Light leakage detection method and mechanism for black-out cloth

Abstract

The invention discloses a black-out cloth light leakage detection method and a black-out cloth light leakage detection mechanism, which do not need a darkroom environment and manual light leakage judgment, thereby liberating workers to work in a special darkroom environment, realizing full-automatic detection and automatic labeling, and being safe, reliable and easy to popularize and use. The black-out cloth light leakage detection method comprises the following steps: irradiating the front surface of a fabric by using a light source emitting invisible light; shooting the back surface of the fabric by using a camera with a visible light filter; and identifying an image obtained by shooting, and identifying a light leakage area in the image.

Description

Technical Field

[0001] The invention belongs to the field of fabric detection in the textile industry, and in particular relates to a method and mechanism for detecting light leakage of a shading cloth. Background Art

[0002] At present, the light leakage detection of blackout fabrics requires the factory to build a completely darkroom environment (shielding natural light), install a strong light source under the fabric inspection machine, and use blackout fabrics to completely block the light source on both sides of the fabric that needs to be tested for light leakage to simulate the user's usage environment, and then rely on the naked eye of the workers to judge whether the fabric is leaking light.

[0003] Working in a dark room, including switching from a dark room to a normal environment, is harmful to workers’ eyes. In addition, working in a dark environment for a long time will also affect workers’ mental state. In addition, in terms of work results and working status, because workers work in a dark room, managers cannot understand their real working status and cannot grasp the quality of their real work. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and mechanism for detecting light leakage of a shading cloth in response to the deficiencies in the above-mentioned prior art, which does not require a darkroom environment and does not require manual judgment of light leakage, thereby freeing workers to work in a special darkroom environment and realizing fully automated detection and automatic labeling, which is safe, reliable, easy to promote and use.

[0005] The first aspect of the present invention discloses a method for detecting light leakage from a shading cloth, comprising the following steps:

[0006] Step 1: Use a light source that emits invisible light to illuminate the front side of the fabric;

[0007] Step 2: Use a camera with a visible light filter to photograph the back of the fabric;

[0008] Step 3: Identify the captured image and identify the light leakage area in the image.

[0009] The above-mentioned light leakage detection method for blackout cloth also includes: step 4, if the light leakage area is identified in step 3, the light leakage area that meets the preset conditions is screened out, and the fabric defect position corresponding to the light leakage area is marked using a marking machine.

[0010] In the above-mentioned method for detecting light leakage from a shading cloth, the step 3 of identifying the image obtained by shooting comprises the following steps:

[0011] Step 3-1, grayscale conversion and Gaussian blur:

[0012] Convert captured color images to grayscale images to simplify data processing;

[0013] Apply Gaussian blur to the grayscale image, using a Gaussian kernel with a standard deviation of 1 to reduce image noise and smooth the edges of the light leakage area;

[0014] Step 3-2, threshold segmentation:

[0015] Apply an adaptive threshold segmentation method (such as Otsu's method) to convert the image into a binary image to highlight potential light leakage areas. This method can better adapt to images under different lighting conditions;

[0016] Step 3-3, feature extraction:

[0017] Use the Canny edge detection algorithm to extract edge features in the image and identify the outline of the light leakage area;

[0018] Gray-level co-occurrence matrix (GLCM) texture analysis is performed to extract the texture features of the light leakage area to distinguish light leakage from other image features.

[0019] Step 3-4: Image recognition:

[0020] Input the image processed by step 3-1 to step 3-3 into the trained convolutional neural network (CNN) model;

[0021] Use the model to automatically learn features in the image and perform classification and recognition to identify light leakage areas;

[0022] Step 3-5: Defect location:

[0023] A sliding window method with a size of 32x32 pixels is used to perform redundant sliding on the original image with a step size of 16 pixels;

[0024] The image of each window is input into the classification network for defect recognition;

[0025] Using image processing technology to determine the coordinates of the light leakage area, including the center point and bounding box of the light leakage area;

[0026] Step 3-6, result output:

[0027] The identified and located light leakage area information is output in the form of image tags, marking the bounding box of each light leakage area; a text report containing the coordinates, size and number of light leakage areas is generated for subsequent processing and quality control.

