Board identification information image recognition system

Abstract

The invention relates to the field of image recognition, and discloses a board identification information image recognition system, which comprises a camera group, an image recognition module, an image analysis module and a service platform, and is characterized in that the image recognition module performs screening and image recognition on a plurality of groups of continuous images in sequence to analyze whether identification information exists in each image or not; an alarm signal is sent out under the condition that no identification information exists; and the image analysis module screens out at least one frame of image from each group of continuous images, carries out image recognition on the screened images, calculates the definition, and sends out an alarm signal under the condition that the definition does not reach the standard. In the invention, the image identification module is used for identifying and detecting whether the identification information is missed on the board surface, the image analysis module is used for carrying out image identification and definition detection, and an alarm signal is sent out in time under the conditions that the identification information is missed and the identification information is not clear, so that subsequent gypsum board processing is facilitated; and automatic identification and alarm of the identification information are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to a board surface marking information image recognition system. Background Art

[0002] During the production process of gypsum board, identification information will be printed on the board surface. If the printing volume is insufficient during the label printing process, resulting in unclear or missed labels, substandard gypsum boards will be produced. Therefore, the gypsum board surface labels need to be inspected after the printing process, mainly to check whether the board surface labels are missed and whether the label clarity meets the standards.

[0003] Currently, staff members mainly use cameras to inspect whether the labels on the gypsum board surface are missing and whether the label clarity is up to standard. Gypsum boards found to have missed labels need to be re-printed, and gypsum boards with label clarity that does not meet the standard will be discharged as waste boards.

[0004] Due to the high camera delay, the system is unable to respond in time and take corresponding measures for the substandard gypsum boards after the staff detects unclear markings. In addition, the lower surface of the gypsum board is usually printed with marking information, but during the transmission process, due to the difficulty in observation, the marking information on the lower surface of the gypsum board is easily overlooked, resulting in the failure to timely discover and deal with the missing or unclear markings on the gypsum board surface. Summary of the invention

[0005] To this end, the present invention provides a board surface identification information image recognition system, which effectively solves the technical problem in the prior art that when the identification on the gypsum board surface is missing or unclear, it cannot be discovered and processed in time.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions: a board surface marking information image recognition system, comprising:

[0007] A camera group, the camera groups are respectively facing the upper and lower board surfaces of the gypsum board, and the camera group collects each frame of the upper and lower board surfaces of the gypsum board during transportation to obtain multiple groups of continuous images;

[0008] An image recognition module is connected to the camera group for sequentially screening and recognizing multiple groups of continuous images to analyze whether there is identification information in each image, and sends an alarm signal if there is no identification information;

[0009] An image analysis module is connected to the camera group for communication. The image analysis module selects at least one frame of image from each group of continuous images, performs image recognition on the selected image and calculates the clarity, and issues an alarm signal when the clarity does not meet the standard;

[0010] A service platform, wherein the service platform is communicatively connected with the camera group and the image recognition module, and the service platform is used to start or shut down the camera group and the image recognition module to perform image recognition work, and receive analysis results of the image recognition module and the image analysis module.

[0011] Furthermore, the camera group includes two sets of relative camera structures, one set of camera structures is facing the upper board surface of the gypsum board, and the other set of camera structures is facing the lower board surface of the gypsum board;

[0012] The camera structure includes at least two groups of cameras facing the gypsum board surface, and the distance between each camera and the gypsum board surface is different.

[0013] Further, the length of the camera area on the transport plane where the gypsum board is located is obtained;

[0014] Corresponding to one of the gypsum boards to be identified, a plurality of frames of images in which at least a portion of the gypsum board surface is within the imaging area are selected based on the length of the imaging area to obtain a first image set;

[0015] At least one frame of image in the first image set is extracted in time sequence to obtain a second image set.

[0016] Further, the length of the camera area on the transport plane where the gypsum board is located is obtained;

[0017] Corresponding to one of the gypsum boards to be identified, a plurality of frames of images whose identification information is within the imaging area are selected based on the length of the imaging area to obtain a first image set;

[0018] At least one frame of image in the first image set is extracted in time sequence to obtain a second image set.

[0019] Furthermore, the second image set includes at least one frame of image whose identification information is located at a position directly facing the camera.

[0020] Furthermore, in the image recognition module, image recognition is performed on each frame of the first image set, and analysis is performed on whether identification information exists in each image;

[0021] When there is no identification information in each image, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0022] Furthermore, in the image analysis module, image recognition is performed on each frame of the second image set, and the clarity is calculated;

[0023] A clarity threshold is set, and when the clarity is lower than the clarity threshold, the alarm device is controlled to output an alarm signal through a preset alarm mode.

