Rapid detection method and system based on big data

Abstract

The invention provides a rapid detection method and system based on big data, and the method comprises the steps: d1, carrying out the first data acquisition, collecting the primary multi-source data of a current image, and collecting the primary multi-source data of a similar image from a big data platform based on the primary multi-source data of the current image; d2, performing first round of data processing, preprocessing primary multi-source data in the current image and the similar image, sorting the preprocessed data into a first data set and a second data set, performing integration calculation on the first data set and the second data set to generate an abnormal coefficient YCZ, and performing second round of data processing on the abnormal coefficient YCZ; and comparing the abnormal coefficient YCZ with a preset first threshold Y to generate a first comparison result, and judging whether the first image is abnormal or not according to the first comparison result. According to the method, the abnormal features in the image can be captured more comprehensively, the data processing efficiency is improved, and the accuracy and the real-time performance of abnormal recognition are enhanced.

Description

Technical Field

[0001] The present application relates to the technical field of detection systems, and specifically, to a rapid detection method and system based on big data. Background Art

[0002] In the current era of rapid development of information technology, big data, as a key field, is profoundly changing the way various industries operate. Its application range is wide, covering many aspects such as data storage, analysis and mining, especially in the field of intelligent recognition and data retrieval.

[0003] In image processing, with the increasing perfection of AI synthesis technology, existing technologies face many challenges in image anomaly detection and recognition. Many traditional methods rely on manual intervention, resulting in low efficiency and prone to human errors. At the same time, existing detection systems are often unable to effectively extract deep features in images, which makes the accuracy of anomaly recognition significantly insufficient. Such limitations not only make it impossible for the system to fully analyze image data, but also affect the timely response to potential problems, thereby limiting the industry's progress in intelligent management and decision-making. Summary of the invention

[0004] In view of this, the present application provides a rapid detection method and system based on big data to improve data processing efficiency and enhance the accuracy and real-time performance of image anomaly detection.

[0005] To achieve the above objectives, the technical solutions adopted in this application are as follows:

[0006] In a first aspect, the present application provides a rapid detection method based on big data, comprising:

[0007] d1, the first round of data acquisition, collecting primary multi-source data of the current image, and based on the primary multi-source data of the current image, collecting primary multi-source data of similar images from the big data platform, wherein the similar images are images having the same feature points as the current image;

[0008] d2, a first round of data processing, preprocessing the primary multi-source data in the current image and the similar images, and organizing the preprocessed data into a first data set and a second data set, integrating and calculating the first data set and the second data set to generate an abnormality coefficient YCZ, comparing the abnormality coefficient YCZ with a preset first threshold value Y to generate a first comparison result, and judging whether the first image has an abnormality according to the first comparison result;

[0009] The first data set includes a first edge feature A1, a second edge feature value A2, a first color space feature value B1, a second color space feature value B2, a first texture feature value C1, a second texture feature value C2, a first image size value D1, and a second image size value D2;

[0010] The second data set includes a first shooting time E1, a second shooting time E2, a first image contrast F1, a second image contrast F2, a first geographical location coordinate G1, a second geographical location coordinate G2, a first image resolution H1, and a second image resolution H2;

[0011] The primary multi-source data of the current image are a first edge feature A1, a first color space feature value B1, a first texture feature value C1, a first image size value D1, a first shooting time E1, a first image contrast F1, a first geographical location coordinate G1 and a first image resolution H1;

[0012] The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

[0013] Optionally, the method further includes: if the first comparison result is abnormal, performing a deep analysis on the current image and executing step d3;

[0014] d3, a second round of data acquisition, collecting deep multi-source data of the current image, and based on the deep multi-source data of the current image, collecting deep multi-source data of similar images from the big data platform;

[0015] d4, a second round of data processing, pre-processing the deep multi-source data of the current image and the similar images, and reorganizing them into a third data set, a fourth data set, a fifth data set and a sixth data set, integrating and calculating the third data set, the fourth data set, the fifth data set and the sixth data set to generate an abnormality identification coefficient YJD, comparing the abnormality identification coefficient YJD with a preset second threshold R to generate a second comparison result, and judging the abnormal magnitude of the current image according to the second comparison result;

[0016] The third data set includes a first region area value I1, a second region area value I2, a first shape complexity J1, a second shape complexity J2, a first contour length K1, and a second contour length K2;

[0017] The fourth data set includes a first pixel value L1, a second pixel value L2, a first grayscale variance value M1, a second grayscale variance value M2, a first grayscale kurtosis value N1, and a second grayscale kurtosis value N2;

[0018] The fifth data set includes the average brightness O1 of the first region, the average brightness O2 of the second region, the first brightness variance P1, the second brightness variance P2, the first illuminance distribution value Q1, and the second illuminance distribution value Q2;

[0019] The sixth data set includes the first amplitude spectrum value R1, the second amplitude spectrum value R2, the first high-low frequency difference U1, the second high-low frequency difference U2, the first frequency domain energy distribution difference T1, and the second frequency domain energy distribution difference T2.

[0020] Optionally, the specific method for determining the abnormality level of the first image according to the second comparison result is as follows:

[0021] When YJD ≤ R, the current picture is a first-level abnormality;

[0022] When R < YJD ≤ R × 110%, the current picture is a second-level abnormality;

[0023] When YJD > R × 110%, the current picture is a third-level abnormality.

[0024] Optionally, the abnormality coefficient YCZ is obtained by calculating according to the following formula;

[0025] YCZ = a1 × S1 - a2 × S2;

[0026]

[0027]

[0028] In the formula: S1 is the first characteristic reference coefficient, S2 is the second characteristic reference coefficient, A1 is the first edge feature, B1 is the first color space feature value, C1 is the first texture feature value, D1 is the first image size value, E1 is the first shooting time, F1 is the first image contrast, G1 is the first geographical location coordinate, H1 is the first image resolution, A2 is the second edge feature, B2 is the second color space feature value, C2 is the second texture feature value, D2 is the second image size value, E2 is the second shooting time, F2 is the second image contrast, G2 is the second geographical location coordinate, H2 is the second image resolution;

[0029] a1, a2, b1, b2, and b3 are weight values, and the values of a1, a2, b1, b2, and b3 are adjusted and set by the user. ln is the logarithmic function, and e is the base function.

