Infant paper diaper stain detection system based on image recognition

Abstract

The invention relates to the technical field of paper diaper stain detection, in particular to a baby paper diaper stain detection system based on image recognition. The system comprises the following steps: S1, building equipment; s2, establishing a thickness compensation coefficient matrix; s3, correcting the ultraviolet signal in real time; s4, carrying out frequency domain band elimination filtering; and S5, visible light auxiliary verification. The UV-CCD camera is arranged to sample the ultraviolet image of the clean paper diaper, the permeability of stains in the paper diaper is considered, the thickness compensation coefficient matrix is added to the ultraviolet image, the detection precision is improved, in addition, due to the reflection effect of SAP molecules on ultraviolet rays, the frequency domain band elimination filtering step is added, interference caused by the SAP molecules is filtered out, and the detection precision is improved. Therefore, abrupt stains can be recognized in the ultraviolet image, and meanwhile, in order to increase the credibility of stain detection, visible light image auxiliary verification is added, and the accuracy of stain recognition is increased through multi-modal fusion judgment.

Description

Technical Field

[0001] The present invention relates to the technical field of diaper stain detection, and in particular to a baby diaper stain detection system based on image recognition. Background Art

[0002] The image stain detection method in the prior art is to set up an industrial camera or a depth camera on the assembly line, establish a target detection module and a deep learning model for the corresponding stains, establish a database and learn according to the input stain samples, so that the target detection module can identify the stain samples during the detection process. However, when this method is applied to the production process of diapers, a mismatch occurs. Since the base color of diapers is often white and has a certain thickness, when some inconspicuous stains drip on the surface of the diaper, the color reflected after the stains are absorbed is not obvious. Therefore, the industrial camera cannot accurately judge the stains and easily skips the stain area. It is considered to add ultraviolet light to the surface of the diaper so that the stains will show high-brightness fluorescence in the image due to the spectral characteristics under the irradiation of ultraviolet light, thereby identifying the location of the stain.

[0003] However, diapers currently on the market are filled with SAP molecules (super absorbent resin particles) in the core layer to increase the water absorption capacity of the diapers. The reflectivity of the SAP molecules to ultraviolet rays is as high as 85%, which can easily cover up the appearance of stains in the image. How to detect stains under ultraviolet light on diapers with SAP molecules becomes a problem. Summary of the invention

[0004] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description and other drawings of the description.

[0005] The purpose of the present invention is to overcome the above-mentioned shortcomings and provide a baby diaper stain detection system based on image recognition. By setting up a pulsed ultraviolet light source on the stain detection assembly line, the corresponding stains on the surface of the diaper can emit high-brightness fluorescence under the action of ultraviolet rays. First, a UV-CCD camera is set to sample the ultraviolet image of the clean diaper, and considering the permeability of the stain in the diaper, a thickness compensation coefficient matrix is ​​added to the ultraviolet image to improve the detection accuracy. In addition, due to the reflection effect of SAP molecules on ultraviolet rays, a frequency domain band-stop filtering step is added to filter out the interference caused by SAP molecules, so that abrupt stains can be identified in the ultraviolet image. At the same time, in order to increase the credibility of stain detection, visible light image auxiliary verification is added, and the accuracy of stain recognition is increased through multimodal fusion judgment.

[0006] The present invention provides a baby diaper stain detection system based on image recognition, comprising: S1. Equipment construction: A visual inspection platform is built on the stain detection conveyor belt. The visual inspection platform includes a pulsed ultraviolet light source, a beam splitter, a UV-CCD camera, a CMOS camera, and a thickness measurement roller linked to the conveyor belt. The computer controls the low-angle pulsed ultraviolet light source to turn on. When the conveyor belt is turned on, n pieces of stain-free diapers pass through the inspection area, and the UV-CCD camera collects ultraviolet images. S2. Establish thickness compensation coefficient matrix: pre-process the UV image and establish UV intensity distribution map , thickness measuring roller synchronously collects thickness data , establish the thickness compensation coefficient matrix ; S3, real-time correction of ultraviolet signal: pixel-by-pixel compensation of each pixel in the currently detected ultraviolet image; S4, frequency domain band-stop filtering: convert the compensated image to the frequency domain to generate an amplitude spectrum and phase spectrum , locate the SAP interference peak in the amplitude spectrum, apply a Gaussian notch filter, and perform IFFT on the filtered amplitude spectrum and the original phase spectrum to obtain the denoised image , the denoised image is a non-contaminated baseline image; S5. Visible light assisted verification: When a suspected stain is detected in the ultraviolet image, a visible light image is extracted through the CMOS camera at the same position, and LBP texture features are extracted from the visible light image. A weighted confidence calculation is performed on the ultraviolet and visible light detection results to confirm whether it is a stain. Otherwise, a second scan is initiated.

