阅读说明：本技术 一种卷装长丝的尾丝缺陷检测方法 (Tail yarn defect detection method for package filament ) 是由 杨崇倡 肖凌云 冯培 张荣根 宋洪征 于 2018-08-24 设计创作，主要内容包括：本发明公开了一种卷装长丝的尾丝缺陷检测方法。该尾丝缺陷检测方法包括：获取卷装长丝的待检测区域的多张采集图像；截取所述采集图像中与所述待检测区域内的受检位置对应的目标图像；过滤目标图像中的连续直线以外的背景图案,获得待分析图像；确定连续直线的直线几何参数和待分析图像的图像几何参数；根据直线几何参数和图像几何参数确定待分析图像中是否存在尾丝缺陷。本发明实施例的卷装长丝的尾丝缺陷检测方法,能够降低人工目测的错误率,提高检测效率,且能够节约生产成本。(The invention discloses a tail fiber defect detection method of a package filament. The fiber defect detection method comprises the following steps: acquiring a plurality of collected images of a to-be-detected area of the coiled filament; intercepting a target image corresponding to a detected position in the region to be detected in the collected image; filtering background patterns except for continuous straight lines in the target image to obtain an image to be analyzed; determining the linear geometric parameters of the continuous straight lines and the image geometric parameters of the image to be analyzed; and determining whether the fiber defect exists in the image to be analyzed according to the linear geometric parameters and the image geometric parameters. The method for detecting the tail yarn defect of the package filament yarn can reduce the error rate of manual visual inspection, improve the detection efficiency and save the production cost.)

1. A method for detecting a tail yarn defect of a package filament is characterized by comprising the following steps:

acquiring a plurality of collected images of the to-be-detected area of the coiled filament;

intercepting a target image corresponding to a detected position in the region to be detected in the collected image;

filtering background patterns except for continuous straight lines in the target image to obtain an image to be analyzed;

determining the linear geometric parameters of the continuous straight line and the image geometric parameters of the image to be analyzed;

and determining whether the fiber defect exists in the image to be analyzed according to the linear geometric parameters and the image geometric parameters.

2. The method of detecting a fiber defect in a wound filament of claim 1, wherein determining whether a fiber defect is present in said image to be analyzed further comprises:

and when the tail yarn defect exists in the image to be analyzed corresponding to at least one acquired image, determining that the tail yarn defect exists in the wound filament yarn.

3. The method of claim 1, wherein the inspection site comprises a first sidewall at an upper end of an ingot of the packaged filament and a second sidewall at a lower end of the ingot, and acquiring the plurality of captured images of the area of the packaged filament to be inspected comprises:

acquiring a plurality of collected images including the first sidewall and the second sidewall along the circumferential direction of the wound filament, respectively.

4. The method of claim 1, wherein capturing the target image corresponding to the inspected location within the area to be inspected in the captured image comprises:

acquiring a copy image of the acquired image;

denoising the duplicate image, and acquiring a plurality of edges in the denoised duplicate image by using an edge detection algorithm;

determining a first edge line and a second edge line corresponding to the detected position in the plurality of edge lines;

obtaining coordinate values of all pixel points of the first edge line and the second edge line;

and intercepting the target image in the acquired image according to the coordinate values of all pixel points of the first edge line and the second edge line.

5. The method of detecting a defect in a wound filament of claim 1, further comprising, before filtering the background pattern other than the continuous straight line in the target image:

and denoising the target image.

6. The method of detecting defects in a wound filament of claim 5 in which filtering background patterns other than continuous lines in said target image to obtain an image to be analyzed comprises:

preliminarily filtering the background of the target image by using a binarization method;

and accurately filtering the background of the primarily filtered target image by using an expansion processing method to obtain the image to be analyzed.

7. A method of detecting a defect in a wound filament tail in accordance with claim 6 wherein initially filtering the background of said target image using a binarization process comprises:

processing the target image by using an edge detection algorithm to obtain a gray level change image;

carrying out binarization processing on the gray level change image to obtain a binarization image;

and preliminarily filtering the first pattern in the binary image according to a first preset length and a first preset area.