[0030] Optionally, the abnormality identification coefficient YJD is obtained by calculating according to the following formula;

[0031] YJD = c1 × S3 + c2 × S4 + c3 × S5 + c4 × S5;

[0032]

[0033]

[0034]

[0035]

[0036] Wherein: c1, c2, c3 and c4 are weight values, and the values ​​of c1, c2, c3 and c4 are adjusted and set by the user;

[0037] I1 is the area value of the first region, I2 is the area value of the second region, J1 is the first shape complexity, J2 is the second shape complexity, K1 is the first contour length, and K2 is the second contour length;

[0038] S3 is the abnormal area characteristic coefficient, S4 is the local statistical characteristic coefficient, S5 is the optical characteristic coefficient, and S6 is the frequency domain characteristic coefficient;

[0039] L1 is the first pixel value, L2 is the second pixel value, M1 is the first grayscale variance value, M2 is the second grayscale variance value, N1 is the first grayscale kurtosis value, and N2 is the second grayscale kurtosis value;

[0040] O1 is the mean brightness of the first area, O2 is the mean brightness of the second area, P1 is the first brightness variance, P2 is the second brightness variance, Q1 is the first illuminance distribution value, and Q2 is the second illuminance distribution value;

[0041] R1 is the first amplitude spectrum value, R2 is the second amplitude spectrum value, U1 is the first high and low frequency difference value, U2 is the second high and low frequency difference value, T1 is the first frequency domain energy distribution difference value, and T2 is the second frequency domain energy distribution difference value.

[0042] In a second aspect, the present application provides a rapid detection system based on big data, comprising:

[0043] A first round of data acquisition module is used to collect primary multi-source data of the current image, and based on the primary multi-source data of the current image, collect primary multi-source data of similar images from the big data platform, wherein the similar images are images having the same feature points as the current image;

[0044] A first round of data processing module, used for preprocessing the primary multi-source data in the current image and the similar images, and arranging the preprocessed data into a first data set and a second data set, generating an abnormality coefficient YCZ after integrating and calculating the first data set and the second data set, generating a first comparison result after comparing the abnormality coefficient YCZ with a preset first threshold value Y, and judging whether the first image has an abnormality according to the first comparison result;

[0045] The first data set includes a first edge feature A1, a second edge feature value A2, a first color space feature value B1, a second color space feature value B2, a first texture feature value C1, a second texture feature value C2, a first image size value D1, and a second image size value D2;

[0046] The second data set includes a first shooting time E1, a second shooting time E2, a first image contrast F1, a second image contrast F2, a first geographical location coordinate G1, a second geographical location coordinate G2, a first image resolution H1, and a second image resolution H2;

[0047] The primary multi-source data of the current image are a first edge feature A1, a first color space feature value B1, a first texture feature value C1, a first image size value D1, a first shooting time E1, a first image contrast F1, a first geographical location coordinate G1 and a first image resolution H1;

[0048] The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

[0049] Optionally, the system further includes:

[0050] A second round of data acquisition module is used to collect the deep multi-source data of the current image, and based on the deep multi-source data of the current image, collect the deep multi-source data of the similar image from the big data platform;

[0051] A second round of data processing module is used to pre-process the deep multi-source data of the current image and the similar images, and reorganize them into a third data set, a fourth data set, a fifth data set and a sixth data set, generate an abnormality identification coefficient YJD after integrating and calculating the third data set, the fourth data set, the fifth data set and the sixth data set, generate a second comparison result after comparing the abnormality identification coefficient YJD with a preset second threshold R, and judge the abnormality magnitude of the current image according to the second comparison result;

[0052] The third data set includes a first region area value I1, a second region area value I2, a first shape complexity J1, a second shape complexity J2, a first contour length K1, and a second contour length K2;

[0053] The fourth data set includes a first pixel value L1, a second pixel value L2, a first gray variance value M1, a second gray variance value M2, a first gray peak value N1, and a second gray peak value N2;

[0054] The fifth data set includes a first region brightness mean O1, a second region brightness mean O2, a first brightness variance P1, a second brightness variance P2, a first illuminance distribution value Q1, and a second illuminance distribution value Q2;

[0055] The sixth data set includes a first amplitude spectrum value R1, a second amplitude spectrum value R2, a first high-low frequency difference U1, a second high-low frequency difference U2, a first frequency domain energy distribution difference T1, and a second frequency domain energy distribution difference T2.

[0056] Optionally, the second-round data processing module includes a judgment module for:

[0057] When YJD ≤ R, determining that the current picture is a first-level anomaly;

[0058] When R < YJD ≤ R × 110%, determining that the current picture is a second-level anomaly;

[0059] When YJD > R × 110%, determining that the current picture is a third-level anomaly.

[0060] Optionally, the first-round data processing module includes a first calculation module for calculating the anomaly coefficient YCZ according to the following formula:

[0061] YCZ = a1 × S1 - a2 × S2;

[0062]

[0063]

[0064] Where: S1 is the first feature reference coefficient, S2 is the second feature reference coefficient, A1 is the first edge feature, B1 is the first color space feature value, C1 is the first texture feature value, D1 is the first image size value, E1 is the first shooting time, F1 is the first image contrast, G1 is the first geographical location coordinate, H1 is the first image resolution, A2 is the second edge feature, B2 is the second color space feature value, C2 is the second texture feature value, D2 is the second image size value, E2 is the second shooting time, F2 is the second image contrast, G2 is the second geographical location coordinate, H2 is the second image resolution;

[0065] a1, a2, b1, b2 and b3 are weight values, and the values ​​of a1, a2, b1, b2 and b3 are adjusted and set by the user, ln is a logarithmic function, and e is a base function.

[0066] Optionally, the second round data acquisition module includes a second calculation module, which is used to calculate the abnormal identification coefficient YJD according to the following formula:

[0067] YJD=c1×S3+c2×S4+c3×S5+c4×S5;

[0068]

[0069]

[0070]

[0071]

[0072] Wherein: c1, c2, c3 and c4 are weight values, and the values ​​of c1, c2, c3 and c4 are adjusted and set by the user;

[0073] I1 is the area value of the first region, I2 is the area value of the second region, J1 is the first shape complexity, J2 is the second shape complexity, K1 is the first contour length, and K2 is the second contour length;

[0074] S3 is the abnormal area characteristic coefficient, S4 is the local statistical characteristic coefficient, S5 is the optical characteristic coefficient, and S6 is the frequency domain characteristic coefficient;

[0075] L1 is the first pixel value, L2 is the second pixel value, M1 is the first grayscale variance value, M2 is the second grayscale variance value, N1 is the first grayscale kurtosis value, and N2 is the second grayscale kurtosis value;

[0076] O1 is the mean brightness of the first area, O2 is the mean brightness of the second area, P1 is the first brightness variance, P2 is the second brightness variance, Q1 is the first illuminance distribution value, and Q2 is the second illuminance distribution value;

[0077] R1 is the first amplitude spectrum value, R2 is the second amplitude spectrum value, U1 is the first high and low frequency difference value, U2 is the second high and low frequency difference value, T1 is the first frequency domain energy distribution difference value, and T2 is the second frequency domain energy distribution difference value.

