System for identifying false information in public health event

Abstract

The invention relates to a system for identifying false information in a public health event, and the system comprises the following parts: a data collection module which is used for obtaining text data of an image format related to public health; the information processing module is used for performing information preprocessing on the text data in the image format to obtain a text associated image of the public health event; the false information identification module is used for acquiring the text associated image input by the information processing module and performing image identification on the text associated image to obtain a false information identification result; and the visual display module is used for associating the false information identification result with the text association image for visual display. According to the method, the false information identification result can be accurately obtained, and then the false information identification result is visually displayed, so that related workers can be greatly assisted in completing false information screening work of public health events, and the screening efficiency of the false information in the public health events is greatly improved on the premise of ensuring the accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field related to computer science, and in particular to a system for identifying false information in public health events. Background Art

[0002] As false information becomes prevalent during public health events, it becomes particularly important to correctly identify and refute it.

[0003] At present, the existing technology for identifying false information in public health events is often to set up a mass reporting mechanism, relying on certain reward and punishment measures to encourage the public to report false information in public health events, and then the relevant staff will screen and verify the information, and finally determine the false information. This method requires the public to report, which may not be able to prevent and control at the source. Secondly, it requires experienced staff to screen a large amount of information, which is a heavy workload and affects the final judgment result. Summary of the invention

[0004] The purpose of the present invention is to solve at least one of the deficiencies of the prior art and to provide a system for identifying false information in public health events.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] Specifically, a system for identifying false information in public health events is proposed, including the following:

[0007] A data acquisition module for acquiring text data in image format related to public health;

[0008] An information processing module, used for performing information preprocessing on the text data in image format to obtain a text-related image of a public health event;

[0009] a false information identification module, used to obtain the text-associated image input from the information processing module, and perform image recognition on the text-associated image to obtain a false information identification result;

[0010] A visualization display module is used to associate the false information identification result with the text-associated image for visualization.

[0011] Further, specifically, obtaining text data in image format related to public health, including,

[0012] Collect multiple text data images related to any public health event uploaded by users from various online platforms to obtain a set of text data images corresponding to each public health event.

[0013] Further, specifically, the text data in image format is preprocessed to obtain a text-related image of a public health event, including:

[0014] For each text data image set corresponding to a public health event, define the image matrix of the kth text data image in the text data image set as Qk(k), and the pixel value of the pixel point in the hkth row and lkth column in Qk(k) is expressed as Qk(k, hk, lk);

[0015] Calculate the ROI contribution rate ROI_le(k, hk, lk) between Qk(k, hk, lk) and its neighboring pixels, where ROI_le(k, hk, lk) is given by exp(Cp(k, hk, lk) 2 -avg(k,hk,lk) 2 ) and exp(max(k, hk, lk) 2 -min(k, hk, lk) 2 ) is determined by the ratio of

[0016] Wherein, max(k, hk, lk) is the maximum value of the pixel value of the neighborhood pixel point of Qk(k, hk, lk), avg(k, hk, lk) and min(k, hk, lk) are the average value and minimum value of the pixel value of the neighborhood pixel point of Qk(k, hk, lk), respectively, wherein the neighborhood pixel point of Qk(k, hk, lk) refers to the 8 neighborhood pixel points in the two-dimensional plane of Qk(k, hk, lk);

[0017] Calculate the ROI contribution areas of all text data images in the text data image set based on the ROI contribution rate;

[0018] Constructing an original image, wherein the original image is obtained by calculating the pixel average value of all the text data images in the text data image set, that is, the pixel value of any pixel point of the original image is the pixel average value of all the text data images at the same pixel point position;

[0019] The ROI contribution areas of all the text data images in the text data image set are used to replace the corresponding areas of the original image in sequence to obtain a text-associated image of the current public health event.

[0020] Further, specifically, the ROI contribution areas of all text data images in the text data image set are calculated based on the ROI contribution rate, including:

[0021] When k=1, that is, for the first text data image in the text data image set, traverse its image matrix Q1(1), find the two pixel points with the maximum ROI contribution rate and the second largest ROI contribution rate in Q1(1), connect the pixel point with the maximum ROI contribution rate and the pixel point with the second largest ROI contribution rate to obtain a line segment lne_sgm(1), then traverse Q1(1), select the pixel point with the maximum ROI contribution rate other than the pixel point in lne_sgm(1) and define it as pixel point H, and the area obtained by connecting H and the two endpoints of the line segment lne_sgm(1) is the ROI contribution area of ​​the first text data image;