8. The method of detecting defects in a wound filament tail according to claim 7 wherein the step of accurately filtering the background of the preliminary filtered target image using a dilation process comprises:

connecting every two adjacent second patterns of the preliminarily filtered binary image, the distance between which is smaller than a preset distance, by using an expansion processing method to obtain the continuous straight line;

and accurately filtering a third pattern except the continuous straight line in the expanded binary image according to a second preset length and a second preset area to obtain the image to be analyzed.

9. The method of claim 1, wherein the geometric parameters of said continuous line include a length of said continuous line, a width of said continuous line, and an angle between said continuous line and a horizontal direction, and wherein the geometric parameters of said image to be analyzed include an image length and an image width of said image to be analyzed.

10. The method of claim 8, wherein determining whether a fiber defect is present in the image to be analyzed based on the line geometry and the image geometry comprises:

determining a first ratio of the straight line length to the image length and a second ratio of the straight line width to the image width;

and determining whether the fiber defects exist in the image to be analyzed according to the number of continuous straight lines of which the first ratio accords with a first preset threshold, the second ratio accords with a second preset threshold and the included angle accords with a third preset threshold.

Technical Field

The invention relates to the technical field of surface detection of packaged filaments, in particular to a tail filament defect detection method of packaged filaments.

Background

The package filament is a package finished product with certain shape and capacity which is made by a winding mechanism in the production process of the filament. The defects of the wound filament are mainly divided into two parts of physical and chemical property defects of the filament and appearance defects of the wound filament. It has been found through practice that appearance defects in the wound filaments have a significant effect on the quality of the fabric, resulting in a reduction in fabric yield. Therefore, in the production process, the detection of appearance defects of the wound filaments is intensified.

The tail defects of the packaged filament are divided into tail-free yarns, tail-rich yarns and top tail yarns, and the tail-free yarns and the tail-rich yarns are all caused by improper operation, no tail yarn joint or more than two joints, and generally appear at the lower end of a yarn spindle of the packaged filament. The upper tail is generated due to a malfunction of the apparatus at the time of winding, and generally occurs at the upper end of the package of the filament.

Because the coiled filament has the characteristics of deformability, multiple curved surfaces and large detected surface, the appearance defect characteristics of the coiled filament are difficult to extract unified standards, the appearance detection of the coiled filament always influences the realization of the intelligent production of a long production flow, and the tail filament defect detection can only be carried out by a manual visual inspection method.

Although some defects can be detected by detecting the fiber defects through manual visual inspection, the manual visual inspection method increases the labor cost of production, and does not have a uniform and strict quality standard, so that the accuracy of the detection result is difficult to ensure.

Disclosure of Invention

In order to solve the above problems, embodiments of the present invention provide a method for detecting a tail yarn defect of a package filament, which can reduce an error rate of manual visual inspection, improve detection efficiency, and save production cost.

In order to achieve the above object, an embodiment of the present invention provides a method for detecting a tail yarn defect of a package filament, including:

acquiring a plurality of collected images of a to-be-detected area of the coiled filament;

intercepting a target image corresponding to a detected position in the region to be detected in the collected image;

filtering background patterns except for continuous straight lines in the target image to obtain an image to be analyzed;

determining the linear geometric parameters of the continuous straight lines and the image geometric parameters of the image to be analyzed;

and determining whether the fiber defect exists in the image to be analyzed according to the linear geometric parameters and the image geometric parameters.

Further, after determining whether the fiber defect exists in the image to be analyzed, the method further comprises the following steps:

and when the tail fiber defect exists in the image to be analyzed corresponding to the at least one collected image, determining that the tail fiber defect exists in the package filament.

Further, the detected position includes a first side wall at the upper end of the spindle for winding the filament and a second side wall at the lower end of the spindle, and acquiring a plurality of collected images of the to-be-detected area for winding the filament includes:

a plurality of collected images including a first sidewall and a second sidewall are acquired along the circumferential direction of the wound filament, respectively.

Further, intercepting a target image corresponding to the detected position in the region to be detected in the collected image comprises:

acquiring a copy image of the acquired image;

denoising the duplicate image, and acquiring a plurality of edge lines in the denoised duplicate image by using an edge detection algorithm;

determining a first edge line and a second edge line corresponding to the detected position in the plurality of edge lines;

obtaining coordinate values of all pixel points of the first edge line and the second edge line;

and intercepting the target image in the collected image according to the coordinate values of all pixel points of the first edge line and the second edge line.