[0078] The beneficial effects of this application are:

[0079] Compared with the prior art, the beneficial effects of this application are:

[0080] 1. This application can more comprehensively capture abnormal features in images, which not only improves the efficiency of data processing, but also enhances the accuracy and real-time performance of abnormality identification;

[0081] 2. This application significantly reduces manual intervention and error rates, improves the scientificity and effectiveness of decision-making, and provides strong technical support for various application scenarios, meeting the needs of modern society for efficient and intelligent analysis;

[0082] 3. This application can flexibly adapt to different application scenarios according to specific abnormal thresholds, effectively improving the response and processing capabilities to abnormal events. It not only makes anomaly detection more accurate, but also helps users quickly identify and handle abnormal situations of different severity.

[0083] 4. This application can reduce the risk of misjudgment and ensure that more appropriate measures are taken during exception handling. In addition, this flexible threshold setting enables the system to better adapt to different environments and data sets, thereby improving the overall applicability and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0085] Figure 1 A flow chart of a rapid detection method based on big data provided in this application;

[0086] Figure 2 Another flow chart of a rapid detection method based on big data provided in this application;

[0087] Figure 3 A structural block diagram of a rapid detection system based on big data provided in this application;

[0088] Figure 4 Another structural block diagram of a rapid detection system based on big data provided for this application. DETAILED DESCRIPTION

[0089] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0090] like Figure 1As shown, the present application provides a rapid detection method based on big data, comprising:

[0091] d1, the first round of data acquisition, collecting primary multi-source data of the current image, and based on the primary multi-source data of the current image, collecting primary multi-source data of similar images from the big data platform, wherein the similar images are images having the same feature points as the current image;

[0092] The similar images are images with the same feature points in the background, such as the same scenic spot, or images taken at a similar time, or images taken at the same location, etc., and the similar images are extracted from authoritative websites to ensure the validity of similar image samples.

[0093] By extracting primary multi-source data, the basic features in the image can be effectively identified, providing the necessary basic data for subsequent analysis, ensuring that subsequent processing is more efficient, reducing the interference of invalid information, and thus improving the performance of the overall system.

[0094] Acquiring primary multi-source data related to the current image from the big data platform is an efficient data collection method that not only speeds up data processing, but also ensures the diversity and breadth of the collected data, providing a richer basis for subsequent analysis. This advantage effectively solves the problem of insufficient recognition ability caused by insufficient data samples in traditional methods, and further improves the adaptability of the system in different application scenarios. In addition, acquiring a large amount of data on similar images also enhances the breadth and representativeness of the data, thereby improving the accuracy of the analysis and ensuring that potential anomalies can be reliably identified in different situations.

[0095] Optionally, the system can be updated in real time while data is being extracted and collected, which provides a guarantee for timely identification of potential anomalies. This dynamic update mechanism not only enhances the system's responsiveness to new data, but also improves the accuracy and timeliness of decision-making. In general, these improvements not only improve the efficiency of image processing, but also provide more reliable technical support for modern intelligent management, meeting the urgent need for efficient and intelligent analysis.

[0096] d2, a first round of data processing, preprocessing the primary multi-source data in the current image and the similar images, and organizing the preprocessed data into a first data set and a second data set, integrating and calculating the first data set and the second data set to generate an abnormality coefficient YCZ, comparing the abnormality coefficient YCZ with a preset first threshold value Y to generate a first comparison result, and judging whether the first image has an abnormality according to the first comparison result;

[0097] This step cleans and formats the extracted and collected data, and then organizes them into the first data set and the second data set. By eliminating noise and irrelevant information, the data quality is improved, laying the foundation for subsequent integrated calculations, ensuring the reliability of the analysis results, and thus enhancing the effectiveness of the system in processing complex data.

[0098] The first data set includes a first edge feature A1, a second edge feature value A2, a first color space feature value B1, a second color space feature value B2, a first texture feature value C1, a second texture feature value C2, a first image size value D1, and a second image size value D2;

[0099] The second data set includes a first shooting time E1, a second shooting time E2, a first image contrast F1, a second image contrast F2, a first geographical location coordinate G1, a second geographical location coordinate G2, a first image resolution H1, and a second image resolution H2;

[0100] The generated first and second data sets not only comprehensively cover key parameters such as edge features, color space features, texture features, and size of the image, but also include shooting time, contrast, geographic location coordinates, and resolution feature data. This multi-dimensional data integration enables the system to comprehensively analyze image features and enhance the depth and accuracy of anomaly detection. This process solves the analysis limitations caused by data isolation in traditional methods, improves the system's ability to understand complex image content, and enables the system to better meet the urgent needs of modern intelligent management for efficiency and reliability.

[0101] The primary multi-source data of the current image are a first edge feature A1, a first color space feature value B1, a first texture feature value C1, a first image size value D1, a first shooting time E1, a first image contrast F1, a first geographical location coordinate G1 and a first image resolution H1;

[0102] The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

[0103] This step preprocesses and dimensionlessly converts the extracted and collected primary multi-source data of various images. By integrating and calculating the first data set and the second data set, an abnormality coefficient YCZ is generated. When YCZ≤Y, it means that the image has no abnormality. When YCZ>Y, it means that the image has an abnormality, thereby achieving a quantitative evaluation of the image status. This clear judgment mechanism makes the abnormality identification process more transparent and reliable, and reduces the errors caused by subjective judgment in traditional methods. When the abnormality coefficient YCZ is less than or equal to the threshold value Y, the system can accurately judge that there is no abnormality; otherwise, it indicates that there is an abnormality. This dual judgment not only improves the response speed of the system, but also enhances its ability to adapt to different application scenarios, ensuring that it can respond to various complex situations in a timely manner.

[0104] In addition, by quickly providing feedback on abnormal conditions, the system can help users take timely measures to optimize management processes and reduce potential losses. This series of improvements has enabled the overall system to significantly surpass traditional technical means in terms of intelligent monitoring and abnormal management, meeting the high-efficiency needs of modern management.

[0105] like Figure 2 As shown, the method further includes: if the first comparison result is abnormal, performing a deep analysis on the current image and executing step d3;

[0106] d3, a second round of data acquisition, collecting deep multi-source data of the current image, and based on the deep multi-source data of the current image, collecting deep multi-source data of similar images from the big data platform;

[0107] This step deeply analyzes the image content and mines more detailed features, providing a richer and more comprehensive source of information for anomaly identification, enhancing the system's ability to adapt to complex scenarios and ensuring high-quality detection results in all situations.