[0024] Furthermore, the image analysis module includes a database, in which standard images and standard image features with identification information are stored;

[0025] When performing image analysis on one of the frames of the image, image features are extracted and compared with standard image features, and the degree of overlap between the image features and the standard image features is calculated;

[0026] The overlap degree threshold is set. When the calculated overlap degree is lower than the overlap degree threshold, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0027] Furthermore, the service platform stores preset personnel identity information and login passwords, and the service platform can accept and display the analysis results of the image recognition module and the image analysis module;

[0028] The service platform obtains personnel identity information and login password;

[0029] Determine whether the personnel identity information meets the preset personnel identity information;

[0030] If the personnel identity information does not meet the preset personnel identity information, a warning signal is generated, and the alarm device is controlled to output an alarm signal through a preset alarm method;

[0031] The preset personnel identity information is the personnel identity information that is allowed to start the camera group.

[0032] Furthermore, an alarm signal is output according to a preset alarm mode, wherein the alarm mode includes a sound alarm mode and a light alarm mode;

[0033] The alarm modes corresponding to different alarm signals output different information.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] In the present invention, each frame of image during the transportation of the upper and lower surfaces of the gypsum board is collected by a camera group, an image recognition module is used to identify and detect whether identification information is missing on the board surface, and an image analysis module is used to perform image recognition and clarity detection on at least one frame of image having identification information. In the case that identification information is missing or unclear, an alarm signal is promptly issued to facilitate subsequent gypsum board processing, thereby realizing automatic identification and alarm of identification information. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0037] Figure 1 A structural block diagram of a board surface identification information image recognition system provided by an embodiment of the present invention;

[0038] Figure 2 is a flowchart of image recognition in an embodiment of the present invention;

[0039] Figure 3 A schematic diagram of the relative position of the gypsum board when the first and last frames of the first image set are collected in the first embodiment;

[0040] Figure 4 This is a schematic diagram of the relative position of the gypsum board when the first and last frames of the first image set are collected in the second embodiment. DETAILED DESCRIPTION

[0041] 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.

[0042] like Figure 1 and Figure 2 As shown, the present invention provides a board surface identification information image recognition system, including a camera group, an image recognition module, an image analysis module and a service platform.

[0043] The camera groups are respectively facing the upper and lower board surfaces of the gypsum board, and the camera groups collect each frame of image of the upper and lower board surfaces of the gypsum board during transportation to obtain multiple groups of continuous images.

[0044] The image recognition module is connected to the camera group for communication. The image recognition module sequentially screens and recognizes multiple groups of continuous images to analyze whether there is identification information in each image, and sends out an alarm signal if there is no identification information.

[0045] The image recognition module mainly detects whether there is any missing printing of identification information on the gypsum board.

[0046] The image analysis module is connected to the camera group for communication. The image analysis module selects at least one frame of image from each group of continuous images, performs image recognition on the selected image and calculates the clarity, and sends out an alarm signal when the clarity does not meet the standard.

[0047] The image analysis module mainly detects whether there is unclear identification information on the gypsum board.

[0048] The service platform is connected to the camera group and the image recognition module in communication. The service platform is used to start or shut down the camera group and the image recognition module to perform image recognition work, and receive analysis results from the image recognition module and the image analysis module.

[0049] In the present invention, each frame of image during the transportation of the upper and lower surfaces of the gypsum board is collected by a camera group, an image recognition module is used to identify and detect whether identification information is missing on the board surface, and an image analysis module is used to perform image recognition and clarity detection on at least one frame of image having identification information. In the case that identification information is missing or unclear, an alarm signal is promptly issued to facilitate subsequent gypsum board processing, thereby realizing automatic identification and alarm of identification information.

[0050] Since identification information is printed on the upper and lower surfaces of the gypsum board, a camera group is set at the upper and lower surfaces of the gypsum board. Specifically, the camera group includes two sets of opposing camera structures, one set of camera structures is facing the upper surface of the gypsum board, and the other set of camera structures is facing the lower surface of the gypsum board.

[0051] The upper and lower positions of the two sets of camera structures do not necessarily have to be facing each other. The position of the camera structure facing the lower surface of the gypsum board is hollowed out. Generally, the gypsum board is transported by transport rollers. At the position facing the camera structure, continuous transport rollers are not set, so the lower surface of the gypsum board is exposed in the camera area of ​​the camera structure, so as to capture the image of the lower surface of the gypsum board.