[0007] In some embodiments, in step S2, the thickness compensation coefficient matrix The specific formula is: in, is the distribution diagram of ultraviolet intensity, is the average thickness.

[0008] In some embodiments, in step S3, the specific compensation formula for pixel compensation is: in, is the compensated pixel, is the UV image pixel, is the thickness compensation for the pixel point, d(t) is the real-time thickness, T(t) is the current temperature, T0 is the reference temperature, α is the temperature coefficient, and the temperature coefficient = 0.02 / ℃.

[0009] In some embodiments, in step S4, the SAP interference peak is at a radial frequency of 0.15-0.3, and a Gaussian notch filter is applied to the amplitude spectrum: in, , is the coordinate of the center point in the frequency domain obtained by FFT transformation, , is the coordinate of the SAP interference signal in the frequency domain, is the decay rate of the Gaussian function.

[0010] In some embodiments, in step S5, the specific process of ultraviolet image and visible light image auxiliary verification is as follows: S51, when a suspected stain is detected in the ultraviolet image, its minimum circumscribed rectangle ROI is extracted, and the corresponding area of ​​the visible light image is located through the spatial mapping matrix; S52, extracting texture features from the ROI region of the visible light image, and calculating the LBP-TOP features of the ROI region: Among them, P is the number of neighborhood pixels, R is the neighborhood radius, is the gray value of the center pixel, is the gray value of the pth neighborhood pixel, is a sign function with only two results: 0 and 1. for Assignment; After calculation, a 256-dimensional feature histogram is generated, and the contrast and energy features of the multi-directional gray-level co-occurrence matrix are extracted to obtain the texture features; S53, color space conversion, converting the RGB image to the HSV space, and calculating the mean and standard deviation of the saturation S and the brightness V of the ROI area; Where N is the total number of pixels in the ROI area, is the saturation value of the i-th pixel, is the brightness value of the i-th pixel, , It is the average value of saturation and brightness in the ROI area, reflecting the concentration and brightness level of the overall color; represents the saturation standard deviation, It indicates the standard deviation of brightness and detects abnormal spots based on the difference in values; S54, dual-modal decision fusion, input UV confidence ,Visible light feature score , the final stain judgment result is output by weighting, and the confidence weighting rule is: is the UV confidence, obtained by normalizing the UV image. is the output probability of the SVM classifier of the visible light feature, is the confidence-weighted score, and is the optimal weight determined by the ROC curve, =0.7, =0.3.

[0011] In some embodiments, in step S1, a pulsed ultraviolet light source is set at a low angle on the side of the stain detection conveyor belt, and the incident angle to the diaper surface is 15°±2°. The incident light is divided into two paths by a dichroic prism, one of which is a narrow-band ultraviolet light of 395nm. A UV-CCD camera is set above the ultraviolet irradiation area, and a 400nm long-pass filter is added to the UV-CCD camera lens. The other path is visible light, and a CMOS camera is set above the visible light irradiation area. The UV-CCD camera and the CMOS camera start acquisition asynchronously, so that the two cameras collect images of the same area during the operation of the stain detection conveyor belt.

[0012] In some embodiments, a fail-safe mechanism is established to first monitor background drift and automatically collect a baseline image every 30 minutes. > Set a threshold value to trigger an alarm and remind production personnel that the background image, which serves as a pollution-free benchmark, has a large error. A compensation failure fallback mechanism is also set up. When the surface temperature of the diaper is greater than 50°C or the thickness fluctuation measured by the thickness measurement roller is greater than 20%, an alarm is triggered to remind production personnel that the current production temperature is too high or the thickness difference of the diapers is too large.