[0108] At the same time, this step method also enhances the ability to deeply analyze image anomalies. This diversified feature extraction method enables the system to fully capture potential abnormal areas in the image and their related attributes, which is in line with the improvement of intelligent recognition capabilities emphasized in the background introduction.

[0109] d4, a second round of data processing, pre-processing the deep multi-source data of the current image and the similar images, and reorganizing them into a third data set, a fourth data set, a fifth data set and a sixth data set, integrating and calculating the third data set, the fourth data set, the fifth data set and the sixth data set to generate an abnormality identification coefficient YJD, comparing the abnormality identification coefficient YJD with a preset second threshold R to generate a second comparison result, and judging the abnormal magnitude of the current image according to the second comparison result;

[0110] The extracted and collected various image deep multi-source data are preprocessed and dimensionless, and then the various image deep multi-source data after preprocessing are sorted to generate the third data set, the fourth data set, the fifth data set and the sixth data set; this step ensures the high quality and high availability of the data through systematic processing, provides a more rigorous data foundation for subsequent analysis and calculation, and further improves the overall processing capability of the system.

[0111] In addition, this step makes the processing of primary to deep data more systematic, and can more comprehensively capture abnormal features in the image. This innovative method greatly reduces manual intervention and error rates, improves the scientificity and effectiveness of decision-making, and provides strong technical support for various application scenarios, meeting the needs of modern society for efficient and intelligent analysis.

[0112] The third data set includes a first region area value I1, a second region area value I2, a first shape complexity J1, a second shape complexity J2, a first contour length K1, and a second contour length K2;

[0113] The fourth data set includes a first pixel value L1, a second pixel value L2, a first grayscale variance value M1, a second grayscale variance value M2, a first grayscale kurtosis value N1, and a second grayscale kurtosis value N2;

[0114] The fifth data set includes a first area brightness mean value O1, a second area brightness mean value O2, a first brightness variance P1, a second brightness variance P2, a first illuminance distribution value Q1, and a second illuminance distribution value Q2;

[0115] The sixth data set includes a first amplitude spectrum value R1, a second amplitude spectrum value R2, a first high-low frequency difference value U1, a second high-low frequency difference value U2, a first frequency domain energy distribution difference value T1, and a second frequency domain energy distribution difference value T2.

[0116] The third, fourth, fifth and sixth datasets cover a variety of important features such as area, shape complexity, contour length, pixel value, variance, brightness mean, etc. This multi-dimensional dataset not only enriches the representation of images and enhances the system's in-depth analysis capabilities in anomaly detection, but also effectively solves the problem of previous methods relying only on a single feature and ignoring global information. This comprehensive feature extraction and organization ensures that the system can more accurately identify and analyze image anomalies in complex scenes.

[0117] Integrate and calculate multiple sorted data sets to generate an anomaly identification coefficient YJD, and compare it with a preset second threshold R. This analysis process can quantify the severity of anomalies, provide a clear basis for subsequent decision-making, improve the accuracy of anomaly recognition, and help users make more effective responses. Through the meticulous processing of deep multi-source image data, the system can provide more accurate anomaly recognition results, providing strong data support for subsequent analysis and decision-making. This series of improvements enhances the overall system's performance in intelligent monitoring and anomaly management, enabling it to exhibit higher efficiency and accuracy when dealing with complex real-world images, thus meeting the higher requirements of modern management and monitoring for technology.

[0118] Optionally, the specific method for determining the anomaly level of the first image according to the second comparison result is as follows:

[0119] When YJD ≤ R, the current picture is a first-level anomaly;

[0120] When R < YJD ≤ R × 110%, the current picture is a second-level anomaly;

[0121] When YJD > R × 110%, the current picture is a third-level anomaly.

[0122] By integrating and calculating to generate the anomaly identification coefficient YJD, it is possible to meticulously evaluate the anomaly degree of the image and provide multi-level anomaly classification. This hierarchical analysis method can flexibly adapt to different application scenarios according to the specific anomaly threshold R, effectively improving the response and handling capabilities for anomaly events. It not only makes anomaly detection more accurate but also helps users quickly identify and handle anomalies of different severity levels. At the same time, it can reduce the risk of misjudgment, ensure more appropriate measures are taken during the anomaly handling process, and thus enhance the overall applicability and stability.

[0123] Optionally, the anomaly coefficient YCZ is obtained by calculating according to the following formula;

[0124] YCZ = a1 × S1 - a2 × S2;

[0125]

[0126]

[0127] Wherein: S1 is the first feature reference coefficient, S2 is the second feature reference coefficient, A1 is the first edge feature, B1 is the first color space feature value, C1 is the first texture feature value, D1 is the first image size value, E1 is the first shooting time, F1 is the first image contrast, G1 is the first geographic location coordinate, H1 is the first image resolution, A2 is the second edge feature, B2 is the second color space feature value, C2 is the second texture feature value, D2 is the second image size value, E2 is the second shooting time, F2 is the second image contrast, G2 is the second geographic location coordinate, H2 is the second image resolution;

[0128] a1, a2, b1, b2 and b3 are weight values, and the values ​​of a1, a2, b1, b2 and b3 are adjusted and set by the user, ln is a logarithmic function, and e is a base function.

[0129] Optionally, the abnormal identification coefficient YJD is calculated by the following formula:

[0130] YJD=c1×S3+c2×S4+c3×S5+c4×S5;

[0131]

[0132]

[0133]

[0134]

[0135] Wherein: c1, c2, c3 and c4 are weight values, and the values ​​of c1, c2, c3 and c4 are adjusted and set by the user;

[0136] I1 is the area value of the first region, I2 is the area value of the second region, J1 is the first shape complexity, J2 is the second shape complexity, K1 is the first contour length, and K2 is the second contour length;

[0137] S3 is the abnormal area characteristic coefficient, S4 is the local statistical characteristic coefficient, S5 is the optical characteristic coefficient, and S6 is the frequency domain characteristic coefficient;

[0138] L1 is the first pixel value, L2 is the second pixel value, M1 is the first grayscale variance value, M2 is the second grayscale variance value, N1 is the first grayscale kurtosis value, and N2 is the second grayscale kurtosis value;

[0139] O1 is the mean brightness of the first area, O2 is the mean brightness of the second area, P1 is the first brightness variance, P2 is the second brightness variance, Q1 is the first illuminance distribution value, and Q2 is the second illuminance distribution value;

[0140] R1 is the first amplitude spectrum value, R2 is the second amplitude spectrum value, U1 is the first high and low frequency difference value, U2 is the second high and low frequency difference value, T1 is the first frequency domain energy distribution difference value, and T2 is the second frequency domain energy distribution difference value.