[0022] When k>1, that is, for any text data image other than the first text data image in the text data image set, calculate the pixel average value avg(k-1) of the pixel points in the ROI contribution area of ​​Qk-1(k-1), traverse Qk(k), select any two pixel points in Qk(k) whose ROI contribution rate is not less than avg(k-1) and connect them to obtain the line segment lne_sgm(k), then traverse Qk(k), select another pixel point with the largest ROI contribution rate except the pixel point on lne_sgm(k) and define it as pixel point G, and the area obtained by connecting G and the two vertices of the line segment lne_sgm(k) is the ROI contribution area of ​​the current text data image.

[0023] Further, specifically, performing image recognition on the text-associated image to obtain a false information recognition result includes:

[0024] Performing image preprocessing on the text-associated image to obtain an image after image enhancement and segmenting the text area into different text lines;

[0025] Perform feature extraction on the image segmented into different text lines to obtain extracted features;

[0026] Performing text recognition based on the extracted features through a pre-trained deep learning model to obtain an initial text recognition result;

[0027] Performing error correction and post-processing on the initial text recognition result through a pre-trained language model to obtain a final text recognition result;

[0028] The false information recognition result is obtained by performing false information judgment on the final text recognition result through the pre-trained LLM model.

[0029] Further, specifically, the image preprocessing operation is implemented based on OkenCV, and the image preprocessing operation includes:

[0030] The image quality is improved by sharpening and denoising, the image is binarized to facilitate the identification of text areas, the image is tilted and corrected to make the text horizontally arranged, the connected domain segmentation is performed to separate the text area from the image, and the separated text area is segmented into different text lines through text line segmentation.

[0031] Furthermore, specifically, the deep learning model is pre-trained based on the TensorFlow deep learning framework, and the language model is pre-trained based on the SkaCy library.

[0032] Further, specifically, the false information identification result is associated with the text-associated image for visual display, including:

[0033] When the false information identification result is false information, the text-associated image associated with the false information identification result is marked as false information, a preset pattern stamp is added, and the text-associated image with the preset pattern stamp is fed back to the user.

[0034] The beneficial effects of the present invention are:

[0035] The present invention proposes a system for identifying false information in public health events. The system obtains text data in image format related to public health, performs information preprocessing on the text data to obtain high-quality text-related images of public health events, ensures the accuracy of information at the identification source, and then uses machine learning models such as LLM to identify false information on the text-related images of public health events. It can accurately obtain false information identification results, and then visualize the false information identification results, which can greatly assist relevant staff in completing the work of identifying false information in public health events, and greatly improve the efficiency of identifying false information in public health events while ensuring accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The above and other features of the present disclosure will become more apparent by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present disclosure represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0037] Figure 1 Shown is a structural principle diagram of a system for identifying false information in public health events according to the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict. The same reference numerals used throughout the drawings indicate the same or similar parts.

[0039] Example 1, reference Figure 1 The present invention proposes a system for identifying false information in public health events, comprising the following:

[0040] The data acquisition module 100 is used to acquire text data in image format related to public health;

[0041] An information processing module 200 is used to perform information preprocessing on the text data in image format to obtain a text-related image of a public health event;

[0042] A false information identification module 300 is used to obtain the text-associated image input from the information processing module, and perform image recognition on the text-associated image to obtain a false information identification result;

[0043] The visualization display module 400 is used to associate the false information identification result with the text-associated image for visualization.

[0044] In this embodiment 1, by acquiring text data in image format related to public health, and then performing information preprocessing on the text data to obtain high-quality text-related images of public health events, the accuracy of information is guaranteed at the identification source, and then the text-related images of public health events are used through machine learning models such as LLM to identify false information, which can accurately obtain false information identification results, and then visualize the false information identification results, which can greatly assist relevant personnel in completing the work of identifying false information in public health events, and greatly improve the efficiency of identifying false information in public health events while ensuring accuracy.

[0045] As a preferred embodiment of the present invention, specifically, obtaining text data in image format related to public health includes:

[0046] Collect multiple text data images related to any public health event uploaded by users from various online platforms to obtain a set of text data images corresponding to each public health event.

