High-precision cloth color difference detection device and method

Abstract

The invention relates to the technical field of color difference detection, in particular to a high-precision cloth color difference detection device and method, and the device comprises a lighting unit, a light filtering unit, a control unit, an image analysis unit and a camera shooting unit; the light filtering unit comprises a light shield, a light filter turntable and a plurality of light filters arranged on the light filter turntable; light emitted by the lighting unit is transmitted to the detection cloth after passing through the optical filter and then is reflected to the camera shooting unit. The method comprises the following steps: acquiring a detection cloth image of an illumination unit irradiated by different optical filters; the optical filter turntable is rotated until detection cloth images under light transmission irradiation corresponding to all optical filters are obtained; a spectrogram under irradiation of the corresponding transmission wave band is obtained based on gray analysis of the image analysis unit; and analyzing the actual color of the to-be-detected cloth according to the spectrogram under the transmission wave band corresponding to each optical filter. By adopting the scheme, the accuracy of chromatic aberration detection can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of color difference detection, and in particular to a high-precision cloth color difference detection device and method. Background Art

[0002] In the textile industry, the color quality of fabrics is one of the key factors that determine their market competitiveness. However, due to various factors in the production process, such as dye batches, dyeing processes, environmental conditions, etc., there are often certain color differences between fabrics. This color difference not only affects the appearance quality of the product, but may also lead to complaints and returns, causing economic losses to the company. Therefore, it is particularly important to accurately detect the color difference of fabrics.

[0003] The traditional color difference detection method mainly uses a color CMOS camera to acquire images of the cloth to be inspected, and then performs color analysis based on the acquired images. However, this method has the problem of low accuracy. The principle is that the filter array on the surface of the current color CMOS is a Bayer filter, which is a mosaic color filter array formed by arranging RGB color filters on the grid of light sensing components. Most of the single-chip digital image sensors used in digital cameras, video recorders, scanners, etc. use this specific arrangement of color filter arrays to produce color images. This color filter arrangement has 50% green, 25% red, and the other 25% blue, so it is also called RGBG, GRGB or RGGB. Because each pixel only filters and records one of the three RGB colors, the information obtained from a single pixel cannot fully express the composition values ​​of red, green, and blue. In order to obtain a full-color image, different demosaicing algorithms can be used to interpolate the red, green, and blue composition values ​​of each pixel. The red value of a green pixel can be calculated by interpolating the two adjacent red pixels; similarly, the blue value can be calculated by interpolating the two adjacent blue pixels. These algorithms use the surrounding pixels of the same color to estimate the composition of a specific pixel. The algorithm has different computational requirements and the final image quality is also different. This simple method performs well when the color is constant or changes evenly, but it will produce noise at sudden changes in color and brightness, such as bleeding, which is particularly obvious at sharp corners.

[0004] With the application of high-end fabrics such as moisture-sensitive color change, temperature-sensitive color change, and the increase in fabric colors, traditional detection methods and accuracy rates are obviously unable to meet the needs. Therefore, a high-precision fabric color difference detection device and method is urgently needed to improve the accuracy of color difference detection. Summary of the invention

[0005] The present invention provides a high-precision cloth color difference detection device and method, which can improve the accuracy of color difference detection.

[0006] The present invention provides a basic solution 1:

[0007] A high-precision cloth color difference detection device, comprising an illumination unit, a filter unit, a control unit, an image analysis unit and a camera unit;

[0008] The filter unit includes a light shield, a filter turntable and a plurality of filters arranged on the filter turntable, wherein the filters are all narrow half-bandwidth filters, and the light transmission wavelengths of any two filters are different from each other; the light shield is placed on one side of the filter turntable, and the light shield is provided with a light-transmitting area, and the size of the light-transmitting area is matched with the size of the filter, so that the light emitted by the lighting unit can only pass through one filter at most after passing through the light-transmitting area; the light emitted by the lighting unit passes through the filter, is transmitted to the detection cloth, and then reflected to the camera unit;

[0009] The camera unit acquires the image of the detected cloth and transmits it to the image analysis unit; the camera unit adopts a black and white CMOS camera;

[0010] The filter turntable is connected to a servo motor; the servo motor drives the filter turntable to rotate to switch the filter; the control end of the servo motor is connected to the output end of the control unit, and the motor output shaft rotation phase signal is fed back to the control unit; the control unit is communicatively connected to the image analysis unit.

