CN102495063A - Method for evaluating washing wear of fabric on basis of machine vision - Google Patents

Abstract

The invention relates to a method for evaluating fabric washing wear based on machine vision, which realizes fine and objective evaluation of washing wear in the fabric washing process. The evaluation method of fabric washing wear based on machine vision will be realized through the cooperation of hardware and software. The hardware should be able to obtain high-precision images of worn cloth samples, and the software should be able to acquire, save, crop, and segment images. And the calculation of washing wear evaluation parameters. In this method, photographs are taken of the worn cloth before and after washing, the area of the worn cloth in the image of the worn cloth sample before and after washing is calculated, and the area change rate is calculated, that is, the wear value is used as a parameter to evaluate the severity of washing wear. The method overcomes the error caused by measuring the size of the worn cloth sample by a manual method and the difficult problem of being unable to accurately calculate the area when the worn cloth is irregular in shape after washing.

Description

A method for evaluating fabric washing wear based on machine vision

technical field

The invention relates to a method for evaluating fabric washing wear based on machine vision, which can be used to accurately calculate "AS/NZS2040.1: 2005 Performance of household electrical appliances-Clothes washing machines Part 1: Methods forming performance, energy and water consumption" The area of the irregularly shaped fabric wear cloth specified in Appendix G after washing, and then calculate the washing wear evaluation index washing wear value. Compared with the manual measurement method commonly used by washing machine manufacturers, the method proposed by the invention is more objective and accurate, and can effectively reduce test errors.

Background technique

Fabric washing wear evaluation is a performance evaluation method for measuring the strength of the washing machine's cleaning effect on fabrics through an accelerated wear test. The wear value of the evaluation index can reflect the wear of the fabric during the washing process. Therefore, the evaluation of fabric washing wear is an indispensable test index for washing machine manufacturers when evaluating the performance of washing machines. The existing test method is to calculate the wear cloth sample after the damaged edge is cut according to the method specified in Appendix G of "AS/NZS 2040.1:2005 Performance of household electrical appliances-Clothes washing machines Part 1: Methods for measuring performance, energy and water consumption" The reduction ratio of the area relative to its initial area is taken as the wear value. Wherein, the length and width of the initial sample of the wear cloth are both 10 cm, which is a regular square. After washing, the edges of the specimen fall off, and the shape of the specimen may become irregular after trimming. Therefore, the calculation of the area of the irregular shape sample becomes a difficult problem in calculating the wear value. The invention adopts the method of machine vision, and can accurately calculate the area of the worn cloth sample after washing through image segmentation, avoiding the error of manual measurement.

Contents of the invention

The invention relates to a method for evaluating fabric washing wear based on machine vision, which realizes fine and objective evaluation of washing wear in the fabric washing process. The evaluation method of fabric washing wear based on machine vision will be realized through the cooperation of hardware and software. The hardware should be able to obtain high-precision images of worn cloth samples, and the software should be able to acquire, save, crop, and segment images. And the calculation of washing wear evaluation parameters. In this method, photographs are taken of the worn cloth before and after washing, and the size of the area of the worn cloth in the image of the worn cloth sample before and after washing is calculated, and then the area change rate is calculated, that is, the wear value is used as a parameter to evaluate the severity of washing wear. The method overcomes the error caused by measuring the size of the worn cloth sample by a manual method and the difficult problem of being unable to accurately calculate the area when the worn cloth is irregular in shape after washing.

The hardware part should be able to achieve high-precision acquisition of the image of the worn cloth sample, which is mainly controlled by an industrial video kit (1), a light source (2), a data transmission wire (3), a computer host (4), a display (5), and a light source (6), and a sample stage (7) for sample placement, industrial video kit and light source installation.

The industrial video kit (1) consists of an area array color CCD camera, a data transmission line and an acquisition card, and is a key device for image acquisition of worn cloths. Among them, the area array CCD resolution is 2048×2048 pixels.

The light source (2) and the light source controller (6) are used together to meet the brightness requirements of the sample surface when the image of the worn cloth sample is collected. The light source controller (6) can adjust the brightness of the light source. Among them, the light source (2) should be an LED strip light source, used in pairs, the length of the light source is 60 cm, and the number of LED rows in a single light source is 6 rows.

The software part is a computer program that can be used in conjunction with the hardware part to obtain the sample image, calculate the size of the area of the worn cloth in the image of the worn cloth sample before and after washing, and calculate the area change rate, that is, the wear value as a parameter to evaluate the washing wear seriousness. In the software part, the washing cloth image analysis function can be realized through the program development environment, such as MATLAB, Microsoft Visual C++, Microsoft VisualBasic and Delphi programming, and can also be realized through secondary development through special digital image processing software.

Description of drawings

The main examples of technical embodiments of the present invention are given schematically in the accompanying drawings, in which,

Figure 1, the image of the worn cloth sample after washing;

Figure 2, the image of the worn cloth sample after washing;

Figure 3, a schematic diagram of the components of the hardware mechanism;

Figure 4, the schematic diagram of the main functions of the software part.