[0013] By adopting the above technical solution, the beneficial effects of the present invention are: The present invention sets up a pulsed ultraviolet light source on the stain detection assembly line so that the corresponding stains on the surface of the diaper can emit high-brightness fluorescence under the action of ultraviolet rays. First, a UV-CCD camera is set to sample the ultraviolet image of the clean diaper, and considering the permeability of the stains in the diaper, a thickness compensation coefficient matrix is ​​added to the ultraviolet image to improve the detection accuracy. In addition, due to the reflection effect of SAP molecules on ultraviolet rays, a frequency domain band-stop filtering step is added to filter out the interference caused by SAP molecules, so that abrupt stains can be identified in the ultraviolet image. At the same time, in order to increase the credibility of stain detection, visible light image auxiliary verification is added, and the accuracy of stain recognition is increased through multimodal fusion judgment.

[0014] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.

[0015] Undoubtedly, these and other objects of the present invention will become more apparent after the following detailed description of the preferred embodiment described with reference to various figures and drawings.

[0016] In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, one or several preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation of the present invention.

[0018] In the drawings, the same reference numerals are used for the same components and the drawings are schematic and not necessarily drawn to scale.

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only one or several embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on such drawings without paying creative work.

[0020] Figure 1 It is a schematic diagram of the overall process of the stain detection system in some embodiments of the present invention; Figure 2 Schematic diagram of the dual-image assisted verification process in some embodiments of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but are not used to limit the present invention.

[0022] In addition, in the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0023] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral body; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. However, if it is indicated as a direct connection, it means that the two connected bodies are not connected through a transition structure, but are connected to form a whole through a connecting structure. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0024] In the present invention, unless otherwise clearly specified and limited, the first feature "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in an appropriate manner in any one or more embodiments or examples.

[0025] Reference Figure 1-Figure 2 , Figure 1 It is a schematic diagram of the overall process of the stain detection system in some embodiments of the present invention; Figure 2 Schematic diagram of the dual-image assisted verification process in some embodiments of the present invention.

[0026] According to some embodiments of the present invention, the present invention provides a baby diaper stain detection system based on image recognition, characterized in that it includes: S1. Equipment construction: A visual inspection platform is built on the stain detection conveyor belt. The visual inspection platform includes a pulsed ultraviolet light source, a beam splitter, a UV-CCD camera, a CMOS camera, and a thickness measurement roller linked to the conveyor belt. The computer controls the low-angle pulsed ultraviolet light source to turn on. When the conveyor belt is turned on, n pieces of stain-free diapers pass through the inspection area, and the UV-CCD camera collects ultraviolet images. The pulsed ultraviolet light source is set at a low angle on the side of the stain detection conveyor belt, and the incident angle to the diaper surface is 15°±2°. The incident light is divided into two paths by a beam splitter prism. One path is 395nm narrow-band ultraviolet light. A UV-CCD camera is set above the ultraviolet irradiation area. A 400nm long-pass filter is added to the UV-CCD camera lens. The other path is visible light. A CMOS camera is set above the visible light irradiation area. The UV-CCD camera and the CMOS camera start acquisition asynchronously, so that the two cameras can collect images of the same area during the operation of the stain detection conveyor belt. The ultraviolet light is made to illuminate the surface of the diaper at an extremely low angle, and the mirror reflection characteristics of the SAP molecules are used to guide most of the reflected light to the non-imaging area, reducing 85% of the direct reflection of SAP entering the camera. At the same time, a 400nm long-pass filter is added to the UV-CCD camera lens, which only allows fluorescent stains with a wavelength of 420-650nm to pass through, completely blocking the 365nm light reflected by the SAP molecules, effectively inhibiting the reflection of ultraviolet light by the SAP molecules, and making the stain image collected by the UV-CCD camera accurate.

