Parcel void ratio identification system, device and method based on computer vision

Abstract

The invention discloses a parcel void ratio recognition system, device and method based on computer vision. The system comprises an image acquisition module, an image preprocessing module, a clustering processing module, an area data output module, a three-dimensional laser scanning module and a void ratio data output module, and the areas of packages and products are obtained through the image acquisition module, the image preprocessing module, the clustering processing module and the area data output module. The height of the package and the height of the product are obtained through the three-dimensional laser scanning module, and finally the void ratio of the package is obtained. According to the invention, real-time identification of the void ratio of the package is realized, and the efficiency and accuracy of package analysis are obviously improved. According to the method, accurate processing of the parcel image is realized through Gaussian filtering denoising and clustering methods, and the background, the package and the product in the image can be effectively distinguished, so that more accurate area information is obtained. The method can effectively evaluate the rationality of the package, and has good expansibility and adaptability.

Description

Technical Field

[0001] The present invention relates to the field of computer vision technology, and in particular to a package void ratio recognition system, device and method based on computer vision. Background Art

[0002] Before products are shipped out of the warehouse, they need to be wrapped with paper, plastic film, corrugated paper and other materials, which are called packaging. Packaging is crucial to the circulation and safety of goods, but it also brings about the problem of overpackaging, which leads to resource waste and environmental pollution. Overpackaging means that the packaging of goods has too many layers, too much material, too large volume, and too high cost, which is inconsistent with the value and functional requirements of the goods, which increases costs and consumer burdens and violates the concept of green and sustainable development.

[0003] At present, the packaging inspection technologies on the market mainly include manual inspection and automated inspection. Manual inspection relies on the experience of the inspectors, is inefficient and easily affected by subjective factors, and cannot meet the needs of large-scale production. Although automated inspection technology has improved the inspection efficiency and accuracy to a certain extent, most of the existing technologies are based on traditional mechanical, electrical and sensor technologies, which are difficult to meet the complex and changing packaging inspection needs. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a package void ratio identification system, device and method based on computer vision.

[0005] The first aspect of the present invention provides a package void ratio recognition system based on computer vision, comprising an image acquisition module, an image preprocessing module, a clustering processing module, an area data output module, a three-dimensional laser scanning module and a void ratio data output module;

[0006] The image acquisition module is used to acquire two-dimensional image information of the package to obtain first package information;

[0007] The image preprocessing module is used to perform image preprocessing on the first package information to obtain second package information;

[0008] The cluster processing module is used to classify the pixels of the second package information, and obtain the first area, the second area and the third area, which respectively represent the background, the packaging and the product of the package;

[0009] The area data output module is used to calculate the area information of the second area and the third area;

[0010] The three-dimensional laser scanning module is used to obtain height information of the second area and the third area;

[0011] The void ratio data output module is used to calculate the void ratio of the package based on the area information and height information.

[0012] Furthermore, the image acquisition module includes at least one camera and a fill light aperture; the camera is used to acquire two-dimensional image information of the package, and the fill light aperture is used to adjust the exposure level of the package.

[0013] Furthermore, the image preprocessing module includes a Gaussian kernel generation module, a Gaussian filtering module and an image integration module;

[0014] The Gaussian kernel generation module is used to generate a Gaussian kernel;

[0015] The Gaussian filtering module is used to perform a Gaussian filtering operation on the first package information using the Gaussian kernel;

[0016] The image integration module is used to perform image integration on the first package information after the Gaussian filtering operation to obtain second package information.

[0017] Furthermore, the cluster processing module classifies pixels of the second package information through the following steps:

[0018] Determining the cluster center of each cluster of the second package information; the clusters include background, packaging and product;

[0019] Calculating the Euclidean distance from each pixel of the second package information to the cluster center;

[0020] According to the Euclidean distance of each pixel, the pixel is assigned to the cluster with the lowest Euclidean distance;

[0021] The pixels under each cluster are combined to form a first region, a second region, and a third region representing the background, packaging, and product of the package.