Further, before filtering the background patterns other than the continuous straight lines in the target image, the method further includes:

and denoising the target image.

Further, filtering background patterns except for continuous straight lines in the target image to obtain an image to be analyzed, wherein the method comprises the following steps:

preliminarily filtering the background of the target image by using a binarization method;

and accurately filtering the background of the primarily filtered target image by using an expansion processing method to obtain an image to be analyzed.

Further, the preliminary filtering of the background of the target image by using a binarization method comprises:

processing the target image by using an edge detection algorithm to obtain a gray level change image;

carrying out binarization processing on the gray level change image to obtain a binarization image;

and preliminarily filtering the first pattern in the binary image according to the first preset length and the first preset area.

Further, the accurately filtering the background of the preliminarily filtered target image by using the dilation method comprises:

connecting every two adjacent second patterns of the preliminarily filtered binary image with a distance interval smaller than a preset distance by using an expansion processing method to obtain a continuous straight line;

and accurately filtering third patterns except for the continuous straight lines in the expanded binary image according to a second preset length and a second preset area to obtain an image to be analyzed.

Further, the linear geometric parameters of the continuous straight line include the linear length, the linear width and the included angle between the continuous straight line and the horizontal direction, and the image geometric parameters of the image to be analyzed include the image length and the image width of the image to be analyzed.

Further, determining whether the fiber defect exists in the image to be analyzed according to the linear geometric parameters and the image geometric parameters comprises the following steps:

determining a first ratio of the length of the straight line to the length of the image and a second ratio of the width of the straight line to the width of the image;

and determining whether the fiber defects exist in the image to be analyzed according to the number of continuous straight lines of which the first ratio accords with a first preset threshold, the second ratio accords with a second preset threshold, and the included angle accords with a third preset threshold.

The method for detecting the tail yarn defect of the packaged filament yarn can be applied to a detection device on a production line, image acquisition is carried out on the packaged filament yarn passing through each detection station, a corresponding acquired image is obtained, then the image is processed, defect characteristics (namely the linear geometric parameters of continuous straight lines and the image geometric parameters of images to be analyzed) for judging the tail yarn defect are extracted, and whether the tail yarn defect exists in the packaged filament yarn is determined according to the properties of the defect characteristics of different tail yarn defects.

The method for detecting the tail yarn defect of the package filament yarn is suitable for detecting the tail yarn defect of the package filament yarn in the chemical fiber field, can quickly analyze the appearance defect of the package filament yarn, reduces the error of manual visual inspection, reduces the false detection rate, improves the detection efficiency and reduces the production cost through high-precision analysis and processing of images.

Drawings

FIG. 1 is a flow chart of a method of detecting a tail defect in a package filament according to one embodiment of the present invention;

FIG. 2 is a flow chart of a method of detecting a tail defect in a wound package filament according to another embodiment of the present invention;

FIG. 3 is a schematic illustration of the structure of a package filament collected according to an embodiment of the present invention;

FIG. 4 is a flowchart illustrating a specific method of step S120 in FIG. 1;

FIG. 5 is a flowchart of a detailed method of one embodiment of step S130 in FIG. 1;

FIG. 6 is a flowchart illustrating another embodiment of the specific method of step S130 in FIG. 1;

fig. 7 is a flowchart illustrating a specific method of step S150 in fig. 1.

Detailed Description

The structure, operation principle, and the like of the embodiments of the present invention will be further described with reference to the accompanying drawings.

As shown in fig. 1, a method for detecting a tail yarn defect of a package filament according to an embodiment of the present invention includes:

s110, acquiring a plurality of collected images of a to-be-detected area of the coiled filament;

s120, intercepting a target image corresponding to the detected position in the region to be detected in the collected image;

s130, filtering background patterns except for continuous straight lines in the target image to obtain an image to be analyzed;

s140, determining linear geometric parameters of continuous straight lines and image geometric parameters of an image to be analyzed;

s150, determining whether the fiber defect exists in the image to be analyzed according to the linear geometric parameters and the image geometric parameters.

In the embodiment of the present invention, as shown in fig. 2, after determining whether there is a fiber defect in the image to be analyzed, the method further includes:

and S160, determining that the package filament has the tail yarn defect when the tail yarn defect exists in the image to be analyzed corresponding to the at least one acquired image.

