformity of the target monitoring area affected by the dyeing concentration based on the comprehensive color difference factor to obtain the uniformity deviation value. Control module: Outputs adjustment warning notification information based on uniformity deviation value, and extracts optimal adjustment parameters from dyeing process optimization parameters based on the adjustment warning notification information, specifically including the following steps: After generating the dyeing process optimization parameters for the target fabric based on the adjustment warning notification information, the candidate optimization parameters are output. The degree of improvement in staining effect corresponding to each candidate optimization parameter is detected to obtain parameter optimization feature data; The baseline process characteristic data is obtained by detecting the baseline value of the dyeing effect to which the current dyeing process parameters belong; The optimal adjustment parameters are obtained by extracting the parameter optimization feature data of the candidate optimization parameters from the baseline process feature data.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the online control method for fabric dyeing uniformity based on machine vision as described in any one of claims 1 to 5.

Images