[0022] Furthermore, before the step of combining the pixels in each cluster, the following steps are also included:

[0023] Calculate the mean of the feature vectors of all pixels in each cluster; the feature vectors include colors;

[0024] According to the feature vector mean, the cluster center of each cluster is moved;

[0025] Return to the step of calculating the Euclidean distance from each pixel of the second package information to the cluster center, and re-perform the pixel allocation operation until a preset number of iterations is reached or the moving distance of the cluster center is less than a preset threshold.

[0026] Furthermore, the method further comprises the following steps:

[0027] The silhouette coefficient is used to judge the clustering effect of the clustering processing module; the silhouette coefficient expression is:

[0028]

[0029] Among them, a(i) is the average distance from the i-th pixel to other pixels in the same cluster, b(i) is the average distance from the i-th sample to all pixels in other clusters, and S(i) represents the silhouette coefficient of the i-th sample.

[0030] Furthermore, the three-dimensional laser scanning module includes a three-dimensional infrared laser scanner, which emits infrared laser to the package and determines the height information of the second area and the third area according to the flight time of the reflected laser.

[0031] Furthermore, the void ratio data output module calculates the package void ratio by the following expression:

[0032]

[0033] Among them, Si and Li represent the area information and height information of the third area; S and L represent the area information and height information of the second area, i represents the i-th package, and n represents the total number of packages.

[0034] The second aspect of the present invention provides an electronic device for implementing the package void ratio identification system based on computer vision as described in the first aspect, comprising a measurement platform, a processor and a display;

[0035] The measuring platform comprises a base, a camera, an adjustable aperture and a three-dimensional infrared laser scanner;

[0036] The base is used to place the package; a rotating platform and a rotating pole are provided on the base; the rotating platform is connected to the rotating pole to rotate the package and / or the rotating pole; a horizontal movable bracket is provided on the rotating pole;

[0037] The camera and the adjustable aperture realize the function of the image acquisition module; the adjustable aperture is connected to the horizontal movable bracket through an aperture clamp; the camera is fixed to the horizontal movable bracket through a camera clamp; the horizontal movable bracket can move the camera and / or the adjustable aperture in the horizontal direction;

[0038] The three-dimensional infrared laser scanner realizes the function of the three-dimensional laser scanning module; the three-dimensional infrared laser scanner is connected to the rotating vertical pole through a fixed slider; the fixed slider can move the three-dimensional infrared laser scanner in the horizontal direction;

[0039] The processor integrates the image preprocessing module, clustering processing module, area data output module and void ratio data output module to realize the corresponding functions of the image preprocessing module, clustering processing module, area data output module and void ratio data output module, and outputs the void ratio of the package;

[0040] The display is used to display the package void ratio.

[0041] The third aspect of the present invention discloses a package void ratio recognition method based on computer vision, which is applied to the package void ratio recognition system based on computer vision, and comprises the following steps:

[0042] Get the two-dimensional image information of the package;

[0043] Determine an area representing a background, an area representing a package, and an area representing a product in the two-dimensional image information;

[0044] Calculate the area information of the area representing the package and the area representing the product;

[0045] Acquire the height information of the area representing the packaging and the area representing the product by using 3D laser scanning technology;

[0046] The void ratio of the package is calculated based on the area information and height information of the area representing the package and the area representing the product.

[0047] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above method.

[0048] The embodiments of the present invention have the following beneficial effects: The present invention provides a package void ratio recognition system, device and method based on computer vision, which realizes real-time recognition of package void ratio and significantly improves the efficiency and accuracy of package analysis. The present invention realizes accurate processing of package images through Gaussian filtering denoising and clustering methods, and can effectively distinguish the background, package and product in the image, so as to obtain more accurate area information. The present invention can effectively evaluate the rationality of packaging, has good scalability and adaptability, and can be widely used in packaging analysis of various products, helping enterprises to optimize packaging design, reduce material waste and promote sustainable development. The present invention is particularly suitable for fast-moving consumer goods, electronic products and other fields.