[0141] The accuracy of detecting abnormalities in image feature data is effectively improved through the precise calculation formula of abnormality coefficient YCZ and abnormality identification coefficient YJD.

[0142] The anomaly coefficient YCZ is calculated comprehensively through multiple reference features to make the basis for anomaly detection more comprehensive. This multi-dimensional analysis method helps to capture potential subtle differences in image data, reduce the possibility of misjudgment, and improve the system's adaptability to complex scenarios.

[0143] The calculation of the abnormality identification coefficient YJD also uses a comprehensive feature coefficient, which provides the system with multi-level abnormality assessment capabilities. By introducing multiple feature dimensions, the system can classify image abnormalities more carefully and clearly indicate the severity of the abnormalities, thereby providing more accurate guidance for subsequent processing. This hierarchical identification mechanism not only improves processing efficiency, but also helps users quickly take appropriate response measures, ensuring higher security and reliability.

[0144] Optionally, the method further includes:

[0145] Feedback: Feedback the first comparison result and the second comparison result to the display panel.

[0146] The first and second comparison results are fed back to the display panel in real time. Through this action, the user can obtain the analysis results instantly, which is convenient for quick response and decision-making, improves the interactivity and user experience of the system, and ensures that the user always understands the operating status of the system.

[0147] Based on the above improved calculation method, the system no longer relies on the analysis of a single feature when processing image data, but through the comprehensive evaluation of multi-source data, it greatly improves the accuracy and response speed of anomaly detection. This advanced analysis method enables the system to maintain sensitivity and adaptability to abnormal situations in a constantly changing environment, meeting the high requirements of modern big data processing for accuracy and real-time performance.

[0148] like Figure 3 As shown, in the second aspect, the present application provides a rapid detection system based on big data, comprising:

[0149] The first round of data acquisition module 310 is used to collect primary multi-source data of the current image, and based on the primary multi-source data of the current image, collect primary multi-source data of similar images from the big data platform, wherein the similar images are images having the same feature points as the current image;

[0150] The similar images are images with the same feature points in the background, such as the same scenic spot, or images taken at a similar time, or images taken at the same location, etc., and the similar images are extracted from authoritative websites to ensure the validity of similar image samples.

[0151] By extracting primary multi-source data, the basic features in the image can be effectively identified, providing the necessary basic data for subsequent analysis, ensuring that subsequent processing is more efficient, reducing the interference of invalid information, and thus improving the performance of the overall system.

[0152] Acquiring primary multi-source data related to the current image from the big data platform is an efficient data collection method that not only speeds up data processing, but also ensures the diversity and breadth of the collected data, providing a richer basis for subsequent analysis. This advantage effectively solves the problem of insufficient recognition ability caused by insufficient data samples in traditional methods, and further improves the adaptability of the system in different application scenarios. In addition, acquiring a large amount of data on similar images also enhances the breadth and representativeness of the data, thereby improving the accuracy of the analysis and ensuring that potential anomalies can be reliably identified in different situations.

[0153] Optionally, the system can be updated in real time while data is being extracted and collected, which provides a guarantee for timely identification of potential anomalies. This dynamic update mechanism not only enhances the system's responsiveness to new data, but also improves the accuracy and timeliness of decision-making. In general, these improvements not only improve the efficiency of image processing, but also provide more reliable technical support for modern intelligent management, meeting the urgent need for efficient and intelligent analysis.

[0154] The first round data processing module 320 is used to pre-process the primary multi-source data in the current image and the similar images, and organize the pre-processed data into a first data set and a second data set, generate an abnormality coefficient YCZ after integrating and calculating the first data set and the second data set, compare the abnormality coefficient YCZ with a preset first threshold value Y to generate a first comparison result, and determine whether the first image has an abnormality according to the first comparison result;

[0155] The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

[0156] This module cleans and formats the extracted and collected data, and then organizes them into the first data set and the second data set. By eliminating noise and irrelevant information, the data quality is improved, laying the foundation for subsequent integrated calculations, ensuring the reliability of the analysis results, and thus enhancing the effectiveness of the system in processing complex data.

[0157] The first data set includes a first edge feature A1, a second edge feature value A2, a first color space feature value B1, a second color space feature value B2, a first texture feature value C1, a second texture feature value C2, a first image size value D1, and a second image size value D2;

[0158] The second data set includes a first shooting time E1, a second shooting time E2, a first image contrast F1, a second image contrast F2, a first geographical location coordinate G1, a second geographical location coordinate G2, a first image resolution H1, and a second image resolution H2;

[0159] The generated first and second data sets not only comprehensively cover key parameters such as edge features, color space features, texture features, and size of the image, but also feature data such as shooting time, contrast, geographic location coordinates, and resolution. This multi-dimensional data integration enables the system to comprehensively analyze image features and enhance the depth and accuracy of anomaly detection. This process solves the analytical limitations of traditional methods caused by data isolation, improves the system's ability to understand complex image content, and enables the system to better meet the urgent needs of modern intelligent management for efficiency and reliability.

[0160] The primary multi-source data of the current image are a first edge feature A1, a first color space feature value B1, a first texture feature value C1, a first image size value D1, a first shooting time E1, a first image contrast F1, a first geographical location coordinate G1 and a first image resolution H1;

[0161] The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

[0162] This module preprocesses and dimensionlessly converts the extracted and collected primary multi-source data of various images. By integrating and calculating the first data set and the second data set, it generates the abnormality coefficient YCZ. When YCZ≤Y, it means that the image has no abnormality. When YCZ>Y, it means that the image has an abnormality, thereby achieving a quantitative evaluation of the image status. This clear judgment mechanism makes the abnormality identification process more transparent and reliable, and reduces the errors caused by subjective judgment in traditional methods. When the abnormality coefficient YCZ is less than or equal to the threshold Y, the system can accurately judge that there is no abnormality; otherwise, it indicates that there is an abnormality. This dual judgment not only improves the response speed of the system, but also enhances its ability to adapt to different application scenarios, ensuring that it can respond to various complex situations in a timely manner.

[0163] In addition, by quickly providing feedback on abnormal conditions, the system can help users take timely measures to optimize management processes and reduce potential losses. This series of improvements has enabled the overall system to significantly surpass traditional technical means in terms of intelligent monitoring and abnormal management, meeting the high-efficiency needs of modern management.

[0164] like Figure 4 As shown, the system also includes:

[0165] The second round of data acquisition module 330 is used to collect the deep multi-source data of the current image, and based on the deep multi-source data of the current image, collect the deep multi-source data of the similar image from the big data platform;

[0166] If the first comparison result is abnormal, the second round of data acquisition module 330 performs a deep analysis on the current image, that is, executes the steps of: collecting deep multi-source data of the current image, and based on the deep multi-source data of the current image, collecting deep multi-source data of similar images from the big data platform;

[0167] This module deeply analyzes the image content and mines more detailed features, providing a richer and more comprehensive source of information for anomaly identification, enhancing the system's adaptability to complex scenarios and ensuring high-quality detection results in all situations.