[0028] In this embodiment, before the threshold segmentation in step 3-2, a morphological operation is also performed on the image to remove noise points; the specific steps include:

[0029] Step a, selecting a suitable structural element according to the image characteristics;

[0030] Step b, using the structural element to corrode the image to eliminate small noise points and smooth the boundaries of larger objects, reducing the small protruding parts of the objects in the image;

[0031] Step c: perform a dilation operation on the eroded image.

[0032] In this embodiment, the invisible light in step 1 is ultraviolet light in the UVA or UVB band.

[0033] The present invention also discloses a light-shielding cloth light leakage detection mechanism, comprising an automatic detection machine and a client host;

[0034] The automatic inspection machine includes a light source emitting invisible light and a camera with a visible light filter;

[0035] The light source is used to illuminate the front side of the fabric entering the automatic inspection machine;

[0036] The camera is used to photograph the back side of the fabric entering the automatic inspection machine;

[0037] The client host is used to identify the captured image and identify the light leakage area in the image.

[0038] It should be noted that the client host

[0039] The above-mentioned light leakage detection mechanism for shading cloth also includes a cloth inspection machine, an automatic marking machine and a PLC controller;

[0040] The fabric inspection machine is used to control the fabric to enter the automatic inspection machine section by section in a cloth feeding manner;

[0041] The PLC controller is used to control the automatic detection machine and the automatic marking machine;

[0042] The automatic marking machine is used to mark the fabric defect position corresponding to the light leakage area that meets the preset conditions.

[0043] The above-mentioned shading cloth light leakage detection mechanism also includes an audible and visual alarm light, which is controlled by a PLC controller. When the client host recognizes a light leakage area in the image, the audible and visual alarm light starts to alarm.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] 1. Improve detection efficiency and accuracy: Through automated detection methods, manual operations are reduced, the continuity and accuracy of detection are improved, and the impact of human factors on the detection results is reduced.

[0046] 2. Improve working environment and safety: No need to work in a completely darkroom environment, which reduces damage to workers' eyes and mental state, and reduces the safety risks that may be caused by working in a darkroom for a long time.

[0047] 3. Reduce costs and facilitate promotion: Reduce dependence on professional darkroom environment, reduce construction and maintenance costs, and the automatic detection machine is easy to promote and use in different environments.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 The present invention is a flow chart of a method for detecting light leakage of a shading cloth.

[0050] Figure 2 This is the schematic diagram of the light leakage detection mechanism of the shading cloth.

[0051] Description of reference numerals:

[0052] 100—Automatic testing machine; 200—Client host; 300—Cloth inspection machine;

[0053] 400—Automatic marking machine; 500—PLC controller; 600—Sound and light alarm lamp. DETAILED DESCRIPTION

[0054] like Figure 1 As shown, a method for detecting light leakage from a shading cloth comprises the following steps:

[0055] Step 1: Use a light source that emits invisible light to illuminate the front side of the fabric;

[0056] Step 2: Use a camera with a visible light filter to photograph the back of the fabric;

[0057] Step 3: Identify the captured image and identify the light leakage area in the image.

[0058] Step 4: If a light leakage area is identified in step 3, the light leakage area that meets the preset conditions is screened out, and the fabric defect position corresponding to the light leakage area is marked using a marking machine.

[0059] In this embodiment, the step 3 of identifying the captured image includes the following steps:

[0060] Step 3-1, grayscale conversion and Gaussian blur:

[0061] Convert the captured color image into a grayscale image;

[0062] Apply Gaussian blur to the grayscale image, using a Gaussian kernel with a standard deviation of 1 to reduce image noise and smooth the edges of the light leakage area;

[0063] Step 3-2, threshold segmentation:

[0064] Apply the adaptive threshold segmentation method to convert the image into a binary image;

[0065] Step 3-3, feature extraction:

[0066] Use the Canny edge detection algorithm to extract edge features in the image and identify the outline of the light leakage area;

[0067] Gray-level co-occurrence matrix (GLCM) texture analysis is performed to extract the texture features of the light leakage area;

[0068] Step 3-4: Image recognition:

[0069] Input the image processed by step 3-1 to step 3-3 into the trained convolutional neural network (CNN) model;