[0052] In order to improve the detection accuracy of the information represented by a single gypsum board, a method of multiple image acquisition and recognition is adopted to increase the number of recognition times and ensure that the identification information is correct. Specifically, the camera structure includes at least two groups of cameras facing the gypsum board surface, such as Figure 4 As shown, the distance from each camera to the gypsum board surface is different.

[0053] The main steps of image recognition include: image preprocessing, edge detection, feature extraction, etc. The extracted features are compared with standard image features to obtain the image recognition results.

[0054] In actual operation, C++ language can be used to write programs using Microsoft Visual Studio software, and image recognition algorithms can be used to identify identification information on images captured by the camera.

[0055] The image recognition process in the program is as follows:

[0056] First, write the necessary header files for the program, define and select the camera, initialize the camera, and loop to capture each frame of the camera. Then, identify and compare each captured frame of the image. Finally, judge the identified and compared information and issue an alarm if it is incorrect.

[0057] Since the gypsum boards are transported in blocks, different images may exist for different gypsum boards. In order to avoid redundant steps in the image recognition process, only a number of frames of images of the gypsum boards entering the camera area are recognized. Specifically, the image recognition module mainly adopts the following two embodiments for the image screening and recognition process;

[0058] First embodiment:

[0059] The image recognition module performs the following steps to screen and identify images:

[0060] Get the length of the camera area on the transport plane where the gypsum board is located;

[0061] Corresponding to one of the gypsum boards to be identified, a plurality of frames of images in which at least a portion of the gypsum board surface is within the imaging area are selected based on the length of the imaging area to obtain a first image set;

[0062] Performing image recognition on each frame of the first image set, and analyzing whether there is identification information in each image;

[0063] When there is no identification information in each image, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0064] Specifically, Figure 3 As shown, the dotted lines extending from the upper and lower cameras can be regarded as the camera area, the gypsum board in the form of a solid line can be regarded as the front end of the gypsum board just entering the camera area, and the gypsum board in the form of a dotted line can be regarded as the end of the gypsum board just leaving the camera area. Corresponding to one of the gypsum boards to be identified, starting from a frame image where the front end of the gypsum board just enters the camera area, until the end of the gypsum board leaves the camera area, each frame image captured by the camera during this period is collected to form a first image set.

[0065] After filtering out the first image set corresponding to the gypsum board, image recognition is performed on each frame of the first image set, and analysis is performed on whether there is identification information in each image. When there is no identification information in each image, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0066] In actual application, image recognition can be used to identify whether the gypsum board surface enters the camera area, so as to perform subsequent screening of the first image set.

[0067] Second embodiment:

[0068] The image recognition module performs the following steps to screen and identify images:

[0069] Get the length of the camera area on the transport plane where the gypsum board is located;

[0070] Corresponding to one of the gypsum boards to be identified, a plurality of frames of images whose identification information is within the imaging area are selected based on the length of the imaging area to obtain a first image set;

[0071] Performing image recognition on each frame of the first image set, and analyzing whether there is identification information in each image;

[0072] When there is no identification information in each image, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0073] Specifically, Figure 4 As shown, corresponding to one of the gypsum boards to be identified, starting from a frame image where the identification information on the gypsum board just enters the camera area, until the identification information on the gypsum board leaves the camera area, each frame image captured by the camera during this period is collected to form a first image set.

[0074] After filtering out the first image set corresponding to the gypsum board, image recognition is performed on each frame of the first image set, and analysis is performed on whether there is identification information in each image. When there is no identification information in each image, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0075] In actual application, image recognition can be used to identify whether the mark on the gypsum board enters the camera area, so as to perform subsequent screening of the first image set.

[0076] In the second embodiment, if the identification information is missing on a certain gypsum board, it means that it is impossible to detect the identification information on the corresponding gypsum board into the camera area. In actual applications, gypsum boards are generally transmitted at equal intervals. In the image screening process, the first image set can be screened out at a certain interval. If the first image set is not screened out within a certain period of time, it means that the identification information on the gypsum board is missing, and an alarm needs to be issued in time.

[0077] Since insufficient ink injection during logo information printing may result in unclear logo information, an image analysis module is used to select at least one frame of image from each group of continuous images, perform image recognition on the selected images and calculate the clarity, and issue an alarm signal if the clarity does not meet the standard.

