Automatic color difference detection system and working method

Abstract

An automatic color difference detection system comprises a detection area, a light source system, a camera shooting assembly, a transmission system, an image acquisition module, a feature extraction module and a classification module, the light source system is used for providing environment illumination for a thin film in the detection process, the camera shooting assembly is used for acquiring a thin film surface image, and the transmission system is used for controlling movement of the thin film in the detection area. The camera component can completely scan the surface of the film; the image acquisition module is used for preprocessing an image acquired from the camera component, and the feature extraction module is used for extracting color features of each pixel point from the preprocessed image; and the classification module carries out chromatic aberration judgment on each pixel point in sequence according to a decision boundary learned by the training sample, and outputs a detection result of each pixel point. By combining the light source system, the camera shooting assembly, the transmission system, the image acquisition module, the feature extraction module and the classification module, automatic film detection is realized, and the color difference of the film can be detected in real time.

Description

Technical Field

[0001] The present invention relates to the field of film detection, and in particular to an automatic color difference detection system and a working method. Background Art

[0002] In the prior art, the color difference detection of the film mainly relies on the standard color difference comparison method, that is, the color of the product is compared with the preset standard color one by one by using a semi-automatic detection method, and the color component of each pixel or area is measured to determine whether its color difference is within an acceptable range. It mainly relies on a fixed standard color difference threshold to determine whether there is a color difference defect. The above method has the following defects: the traditional color difference detection method is very sensitive to the deviation of a certain color component, resulting in that even if the overall color difference is within an acceptable range, as long as a certain component has a large deviation, the system will determine that the point or area is unqualified. This method ignores the relationship between multiple color components, resulting in a high misjudgment rate. At the same time, traditional color difference detection usually only focuses on simple numerical differences, such as the difference between a single color component and the standard value. The multidimensionality of color features is not fully considered, resulting in the inability to accurately distinguish subtle color changes. Therefore, in order to solve the above problems, an automated color difference detection system and working method are needed. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention proposes an automatic color difference detection system and a working method, wherein a specific technical solution of an automatic color difference detection system is as follows:

[0004] An automated color difference detection system, characterized in that:

[0005] It includes a detection area, a light source system, a camera assembly, a transmission system, an image acquisition module, a feature extraction module and a classification module;

[0006] The light source system, the camera assembly and the transmission system are arranged in the detection area;

[0007] The light source system is used to provide environmental lighting for the film during the inspection process, the camera assembly is used to collect the film surface image, and the transmission system is used to control the movement of the film in the inspection area so that the camera assembly can completely scan the film surface;

[0008] The image acquisition module is used to preprocess the image acquired from the camera assembly, and the feature extraction module is used to extract the color feature of each pixel from the preprocessed image;

[0009] The classification module performs color difference judgment on each pixel point in turn according to the decision boundary learned from the training samples, and outputs the detection result of each pixel point.

[0010] To better realize the present invention, it can be further:

[0011] A visualization analysis module is provided, and the visualization module is used to analyze the classification module. The visualization analysis module graphically displays the areas on the film where the color difference exceeds the standard and the specific deviation values.

[0012] Further:

[0013] The classification module adopts a support vector machine model.

[0014] Further:

[0015] An early warning module is provided, which is connected to the classification module. The early warning module is used to send out an alarm when the color difference exceeds the standard area, and record the abnormal data in the database.

[0016] Among them, a specific technical scheme of a working method of an automated color difference detection system is as follows:

[0017] A working method of an automated color difference detection system, characterized in that:

[0018] The steps include:

[0019] S1: The light source system is arranged in the detection area, and the light source system is started to provide uniform illumination for the film entering the detection area, and the light source system provides a light source for each detection point with uniform brightness;

[0020] S2: The transmission system drives the film to smoothly enter the detection area, and the moving speed of the transmission system matches the acquisition speed of the camera assembly, ensuring that the camera assembly seamlessly captures the film surface image;

[0021] S3: When the film moves to below the camera assembly, the camera assembly collects a surface image of the film in real time;

[0022] S4: the image acquisition module acquires the original image from the camera assembly, and the image acquisition module preprocesses the original image to obtain preprocessed data;

[0023] S5: the image acquisition module transmits the preprocessed data to the feature extraction module, and the feature extraction module extracts the color feature vector of each pixel;

[0024] S6: the feature extraction module inputs the color feature vector into the classification module, and the classification module classifies the pixel point as qualified or unqualified according to the learned decision boundary;

[0025] S7: the classification module outputs the detection value of each pixel to the visualization analysis module;

[0026] S8: The visual analysis module records the color difference value of each pixel point on the surface area of ​​the film, and the visual analysis module graphically displays the area on the film where the color difference exceeds the standard and the specific deviation value;

[0027] S9: When the visual analysis module detects that there is color abnormality in a key area of ​​the film, the visual analysis module activates the early warning module.