[0049] Additional aspects and advantages of the present invention will be given in part in the following description, and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 It is a schematic diagram of the implementation process of a package void ratio recognition system based on computer vision of the present invention.

[0052] Figure 2 It is a schematic diagram of the electronic device architecture of the present invention.

[0053] Figure 3 It is a schematic diagram of the measurement platform architecture in the electronic device of the present invention.

[0054] Figure 4 It is a schematic flow chart of the steps of the package void ratio identification method based on computer vision of the present invention.

[0055] Figure numerals: 1-fixed slider, 2-three-dimensional infrared laser scanner, 3-rotating vertical pole, 4-camera holder, 5-camera, 6-aperture holder, 7-adjustable aperture, 8-rotating platform, 9-base, 10-horizontally movable bracket. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] Before products are shipped out of the warehouse, they need to be wrapped with paper, plastic film, corrugated paper and other materials, which are called packaging. Packaging is crucial to the circulation and safety of goods, but it also brings about the problem of overpackaging, which leads to resource waste and environmental pollution. Overpackaging means that the packaging of goods has too many layers, too much material, too large volume, and too high cost, which is inconsistent with the value and functional requirements of the goods, which increases costs and consumer burdens and violates the concept of green and sustainable development.

[0058] At present, the packaging inspection technologies on the market mainly include manual inspection and automated inspection. Manual inspection relies on the experience of the inspectors, is inefficient and easily affected by subjective factors, and cannot meet the needs of large-scale production. Although automated inspection technology has improved the inspection efficiency and accuracy to a certain extent, most of the existing technologies are based on traditional mechanical, electrical and sensor technologies, which are difficult to meet the complex and changing packaging inspection needs.

[0059] As an emerging automated detection technology, computer vision technology has shown great potential in the field of packaging detection with its high precision, speed and flexibility. To this end, the embodiment of the present invention proposes a package void ratio recognition system, device and method based on computer vision, aiming to solve the problems existing in the above situation.

[0060] like Figure 1 As shown, the first embodiment of the present invention provides a package void ratio recognition system based on computer vision, including an image acquisition module, an image preprocessing module, a clustering processing module, an area data output module, a three-dimensional laser scanning module and a void ratio data output module. Among them, the image acquisition module is used to acquire two-dimensional image information of the package to obtain first package information; the image preprocessing module is used to perform image preprocessing on the first package information to obtain second package information; the clustering processing module is used to classify pixels of the second package information to obtain a first area, a second area and a third area representing the background, packaging and product of the package respectively; the area data output module is used to calculate the area information of the second area and the third area; the three-dimensional laser scanning module is used to obtain the height information of the second area and the third area; the void ratio data output module is used to calculate the void ratio of the package according to the area information and the height information.

[0061] The embodiment of the present invention realizes real-time identification of the void ratio of the package, and significantly improves the efficiency and accuracy of packaging analysis.

[0062] The implementation effects of each module of the system of the present invention are described in detail below:

[0063] Image acquisition module: In the embodiment of the present invention, the image acquisition module includes at least one camera and a fill light aperture; the camera is used to collect two-dimensional image information of the package, and the fill light aperture is used to adjust the exposure level of the package. In the embodiment of the present invention, the camera is set at the top of the package, and the collected two-dimensional image information is a top view of the package. As a preferred embodiment, multiple cameras can be set, and multiple cameras at different positions can collect two-dimensional image information of the package from different angles to offset the deviation caused by viewing angle and light. The adjustable aperture can control the light environment around the package so that the package can be photographed in an environment with appropriate exposure. The image acquisition module of the present invention can obtain clear images of the package under different environmental conditions through the cooperation of the camera and the fill light aperture, providing a high-quality image data foundation for subsequent image processing.

[0064] Image preprocessing module: Since the two-dimensional image information collected by the image acquisition module may be affected by factors such as uneven illumination and noise interference, it is necessary to preprocess the image to improve the accuracy of the subsequent clustering algorithm. The image preprocessing module of the embodiment of the present invention includes a Gaussian kernel generation module, a Gaussian filter module and an image integration module.