In the embodiment of the present invention, as shown in fig. 3, the detected position includes a first side wall of an upper end 1 of the package filament and a second side wall of a lower end 2 of the package filament, and acquiring a plurality of collected images of the area to be detected of the package filament includes:

a plurality of collected images including a first sidewall and a second sidewall are acquired along the circumferential direction of the wound filament, respectively.

In the embodiment of the invention, the area to be detected of the coiled filament, namely the corresponding camera and the light source, can be selected according to the characteristics of the defect of the tail fiber so as to collect images.

First, the wound filament can be divided into two regions with filaments including the upper end 1 and the lower end 2 of the spindle as regions to be detected, according to the position where the defect of the tail filament may occur. Then, an area array camera is respectively arranged at the positions corresponding to the upper end 1 and the lower end 2 of the spindle and facing the packaged filament, so as to respectively acquire images of the area to be detected of the upper end 1 of the spindle and the area to be detected of the lower end 2 of the spindle as acquired images. The area-array camera can reduce the shooting times and save time. And according to the characteristics of fixed position and limited shooting angle of the area-array camera, the coiled filament can be rotated, and collected images of the area to be detected at the upper end 1 of the filament spindle and the area to be detected at the lower end 2 of the filament spindle are collected while the coiled filament is rotated, so that the image collection of the whole circumference of the coiled filament is completed. And finally, continuously shooting the area to be detected at the upper end 1 of the ingot and the area to be detected at the lower end 2 of the ingot by the two area-array cameras according to the received control signals respectively, and obtaining a plurality of collected images. In one embodiment, the photographing time may be 2.2s, and the number of the collected images of the region to be detected of the upper end 1 of the ingot and the region to be detected of the lower end 2 of the ingot may be 16 pieces, respectively.

Because the defect of the tail yarn with the flaw characteristic higher than that of the surface of the detected object can be highlighted by using low-angle illumination, in the embodiment of the invention, the upper end 1 and the lower end 2 of the filament spindle are respectively illuminated by using low-angle illumination light sources, and the light of the light sources is vertical to the area to be detected of the wound filament yarn.

In the embodiment of the present invention, as shown in fig. 4, intercepting a target image corresponding to a detected position in the to-be-detected region in the collected image includes:

s121, acquiring a copy image of the acquired image;

s122, denoising the duplicate image, and acquiring a plurality of edge lines in the denoised duplicate image by using an edge detection algorithm;

s123, determining a first edge line and a second edge line corresponding to the detected position in the plurality of edge lines;

s124, obtaining coordinate values of all pixel points of the first edge line and the second edge line;

and S125, intercepting the target image in the collected image according to the coordinate values of all the pixel points of the first edge line and the second edge line.

In the embodiment of the invention, the interference of a non-processing object in the image can be eliminated by intercepting the target image in the acquired image. By adopting the steps, the situation that the detected position has deviation or the detected position ranges are different can be processed, and the image capturing precision is improved. Wherein, the detected position is the position without silk thread at the upper end and the lower end of the silk ingot.

In the embodiment of the invention, the acquired image can be copied firstly, the copied image can be obtained, and the first side line and the second side line can be obtained by processing the copied image, so that the influence on the subsequent processing of the original acquired image caused by processing the original acquired image can be prevented.

Then, denoising processing may be performed on the copied image through a filtering method, and in this embodiment, the filtering method may preferably be gaussian filtering. The denoised copy image can obtain a plurality of edge lines through an edge detection algorithm. Wherein, the sideline is the position of the gray scale light and dark mutation of the edge of the package filament.

Secondly, an edge line closest to the upper edge of the image in the copied image can be selected as a first edge line, an edge line closest to the bottom edge of the image in the copied image can be selected as a second edge line, and coordinate values of all pixel points of the first edge line and the second edge line in the copied image are extracted.

And thirdly, the coordinate values of all the pixel points of the first sideline and the second sideline in the copied image are brought into the collected image, the same pixel point coordinate value extraction is carried out on the collected image, so that the pixel point coordinate values are utilized to obtain the segmentation sidelines of the collected image, namely the first sideline and the second sideline corresponding to the collected image. And finally, intercepting the image in the range between the first edge line and the second edge line to obtain a target image.