[0028] To better realize the present invention, it can be further:

[0029] The S5 comprises the following steps:

[0030] S51: The feature module converts the pixel points of the image from the RGB color space to the Lab color space;

[0031] S52: For each pixel, the color feature vector of the pixel in the Lab color space is expressed as:

[0032] x=(L,a,b)

[0033] S53: Normalize the color feature vector of each pixel to obtain a normalized color feature vector x′=(L′, a′, b′).

[0034] To better realize the present invention, it can be further:

[0035] The S6 comprises the following steps:

[0036] S61: The feature extraction module inputs the normalized color feature vector x′=(L′, a′, b′) into the classification module;

[0037] S62: The classification module adopts a support vector machine model, which is expressed as:

[0038]

[0039] Among them, α i is the Lagrange multiplier;

[0040] y i is the training sample label;

[0041] K(x i ,x) is the kernel function;

[0042] b is the bias term;

[0043]

[0044] σ is the width parameter of the kernel function, x i-x is the Euclidean distance between two color feature vectors;

[0045] S63: The classification module uses the output result of f(x) to classify the color difference of each pixel in turn. If f(x)>0, the pixel is judged to be qualified; if f(x)<0, the pixel is judged to be unqualified.

[0046] The beneficial effects of the present invention are as follows: the present invention realizes automated film detection by combining a light source system, a camera assembly, a transmission system, an image acquisition module, a feature extraction module and a classification module, and can detect film color difference in real time. By introducing a support vector module, a comprehensive analysis can be performed in a high-dimensional feature space based on color features to improve the accuracy and robustness of color difference detection. Through the learning ability of the classification module, the system can dynamically adapt to the color changes of the film and improve the flexibility and accuracy of detection. By setting a visual analysis module, the system can generate intuitive detection results, mark the color difference distribution on the film surface, and help operators quickly locate problem areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is the overall block diagram of the present invention;

[0048] Figure 2 It is the work flow chart of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In the description of the present invention, it should be noted that the terms "vertical", "upper", "lower", "horizontal", etc. indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific position, be constructed and operated in a specific position, and therefore cannot be understood as limiting the present invention. In addition, "first", "second", "third", and "fourth" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0051] like Figure 1As shown: an automated color difference detection system, comprising a detection area, a light source system, a camera assembly, a transmission system, an image acquisition module, a feature extraction module, a classification module, a visualization analysis module and an early warning module. The light source system, the camera assembly and the transmission system are arranged in the detection area. The light source system is used to provide environmental lighting for the film during the detection process, the camera assembly is used to collect the surface image of the film, and the transmission system is used to control the movement of the film in the detection area so that the camera assembly can completely scan the surface of the film.

[0052] The image acquisition module is used to preprocess the image acquired from the camera assembly, and the feature extraction module is used to extract the color feature of each pixel from the preprocessed image.

[0053] The classification module adopts a support vector machine model. The classification module determines the color difference of each pixel in turn according to the decision boundary learned from the training sample, and outputs the detection result of each pixel. The visualization module is used to analyze the classification module. The visualization analysis module graphically displays the area on the film where the color difference exceeds the standard and the specific deviation value. The early warning module is connected to the classification module. When the early warning module is used in an area where the color difference exceeds the standard, it will issue an alarm and record the abnormal data in the database.

[0054] like Figure 2 As shown, a specific technical scheme of a working method of an automated color difference detection system is as follows:

[0055] A working method of an automated color difference detection system comprises the following steps:

[0056] S1: The light source system is arranged in the detection area, and the light source system is started to provide uniform illumination for the film entering the detection area, and the light source system provides a light source for each detection point with uniform brightness;

[0057] S2: The transmission system drives the film to smoothly enter the detection area, and the moving speed of the transmission system matches the acquisition speed of the camera assembly, ensuring that the camera assembly seamlessly captures the film surface image;

[0058] S3: When the film moves to below the camera assembly, the camera assembly collects a surface image of the film in real time;

[0059] S4: the image acquisition module acquires the original image from the camera assembly, and the image acquisition module preprocesses the original image to obtain preprocessed data;

[0060] S5: the image acquisition module transmits the preprocessed data to the feature extraction module, and the feature extraction module extracts the color feature vector of each pixel;

[0061] Specifically: S5 includes the following steps:

[0062] S51: The feature module converts the pixel points of the image from the RGB color space to the Lab color space;

[0063] S52: For each pixel, the color feature vector of the pixel in the Lab color space is expressed as:

[0064] x=(L,a,b)

[0065] S53: Normalize the color feature vector of each pixel to obtain a normalized color feature vector x′=(L′, a′, b′).