[0065] The Gaussian kernel generation module is used to generate a Gaussian kernel. The Gaussian kernel size determined in the embodiment of the present invention is (2×1+1)×(2×1+1), which is represented as a 3x3 matrix. The center point of the matrix is ​​(0, 0), and the radius r=1, then the x and y value ranges of the Gaussian kernel are [-1, 1].

[0066] The Gaussian filter module is used to perform a Gaussian filter operation on the first package information using a Gaussian kernel. For each pixel point (x, y) in the Gaussian kernel, the corresponding Gaussian value is calculated by the following formula:

[0067]

[0068] Here, σ is the standard deviation of the Gaussian distribution.

[0069] The expression for filtering the Gaussian kernel is:

[0070] l′(x,y)=(l*G)(x,y)=∑ m＝-k ∑ n＝-k l(x+m,y+n)·G(m,n);

[0071] Wherein, l(x,y) is the pixel value of the original image at point (x,y)(x,y), l'(x,y) is the pixel value of the image after Gaussian filtering at point (x,y)(x,y), G(m,n) is the value of the Gaussian kernel at position (m,n)(m,n), and k is the radius of the Gaussian kernel, which is 1 in the present invention. m represents the horizontal offset of the convolution kernel, and its value range is [-k,k]. n represents the vertical offset of the convolution kernel, and its value range is also [-k,k].

[0072] The image integration module is used to perform image integration on the first package information after the Gaussian filter operation to obtain the second package information. If the previous image acquisition module is set with multiple cameras, multiple two-dimensional image information wrapped after the Gaussian filter operation will be obtained here. The image integration module integrates the multiple two-dimensional images into one two-dimensional image as the second package information, which can eliminate the influence of factors such as light and viewing angle on image recognition.

[0073] Clustering processing module: In the embodiment of the present invention, RGB color value is selected as the feature vector, and the second package information is placed in the RGB color space for clustering operation. In another embodiment, parameters such as texture features, geometric moment features and their combinations can also be selected as feature vectors for clustering operation. After the second package information is placed in the RGB color space, each pixel in the image will be represented in the form of a two-dimensional array.

[0074] In the embodiment of the present invention, a K-Means clustering method is used to perform a clustering operation on pixel points, which specifically includes the following steps:

[0075] First, the cluster centers of each cluster of the second package information are determined. The clusters set in the embodiment of the present invention include background, packaging and product, which respectively represent the background area, packaging area and product area in the image.

[0076] Next, the Euclidean distance from each pixel of the second package information to the cluster center is calculated. Specifically, the Euclidean distance d(x i ,c k ) is used to measure the similarity between each pixel (feature point) in the image and the current cluster center. Specifically, the clustering algorithm calculates the distance from each pixel to all cluster centers and assigns the pixel to the cluster represented by the cluster center with the shortest distance.

[0077] The Euclidean distance expression is:

[0078]

[0079] Where x i is the feature vector of the current pixel (RGB color value in the embodiment of the present invention), c k is the kth cluster center. In this way, the clustering algorithm can effectively classify each pixel in the image into product, background or packaging, and then form background area, packaging area and product area.

[0080] In some embodiments, the clustering algorithm further includes an iterative optimization step. Specifically, the following steps are included:

[0081] Calculate the mean of the feature vectors of all pixels in each cluster; the feature vector includes color;

[0082] Move the cluster center of each cluster according to the mean of the eigenvector;

[0083] Return to the step of calculating the Euclidean distance from each pixel of the second package information to the cluster center, and re-perform the pixel allocation operation until a preset number of iterations is reached or the moving distance of the cluster center is less than a preset threshold.

[0084] In each iteration, the step of updating the cluster centers aims to make the clustering more precise by calculating the mean of all pixels belonging to each cluster. This process ensures that the cluster center constantly moves to the "center" of the pixel distribution of the category to which it belongs, so as to better represent the category.