[0027] S2. Establish thickness compensation coefficient matrix: pre-process the UV image and establish UV intensity distribution map , thickness measuring roller synchronously collects thickness data , establish the thickness compensation coefficient matrix ,The thickness compensation coefficient matrix is ​​used to compensate for the problem of unclear fluorescence reflection of stains due to depth; Thickness compensation coefficient matrix The specific formula is: in, is the distribution diagram of ultraviolet intensity, is the average thickness.

[0028] S3, real-time correction of ultraviolet signal: pixel-by-pixel compensation is performed on each pixel in the currently detected ultraviolet image to form a complete image; The specific compensation formula for pixel compensation is: in, is the compensated pixel, is the UV image pixel, is the thickness compensation for the pixel point, d(t) is the real-time thickness, T(t) is the current temperature, T0 is the reference temperature, α is the temperature coefficient, and the temperature coefficient = 0.02 / ℃.

[0029] S4, frequency domain band-stop filtering: convert the compensated image to the frequency domain to generate an amplitude spectrum and phase spectrum , locate the SAP interference peak in the amplitude spectrum, apply a Gaussian notch filter, and perform IFFT on the filtered amplitude spectrum and the original phase spectrum to obtain the denoised image , the denoised image is a non-contaminated baseline image; The SAP interference peak is at radial frequency 0.15-0.3, and a Gaussian notch filter is applied to the amplitude spectrum: in, , is the coordinate of the center point in the frequency domain obtained by FFT transformation, , is the coordinate of the SAP interference signal in the frequency domain, is the decay rate of the Gaussian function; Example , The method for determining the parameters is to first perform a two-dimensional FFT on the UV image of the SAP molecular clean sample to generate an amplitude spectrum, observe the bright rings or bright lines in the amplitude spectrum, and record the coordinates of the center of the bright area. , if the SAP particles vibrate at 120 Hz, the corresponding spatial frequency is: pass You can determine .

[0030] S5, Visible light assisted verification: When a suspected stain is detected in the ultraviolet image, a visible light image is extracted by a CMOS camera at the same position, LBP texture features are extracted from the visible light image, and weighted confidence calculation is performed on the ultraviolet and visible light detection results to confirm whether it is a stain. Otherwise, a secondary scan is initiated; The specific process of UV image and visible light image assisted verification is as follows: S51, when a suspected stain is detected in the ultraviolet image, its minimum circumscribed rectangle ROI is extracted, and the corresponding area of ​​the visible light image is located through the spatial mapping matrix; Use the chessboard calibration plate to calibrate the external parameters of the UV-CCD camera and the CMOS camera, establish a pixel-level mapping relationship, and automatically perform the calibration process every time the camera is turned on: in, , is the coordinate of the upper left corner of the rectangle, , is the visible light ROI coordinate, , , , Scale and rotation coefficients calculated for the calibration plate, , Translation compensation, used to correct dual-camera parallax and expand the visible light ROI by ±3 pixels to avoid mapping errors; S52, extracting texture features from the ROI region of the visible light image, and calculating the LBP-TOP features of the ROI region: Among them, P is the number of neighborhood pixels, R is the neighborhood radius, is the gray value of the center pixel, is the gray value of the pth neighborhood pixel, is a sign function with only two results: 0 and 1. for Assignment; After calculation, a 256-dimensional feature histogram is generated, and the contrast and energy features of the multi-directional gray-level co-occurrence matrix are extracted to obtain the texture features; S53, color space conversion, converting the RGB image to the HSV space, and calculating the mean and standard deviation of the saturation S and the brightness V of the ROI area; Where N is the total number of pixels in the ROI area, is the saturation value of the i-th pixel, is the brightness value of the i-th pixel, , It is the average value of saturation and brightness in the ROI area, reflecting the concentration and brightness level of the overall color; represents the saturation standard deviation, It indicates the standard deviation of brightness and detects abnormal spots based on the difference in values; An example of detecting abnormal color spots is given, such as the lubricating oil used to lubricate equipment on the production line accidentally dripping onto the surface of diapers during the production process. Although it is transparent and colorless, it will reduce the saturation and increase the brightness fluctuation; First, convert the visible light ROI image from RGB to HSV and calculate the statistics: , while the normal area , saturation decreased by 21%; , while the normal area , the standard deviation doubles. According to the judgment rule, if It is determined that stains appear, indicating that stains appear in the current image. It should be understood that the determination rule can be adaptively adjusted according to the actual application production line and the specific type of diapers, and is determined according to the production situation, not to determine the value. This is only an example; S54, dual-modal decision fusion, input UV confidence ,Visible light feature score , the final stain judgment result is output by weighting, and the confidence weighting rule is: is the UV confidence, obtained by normalizing the UV image. is the output probability of the SVM classifier of the visible light feature, is the confidence-weighted score, and is the optimal weight determined by the ROC curve, =0.7, =0.3.