[0168] At the same time, the module also enhances the ability to deeply analyze image anomalies. This diversified feature extraction method enables the system to fully capture potential abnormal areas in the image and their related attributes, which is in line with the improvement of intelligent recognition capabilities emphasized in the background introduction.

[0169] The second round of data processing module 340 is used to pre-process the deep multi-source data of the current image and the similar images, and reorganize them into a third data set, a fourth data set, a fifth data set and a sixth data set, generate an abnormality identification coefficient YJD after integrating and calculating the third data set, the fourth data set, the fifth data set and the sixth data set, generate a second comparison result after comparing the abnormality identification coefficient YJD with a preset second threshold R, and judge the abnormality magnitude of the current image according to the second comparison result;

[0170] The extracted and collected various image deep multi-source data are preprocessed and dimensionless, and then the various image deep multi-source data after preprocessing are sorted to generate the third data set, the fourth data set, the fifth data set and the sixth data set; this step ensures the high quality and high availability of the data through systematic processing, provides a more rigorous data foundation for subsequent analysis and calculation, and further improves the overall processing capability of the system.

[0171] In addition, this module makes the processing of primary to deep data more systematic, and can more comprehensively capture abnormal features in images. This innovative method greatly reduces manual intervention and error rates, improves the scientificity and effectiveness of decision-making, and provides strong technical support for various application scenarios, meeting the needs of modern society for efficient and intelligent analysis.

[0172] The third data set includes a first region area value I1, a second region area value I2, a first shape complexity J1, a second shape complexity J2, a first contour length K1, and a second contour length K2;

[0173] The fourth data set includes a first pixel value L1, a second pixel value L2, a first grayscale variance value M1, a second grayscale variance value M2, a first grayscale kurtosis value N1, and a second grayscale kurtosis value N2;

[0174] The fifth data set includes a first area brightness mean value O1, a second area brightness mean value O2, a first brightness variance P1, a second brightness variance P2, a first illuminance distribution value Q1, and a second illuminance distribution value Q2;

[0175] The sixth data set includes a first amplitude spectrum value R1, a second amplitude spectrum value R2, a first high-low frequency difference value U1, a second high-low frequency difference value U2, a first frequency domain energy distribution difference value T1, and a second frequency domain energy distribution difference value T2.

[0176] The third dataset, fourth dataset, fifth dataset, and sixth dataset cover various important features such as regional area, shape complexity, contour length, pixel value, variance, and brightness mean. This multi-dimensional dataset not only enriches the representation of the image, enhances the system's in-depth analysis ability in anomaly detection, but also effectively solves the problem in previous methods that only rely on a single feature and ignore global information. This comprehensive feature extraction and collation ensure that the system can more accurately identify and analyze image anomalies in complex scenarios.

[0177] Integrate and calculate the multiple collated datasets to generate an anomaly identification coefficient YJD, and compare it with a preset second threshold R. This analysis process can quantify the severity of the anomaly, provide a clear basis for subsequent decision-making, improve the accuracy of anomaly recognition, and help users make more effective responses. Through the meticulous processing of the deep multi-source data of the image, the system can provide more accurate anomaly recognition results, providing strong data support for subsequent analysis and decision-making. This series of improvements enhance the overall performance of the system in intelligent monitoring and anomaly management, making it more efficient and accurate when dealing with complex images in the real world, thus meeting the higher requirements of modern management and monitoring for technology.

[0178] Optionally, the second-round data processing module 340 includes a judgment module 341 for:

[0179] When YJD ≤ R, determine that the current picture is a first-level anomaly;

[0180] When R < YJD ≤ R × 110%, determine that the current picture is a second-level anomaly;

[0181] When YJD > R × 110%, determine that the current picture is a third-level anomaly.

[0182] By integrating and calculating to generate the anomaly identification coefficient YJD, it can meticulously evaluate the anomaly degree of the image and provide multi-level anomaly classification. This hierarchical analysis method can flexibly adapt to different application scenarios according to the specific anomaly threshold R, effectively improving the response and processing ability to anomaly events. It not only makes anomaly detection more accurate, but also helps users quickly identify and process anomalies of different severity levels. At the same time, it can reduce the risk of misjudgment, ensure more appropriate measures are taken during the anomaly handling process, and thus improve the overall applicability and stability.

[0183] Optionally, the first-round data processing module 320 includes a first calculation module 321 for calculating the anomaly coefficient YCZ according to the following formula:

[0184] YCZ = a1 × S1 - a2 × S2;

[0185]

[0186]

[0187] Wherein: S1 is the first feature reference coefficient, S2 is the second feature reference coefficient, A1 is the first edge feature, B1 is the first color space feature value, C1 is the first texture feature value, D1 is the first image size value, E1 is the first shooting time, F1 is the first image contrast, G1 is the first geographic location coordinate, H1 is the first image resolution, A2 is the second edge feature, B2 is the second color space feature value, C2 is the second texture feature value, D2 is the second image size value, E2 is the second shooting time, F2 is the second image contrast, G2 is the second geographic location coordinate, H2 is the second image resolution;

[0188] a1, a2, b1, b2 and b3 are weight values, and the values ​​of a1, a2, b1, b2 and b3 are adjusted and set by the user, ln is a logarithmic function, and e is a base function.

[0189] Optionally, the second round data processing module 340 further includes a second calculation module 342, which is used to calculate the abnormal identification coefficient YJD according to the following formula:

[0190] YJD=c1×S3+c2×S4+c3×S5+c4×S5;

[0191]

[0192]

[0193]

[0194]

[0195] Wherein: c1, c2, c3 and c4 are weight values, and the values ​​of c1, c2, c3 and c4 are adjusted and set by the user;

[0196] I1 is the area value of the first region, I2 is the area value of the second region, J1 is the first shape complexity, J2 is the second shape complexity, K1 is the first contour length, and K2 is the second contour length;

[0197] S3 is the abnormal area characteristic coefficient, S4 is the local statistical characteristic coefficient, S5 is the optical characteristic coefficient, and S6 is the frequency domain characteristic coefficient;

[0198] L1 is the first pixel value, L2 is the second pixel value, M1 is the first grayscale variance value, M2 is the second grayscale variance value, N1 is the first grayscale kurtosis value, and N2 is the second grayscale kurtosis value;

[0199] O1 is the mean brightness of the first area, O2 is the mean brightness of the second area, P1 is the first brightness variance, P2 is the second brightness variance, Q1 is the first illuminance distribution value, and Q2 is the second illuminance distribution value;

[0200] R1 is the first amplitude spectrum value, R2 is the second amplitude spectrum value, U1 is the first high and low frequency difference value, U2 is the second high and low frequency difference value, T1 is the first frequency domain energy distribution difference value, and T2 is the second frequency domain energy distribution difference value.