[0078] In order to avoid redundant steps in the image clarity analysis, based on the first image set, only a plurality of frames of images are analyzed for clarity. Specifically, the image analysis module mainly adopts the following two embodiments for the clarity analysis:

[0079] First embodiment:

[0080] In the image analysis module, corresponding to one of the gypsum boards to be identified, a plurality of frames of images in which at least a portion of the gypsum board surface is within the camera area are screened out based on the length of the camera area to obtain a first image set, and at least one frame of the image in the first image set is extracted in chronological order to obtain a second image set;

[0081] Performing image recognition on each frame of the second image set and calculating the clarity;

[0082] Set a clarity threshold and, when the clarity is lower than the clarity threshold, control the alarm device to output an alarm signal through a preset alarm method.

[0083] Specifically, Figure 4 As shown, the dotted lines extending from the upper and lower cameras can be regarded as the camera area, the gypsum board in the form of a solid line can be regarded as the identification information on the gypsum board just entering the camera area, and the gypsum board in the form of a dotted line can be regarded as the identification information on the gypsum board just leaving the camera area. Corresponding to one of the gypsum boards to be identified, starting from a frame image where the front end of the gypsum board just enters the camera area, until the rear end of the gypsum board leaves the camera area, each frame image captured by the camera during this period is collected to form a first image set, and at least one frame image is extracted from the first image set to perform clarity analysis on the image.

[0084] The second image set includes at least one frame of image whose identification information is located at a position directly facing the camera.

[0085] The second image set contains at least one frame of image with identification information in the center. By performing clarity analysis on the image, an accurate clarity analysis result can be obtained, and then an alarm can be issued based on the clarity analysis result.

[0086] Second embodiment:

[0087] In the image analysis module, corresponding to one of the gypsum boards to be identified, a number of frames of images whose identification information is within the camera area are screened out based on the length of the camera area to obtain a first image set, and at least one frame of image in the first image set is extracted in chronological order to obtain a second image set;

[0088] Performing image recognition on each frame of the second image set and calculating the clarity;

[0089] Set a clarity threshold and, when the clarity is lower than the clarity threshold, control the alarm device to output an alarm signal through a preset alarm method.

[0090] Specifically, corresponding to one of the gypsum boards to be identified, starting from a frame image when the identification information on the gypsum board just enters the camera area, until the identification information on the gypsum board leaves the camera area, each frame image captured by the camera during this period is collected to form a first image set, and at least one frame image is extracted from the first image set to perform clarity analysis of the image.

[0091] The second image set includes at least one frame of image whose identification information is located at a position directly facing the camera.

[0092] The second image set contains at least one frame of image with identification information in the center. By performing clarity analysis on the image, an accurate clarity analysis result can be obtained, and then an alarm can be issued based on the clarity analysis result.

[0093] The clarity of the identification information in the image analysis module can be calculated and characterized by the Tenengrad gradient method, Laplacian gradient method and variance method. Taking the Tenengrad gradient algorithm as an example, it mainly includes: using the Sobel operator to calculate the gradients in the horizontal and vertical directions respectively. In the same image, the higher the gradient value, the clearer the image. After processing with the Sobel operator, the average grayscale value is obtained. The larger the grayscale value, the clearer the image and the higher the clarity. The grayscale value is used as a parameter to characterize the clarity.

[0094] In the image analysis module, in addition to identifying the clarity of the identification information, it is also possible to identify whether the identification information is correct. Specifically, the image analysis module includes a database, in which standard images and standard image features of the identification information are stored;

[0095] When performing image analysis on one of the frames of the image, image features are extracted and compared with standard image features, and the degree of overlap between the image features and the standard image features is calculated;

[0096] The overlap degree threshold is set. When the calculated overlap degree is lower than the overlap degree threshold, the alarm device is controlled to output an alarm signal through a preset alarm method.

[0097] In the present invention, the service platform can be a user terminal, and the user can control the camera group and the image recognition module to perform image recognition through the service platform, and can also observe the image information collected by the camera group and the analysis results of the image recognition module and the image analysis module on the service platform.

[0098] Specifically, the service platform stores preset personnel identity information and login passwords, and the service platform can accept and display the analysis results of the image recognition module and the image analysis module.

[0099] The specific steps for logging into the service platform are as follows:

[0100] The service platform obtains the personnel's identity information and login password;

[0101] Determine whether the personnel identity information meets the preset personnel identity information;

[0102] If the personnel identity information does not meet the preset personnel identity information, a warning signal is generated, and the alarm device is controlled to output an alarm signal through a preset alarm method;

[0103] Among them, the preset personnel identity information is the personnel identity information that is allowed to start the camera group.