[0031] Preferably, a conflict arbitration mechanism is also set for the ultraviolet results and the visible light results. The established conflict judgment mechanism is shown in Table 1: Table 1 UV results Visible light results action significance Positive Positive Confirm the stain and trigger sorting Bimodal consistency, high confidence Positive Negative Start a secondary scan UV may be a false alarm, and a second scan is required for verification Negative Positive Record logs and manually review Visible light detection of UV leakage requires algorithm iteration Preferably, a fail-safe mechanism is also established. First, the background drift is monitored and a reference image is automatically acquired every 30 minutes. > Set a threshold value, trigger an alarm, and remind production personnel that the background image used as the pollution-free benchmark has a large error. The specific calculation formula for updating the benchmark image is: In the process of automatically collecting the reference image every 30 minutes, the current image and the background image are automatically fused, and the weight ratio of the current image to the background image is 1:9. is the pixel value of the current image, is the pixel value of the background image, is the pixel value of the newly generated background image; There is also a compensation failure fallback mechanism. When the surface temperature of the diaper is greater than 50°C or the thickness fluctuation measured by the thickness measuring roller is greater than 20%, an alarm is triggered to remind production personnel that the current production temperature is too high or the thickness difference of the diapers is too large.

[0032] It should be understood that the embodiments disclosed in the present invention are not limited to the specific processing steps or materials disclosed herein, but should be extended to equivalent substitutions of such features understood by ordinary technicians in the relevant field. It should also be understood that the terms used herein are only used for the purpose of describing specific embodiments and are not meant to be limiting.

[0033] The "embodiment" mentioned in the specification means that the specific features or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases or "embodiment" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0034] In addition, the described features or characteristics may be combined in one or more embodiments in any other suitable manner. In the above description, some specific details, such as thickness, quantity, etc., are provided to provide a comprehensive understanding of the embodiments of the present invention. However, those skilled in the relevant art will understand that the present invention can be implemented without one or more of the above specific details or can also be implemented using other methods, components, materials, etc.

Claims

1. A baby diaper stain detection system based on image recognition, characterized in that: include: S1. Equipment construction: A visual inspection platform is built on the stain detection conveyor belt. The visual inspection platform includes a pulsed ultraviolet light source, a beam splitter, a UV-CCD camera, a CMOS camera, and a thickness measurement roller linked to the conveyor belt. The computer controls the low-angle pulsed ultraviolet light source to turn on. When the conveyor belt is turned on, n pieces of stain-free diapers pass through the inspection area, and the UV-CCD camera collects ultraviolet images. S2. Establish thickness compensation coefficient matrix: pre-process the UV image and establish UV intensity distribution map , thickness measuring roller synchronously collects thickness data , establish the thickness compensation coefficient matrix ; S3, real-time correction of ultraviolet signal: pixel-by-pixel compensation of each pixel in the currently detected ultraviolet image; S4, frequency domain band-stop filtering: convert the compensated image to the frequency domain to generate an amplitude spectrum and phase spectrum , locate the SAP interference peak in the amplitude spectrum, apply a Gaussian notch filter, and perform IFFT on the filtered amplitude spectrum and the original phase spectrum to obtain the denoised image , the denoised image is a non-contaminated baseline image; S5. Visible light assisted verification: When a suspected stain is detected in the ultraviolet image, a visible light image is extracted through the CMOS camera at the same position, and LBP texture features are extracted from the visible light image. A weighted confidence calculation is performed on the ultraviolet and visible light detection results to confirm whether it is a stain. Otherwise, a second scan is initiated.