In the embodiment of the present invention, before filtering the background patterns other than the continuous straight lines in the target image, the method further includes:

and denoising the target image.

In the embodiment of the present invention, the noise point in the target image may be analyzed first, and then the target image may be filtered according to the existing noise point. Since the noise points of the target image are concentrated in a high-frequency region, the filtering process may employ a median filtering method. The median filtering method can make the surrounding pixels close to the true value to obtain the image with the isolated noise points eliminated, thereby improving the deterioration of the imaging quality of the image caused by various interference factors in the process of acquiring the original signal, transmitting the signal, and transmitting the signal by the area-array camera.

In the embodiment of the present invention, filtering background patterns other than continuous straight lines in a target image to obtain an image to be analyzed includes:

preliminarily filtering the background of the target image by using a binarization method;

and accurately filtering the background of the primarily filtered target image by using an expansion processing method to obtain an image to be analyzed.

In the embodiment of the present invention, as shown in fig. 5, the preliminary filtering of the background of the target image by using the binarization method includes:

s131, processing the target image by using an edge detection algorithm to obtain a gray level change image;

s132, carrying out binarization processing on the gray level change image to obtain a binarized image;

and S133, preliminarily filtering the first pattern in the binary image according to the first preset length and the first preset area.

In the embodiment of the invention, the target image after denoising can be analyzed for gray level change through an edge detection algorithm to obtain a gray level number image. The edge detection processing can be performed by selecting a Laplace operator and a logarithm type algorithm or a Roberts operator and a logarithm type algorithm, so that the interference of different patterned paper tubes can be weakened, the original details of a target image are not lost, and the effect of the obtained gray level change image is optimal.

Then, the gray level change image may be subjected to binarization processing to obtain a binarized image having only two colors. Specifically, the adaptive segmentation threshold may be selected in the binarization processing according to the different color backgrounds of the wound paper tube of the wound filament. The self-adaptive segmentation threshold value can be based on the characteristic that the background of a rolled paper tube has diversification, such as a red original point background and a dark green square background, and the target false extraction caused by the fixed threshold value to an image is eliminated.

In the embodiment of the present invention, the adaptive segmentation threshold is preferably calculated by using an OTSU algorithm, which can calculate the gray level variance of the foreground of the wound filament and the background of the wound paper tube, and the weighted minimum sum of the variances is the optimal threshold for different backgrounds.

Finally, the preliminary process can be carried out according to a preset first preset length and a preset first areaAnd filtering the first pattern in the binary image to obtain an image with partial pattern interference removed. In one embodiment, the first predetermined area can be set to be less than or equal to 0.03mm²The first preset length is less than or equal to 1 mm.

In the embodiment of the present invention, as shown in fig. 6, the accurately filtering the background of the preliminarily filtered target image by using the dilation method includes:

s134, connecting every two adjacent second patterns of the preliminarily filtered binary image with the distance interval smaller than the preset distance by using an expansion processing method to obtain a continuous straight line;

and S135, accurately filtering third patterns except for the continuous straight lines in the expanded binary image according to a second preset length and a second preset area to obtain an image to be analyzed.

In the embodiment of the invention, the expansion processing method is utilized to connect the second patterns to obtain continuous straight lines, so that unnecessary interference on the preliminarily filtered binary image can be avoided.

In the embodiment of the present invention, the second predetermined area may be set to be less than or equal to 0.05mm²And the second preset length is less than or equal to 2mm, and the image to be analyzed with pattern interference completely removed is obtained.

In the embodiment of the invention, the linear geometric parameters of the continuous straight line comprise the linear length and the linear width of the continuous straight line and the included angle between the continuous straight line and the horizontal direction, and the image geometric parameters of the image to be analyzed comprise the image length and the image width of the image to be analyzed.

As shown in fig. 7, in the embodiment of the present invention, determining whether a fiber defect exists in the image to be analyzed according to the geometric parameters of the straight line and the geometric parameters of the image includes:

s151, determining a first ratio of the length of the straight line to the length of the image and a second ratio of the width of the straight line to the width of the image;

s152, determining whether the fiber defects exist in the image to be analyzed according to the number of continuous straight lines of which the first ratio accords with a first preset threshold, the second ratio accords with a second preset threshold, and the included angle accords with a third preset threshold.