[0047] As a preferred embodiment of the present invention, specifically, the text data in image format is preprocessed to obtain a text-related image of a public health event, including:

[0048] For each text data image set corresponding to a public health event, define the image matrix of the kth text data image in the text data image set as Qk(k), and the pixel value of the pixel point in the hkth row and lkth column in Qk(k) is expressed as Qk(k, hk, lk);

[0049] Calculate the ROI contribution rate ROI_le(k, hk, lk) between Qk(k, hk, lk) and its neighboring pixels, where ROI_le(k, hk, lk) is given by exp(Cp(k, hk, lk) 2 -avg(k,hk,lk) 2 ) and exp(max(k, hk, lk) 2 -min(k, hk, lk) 2 ) is determined by the ratio of

[0050] Wherein, max(k, hk, lk) is the maximum value of the pixel value of the neighborhood pixel point of Qk(k, hk, lk), avg(k, hk, lk) and min(k, hk, lk) are the average value and minimum value of the pixel value of the neighborhood pixel point of Qk(k, hk, lk), respectively, wherein the neighborhood pixel point of Qk(k, hk, lk) refers to the 8 neighborhood pixel points in the two-dimensional plane of Qk(k, hk, lk);

[0051] Calculate the ROI contribution areas of all text data images in the text data image set based on the ROI contribution rate;

[0052] Constructing an original image, wherein the original image is obtained by calculating the pixel average value of all the text data images in the text data image set, that is, the pixel value of any pixel point of the original image is the pixel average value of all the text data images at the same pixel point position;

[0053] The ROI contribution areas of all the text data images in the text data image set are used to replace the corresponding areas of the original image in sequence to obtain a text-associated image of the current public health event.

[0054] In this preferred embodiment, taking into account that the quality of text data images collected by various online platforms may be poor, for the text data images of the same public health event, several more images are collected to form a text data image set, and then an ROI area is extracted from the text data image set. The area that best represents the ROI area is selected from each image and extracted, and then the corresponding area of ​​the original image is replaced. This can ensure the image quality of the text data image, and then ensure that the subsequent machine learning model can obtain the best quality image data, thereby ensuring the accuracy of subsequent results from the data source level.

[0055] As a preferred embodiment of the present invention, specifically, the ROI contribution area of ​​all text data images in the text data image set is calculated based on the ROI contribution rate, including:

[0056] When k=1, that is, for the first text data image in the text data image set, traverse its image matrix Q1(1), find the two pixel points with the maximum ROI contribution rate and the second largest ROI contribution rate in Q1(1), connect the pixel point with the maximum ROI contribution rate and the pixel point with the second largest ROI contribution rate to obtain a line segment lne_sgm(1), then traverse Q1(1), select the pixel point with the maximum ROI contribution rate other than the pixel point in lne_sgm(1) and define it as pixel point H, and the area obtained by connecting H and the two endpoints of the line segment lne_sgm(1) is the ROI contribution area of ​​the first text data image;

[0057] When k>1, that is, for any text data image other than the first text data image in the text data image set, calculate the pixel average value avg(k-1) of the pixel points in the ROI contribution area of ​​Qk-1(k-1), traverse Qk(k), select any two pixel points in Qk(k) whose ROI contribution rate is not less than avg(k-1) and connect them to obtain the line segment lne_sgm(k), then traverse Qk(k), select another pixel point with the largest ROI contribution rate except the pixel point on lne_sgm(k) and define it as pixel point G, and the area obtained by connecting G and the two vertices of the line segment lne_sgm(k) is the ROI contribution area of ​​the current text data image.

[0058] In this preferred embodiment, by calculating the ROI contribution area of ​​the text data image in the above manner, the ROI contribution areas of all text data images can be accurately obtained.

[0059] As a preferred embodiment of the present invention, specifically, performing image recognition on the text-associated image to obtain a false information recognition result includes:

[0060] Performing image preprocessing on the text-associated image to obtain an image after image enhancement and segmenting the text area into different text lines;

[0061] Perform feature extraction on the image segmented into different text lines to obtain extracted features;

[0062] Performing text recognition based on the extracted features through a pre-trained deep learning model to obtain an initial text recognition result;

[0063] Performing error correction and post-processing on the initial text recognition result through a pre-trained language model to obtain a final text recognition result;

[0064] The false information recognition result is obtained by performing false information judgment on the final text recognition result through the pre-trained LLM model.

[0065] In this preferred embodiment, a mature machine learning model is used to perform OCR recognition on text-related images and then perform semantic recognition to obtain false information recognition results, which can automatically, efficiently and accurately obtain false information recognition results for text-related images.