[0085] The updated cluster center expression is:

[0086]

[0087] Where N k is assigned to cluster C k The number of pixels, x i is the feature vector of these pixels. After updating the cluster center, the clustering algorithm will calculate the distance from each pixel to the new cluster center again and redistribute the cluster. This process will continue to iterate until the moving distance of the cluster center is less than the preset threshold or the predetermined number of iterations is reached.

[0088] In some embodiments, after completing the clustering operation on the pixel points, a step of evaluating the clustering effect is also included. The embodiment of the present invention uses the silhouette coefficient to judge the clustering effect of the clustering processing module; the silhouette coefficient expression is:

[0089]

[0090] Among them, a(i) is the average distance from the i-th pixel to other pixels in the same cluster, b(i) is the average distance from the i-th sample to all pixels in other clusters, and S(i) represents the silhouette coefficient of the i-th sample. The larger the silhouette coefficient, the better the effect of the clustering algorithm, and a negative value indicates that the clustering effect is very poor.

[0091] Area data output module: The area data output module of the embodiment of the present invention outputs the area information of the actual packaging area and the product area after K-Means clustering processing, and visualizes the clustering, converts the area in pixel units into actual physical units, and obtains the area S of the third area representing the product. i =A1, indicating that the area of ​​the second region of the package S=A2.

[0092] 3D laser scanning module: The embodiment of the present invention uses a 3D infrared laser scanner to obtain depth information of the collected packaging image, and determines the distance of the light by emitting infrared laser pulses and measuring the flight time (Time of Flight) required for them to reflect back from the package, thereby obtaining the depth information of each position of the package. The height of the second area representing the package is L, and the depth of the third area representing the product is L i , providing basic data for subsequent void fraction calculation.

[0093] Void ratio data output module: The void ratio data output module of the embodiment of the present invention calculates the package void ratio by the following expression:

[0094]

[0095] Among them, Si and Li represent the area information and height information of the third area; S and L represent the area information and height information of the second area, i represents the i-th package, and n represents the total number of packages.

[0096] The beneficial effects of the embodiments of the present invention are:

[0097] 1. Automatically identify and calculate packaging void ratio, significantly improving the efficiency and accuracy of packaging analysis.

[0098] 2. Through Gaussian filtering denoising and K-Means clustering method, accurate processing of packaging images is achieved, which can effectively distinguish the background area, packaging area and product area, thereby obtaining more accurate area information.

[0099] 3. It can output the actual packaging area and product area information according to the height of the image sensor and the relevant ratio of the platform area, providing reliable basic data for the subsequent packaging void ratio calculation.

[0100] 4. It realizes real-time identification of packaging void ratio and can also provide data support for packaging design optimization. By analyzing the area and height information of products and packaging, it can effectively evaluate the rationality of packaging, help enterprises optimize packaging design, reduce material waste, and promote sustainable development.

[0101] The second embodiment of the present invention provides an electronic device for implementing the package void ratio recognition system based on computer vision of the first embodiment. Figure 2 As shown, the device includes a measuring platform, a processor and a display, and the measuring platform, the processor and the display are connected via a communication serial bus.

[0102] like Figure 3 As shown, the measurement platform includes a base, a camera, an adjustable aperture and a three-dimensional infrared laser scanner.

[0103] The base is used to place the package; a rotating platform and a rotating pole are provided on the base; the rotating platform is connected to the rotating pole to rotate the package and / or the rotating pole; a horizontal movable bracket is provided on the rotating pole.

[0104] The camera and the adjustable aperture realize the function of the image acquisition module; the adjustable aperture is connected to the horizontal movable bracket through the aperture clamp; the camera is fixed to the horizontal movable bracket through the camera clamp; the horizontal movable bracket can move the camera and / or the adjustable aperture in the horizontal direction to realize multi-angle package recognition.

[0105] The three-dimensional infrared laser scanner realizes the function of a three-dimensional laser scanning module; the three-dimensional infrared laser scanner is connected to the rotating vertical pole through a fixed slider; the fixed slider can enable the three-dimensional infrared laser scanner to move in the horizontal direction.