[0011] The principle and advantage of the present invention are: in this scheme, by rotating the filter wheel, different filters can be quickly switched to obtain spectral information under different bands. Compared with the prior art that uses a color camera and calculates the color information of certain pixels through an interpolation algorithm, in this scheme, multiple actual color information can be directly obtained by rotating the filter wheel, and its accuracy is higher. Compared with color cameras, black and white CMOS cameras are more efficient in processing specific color detection tasks because they avoid the interference of color information and focus on changes in brightness and contrast. Therefore, black and white CMOS cameras can provide high-sensitivity and high-resolution image acquisition capabilities, especially in low-light conditions. Good imaging effects can still be maintained. In addition, the cost of black and white CMOS cameras is relatively low, which helps to reduce the manufacturing cost of the entire system. The use of narrow-band filters can effectively reduce spectral overlap, improve spectral resolution, and make color difference detection more refined. Compared with the common wide-band filters in the prior art, more accurate color information can be provided, which helps to reduce misjudgment and improve detection efficiency.

[0012] In summary, compared with the traditional detection device using a color camera, this solution significantly improves the flexibility and accuracy of cloth color detection by using a black and white CMOS camera and integrating multiple different filters. It can more comprehensively capture the color changes of cloth under different lighting conditions, effectively reduce the detection error caused by single light, and thus improve the accuracy of color difference detection.

[0013] Furthermore, the filter wheel includes filters with transmission bands of 405 nanometers, 500 nanometers, 550 nanometers, 600 nanometers, 700 nanometers and 800 nanometers respectively.

[0014] Beneficial effects: Using filters with different transmission bands can cover a wider spectrum range, thereby more accurately capturing the color changes of cloth at different wavelengths. This helps to improve the accuracy of color difference detection, especially when dealing with complex colors or subtle color differences, and can provide more accurate data support.

[0015] Furthermore, the lighting unit adopts a halogen lamp light source and / or an LED light source.

[0016] Beneficial effects: LED light sources have the characteristics of long life and low energy consumption, while halogen light sources can provide more uniform light distribution. The selection of both can meet different detection needs and improve the service life and energy efficiency of the equipment.

[0017] Furthermore, the grayscale bit number of the image analysis unit is 10 bits or 12 bits.

[0018] The traditional 8-bit grayscale analysis unit can achieve 256 levels of grayscale resolution, and multi-bit grayscale helps to achieve finer grayscale resolution.

[0019] Furthermore, the image analysis unit or control unit includes a preset data table, which contains cloth color parameter values ​​corresponding to different temperature and humidity environments; correspondingly, the high-precision cloth color difference detection device also includes a temperature sensor and / or a humidity sensor, which is used to obtain the temperature and humidity conditions of the cloth detection environment and feed them back to the control unit.

[0020] Combined with the standard reference cloth color parameter values ​​containing temperature and humidity information, on the one hand, it helps to compare with the accurate target color parameters, making color difference detection more accurate. On the other hand, it helps to adjust the temperature / humidity in combination with the temperature and humidity adjustment device, facilitating the accurate detection of temperature-sensitive cloth and color-sensitive cloth.

[0021] The present invention provides a second basic solution:

[0022] A high-precision cloth color difference detection method uses the above-mentioned high-precision cloth color difference detection device.

[0023] Further, the method for analyzing the color difference of the cloth to be tested includes the following steps:

[0024] Obtain the images of the fabric to be inspected when the lighting unit is illuminated through different filters; rotate the filter turntable until the images of the fabric to be inspected under the corresponding light transmission of all filters are obtained;

[0025] Based on the grayscale analysis of the image analysis unit, a spectrum diagram under the illumination of the corresponding transmission band is obtained;

[0026] Analyze the actual color of the cloth to be tested according to the spectrum diagram under the corresponding transmission band of each filter;

[0027] Compare the actual color of the cloth to be tested with the preset target color to generate color difference analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram of the principle of Example 1 of the present invention.

[0029] Figure 2 Schematic diagram of the structure of the filter unit in Example 1 of the present invention.

[0030] Figure 3 The present invention is a flowchart of an embodiment of a high-precision cloth color difference detection method. DETAILED DESCRIPTION

[0031] The following is further described in detail through specific implementation methods:

[0032] Embodiment 1:

[0033] A high-precision cloth color difference detection device, such as Figure 1 , Figure 2As shown, it includes an illumination unit 1, a filter unit, a control unit, an image analysis unit and a camera unit 6; the filter unit includes a light shield 2, a filter turntable 3 and a plurality of filters arranged on the filter turntable, the filters are all narrow half-bandwidth filters, and the transmittance wavelengths of any two filters are different; in this embodiment, the number of filters on the filter turntable is 6, and the filters are equidistantly distributed in a circular array on the filter turntable, and the filter turntable is provided with 6 filters with transmission bands of 405 nanometers (ultraviolet light), 500 nanometers (green light), 550 nanometers (yellow light), 600 nanometers (red light), 700 nanometers (deep red light) and 800 nanometers (infrared light), and the half-bandwidth of each filter is 10 nanometers. The light shield 2 is placed on one side of the filter turntable, and the light shield is provided with a light-transmitting area, the size of which is matched with the size of the filter, so that the light emitted by the lighting unit can only pass through one filter at most after passing through the light-transmitting area; the light-transmitting area can be a light-transmitting hole, or it can be a recessed portion arranged on the outer edge of the light shield; the light emitted by the lighting unit 1 passes through the filter, is transmitted to the detected cloth, and then reflected to the camera unit 6; the camera unit 6 obtains the image of the detected cloth and transmits it to the image analysis unit; the camera unit adopts a black and white CMOS camera; the filter turntable is connected to a servo motor; the servo motor drives the filter turntable to rotate to switch the filter; the control end of the servo motor is connected to the output end of the control unit, and feeds back the motor output shaft rotation phase signal to the control unit; the control unit is communicatively connected with the image analysis unit.

[0034] The lighting unit 1 uses a halogen lamp light source and / or an LED light source; in this embodiment, an array-type LED light source is used. The CMOS sensor built into the CMOS black-and-white camera has a sensitive wavelength between 380 nanometers and 850 nanometers, and a resolution of 1920×1080. This camera can capture light from ultraviolet to near-infrared, but it does not distinguish between colors, and only records the intensity and grayscale of light.

[0035] The filter unit switches the filter by rotating the filter dial, so that the CMOS black and white camera can record the grayscale information of multiple different bands, and then more accurately judge the color of the cloth to be detected by analyzing the image data under each band. With this solution, the number and parameters of filters can be flexibly adjusted according to needs, so as to obtain information of multiple visible light bands, rather than just the traditional RGB three channels. Six filters can provide six spectral channels. Even if each channel has 256 grayscale levels (8 bits), theoretically, it can distinguish 256 6 ≈2.8×10 14Although the number of colors actually distinguished is less than this value due to spectral overlap, noise, and the dynamic range limitation of the device, in this embodiment, the grayscale bit number of the image analysis unit is 10 bits or 12 bits, which can also achieve higher-precision cloth color difference detection.

[0036] The color difference detection device in this solution uses 5 visible light color channels, each of which provides additional information, greatly increasing the dimension of the color space. Traditional color cameras can only capture visible light images, while this method provides an infrared band image. Infrared light can detect color differences or defects that visible light cannot detect in some cases, and the influence of moisture on the absorption and reflection of infrared light under different humidity conditions makes the detection of moisture-sensitive fabrics more accurate, so it has great advantages in specific applications. In summary, the use of this solution can significantly improve the accuracy of color detection.

[0037] Embodiment 2:

[0038] Compared with Example 1, in this embodiment, the image analysis unit or control unit includes a preset data table, and the preset data table includes cloth color parameter values ​​corresponding to different temperature and humidity environments; the cloth color parameter values ​​can be grayscale values ​​corresponding to different light bands, so that the image grayscale values ​​of different filter bands can be obtained while combining the parameters in the data table for comparison, so that there is no need to obtain grayscale images in different bands and then reproduce the detected color parameter values ​​through the spectral reconstruction algorithm, thereby achieving faster comparison and detection; of course, spectral reconstruction can also be performed to perform corresponding color difference comparison. Correspondingly, the high-precision cloth color difference detection device also includes a temperature sensor and / or a humidity sensor, and the temperature sensor and / or humidity sensor is used to obtain the temperature and humidity conditions of the cloth detection environment and feed it back to the control unit; in addition, this embodiment also includes a temperature and humidity adjustment device; the temperature and humidity adjustment device is arranged on one side of the cloth transmission area for detection, and the control end of the temperature and humidity adjustment device is connected to the control unit; the temperature and humidity adjustment device includes a heating device and / or an electrically controlled spray device. The heating device can be such as Figure 1 The heating plate 5 shown can also be a non-contact heating lamp, in particular, an infrared lamp, which directly irradiates the cloth. Combined with the existing infrared humidity detection algorithm, the humidity sensor can be saved. At the same time, the temperature can be adjusted within a certain temperature range to achieve the detection of temperature-sensitive cloth. The electric-controlled spray device can be set to a certain height and a fixed spray volume. Combined with the spray area and actual testing, the humidity can be controlled and adjusted. When it is necessary to detect moisture-sensitive cloth, combined with the initial humidity and the number of sprays, the cloth color detection at the target humidity can be achieved.