2. The baby diaper stain detection system based on image recognition according to claim 1, characterized in that: In step S2, the thickness compensation coefficient matrix The specific formula is: in, is the distribution diagram of ultraviolet intensity, The average thickness.

3. The baby diaper stain detection system based on image recognition according to claim 1, characterized in that: In step S3, the specific compensation formula for pixel compensation is: in, is the compensated pixel, is the UV image pixel, is the thickness compensation for the pixel point, d(t) is the real-time thickness, T(t) is the current temperature, T0 is the reference temperature, α is the temperature coefficient, and the temperature coefficient = 0.02 / ℃.

4. The baby diaper stain detection system based on image recognition according to claim 1, characterized in that: In step S4, the SAP interference peak is at radial frequency 0.15-0.3, and a Gaussian notch filter is applied to the amplitude spectrum: in, , is the coordinate of the center point in the frequency domain obtained by FFT transformation, , is the coordinate of the SAP interference signal in the frequency domain, is the decay rate of the Gaussian function.

5. The baby diaper stain detection system based on image recognition according to claim 1, characterized in that: In step S5, the specific process of auxiliary verification of ultraviolet image and visible light image is as follows: S51, when a suspected stain is detected in the ultraviolet image, its minimum circumscribed rectangle ROI is extracted, and the corresponding area of ​​the visible light image is located through the spatial mapping matrix; S52, extracting texture features from the ROI region of the visible light image, and calculating the LBP-TOP features of the ROI region: Among them, P is the number of neighborhood pixels, R is the neighborhood radius, is the gray value of the center pixel, is the gray value of the pth neighborhood pixel, is a sign function with only two results: 0 and 1. for Assignment; After calculation, a 256-dimensional feature histogram is generated, and the contrast and energy features of the multi-directional gray-level co-occurrence matrix are extracted to obtain the texture features; S53, color space conversion, converting the RGB image to the HSV space, and calculating the mean and standard deviation of the saturation S and the brightness V of the ROI area; Where N is the total number of pixels in the ROI area, is the saturation value of the i-th pixel, is the brightness value of the i-th pixel, , It is the average value of saturation and brightness in the ROI area, reflecting the concentration and brightness level of the overall color; represents the standard deviation of saturation, It indicates the standard deviation of brightness and detects abnormal spots based on the difference in values; S54, dual-modal decision fusion, input UV confidence ,Visible light feature score , the final stain judgment result is output by weighting, and the confidence weighting rule is: is the UV confidence, obtained by normalizing the UV image. is the output probability of the SVM classifier of the visible light feature, is the confidence-weighted score, and is the optimal weight determined by the ROC curve, =0.7, =0.

3.

6. The baby diaper stain detection system based on image recognition according to claim 1, characterized in that: In step S1, a pulsed ultraviolet light source is set at a low angle on the side of the stain detection conveyor belt, and the incident angle to the diaper surface is 15°±2°. The incident light is divided into two paths by a dichroic prism. One path is 395nm narrow-band ultraviolet light. A UV-CCD camera is set above the ultraviolet irradiation area, and a 400nm long-pass filter is added to the UV-CCD camera lens. The other path is visible light, and a CMOS camera is set above the visible light irradiation area. The UV-CCD camera and the CMOS camera start acquisition asynchronously, so that the two cameras collect images of the same area during the operation of the stain detection conveyor belt.

7. The baby diaper stain detection system based on image recognition according to claim 1, characterized in that: A fail-safe mechanism was established. First, background drift was monitored and a baseline image was automatically acquired every 30 minutes. > Set a threshold value to trigger an alarm and remind production personnel that the background image, which serves as a pollution-free benchmark, has a large error. A compensation failure fallback mechanism is also set up. When the diaper surface temperature is greater than 50°C or the thickness fluctuation measured by the thickness measurement roller is greater than 20%, the production personnel are reminded that the current production temperature is too high or the thickness difference of the diapers is too large.

Images