[0106] The processor integrates an image preprocessing module, a clustering processing module, an area data output module and a void ratio data output module to realize the corresponding functions of the image preprocessing module, the clustering processing module, the area data output module and the void ratio data output module, and outputs the package void ratio.

[0107] The display is used to show the void ratio of the package, and can also show intermediate information such as area data and depth data.

[0108] The third embodiment of the present invention discloses a package void ratio recognition method based on computer vision, which is applied to the package void ratio recognition system based on computer vision of the first embodiment. Figure 4 As shown, the following steps are included:

[0109] S1. Obtaining two-dimensional image information of the package;

[0110] S2. Determine the area representing the background, the area representing the packaging, and the area representing the product in the two-dimensional image information;

[0111] S3. Calculate the area information of the packaging area and the product area;

[0112] S4. Obtaining height information of the respective areas representing the packaging and the product through three-dimensional laser scanning technology;

[0113] S5. Calculate the void ratio of the package based on the area information and height information of the area representing the package and the area representing the product.

[0114] The functions specifically implemented by the embodiment of the method are the same as those of the above-mentioned system embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned system.

[0115] This embodiment also provides a computer program product. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement the package void ratio identification method based on computer vision provided in the above embodiment.

[0116] Those skilled in the art will appreciate that the modules in the devices in the embodiments of the present invention may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments of the present invention may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including corresponding claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including corresponding claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0117] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0118] In addition, each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. In particular, for embodiments such as devices and equipment, since they are basically similar to method embodiments, the relevant parts can refer to the partial description of the method embodiments. The embodiments of the devices and equipment described above are merely schematic, wherein the modules, units, etc. described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed in multiple places, such as nodes of a system network. Specifically, some or all of the modules and units may be selected according to actual needs to achieve the purpose of the above-mentioned embodiment scheme. Those skilled in the art can understand and implement it without paying creative labor.

[0119] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", 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 any one or more embodiments or examples in a suitable manner.

[0120] In addition, the terms "first", "second", etc. used in the embodiments of the present invention are only used for descriptive purposes and should not be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated in the present embodiment. Therefore, the features defined by the terms "first", "second", etc. in the embodiments of the present invention can explicitly or implicitly indicate that the embodiment includes at least one of the features. In the description of the present invention, the word "multiple" means at least two or two or more, such as two, three, four, etc., unless otherwise clearly and specifically defined in the embodiments.

[0121] In the embodiments of the present invention, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present invention may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context in the specific embodiment.

[0122] Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and cannot be construed as limitations of the present invention, and those of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention. Those skilled in the art will readily come to think of other embodiments of the present invention after considering the specification and practicing the present invention. This application is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art that are not disclosed in the present invention. The specification and embodiments are intended to be exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

Claims

1. A package void ratio recognition system based on computer vision, characterized in that: It includes an image acquisition module, an image preprocessing module, a clustering processing module, an area data output module, a three-dimensional laser scanning module and a void ratio data output module; The image acquisition module is used to acquire two-dimensional image information of the package to obtain first package information; The image preprocessing module is used to perform image preprocessing on the first package information to obtain second package information; The cluster processing module is used to classify the pixels of the second package information, and obtain the first area, the second area and the third area, which respectively represent the background, the packaging and the product of the package; The area data output module is used to calculate the area information of the second area and the third area; The three-dimensional laser scanning module is used to obtain height information of the second area and the third area; The void ratio data output module is used to calculate the void ratio of the package based on the area information and height information.

2. A package void ratio recognition system based on computer vision according to claim 1, characterized in that: The image acquisition module includes at least one camera and a fill light aperture; the camera is used to acquire two-dimensional image information of the package, and the fill light aperture is used to adjust the exposure level of the package.

3. The package void ratio recognition system based on computer vision according to claim 1, characterized in that: The image preprocessing module includes a Gaussian kernel generation module, a Gaussian filtering module and an image integration module; The Gaussian kernel generation module is used to generate a Gaussian kernel; The Gaussian filtering module is used to perform a Gaussian filtering operation on the first package information using the Gaussian kernel; The image integration module is used to perform image integration on the first package information after the Gaussian filtering operation to obtain second package information.

