KR19990083893A - Automatic Evaluation System of Fabric Wrinkles and Seam Puckers and Their Method - Google Patents

Abstract

The present invention seeks to establish a new standard for automating the wrinkling of textile products and the quality assessment of simperkers and for ranking the properties of the fabrics. In addition, the present invention relates to an automated apparatus and method for easily and accurately measuring various factors of textile products.

Factors that determine the quality and merchandise of textile products have a strong emotional aspect, and are mainly compared to people's senses. As it is evaluated by very uncomfortable test method, it is difficult to obtain consistent results by subject's subjectivity and emotion. Therefore, new standardized test methods are needed.

To this end, the present invention obtains the information on the surface of the fabric through the development of an image processing technique and an algorithm for obtaining three-dimensional information of the surface of the fabric after obtaining information on the surface of the fabric with a CCD camera. The obtained data is an invention that grades the wrinkles and simperkers of the fabric and automatically evaluates the quality of the wrinkles and the simperkers through a fractal dimension algorithm that can grade the roughness of the fabric surface.

Description

Automatic Evaluation System of Fabric Wrinkles and Seam Puckers and Their Method}

The present invention relates to an apparatus and method for automatically evaluating wrinkles and seam puckers of fabrics. In particular, the present invention relates to an apparatus and method for automatic evaluation of fabric wrinkles and simperkers, which are composed of CCD cameras, image processing for analyzing images obtained from cameras, and algorithms for objectively assigning grades.

In general, the wrinkling of fabrics and the evaluation of the symperker depend mainly on the senses and subjectivity of the person. For example, five plastic plates (AATCC Standard Wrinkle Replica Grade 1-5), which have different surface wrinkles (unevenness) as shown in FIG. It is given as an integer value. The method of Figure 2 is a method that can give a piece of information about the fabric wrinkles Monsanto (Monsanto) method of measuring the angle recovered by removing the load after pressing a certain load for a certain time by folding the fabric specimen.

Both methods are difficult to overcome limitations such as subjectivity, low reliability, and observer expertise, which are disadvantages of visual fabric evaluation by humans. When evaluating wrinkles with AATCC replicas, there are disadvantages in that the wrinkle difference between grades is not constant.

And because the grade is expressed in the integer dimension, it does not provide a detailed explanation of wrinkles.

Fabric evaluation of the above two methods is subjective and the problem of data reliability continues to arise.

In order to solve the above problems, the present invention obtains image information about wrinkles of a fabric by using a commercially available CCD camera in an optical non-contact three-dimensional manner, and then image-processes the 3D information of the fabric required in the image. And an automatic evaluation apparatus and method for fabric wrinkle and simperker by applying a fractal algorithm from three-dimensional information of a fabric obtained by the development of an optical algorithm and obtained from an image.

In order to achieve the above object, the present invention, in obtaining and grading the three-dimensional information of the fabric sample to be evaluated and analyzed, the image input means for sending an electrical signal of the surface information of the fabric sample to the computer, the laser to the fabric sample Laser light irradiation means for forming a profile, transferring the fabric at regular intervals by horizontal conveying means to obtain the entire information of the fabric sample, halogen light required to obtain an image of the correction plate when calibrating the CCD camera as an image input means Irradiation means, darkroom means necessary for clearly obtaining a profile of laser light irradiation with a CCD camera, computer means for converting an electrical signal from the CCD camera into a digital signal and storing and processing the digital signal, after which Automatically take all steps to obtain and grade three-dimensional information It is characterized by consisting of an integrated device and method that can be processed by.

1 is a diagram showing images of grades 1-5 of the American Association of Textile Chemists and Colorists (hereinafter referred to as AATCC) Standard Wrinkle Replica

2 is a view showing a Monsanto recovery angle measuring device

3 is a conceptual diagram of an optical three-dimensional non-contact measurement system using a laser

4 shows a laser profile on a monitor according to camera position.

5 is a view showing a black and white image of a correction plate

6 shows a camera correction means using a correction plate and a laser;

Fig. 7 is a diagram showing the relationship between the plane made by the laser and the straight line of the optical axis of the camera.

8 is a flow chart of a method for automatic evaluation of fabric wrinkles and simperkers

<Description of the symbols for the main parts of the drawings>

1: Pentium 2 personal computer 2, 10: monochrome CCD camera

3: slit light laser 4: driving aluminum plate to place textile sample

5: square metal box 6: subject fabric

7: general halogen lamp 8, 9: monitor of personal computer

11: correction plate 12: rectangular grid on the correction plate

13: height to move the correction plate 14: image coordinate system

15: Straight line formed by camera optical axis 16: Plane made by laser

Hereinafter, described in detail by the accompanying drawings as follows.

