CN114782711A - Intelligent risk detection method and system based on image recognition - Google Patents
Intelligent risk detection method and system based on image recognition Download PDFInfo
- Publication number
- CN114782711A CN114782711A CN202210696467.9A CN202210696467A CN114782711A CN 114782711 A CN114782711 A CN 114782711A CN 202210696467 A CN202210696467 A CN 202210696467A CN 114782711 A CN114782711 A CN 114782711A
- Authority
- CN
- China
- Prior art keywords
- image
- brightness
- features
- target image
- shape
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Abstract
The invention relates to the field related to information risk retrieval, and discloses an intelligent risk detection method and system based on image identification.
Description
Technical Field
The invention relates to the related field of information risk retrieval, in particular to an intelligent risk detection method and system based on image identification.
Background
With the rapid development of networks, the expansion of information resources brings great convenience, and simultaneously makes the management of information very difficult, in some industries, the situations of stealing and repurposing of image materials often occur, if the images are not effectively and reliably detected and evaluated before being put into use, a series of problems may occur after being used, and adverse effects are caused.
In the prior art, most of risk retrieval and evaluation of images adopt overall characteristics, colors and the like of the images to retrieve, so that network image objects with consistency or high coincidence degree are screened out.
Disclosure of Invention
The invention aims to provide an intelligent risk detection method and system based on image recognition, so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
an intelligent risk detection system based on image recognition, comprising:
the spatial feature extraction module is used for carrying out feature segmentation processing on the target image, acquiring a plurality of shape features of the target image and establishing spatial relationship features among the shape features, wherein the shape features are used for representing basic graphic line composition elements of the target image;
the characteristic brightness establishing module is used for acquiring color channel information of the target image in a plurality of shape characteristic demarcating areas, calculating a change value of the color channel information of adjacent areas, and performing brightness marking on the shape characteristics according to a preset brightness setting method and the change value, wherein the color channel information is used for representing color mixing information and color brightness information;
the retrieval feature reduction module is used for performing resampling processing on the shape features according to the brightness marks to obtain screening relation features, the resampling processing is used for extracting a plurality of shape features of which the variation values are larger than a preset standard to establish the screening relation features, and the spatial relation features comprise the screening relation features;
and the risk detection judging module is used for screening the comparison library according to the screening relation characteristics, acquiring a plurality of comparison images reaching the preset contact ratio, comparing and screening the plurality of comparison images according to the spatial relation characteristics, and generating a risk comparison result, wherein the risk comparison result is used for representing the possibility of infringement or embezzlement of the target image and the corresponding comparison images.
As a further scheme of the invention: still include the segmentation of region module, the segmentation of region module specifically includes:
an object identification unit, configured to identify a feature subject and a background environment of the target image and generate a range label, where if the feature subject is not identified, the step of generating the range label is not performed, where the feature subject is used to represent non-background environment picture content in the target image and may be a person or an animal;
and the object segmentation unit is used for segmenting the target image according to the range mark and respectively acquiring a target main body image and a target environment image, wherein the target main body image and the target environment are both used for establishing spatial relationship characteristics and screening to generate a risk comparison result.
As a further scheme of the invention: the feature brightness establishing module includes a brightness flag unit for performing the brightness setting method, the brightness flag unit including:
the span analysis subunit is used for acquiring the change values of a plurality of adjacent areas, acquiring a maximum value and a minimum value of the change values by sorting the size of the change values, and segmenting the span areas of the minimum value and the maximum value to acquire brightness levels of a plurality of brightness marks;
and the marking subunit is used for judging the brightness level of the change value of the shape characteristic at the division position of the adjacent area and marking the brightness according to the brightness level of the judgment result.
As a further scheme of the invention: the image feature extraction module includes:
the shape feature acquisition unit is used for automatically segmenting the target image according to the color of the target image to acquire a plurality of image segmentation lines, wherein the image segmentation lines are the shape features and are used for representing basic graphic lines forming the target image, and the automatic segmentation step is only used for segmenting adjacent color blocks with discontinuous change;
and the spatial feature establishing unit is used for analyzing the shape features, randomly selecting a preset number of shape features from various positions in the target image, analyzing the position relationship among the shape features, and generating spatial relationship features based on the position relationship and the shape features.