4. The computer vision-based package void ratio recognition system according to claim 1, characterized in that: The cluster processing module classifies pixels of the second package information through the following steps: Determining the cluster center of each cluster of the second package information; the clusters include background, packaging and product; Calculating the Euclidean distance from each pixel of the second package information to the cluster center; According to the Euclidean distance of each pixel, the pixel is assigned to the cluster with the lowest Euclidean distance; The pixels under each cluster are combined to form a first region, a second region, and a third region representing the background, packaging, and product of the package.

5. The computer vision-based package void ratio recognition system according to claim 4, characterized in that: Before the step of combining the pixel points in each cluster, the following steps are also included: Calculate the mean of the feature vectors of all pixels in each cluster; the feature vectors include colors; According to the feature vector mean, the cluster center of each cluster is moved; Return to the step of calculating the Euclidean distance from each pixel of the second package information to the cluster center, and re-perform the pixel allocation operation until a preset number of iterations is reached or the moving distance of the cluster center is less than a preset threshold.

6. A computer vision-based package void ratio identification system according to claim 5, characterized in that: The following steps are also included: The silhouette coefficient is used to judge the clustering effect of the clustering processing module; the silhouette coefficient expression is: Among them, a(i) is the average distance from the i-th pixel to other pixels in the same cluster, b(i) is the average distance from the i-th sample to all pixels in other clusters, and S(i) represents the silhouette coefficient of the i-th sample.

7. The package void ratio identification system based on computer vision according to claim 1, characterized in that: The three-dimensional laser scanning module includes a three-dimensional infrared laser scanner, which emits infrared laser to the package and determines the height information of the second area and the third area according to the flight time of the reflected laser.

8. The package void ratio identification system based on computer vision according to claim 1, characterized in that: The void ratio data output module calculates the package void ratio by the following expression: Among them, Si and Li represent the area information and height information of the third area; S and L represent the area information and height information of the second area, i represents the i-th package, and n represents the total number of packages.

9. An electronic device for implementing the package void ratio identification system based on computer vision according to any one of claims 1 to 8, characterized in that: Includes a measurement platform, a processor, and a display; The measuring platform comprises a base, a camera, an adjustable aperture and a three-dimensional infrared laser scanner; The base is used to place the package; a rotating platform and a rotating pole are provided on the base; the rotating platform is connected to the rotating pole to rotate the package and / or the rotating pole; a horizontal movable bracket is provided on the rotating pole; The camera and the adjustable aperture realize the function of the image acquisition module; the adjustable aperture is connected to the horizontal movable bracket through an aperture clamp; the camera is fixed to the horizontal movable bracket through a camera clamp; the horizontal movable bracket can move the camera and / or the adjustable aperture in the horizontal direction; The three-dimensional infrared laser scanner realizes the function of the three-dimensional laser scanning module; the three-dimensional infrared laser scanner is connected to the rotating vertical pole through a fixed slider; the fixed slider can move the three-dimensional infrared laser scanner in the horizontal direction; The processor integrates the image preprocessing module, clustering processing module, area data output module and void ratio data output module to realize the corresponding functions of the image preprocessing module, clustering processing module, area data output module and void ratio data output module, and outputs the void ratio of the package; The display is used to display the package void ratio.

10. A package void ratio recognition method based on computer vision, applied to a package void ratio recognition system based on computer vision as claimed in any one of claims 1 to 8, characterized in that: The following steps are involved: Get the two-dimensional image information of the package; Determine an area representing a background, an area representing a package, and an area representing a product in the two-dimensional image information; Calculate the area information of the area representing the package and the area representing the product; Acquire the height information of the area representing the packaging and the area representing the product by using 3D laser scanning technology; The void ratio of the package is calculated based on the area information and height information of the area representing the package and the area representing the product.

Images