[0066] Specifically, when applying, as a preferred method, image preprocessing: OpenCV and other libraries can be used to implement image enhancement, binarization, tilt correction, segmentation and other operations. Text recognition: Deep learning frameworks such as TensorFlow and PyTorch can be used to train text recognition models. Open source libraries such as Tesseract OCR and EasyOCR can also be used for text recognition. Post-processing: NLTK, SpaCy and other libraries can be used for text error correction, formatting, etc. LLM application: Hugging Face and other platforms provide various pre-trained LLMs.

[0067] The following is an example application code.

[0068] import cv2

[0069] import pytesseract

[0070] # Load the image

[0071] image=cv2.imread("image.jpg")

[0072] #Preprocess the image

[0073] gray=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)

[0074] thresh=cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)[1]

[0075] #Identify false information

[0076] text=pytesseract.image_to_string(thresh)

[0077] #Print false information identification results

[0078] print(text).

[0079] As a preferred embodiment of the present invention, specifically, the image preprocessing operation is implemented based on OkenCV, and the image preprocessing operation includes:

[0080] The image quality is improved by sharpening and denoising, the image is binarized to facilitate the identification of text areas, the image is tilted and corrected to make the text horizontally arranged, the connected domain segmentation is performed to separate the text area from the image, and the separated text area is segmented into different text lines through text line segmentation.

[0081] In this preferred embodiment, image enhancement: image processing techniques such as sharpening and denoising are used to improve image quality and facilitate subsequent recognition. Binarization: converting an image into a black and white image to facilitate identification of text areas. Tilt correction: correcting tilted text areas so that the text is arranged horizontally. Segmentation: using image segmentation techniques such as connected domain analysis to separate text areas from images. Text line segmentation: segmenting text areas into different text lines.

[0082] As a preferred implementation manner of the present invention, specifically, the deep learning model is pre-trained based on the TensorFlow deep learning framework, and the language model is pre-trained based on the SkaCy library.

[0083] In this preferred embodiment, feature extraction: extract features from text images, such as shape, direction, color, etc. Model training: use deep learning models (such as CNN, RNN, Transformer, etc.) for training to learn the mapping relationship between text images and text. Text recognition: use the trained model to identify the text area in the image and output the corresponding text information. Post-processing: Error correction: use the language model to correct the recognized text to improve recognition accuracy. Formatting: format the recognized text as needed, such as adding punctuation, adjusting capitalization, etc. LLM application: semantic understanding: use LLM to perform semantic understanding of the recognized text, and then obtain false information recognition results.

[0084] As a preferred embodiment of the present invention, specifically, the false information identification result is associated with the text-associated image for visual display, including:

[0085] When the false information identification result is false information, the text-associated image associated with the false information identification result is marked as false information, a preset pattern stamp is added, and the text-associated image with the preset pattern stamp is fed back to the user.

[0086] Specifically, during actual application, after the text content in the acquired public health information picture is extracted, it is identified by the method proposed in the present invention, and it is found that part of the content is true and part of the content is false information. Therefore, a special logo of the platform applying the method of the present invention is marked, and the result is fed back to the user.

[0087] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0088] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or system that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0089] Although the description of the present invention has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims in view of the prior art by reference to the appended claims, thereby effectively covering the intended scope of the present invention. In addition, the above description of the present invention is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that have not yet been foreseen may still represent equivalent changes to the present invention.

[0090] The above is only a preferred embodiment of the present invention. The present invention is not limited to the above implementation. As long as the technical effect of the present invention is achieved by the same means, it should belong to the protection scope of the present invention. Within the protection scope of the present invention, its technical scheme and / or implementation method can have various modifications and changes.

Claims

1. A system for identifying false information in public health events, characterized in that: Includes the following: A data acquisition module for acquiring text data in image format related to public health; An information processing module, used for performing information preprocessing on the text data in image format to obtain a text-related image of a public health event; a false information identification module, used to obtain the text-associated image input from the information processing module, and perform image recognition on the text-associated image to obtain a false information identification result; A visualization display module is used to associate the false information identification result with the text-associated image for visualization.

2. A system for identifying false information in public health events according to claim 1, characterized in that: Specifically, obtain text data in image format related to public health, including: Collect multiple text data images related to any public health event uploaded by users from various online platforms to obtain a set of text data images corresponding to each public health event.