FIG. 3 is a structural diagram showing an automatic fabric wrinkle and simperker evaluation device and a method thereof, wherein a slit laser of red wavelength (Slit Laser) 3 is irradiated on the vertical of the fabric 6 to be measured and the CCD camera 2 is lasered. Maintain a constant angle with incident light and obtain an image. The reason for maintaining the above-mentioned constant angle is that as shown in the figure of FIG. 4, by maintaining the constant angle between the CCD camera and the laser line, information on the unevenness of the fabric surface can be obtained. In other words, it can be seen that the three-dimensional information of the subject is well represented on the monitor screen 8 representing the electrical signal by the CCD camera maintained at a constant angle than the monitor screen 9 by the CCD camera on the vertical plane. Since the image of the object is obtained by tilting the CCD camera at an angle to obtain three-dimensional information of the subject, as shown in FIG. 5, perspective is clearly generated, and the camera lens itself has a combination of physical aberration and distortion. The coupling is eliminated through the calibration method of the CCD camera before the three-dimensional information of the fabric sample is obtained. The process of the calibration method is as follows.

As shown in FIG. 6, an image of a compensating plate 11 having dots marked in a constant rectangular grid on a table for transferring fabric samples is obtained through a CCD camera. At this time, the actual absolute coordinate values (white points in FIG. 5) of the square grid point of the correction plate are already known, and image coordinate values 14 corresponding to these points are found through image processing, and the corresponding relations of the same points are set. Thus, all coordinate values of the image can be converted into actual coordinate values of the correction plate. The expression that sets the correspondence of coordinate values can be expressed as follows.

X = HY

Where X is the plane absolute coordinate value of the compensator, Y is the plane image coordinate value of the image, and H is the conversion constant value. The H value can be obtained by knowing the coordinate values of the four points of the correction plate and the four corresponding coordinate values of the image. When the H value is obtained, any image coordinate value on the image can find an absolute coordinate value corresponding to the correction plate. By correcting each constant H value in all the rectangular grids in FIG. Then, after moving the correction plate by a certain height (h, 13) as shown in FIG. After calculating H _{z = 0} and H _{z = h} through calculation, laser lines 15 are formed on the planes of Z = 0 and Z = h, respectively. The absolute position coordinates in space of the formed line profile can be obtained by applying the H value obtained above. Through this method, the equation (16) of the plane formed by the laser line can be obtained as shown in FIG. That is, since the laser lines are linear, the laser lines formed on the Z = 0 correction plate and the laser lines formed at the Z = h points are lines on one plane. Therefore, the spatial position coordinate of each line formed in each image is obtained through the conversion constant H, and then the equation of the plane consisting of two lines is obtained.

After obtaining the H conversion constant value and the equation of the plane, if any object to be measured is placed in the position where the laser line passes, the laser profile is generated at the point O as shown in FIG. This profile is imaged by the camera and as shown in Fig. 7, the laser profile O point can obtain points O1 and O2 of each correction plate which meets the camera optical axis, so that a straight line equation can be derived. Coordinate values in space can be obtained through the point where the equation of the plane and the equation of the straight line meet. This algorithm is much more precise than conventional methods and can be applied to the non-contact metrology industry, which requires high precision, because it can perform the calibration process easily and automatically in a short time.

In order to obtain the overall surface information of the fabric sample to be measured, each image is obtained by passing the fabric sample through the CCD camera and the laser irradiation device by the transfer means 4. That is, since only the height information of the subject where the laser line passes can be obtained, the measurement sample is moved by the transfer means at regular intervals. As a result, height information can be obtained for a certain size of fabric sample.

The obtained height information is programmed to obtain fast, stable and reliable results by incorporating fractal order, a technique for analyzing irregular shapes and behaviors of natural systems, into the quality evaluation of textile products. The algorithm has been developed and applied to the three-dimensional shape data of the fabric surface in addition to the box counting method (Cube Counting method) can also be applied to practical objects such as textile products. Flowchart of the method)

As described above, the present invention can establish a new standard by automating the evaluation of the quality of a textile product using an image processing technique, and it is effective to measure various factors of the textile product easily and accurately even by an inexperienced person.

In the present invention, since the quality evaluation of the existing textile products is made by visual evaluation by humans, it overcomes the disadvantage of not being able to give reliability to the quality of the textile products and ensures reliability and cost savings by automatically testing without depending on the measurer. It can work. In addition, by managing the data of the product can be expected to improve the quality of the textile products.

In addition, 3D measurement of textile products can be performed using optical correction method using CCD camera using image processing technique in textile industry. The image processing and analysis technology applied to the invention may be applicable to various other industries. Image processing and analysis technology is an advanced technology that is expected to contribute to the development of the hardware and software industry.

Claims (1)

In obtaining and grading the three-dimensional information of the fabric sample to be evaluated and analyzed, Image input means for sending an electrical signal of surface information of the fabric sample to a computer, Laser light irradiation means for forming a laser profile on a fabric sample, Transporting the fabric at regular intervals by horizontal conveying means to obtain complete information of the fabric sample, Halogen light irradiation means necessary to obtain an image of a correction plate when correcting a CCD camera as an image input means, Darkroom means necessary for clearly obtaining a profile of laser light irradiation with a CCD camera, Computer means for converting an electrical signal from a CCD camera into a digital signal and storing and processing the digital signal, An integrated apparatus and method capable of automatically processing all steps for obtaining and grading three-dimensional information of a fabric sample after said step.