[0039] Embodiment 3:

[0040] A high-precision cloth color difference detection method is used to analyze the color difference of the cloth to be detected, using the above-mentioned high-precision cloth color difference detection device, such as Figure 3 As shown, the following steps are included:

[0041] Aim the camera unit's lens at the cloth to be inspected.

[0042] Acquire the detected cloth images under the illumination of the lighting unit through different filters; rotate the filter dial until the detected cloth images under the corresponding transmitted light illumination of all filters are obtained; based on the grayscale analysis of the image analysis unit, obtain the spectrum diagram under the illumination of the corresponding transmission band; in this embodiment, image data of 6 different bands (405 nanometers, 500 nanometers, 550 nanometers, 600 nanometers, 700 nanometers, and 800 nanometers) can be obtained.

[0043] According to the spectrum diagram under the corresponding transmission band of each filter, the actual color of the cloth to be detected is analyzed; in this embodiment, the hue value of the image data captured under different bands is analyzed to analyze the actual color of the cloth to be detected. This is a prior art and will not be described in detail here; in other embodiments of the present application, the RGB model and the HSV model can also be used to determine the actual color of the cloth to be detected.

[0044] Compare the actual color of the cloth to be tested with the preset target color to generate color difference analysis results.

[0045] In this way, the color changes of cloth under different lighting conditions can be fully captured, effectively reducing the detection errors caused by single light, thereby improving the accuracy of color difference detection.

[0046] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A high-precision cloth color difference detection device, characterized in that: It includes an illumination unit, a filter unit, a control unit, an image analysis unit and a camera unit; The filter unit includes a light shield, a filter turntable and a plurality of filters arranged on the filter turntable, wherein the filters are all narrow half-bandwidth filters, and the light transmission wavelengths of any two filters are different from each other; the light shield is placed on one side of the filter turntable, and the light shield is provided with a light-transmitting area, and the size of the light-transmitting area is matched with the size of the filter, so that the light emitted by the lighting unit can only pass through one filter at most after passing through the light-transmitting area; the light emitted by the lighting unit passes through the filter, is transmitted to the detection cloth, and then reflected to the camera unit; The camera unit acquires the image of the detected cloth and transmits it to the image analysis unit; the camera unit adopts a black and white CMOS camera; The filter turntable is connected to a servo motor; the servo motor drives the filter turntable to rotate to switch the filter; the control end of the servo motor is connected to the output end of the control unit, and the motor output shaft rotation phase signal is fed back to the control unit; the control unit is communicatively connected to the image analysis unit.

2. The high-precision cloth color difference detection device according to claim 1, characterized in that: The number of filters on the filter turntable is 6.

3. The high-precision cloth color difference detection device according to claim 2, characterized in that: The filter wheel comprises filters with transmission bands of 405 nanometers, 500 nanometers, 550 nanometers, 600 nanometers, 700 nanometers and 800 nanometers respectively.

4. The high-precision cloth color difference detection device according to claim 1, characterized in that: The grayscale bit number of the image analysis unit is 10 bits or 12 bits.

5. The high-precision cloth color difference detection device according to claim 4, characterized in that: The image analysis unit or control unit includes a preset data table, which includes cloth color parameter values ​​corresponding to different temperature and humidity environments; correspondingly, the high-precision cloth color difference detection device also includes a temperature sensor and / or a humidity sensor, which is used to obtain the temperature and humidity conditions of the cloth detection environment and feed them back to the control unit.

6. The high-precision cloth color difference detection device according to claim 1, characterized in that: The lighting unit adopts a halogen lamp light source and / or an LED light source.

7. The high-precision cloth color difference detection device according to claim 1, characterized in that: It also includes a temperature and humidity regulating device; the temperature and humidity regulating device is arranged on one side of the cloth transmission detection area, and the control end of the temperature and humidity regulating device is connected to the control unit.

8. The high-precision cloth color difference detection device according to claim 7, characterized in that: The temperature and humidity regulating device comprises a heating device and / or an electrically controlled spraying device.

9. A high-precision cloth color difference detection method, characterized in that: A high-precision cloth color difference detection device as described in any one of claims 1 to 8 is used.

10. The high-precision cloth color difference detection method according to claim 9 is used to analyze the color difference of the cloth to be detected, characterized in that: The following steps are involved: Acquire the inspection cloth image when the lighting unit is illuminated through different filters; Rotate the filter turntable until the images of the tested cloth under the light transmission corresponding to all filters are obtained; Based on the grayscale analysis of the image analysis unit, a spectrum diagram under the illumination of the corresponding transmission band is obtained; Analyze the actual color of the cloth to be tested according to the spectrum diagram under the corresponding transmission band of each filter; Compare the actual color of the cloth to be tested with the preset target color to generate color difference analysis results.

Images