[0070] Use the model to automatically learn features in the image and perform classification and recognition to identify light leakage areas;

[0071] Step 3-5: Defect location:

[0072] A sliding window method with a size of 32x32 pixels is used to perform redundant sliding on the original image with a step size of 16 pixels;

[0073] The image of each window is input into the classification network for defect recognition;

[0074] Using image processing technology to determine the coordinates of the light leakage area, including the center point and bounding box of the light leakage area;

[0075] Step 3-6, result output:

[0076] The identified and located light leakage area information is output in the form of image tags, marking the bounding box of each light leakage area; and a text report containing the coordinates, size and number of light leakage areas is generated.

[0077] Preferably, before the threshold segmentation in step 3-2, the image is further subjected to morphological operations, and the specific steps include:

[0078] Step a, selecting a suitable structural element according to the image characteristics;

[0079] Step b, using the structural element to corrode the image to eliminate small noise points and smooth the boundaries of larger objects, reducing the small protruding parts of the objects in the image;

[0080] Step c: perform a dilation operation on the eroded image.

[0081] It should be noted that the invisible light in step 1 is ultraviolet light in the UVA or UVB band.

[0082] like Figure 2 As shown, a light leakage detection mechanism for a shading cloth includes an automatic detection machine 100, a client host 200, a cloth inspection machine 300, an automatic marking machine 400, a PLC controller 500 and an audible and visual alarm lamp 600;

[0083] The automatic inspection machine 100 includes a light source emitting invisible light and a camera with a visible light filter;

[0084] The light source is used to illuminate the front side of the fabric entering the automatic inspection machine 100;

[0085] The camera is used to photograph the back side of the fabric entering the automatic inspection machine 100;

[0086] The client host 200 is used to identify the captured image and identify the light leakage area in the image;

[0087] The fabric inspection machine 300 is used to control the fabric to enter the automatic inspection machine 100 section by section in a cloth feeding manner;

[0088] The PLC controller 500 is used to control the automatic inspection machine 100 and the automatic marking machine 400;

[0089] The automatic marking machine 400 is used to mark the fabric defect position corresponding to the light leakage area that meets the preset conditions;

[0090] The sound and light alarm lamp 600 is controlled by the PLC controller 500. When the client host 200 identifies the light leakage area in the image, the sound and light alarm lamp 600 starts to alarm.

[0091] When the shading cloth light leakage detection mechanism of the present invention is used, the following working steps are included:

[0092] Step 1: Workers pass the fabrics to be tested from the automatic testing machine 100 to 300 traditional fabric inspection machines.

[0093] Step 2: Turn on the client host 200, start the detection program, and click Start Detection;

[0094] Step 3: The automatic inspection machine 100 starts the light source (invisible light) and triggers the industrial camera to shoot (special light) according to the shooting instruction of the PLC controller 500 (PLC records the distance of the fabric through the code table and triggers shooting every 150 mm). Because the camera lens has a visible light filter (380 nanometers to 750 nanometers are filtered), the camera will not be disturbed by the natural light in the gap between the equipment and the fabric, and only shoots the special light projected from the back of the fabric (if there is light leakage, the light source intensity must be sufficient to penetrate the light leakage point of the fabric);

[0095] Step 4: The client host 200 connects to the industrial camera of the automatic inspection machine 100 through the network, obtains relevant pictures, calls relevant algorithms for detection, and determines whether there are light leakage points.

[0096] Step 5: The client host 200 filters out the noise caused by camera exposure in the image, and filters out special exposure imaging in the image, such as seams, holes, and light leakage from the edge of the fabric; the above steps 3-1 and 3-2 are used here; and a corresponding image preprocessing algorithm can also be used according to actual needs;

[0097] Step 7: The client host 200 can identify the corresponding light leakage defects according to customer requirements (filter according to different standards, for some fabrics, light leakage intensity and light leakage point size are filtering conditions, and the above steps 3-3 to 3-6 can be used for identification and further screening).