In the embodiment of the present invention, the criterion for determining whether there is a fiber defect in the image to be analyzed can be divided into two types according to the detected position corresponding to the image to be analyzed:

first one

When the detected position corresponding to the image to be analyzed is the upper end 1 of the wire ingot, when no first ratio in the image to be analyzed accords with a first preset threshold value, the second ratio accords with a second preset threshold value, and the included angle accords with a continuous straight line of a third preset threshold value, no tail wire defect exists in the image to be analyzed. When one or more than one continuous straight line with the first ratio meeting a first preset threshold, the second ratio meeting a second preset threshold and the included angle meeting a third preset threshold exists in the image to be analyzed, the fiber defect exists in the image to be analyzed, and the fiber defect is the multi-fiber defect. In one embodiment, the first preset threshold may be set to be greater than or equal to 0.8, the second preset threshold may be greater than or equal to 0.4, and the third preset threshold may be set to be 5 ° to 20 °.

In the embodiment of the invention, when the detected position is a plurality of collected images corresponding to the upper end 1 of the spindle, if one or more images to be analyzed corresponding to all the collected images have a plurality of tail fiber defects, the upper end 1 of the spindle is judged to have the tail fiber defects; if all the images to be analyzed do not have tail fiber defects, judging that the tail fibers at the upper end 1 of the silk ingot are normal.

Second kind

When the detected position corresponding to the image to be analyzed is the lower end 2 of the wire ingot, when a first ratio in the image to be analyzed meets a first preset threshold value, a second ratio in the image to be analyzed meets a second preset threshold value, and an included angle in the image to be analyzed meets a continuous straight line of a third preset threshold value, the image to be analyzed does not have tail wire defects. And when the image to be analyzed does not have or has more than one continuous straight line with the first ratio meeting the first preset threshold, the second ratio meeting the second preset threshold and the included angle meeting the third preset threshold, the tail fiber defect exists in the image to be analyzed, namely the multi-tail fiber defect exists. Wherein, when the continuous straight line which does not accord with the setting is not provided, the defect of the tail fiber is avoided, and when more than one continuous straight line which accords with the setting is provided, the defect of the multi-tail fiber is avoided.

In one embodiment, the first preset threshold may be set to be greater than or equal to 0.8, the second preset threshold may be greater than or equal to 0.4, and the third preset threshold may be set to be 5-20 °

In the embodiment of the invention, when the detected position is a plurality of collected images corresponding to the lower end 2 of the silk ingot, if one or more images to be analyzed corresponding to all the collected images have a plurality of tail defects, the lower end 2 of the silk ingot is judged to have the tail defects; if all the images to be analyzed have no tail fiber defect, judging that the tail fiber defect exists at the lower end 2 of the spindle; if at least one image to be analyzed has no tail fiber defect, and other images have no tail fiber defect, judging that the tail fibers at the lower end 2 of the spindle are normal.

In the embodiment of the invention, if the tail yarn defects appear at the lower end 2 of the spindle and the lower end 2 of the spindle, the whole packaged filament is determined to be normal; if at least one of the lower end 2 of the filament ingot and the lower end 2 of the filament ingot has a tail filament defect, judging that the tail filament defect exists at the lower end 2 of the whole filament ingot and the lower end 2 of the filament ingot.

According to the tail yarn defect detection method of the packaged filament, the defect rate of correctly detected tail yarn can reach 98%, the efficiency is 4 seconds per ingot, the method is superior to a manual visual inspection method, the online automatic detection of the appearance quality of the packaged filament is realized, and the manual workload is reduced by 95%.

In conclusion, the method for detecting the tail yarn defect of the packaged filament yarn can be widely applied to online detection of the tail yarn defect of the packaged filament yarn in the chemical fiber field, and is easy to analyze the appearance defect. And the fiber defects are detected by a method for extracting and judging the fiber defect characteristics, so that the detection precision can be improved, and the errors caused by manual visual inspection can be reduced. Meanwhile, in the image processing process, interference can be eliminated, the fiber defects can be rapidly judged, and the false detection rate is reduced.

The foregoing is merely illustrative of the present invention, and it will be appreciated by those skilled in the art that various modifications may be made without departing from the principles of the invention, and the scope of the invention is to be determined accordingly.