[0201] The accuracy of detecting abnormalities in image feature data is effectively improved through the precise calculation formula of abnormality coefficient YCZ and abnormality identification coefficient YJD.

[0202] The anomaly coefficient YCZ is calculated comprehensively through multiple reference features to make the basis for anomaly detection more comprehensive. This multi-dimensional analysis method helps to capture potential subtle differences in image data, reduce the possibility of misjudgment, and improve the system's adaptability to complex scenarios.

[0203] The calculation of the abnormality identification coefficient YJD also uses a comprehensive feature coefficient, which provides the system with multi-level abnormality assessment capabilities. By introducing multiple feature dimensions, the system can classify image abnormalities more carefully and clearly indicate the severity of the abnormalities, thereby providing more accurate guidance for subsequent processing. This hierarchical identification mechanism not only improves processing efficiency, but also helps users quickly take appropriate response measures, ensuring higher security and reliability.

[0204] Optionally, the system further includes:

[0205] A feedback module is used to feed back the first comparison result and the second comparison result to the display panel.

[0206] The first and second comparison results are fed back to the display panel in real time. Through this action, the user can obtain the analysis results instantly, which is convenient for quick response and decision-making, improves the interactivity and user experience of the system, and ensures that the user always understands the operating status of the system.

[0207] Based on the above improved calculation method, the system no longer relies on the analysis of a single feature when processing image data, but through the comprehensive evaluation of multi-source data, it greatly improves the accuracy and response speed of anomaly detection. This advanced analysis method enables the system to maintain sensitivity and adaptability to abnormal situations in a constantly changing environment, meeting the high requirements of modern big data processing for accuracy and real-time performance.

[0208] In the embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the node embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0209] The modules described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0210] In addition, each functional module in the embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0211] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A rapid detection method based on big data, characterized in that: include: d1, the first round of data acquisition, collecting primary multi-source data of the current image, and based on the primary multi-source data of the current image, collecting primary multi-source data of similar images from the big data platform, wherein the similar images are images having the same feature points as the current image; d2, a first round of data processing, preprocessing the primary multi-source data in the current image and the similar images, and organizing the preprocessed data into a first data set and a second data set, integrating and calculating the first data set and the second data set to generate an abnormality coefficient YCZ, comparing the abnormality coefficient YCZ with a preset first threshold value Y to generate a first comparison result, and judging whether the first image has an abnormality according to the first comparison result; The first data set includes a first edge feature A1, a second edge feature value A2, a first color space feature value B1, a second color space feature value B2, a first texture feature value C1, a second texture feature value C2, a first image size value D1, and a second image size value D2; The second data set includes a first shooting time E1, a second shooting time E2, a first image contrast F1, a second image contrast F2, a first geographical location coordinate G1, a second geographical location coordinate G2, a first image resolution H1, and a second image resolution H2; The primary multi-source data of the current image are a first edge feature A1, a first color space feature value B1, a first texture feature value C1, a first image size value D1, a first shooting time E1, a first image contrast F1, a first geographical location coordinate G1 and a first image resolution H1; The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

2. A rapid detection method based on big data as claimed in claim 1, characterized in that: The method further comprises: if the first comparison result is abnormal, performing a deep analysis on the current image and executing step d3; d3, a second round of data acquisition, collecting deep multi-source data of the current image, and based on the deep multi-source data of the current image, collecting deep multi-source data of similar images from the big data platform; d4, a second round of data processing, pre-processing the deep multi-source data of the current image and the similar images, and reorganizing them into a third data set, a fourth data set, a fifth data set and a sixth data set, integrating and calculating the third data set, the fourth data set, the fifth data set and the sixth data set to generate an abnormality identification coefficient YJD, comparing the abnormality identification coefficient YJD with a preset second threshold R to generate a second comparison result, and judging the abnormal magnitude of the current image according to the second comparison result; The third data set includes a first region area value I1, a second region area value I2, a first shape complexity J1, a second shape complexity J2, a first contour length K1, and a second contour length K2; The fourth data set includes a first pixel value L1, a second pixel value L2, a first gray variance value M1, a second gray variance value M2, a first gray kurtosis value N1, and a second gray kurtosis value N2; The fifth data set includes a first regional brightness mean O1, a second regional brightness mean O2, a first brightness variance P1, a second brightness variance P2, a first illuminance distribution value Q1, and a second illuminance distribution value Q2; The sixth data set includes a first amplitude spectrum value R1, a second amplitude spectrum value R2, a first high-low frequency difference U1, a second high-low frequency difference U2, a first frequency domain energy distribution difference T1, and a second frequency domain energy distribution difference T2.

3. A rapid detection method based on big data as claimed in claim 2, characterized in that: The specific determination of the abnormality level of the first image according to the second comparison result is as follows: When YJD ≤ R, the current picture is a first-level abnormality; When R < YJD ≤ R × 110%, the current picture is a second-level abnormality; When YJD > R × 110%, the current picture is a third-level abnormality.

4. A rapid detection method based on big data as described in any one of claims 1 to 3, characterized in that: The abnormality coefficient YCZ is obtained by calculating through the following formula; YCZ = a1 × S1 - a2 × S2; In the formula: S1 is the first feature reference coefficient, S2 is the second feature reference coefficient, A1 is the first edge feature, B1 is the first color space feature value, C1 is the first texture feature value, D1 is the first image size value, E1 is the first shooting time, F1 is the first image contrast, G1 is the first geographical location coordinate, H1 is the first image resolution, A2 is the second edge feature, B2 is the second color space feature value, C2 is the second texture feature value, D2 is the second image size value, E2 is the second shooting time, F2 is the second image contrast, G2 is the second geographical location coordinate, H2 is the second image resolution; a1, a2, b1, b2, and b3 are weight values, and the values of a1, a2, b1, b2, and b3 are adjusted and set by the user, ln is the logarithmic function, and e is the base function.