[0104] In the above embodiments, if the person identity information satisfies the preset person identity information, the camera group and the image recognition module can be started to perform image recognition.

[0105] All of the above processes require the use of alarms to remind staff and output alarm signals according to preset alarm methods, including sound alarms and light alarms.

[0106] Among them, the alarm modes corresponding to different alarm signals output different information. For example, in the case of missing identification information and unclear identification information, the alarm signals issued are different, which can be specific voice signals such as "missing printing" and "unclear identification information" to distinguish different identification information errors.

[0107] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A board surface identification information image recognition system, characterized in that: include: A camera group, the camera groups are respectively facing the upper and lower board surfaces of the gypsum board, and the camera group collects each frame of the upper and lower board surfaces of the gypsum board during transportation to obtain multiple groups of continuous images; An image recognition module is connected to the camera group for sequentially screening and recognizing multiple groups of continuous images to analyze whether there is identification information in each image, and sends an alarm signal if there is no identification information; An image analysis module is connected to the camera group for communication. The image analysis module selects at least one frame of image from each group of continuous images, performs image recognition on the selected image and calculates the clarity, and issues an alarm signal when the clarity does not meet the standard; A service platform, wherein the service platform is communicatively connected with the camera group and the image recognition module, and the service platform is used to start or shut down the camera group and the image recognition module to perform image recognition work, and receive analysis results of the image recognition module and the image analysis module.

2. The board surface marking information image recognition system according to claim 1, characterized in that: The camera group includes two sets of opposing camera structures, one set of camera structures facing the upper surface of the gypsum board, and the other set of camera structures facing the lower surface of the gypsum board; The camera structure includes at least two groups of cameras facing the gypsum board surface, and the distance between each camera and the gypsum board surface is different.

3. The board surface marking information image recognition system according to claim 2, characterized in that: Obtaining the length of the camera area on the transport plane where the gypsum board is located; Corresponding to one of the gypsum boards to be identified, a plurality of frames of images in which at least a portion of the gypsum board surface is within the imaging area are selected based on the length of the imaging area to obtain a first image set; At least one frame of image in the first image set is extracted in time sequence to obtain a second image set.

4. The board surface marking information image recognition system according to claim 2, characterized in that: Obtaining the length of the camera area on the transport plane where the gypsum board is located; Corresponding to one of the gypsum boards to be identified, a plurality of frames of images whose identification information is within the imaging area are selected based on the length of the imaging area to obtain a first image set; At least one frame of image in the first image set is extracted in time sequence to obtain a second image set.

5. The board surface marking information image recognition system according to claim 3 or 4, characterized in that: The second image set includes at least one frame of image whose identification information is located at a position directly facing the camera.

6. The board surface marking information image recognition system according to claim 3 or 4, characterized in that: In the image recognition module, image recognition is performed on each frame of the first image set, and each image is analyzed to determine whether there is identification information; When there is no identification information in each image, the alarm device is controlled to output an alarm signal through a preset alarm method.

7. The board surface marking information image recognition system according to claim 3 or 4, characterized in that: In the image analysis module, image recognition is performed on each frame of the second image set, and the clarity is calculated; A clarity threshold is set, and when the clarity is lower than the clarity threshold, the alarm device is controlled to output an alarm signal in a preset alarm mode.

8. The board surface marking information image recognition system according to claim 7, characterized in that: The image analysis module includes a database, in which standard images and standard image features with identification information are stored; When performing image analysis on one of the frames of the image, image features are extracted and compared with standard image features, and the degree of overlap between the image features and the standard image features is calculated; The overlap degree threshold is set. When the calculated overlap degree is lower than the overlap degree threshold, the alarm device is controlled to output an alarm signal through a preset alarm method.

9. The board surface marking information image recognition system according to claim 1, characterized in that: The service platform stores preset personnel identity information and login passwords, and the service platform can accept and display the analysis results of the image recognition module and the image analysis module; The service platform obtains personnel identity information and login password; Determine whether the personnel identity information meets the preset personnel identity information; If the personnel identity information does not meet the preset personnel identity information, a warning signal is generated, and the alarm device is controlled to output an alarm signal through a preset alarm method; The preset personnel identity information is the personnel identity information that is allowed to start the camera group.

10. The board surface marking information image recognition system according to claim 8, characterized in that: Output an alarm signal according to a preset alarm mode, wherein the alarm mode includes a sound alarm mode and a light alarm mode; The alarm modes corresponding to different alarm signals output different information.

Images