[0098] Step 8: According to the customer's set standards, the light leakage points that need to be alarmed and labeled, the client host 200 will notify the sound and light alarm and the automatic marking machine 400 through the PLC controller 500, and the PLC controller 500 will trigger the marking according to the set distance (detection position to marking position) to ensure the correct marking position. (The marking pattern can be configured in the automatic marking machine 400 according to customer requirements);

[0099] Step 9: After the whole batch of fabrics is tested, the testing program can be closed on the client host 200, and the system will generate a test report for subsequent customer inquiries.

[0100] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any simple modification, change and equivalent structural change made to the above embodiment based on the technical essence of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for detecting light leakage of a shading cloth, characterized in that: The following steps are involved: Step 1: Use a light source that emits invisible light to illuminate the front side of the fabric; Step 2: Use a camera with a visible light filter to photograph the back of the fabric; Step 3: Identify the captured image and identify the light leakage area in the image.

2. A method for detecting light leakage of a shading cloth according to claim 1, characterized in that: The method further includes: step 4, if a light leakage area is identified in step 3, screening out the light leakage area that meets the preset conditions, and using a marking machine to mark the fabric defect position corresponding to the light leakage area.

3. A method for detecting light leakage of a shading cloth according to claim 1, characterized in that: The step 3 of identifying the image obtained by shooting includes the following steps: Step 3-1, grayscale conversion and Gaussian blur: Convert the captured color image into a grayscale image; Apply Gaussian blur to the grayscale image, using a Gaussian kernel with a standard deviation of 1 to reduce image noise and smooth the edges of the light leakage area; Step 3-2, threshold segmentation: Apply the adaptive threshold segmentation method to convert the image into a binary image; Step 3-3, feature extraction: Use the Canny edge detection algorithm to extract edge features in the image and identify the outline of the light leakage area; Gray-level co-occurrence matrix (GLCM) texture analysis is performed to extract the texture features of the light leakage area; Step 3-4: Image recognition: Input the image processed by step 3-1 to step 3-3 into the trained convolutional neural network (CNN) model; Use the model to automatically learn features in the image and perform classification and recognition to identify light leakage areas; Step 3-5: Defect location: A sliding window method with a size of 32x32 pixels is used to perform redundant sliding on the original image with a step size of 16 pixels; The image of each window is input into the classification network for defect recognition; Using image processing technology to determine the coordinates of the light leakage area, including the center point and bounding box of the light leakage area; Step 3-6, result output: The identified and located light leakage area information is output in the form of image markings, marking the boundary box of each light leakage area.

4. A method for detecting light leakage of a shading cloth according to claim 3, characterized in that: Steps 3-6 also include generating a text report containing the coordinates, size, and number of light leak areas.

5. A method for detecting light leakage of a shading cloth according to claim 3, characterized in that: Before the threshold segmentation in step 3-2, the image is further subjected to morphological operation, and the specific steps include: Step a, selecting a suitable structural element according to the image characteristics; Step b, using the structural element to corrode the image to eliminate small noise points and smooth the boundaries of larger objects, reducing the small protruding parts of the objects in the image; Step c: perform a dilation operation on the eroded image.

6. A method for detecting light leakage of a shading cloth according to claim 1, characterized in that: In step 1, the invisible light is ultraviolet light in the UVA or UVB band.

7. A light leakage detection mechanism for a shading cloth, characterized in that: Including automatic detection machine and client host; The automatic inspection machine includes a light source emitting invisible light and a camera with a visible light filter; The light source is used to illuminate the front side of the fabric entering the automatic inspection machine; The camera is used to photograph the back side of the fabric entering the automatic inspection machine; The client host is used to identify the captured image and identify the light leakage area in the image.

8. A light leakage detection mechanism for a shading cloth according to claim 7, characterized in that: It also includes a fabric inspection machine, an automatic marking machine and a PLC controller; The fabric inspection machine is used to control the fabric to enter the automatic inspection machine section by section in a cloth feeding manner; The PLC controller is used to control the automatic detection machine and the automatic marking machine; The automatic marking machine is used to mark the fabric defect position corresponding to the light leakage area that meets the preset conditions.

9. A shading cloth light leakage detection mechanism according to claim 8, characterized in that: It also includes an audible and visual alarm light, which is controlled by a PLC controller. When the client host recognizes a light leakage area in the image, the audible and visual alarm light starts to alarm.

Images