5. A rapid detection method based on big data as described in any one of claims 2-3, characterized in that: The abnormality identification coefficient YJD is obtained by calculating through the following formula; YJD = c1 × S3 + c2 × S4 + c3 × S5 + c4 × S5; In the formula: c1, c2, c3, and c4 are weight values, and the values of c1, c2, c3, and c4 are adjusted and set by the user; I1 is the first regional area value, I2 is the second regional area value, J1 is the first shape complexity, J2 is the second shape complexity, K1 is the first contour length, and K2 is the second contour length; S3 is the abnormal area feature coefficient, S4 is the local statistical feature coefficient, S5 is the light feature coefficient, and S6 is the frequency domain feature coefficient; L1 is the first pixel value, L2 is the second pixel value, M1 is the first gray variance value, M2 is the second gray variance value, N1 is the first gray kurtosis value, and N2 is the second gray kurtosis value; O1 is the first regional brightness mean, O2 is the second regional brightness mean, P1 is the first brightness variance, P2 is the second brightness variance, Q1 is the first illuminance distribution value, and Q2 is the second illuminance distribution value; R1 is the first amplitude spectrum value, R2 is the second amplitude spectrum value, U1 is the first high and low frequency difference value, U2 is the second high and low frequency difference value, T1 is the first frequency domain energy distribution difference value, and T2 is the second frequency domain energy distribution difference value.

6. A rapid detection system based on big data, characterized in that: include: A first round of data acquisition module is used to collect primary multi-source data of the current image, and based on the primary multi-source data of the current image, collect primary multi-source data of similar images from the big data platform, wherein the similar images are images having the same feature points as the current image; A first round of data processing module, used for preprocessing the primary multi-source data in the current image and the similar images, and arranging the preprocessed data into a first data set and a second data set, generating an abnormality coefficient YCZ after integrating and calculating the first data set and the second data set, generating a first comparison result after comparing the abnormality coefficient YCZ with a preset first threshold value Y, and judging whether the first image has an abnormality according to the first comparison result; The first data set includes a first edge feature A1, a second edge feature value A2, a first color space feature value B1, a second color space feature value B2, a first texture feature value C1, a second texture feature value C2, a first image size value D1, and a second image size value D2; The second data set includes a first shooting time E1, a second shooting time E2, a first image contrast F1, a second image contrast F2, a first geographical location coordinate G1, a second geographical location coordinate G2, a first image resolution H1, and a second image resolution H2; The primary multi-source data of the current image are a first edge feature A1, a first color space feature value B1, a first texture feature value C1, a first image size value D1, a first shooting time E1, a first image contrast F1, a first geographical location coordinate G1 and a first image resolution H1; The primary multi-source data of the similar image are a second edge feature A2, a second color space feature value B2, a second texture feature value C2, a second image size value D2, a second shooting time E2, a second image contrast F2, a second geographic location coordinate G2 and a second image resolution H2.

7. A rapid detection system based on big data as claimed in claim 6, characterized in that: The system further comprises: A second round of data acquisition module is used to collect the deep multi-source data of the current image, and based on the deep multi-source data of the current image, collect the deep multi-source data of the similar image from the big data platform; A second round of data processing module is used to pre-process the deep multi-source data of the current image and the similar images, and reorganize them into a third data set, a fourth data set, a fifth data set and a sixth data set, generate an abnormality identification coefficient YJD after integrating and calculating the third data set, the fourth data set, the fifth data set and the sixth data set, generate a second comparison result after comparing the abnormality identification coefficient YJD with a preset second threshold R, and judge the abnormality magnitude of the current image according to the second comparison result; The third data set includes a first region area value I1, a second region area value I2, a first shape complexity J1, a second shape complexity J2, a first contour length K1, and a second contour length K2; The fourth data set includes a first pixel value L1, a second pixel value L2, a first gray variance value M1, a second gray variance value M2, a first gray kurtosis value N1, and a second gray kurtosis value N2; The fifth data set includes a first regional brightness mean O1, a second regional brightness mean O2, a first brightness variance P1, a second brightness variance P2, a first illuminance distribution value Q1, and a second illuminance distribution value Q2; The sixth data set includes a first amplitude spectrum value R1, a second amplitude spectrum value R2, a first high-low frequency difference U1, a second high-low frequency difference U2, a first frequency domain energy distribution difference T1, and a second frequency domain energy distribution difference T2.

8. A rapid detection system based on big data as claimed in claim 7, characterized in that: The second-round data processing module includes a judgment module for: When YJD ≤ R, determining that the current picture is a first-level anomaly; When R < YJD ≤ R × 110%, determining that the current picture is a second-level anomaly; When YJD > R × 110%, determining that the current picture is a third-level anomaly.

9. A rapid detection system based on big data as described in any one of claims 6 to 8, characterized in that: The first-round data processing module includes a first calculation module for calculating the anomaly coefficient YCZ according to the following formula: YCZ = a1 × S1 - a2 × S2; Where: S1 is the first feature reference coefficient, S2 is the second feature reference coefficient, A1 is the first edge feature, B1 is the first color space feature value, C1 is the first texture feature value, D1 is the first image size value, E1 is the first shooting time, F1 is the first image contrast, G1 is the first geographical location coordinate, H1 is the first image resolution, A2 is the second edge feature, B2 is the second color space feature value, C2 is the second texture feature value, D2 is the second image size value, E2 is the second shooting time, F2 is the second image contrast, G2 is the second geographical location coordinate, H2 is the second image resolution; a1, a2, b1, b2, and b3 are weight values, and the values of a1, a2, b1, b2, and b3 are adjusted and set by the user, ln is the logarithmic function, and e is the base function.

10. A rapid detection system based on big data as described in any one of claims 7-8, characterized in that: The second-round data acquisition module includes a second calculation module for calculating the anomaly identification coefficient YJD according to the following formula: YJD = c1 × S3 + c2 × S4 + c3 × S5 + c4 × S5; Where: c1, c2, c3, and c4 are weight values, and the values of c1, c2, c3, and c4 are adjusted and set by the user; I1 is the first regional area value, I2 is the second regional area value, J1 is the first shape complexity, J2 is the second shape complexity, K1 is the first contour length, and K2 is the second contour length; S3 is the anomaly region feature coefficient, S4 is the local statistical feature coefficient, S5 is the light feature coefficient, and S6 is the frequency domain feature coefficient; L1 is the first pixel value, L2 is the second pixel value, M1 is the first gray variance value, M2 is the second gray variance value, N1 is the first gray kurtosis value, and N2 is the second gray kurtosis value; O1 is the first regional brightness mean, O2 is the second regional brightness mean, P1 is the first brightness variance, P2 is the second brightness variance, Q1 is the first illuminance distribution value, and Q2 is the second illuminance distribution value; R1 is the first amplitude spectrum value, R2 is the second amplitude spectrum value, U1 is the first high and low frequency difference value, U2 is the second high and low frequency difference value, T1 is the first frequency domain energy distribution difference value, and T2 is the second frequency domain energy distribution difference value.

Images