[0066] S6: the feature extraction module inputs the color feature vector into the classification module, and the classification module classifies the pixel point as qualified or unqualified according to the learned decision boundary;

[0067] Specifically: S6 includes the following steps:

[0068] S61: The feature extraction module inputs the normalized color feature vector x′=(L′, a′, b′) into the classification module;

[0069] S62: The classification module adopts a support vector machine model, which is expressed as:

[0070]

[0071] Among them, α i is the Lagrange multiplier;

[0072] y i is the training sample label;

[0073] K(x i ,x) is the kernel function;

[0074] b is the bias term;

[0075]

[0076] σ is the width parameter of the kernel function, x i -x is the Euclidean distance between two color feature vectors;

[0077] S63: The classification module uses the output result of f(x) to classify the color difference of each pixel in turn. If f(x)>0, the pixel is judged to be qualified; if f(x)<0, the pixel is judged to be unqualified.

[0078] S7: the classification module outputs the detection value of each pixel to the visualization analysis module;

[0079] S8: The visual analysis module records the color difference value of each pixel point on the surface area of ​​the film, and the visual analysis module graphically displays the area on the film where the color difference exceeds the standard and the specific deviation value;

[0080] S9: When the visual analysis module detects that there is color abnormality in a key area of ​​the film, the visual analysis module activates the early warning module.

[0081] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0082] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. An automated color difference detection system, characterized in that: It includes a detection area, a light source system, a camera assembly, a transmission system, an image acquisition module, a feature extraction module and a classification module; The light source system, the camera assembly and the transmission system are arranged in the detection area; The light source system is used to provide environmental lighting for the film during the inspection process, the camera assembly is used to collect the film surface image, and the transmission system is used to control the movement of the film in the inspection area so that the camera assembly can completely scan the film surface; The image acquisition module is used to preprocess the image acquired from the camera assembly, and the feature extraction module is used to extract the color feature of each pixel from the preprocessed image; The classification module performs color difference judgment on each pixel point in turn according to the decision boundary learned from the training samples, and outputs the detection result of each pixel point.

2. The automatic color difference detection system according to claim 1, characterized in that: A visualization analysis module is provided, and the visualization module is used to analyze the classification module. The visualization analysis module graphically displays the areas on the film where the color difference exceeds the standard and the specific deviation values.

3. The automatic color difference detection system according to claim 2, characterized in that: The classification module adopts a support vector machine model.

4. The automatic color difference detection system according to claim 3, characterized in that: An early warning module is provided, which is connected to the classification module. The early warning module is used to send out an alarm when the color difference exceeds the standard area, and record the abnormal data in the database.

5. A working method of an automated color difference detection system according to any one of claims 1 to 4, characterized in that: The steps include: S1: The light source system is arranged in the detection area, and the light source system is started to provide uniform illumination for the film entering the detection area, and the light source system provides a light source for each detection point with uniform brightness; S2: The transmission system drives the film to smoothly enter the detection area, and the moving speed of the transmission system matches the acquisition speed of the camera assembly, ensuring that the camera assembly seamlessly captures the film surface image; S3: When the film moves to below the camera assembly, the camera assembly collects a surface image of the film in real time; S4: the image acquisition module acquires the original image from the camera assembly, and the image acquisition module preprocesses the original image to obtain preprocessed data; S5: the image acquisition module transmits the preprocessed data to the feature extraction module, and the feature extraction module extracts the color feature vector of each pixel; S6: the feature extraction module inputs the color feature vector into the classification module, and the classification module classifies the pixel point as qualified or unqualified according to the learned decision boundary; S7: the classification module outputs the detection value of each pixel to the visualization analysis module; S8: The visual analysis module records the color difference value of each pixel point on the surface area of ​​the film, and the visual analysis module graphically displays the area on the film where the color difference exceeds the standard and the specific deviation value; S9: When the visual analysis module detects that there is color abnormality in a key area of ​​the film, the visual analysis module activates the early warning module.

6. A working method of an automated color difference detection system according to claim 5, characterized in that: The S5 comprises the following steps: S51: The feature module converts the pixel points of the image from the RGB color space to the Lab color space; S52: For each pixel, the color feature vector of the pixel in the Lab color space is expressed as: x=(L,a,b) S53: Normalize the color feature vector of each pixel to obtain a normalized color feature vector x′=(L′, a′, b′).

7. The working method of the automatic color difference detection system according to claim 6, characterized in that: The S6 comprises the following steps: S61: The feature extraction module inputs the normalized color feature vector x′=(L′, a′, b′) into the classification module; S62: The classification module adopts a support vector machine model, which is expressed as: Among them, α i is the Lagrange multiplier; y i is the training sample label; K(x i ,x) is the kernel function; b is the bias term; σ is the width parameter of the kernel function, x i -x is the Euclidean distance between two color feature vectors; S63: The classification module uses the output result of f(x) to classify the color difference of each pixel in turn. If f(x)>0, the pixel is judged to be qualified; if f(x)<0, the pixel is judged to be unqualified.

Images