3. A system for identifying false information in public health events according to claim 2, characterized in that: Specifically, the text data in image format is preprocessed to obtain a text-related image of a public health event, including: For each text data image set corresponding to a public health event, define the image matrix of the kth text data image in the text data image set as Qk(k), and the pixel value of the pixel point in the hkth row and lkth column in Qk(k) is expressed as Qk(k, hk, lk); Calculate the ROI contribution rate ROI_le(k, hk, lk) between Qk(k, hk, lk) and its neighboring pixels, where ROI_le(k, hk, lk) is given by exp(Cp(k, hk, lk) 2 -avg(k,hk,lk) 2 ) and exp(max(k, hk, lk) 2 -min(k, hk, lk) 2 ) is determined by the ratio of Wherein, max(k, hk, lk) is the maximum value of the pixel value of the neighborhood pixel point of Qk(k, hk, lk), avg(k, hk, lk) and min(k, hk, lk) are the average value and minimum value of the pixel value of the neighborhood pixel point of Qk(k, hk, lk), respectively, wherein the neighborhood pixel point of Qk(k, hk, lk) refers to the 8 neighborhood pixel points in the two-dimensional plane of Qk(k, hk, lk); Calculate the ROI contribution areas of all text data images in the text data image set based on the ROI contribution rate; Constructing an original image, wherein the original image is obtained by calculating the pixel average value of all text data images in the text data image set, that is, the pixel value of any pixel point of the original image is the pixel average value of all text data images at the same pixel point position; The ROI contribution areas of all the text data images in the text data image set are used to replace the corresponding areas of the original image in sequence to obtain a text-associated image of the current public health event.

4. A system for identifying false information in public health events according to claim 3, characterized in that: Specifically, the ROI contribution areas of all text data images in the text data image set are calculated based on the ROI contribution rate, including: When k=1, that is, for the first text data image in the text data image set, traverse its image matrix Q1(1), find the two pixel points with the maximum ROI contribution rate and the second largest ROI contribution rate in Q1(1), connect the pixel point with the maximum ROI contribution rate and the pixel point with the second largest ROI contribution rate to obtain a line segment lne_sgm(1), then traverse Q1(1), select the pixel point with the maximum ROI contribution rate other than the pixel point in lne_sgm(1) and define it as pixel point H, and the area obtained by connecting H and the two endpoints of the line segment lne_sgm(1) is the ROI contribution area of ​​the first text data image; When k>1, that is, for any text data image other than the first text data image in the text data image set, calculate the pixel average value avg(k-1) of the pixel points in the ROI contribution area of ​​Qk-1(k-1), traverse Qk(k), select any two pixel points in Qk(k) whose ROI contribution rate is not less than avg(k-1) and connect them to obtain the line segment lne_sgm(k), then traverse Qk(k), select another pixel point with the largest ROI contribution rate except the pixel point on lne_sgm(k) and define it as pixel point G, and the area obtained by connecting G and the two vertices of the line segment lne_sgm(k) is the ROI contribution area of ​​the current text data image.

5. A system for identifying false information in public health events according to claim 1, characterized in that: Specifically, performing image recognition on the text-associated image to obtain a false information recognition result includes: Performing image preprocessing on the text-associated image to obtain an image after image enhancement and segmenting the text area into different text lines; Perform feature extraction on the image segmented into different text lines to obtain extracted features; Performing text recognition based on the extracted features through a pre-trained deep learning model to obtain an initial text recognition result; Performing error correction and post-processing on the initial text recognition result through a pre-trained language model to obtain a final text recognition result; The false information recognition result is obtained by performing false information judgment on the final text recognition result through the pre-trained LLM model.

6. A system for identifying false information in public health events according to claim 5, characterized in that: Specifically, the image preprocessing operation is implemented based on OkenCV, and the image preprocessing operation includes: The image quality is improved by sharpening and denoising, the image is binarized to facilitate the identification of text areas, the image is tilted and corrected to make the text horizontally arranged, the connected domain segmentation is performed to separate the text area from the image, and the separated text area is segmented into different text lines through text line segmentation.

7. A system for identifying false information in public health events according to claim 5, characterized in that: Specifically, the deep learning model is pre-trained based on the TensorFlow deep learning framework, and the language model is pre-trained based on the SkaCy library.

8. A system for identifying false information in public health events according to claim 1, characterized in that: Specifically, the false information identification result is associated with the text-associated image for visual display, including: When the false information identification result is false information, the text-associated image associated with the false information identification result is marked as false information, a preset pattern stamp is added, and the text-associated image with the preset pattern stamp is fed back to the user.

Images