Detailed ways

According to "AS/NZS 2040.1: 2005 Performance of household electrical appliances-Clothes washing machines Part1: Methods for measuring performance, energy and water consumption" appendix A, B, C requirements to determine the washing test conditions and material specification requirements, formulate the test plan . Wherein, each wear cloth is photographed by the machine vision system shown in Figure 3 before washing. The worn cloth samples were washed and dried naturally, ironed with an iron, and then photographed with the machine vision system shown in Figure 3. Assuming that the worn cloth before washing is an ideal model, the length and width of the initial sample are both 10 cm, which is a square with an area of 10,000 square millimeters. Images were collected and segmented from the worn cloth samples before washing. Assume that the gray value of the pixel in the sample area is 1, and the gray value of the pixel in the non-sample area is 0. Then, the number of pixels with a pixel value of 1 in the segmented image can be calculated as the area A ₀ of the worn cloth before washing. Assuming that the sample after washing is segmented as shown in Figure 2, calculate the number of pixels with a value of 1 contained in the sample area in the segmented image as A ₁ , then the area of the washed cloth sample can be represented by A ₁ . Then the wear value η can be expressed as formula (1)

$\mathrm{<span\; class="notranslate">\; \&eta;</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

In formula (1), the greater η of the same sample before and after washing, the more serious the wear of the sample.

In the machine vision system shown in Figure 3, after the camera is calibrated, when the length of the line is 1 cm, its length direction contains 72 pixels, that is, the actual size scale is equivalent to 1 cm and 72 pixels. According to the equivalent of 1 cm and 72 pixels, the actual area A ₀ and A ₁ of the worn cloth sample before and after washing can be obtained from the number of pixels contained in the worn cloth sample, and the actual area of the sample and the wear value η can be calculated.

According to "AS/NZS 2040.1: 2005 Performance of household electrical appliances-Clothes washing machines Part1: Methods for measuring performance, energy and water consumption" appendix A, B, C requirements to determine the washing test conditions and material specification requirements, to develop a test plan . In the test plan, a household drum washing machine is selected, with a maximum washing capacity of 8KG and a washing mode of 30 minutes of quick washing. For the load type, see "AS/NZS 2040.1: 2005 Performance of household electrical appliances-Clothes washing machines Part 1: Methods for measuring performance , energy and waterconsumption" appendix C requirements. When the load weight is selected as 4KG, the number of samples of washing wear cloth is 15 pieces, and their numbers are shown in Table 1. The specifications of the wear cloth are: "India アンクロス COSMO 1100 cotton 100% color no. 11 width 91.4cm length 30m a piece of thorn しゆう cloth". The specific washing parameters are as follows: washing temperature is 30°C, washing once, rinsing three times, dehydration once, dehydration speed is 800rpm, and the total washing time is 34min. Using the machine vision system shown in Figure 3, the image of the worn cloth sample before washing is collected and segmented to calculate the area of the sample before washing. Similarly, calculate the sample area after washing, and calculate the wear value η according to formula (1), as shown in Table 1.

Table 1 Abrasion test data of household washing machine quick wash 30 rinse 3 times

Claims (7)

1. A machine vision-based fabric washing wear evaluation method will be realized through the collaboration of hardware and software. The hardware should be able to achieve high-precision acquisition of the image of the worn cloth sample. It is mainly composed of an industrial video kit (1), a light source ( 2), data transmission wire (3), computer mainframe (4), display (5), light source controller (6), and sample stage (7) for sample placement, industrial video kit and light source installation. The software part can realize image acquisition, storage, cropping, image segmentation and calculation of washing wear evaluation parameters. 2. According to claim 1, the industrial video kit (1) should at least include an area array color CCD camera, a data transmission line and a capture card. Among them, the area array CCD resolution is not less than 2048×2048 pixels. 3. According to claim 1, the light source (2) is used in conjunction with the light source controller (3), and should meet the brightness requirements of the surface of the sample when collecting images of the washed and worn cloth sample. The light source controller (6) should be able to adjust the brightness of the light source. Among them, the light source (2) should be an LED strip light source, used in pairs, the length of the light source is not less than 50 cm, and the LED in a single light source is not less than 4 rows. 4. according to claim 1, host computer (4) should be equipped with industrial video kit image acquisition card slot or interface, at least can pass through VGA interface, DVI interface, USB interface, 1394 interface or PCI slot, PCI-E slot , ISA slot, and AGP slot to connect the image acquisition card with the computer motherboard to realize image acquisition and display. 5. According to claim 1 and claim 4, the industrial video kit (1) and the main computer (4) should be connected to each other with the display (5) by the data transmission wire (3), wherein the display (5) should be a color display, Can display RGB color images. 6. According to claim 1 and claim 5, in the hardware system, there are at least two data transmission wires (2), which can respectively realize the image from the industrial video kit (1) to the computer host (4), and the image information from the computer Host (4) to display (5) transmission. 7. According to claim 1, the hardware part should include a sample stand (7), which can be used to install the industrial video kit (1) and light source (2), and to place samples of washing wear cloths that need to be photographed. Among them, the sample table where the sample is placed is a square, and the size should not be smaller than 3 times the sample area.

Images