As a still further scheme of the invention: the discontinuous change of the color blocks depends on the sampling color depth of the shape feature acquisition unit and the sampling resolution of the target image, the sampling color depth is in direct proportion to the color types which can be identified by the shape feature acquisition unit, and therefore when the color blocks are of continuous transition gradual change colors, the sampling depth is in direct proportion to the number of generated image dividing lines under the same sampling resolution.
The embodiment of the invention aims to provide an intelligent risk detection method based on image identification, which comprises the following steps:
performing feature segmentation processing on the target image to obtain a plurality of shape features of the target image, and establishing spatial relationship features among the shape features, wherein the shape features are used for representing basic figure line composition elements of the target image;
acquiring color channel information of the target image in a plurality of shape feature demarcated areas, calculating a change value of the color channel information of adjacent areas, and performing brightness marking on the shape features according to a preset brightness setting method and the change value, wherein the color channel information is used for representing color mixing information and color brightness information;
resampling the shape features according to the brightness marks to obtain screening relation features, wherein the resampling process is used for extracting a plurality of shape features of which the variation values are larger than a preset standard to establish the screening relation features, and the spatial relation features comprise the screening relation features;
and screening a comparison library according to the screening relation characteristics to obtain a plurality of comparison images reaching a preset contact ratio, comparing and screening the plurality of comparison images according to the spatial relation characteristics to generate a risk comparison result, wherein the risk comparison result is used for representing the possibility of infringement or embezzlement of the target image and the corresponding comparison images.
As a further scheme of the invention, the method also comprises the following steps:
identifying a characteristic subject and a background environment of the target image to generate a range mark, and if the characteristic subject is not identified, not executing the step of generating the range mark, wherein the characteristic subject is used for representing the non-background environment picture content in the target image and can be a person or an animal;
and segmenting the target image according to the range mark, and respectively obtaining a target main body image and a target environment image, wherein the target main body image and the target environment are used for establishing spatial relationship characteristics and screening to generate a risk comparison result.
As a still further scheme of the invention: the step of brightness marking the shape feature according to a preset brightness setting method and the variation value specifically includes:
acquiring the change values of a plurality of adjacent areas, acquiring a maximum value and a minimum value of the change values by sequencing the size of the change values, and segmenting a span area of the minimum value and the maximum value to acquire brightness marks of a plurality of levels;
and judging the brightness level of the change value of the shape feature at the division position of the adjacent region, and marking brightness according to the brightness level of the judgment result.
Compared with the prior art, the invention has the beneficial effects that: through the arrangement of the spatial feature extraction module, the feature brightness establishment module, the retrieval feature reduction module and the risk detection judgment module, the risk retrieval based on the image shape features and the spatial relationship features is realized, meanwhile, based on the establishment of brightness marks, the traditional image retrieval traceability retrieval mode based on color judgment can be effectively replaced, the avoidance of image embezzlement behaviors through the modes of modifying the original image color structure and the like to risk assessment can be reliably avoided, and the reliability of risk detection is greatly improved.
Drawings
Fig. 1 is a block diagram of an intelligent risk detection system based on image recognition.
FIG. 2 is a block diagram of brightness mark units in an intelligent risk detection system based on image recognition.
Fig. 3 is a block diagram of an image feature extraction module in an intelligent risk detection system based on image recognition.
Fig. 4 is a flow chart of an intelligent risk detection method based on image recognition.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following detailed description of specific embodiments of the present invention is provided in connection with specific embodiments.
As shown in fig. 1, an intelligent risk detection system based on image recognition provided for an embodiment of the present invention includes:
the spatial feature extraction module 100 is configured to perform feature segmentation processing on the target image, acquire a plurality of shape features of the target image, and establish a spatial relationship feature between the plurality of shape features, where the shape features are used to represent basic graphic line composition elements of the target image.
A feature brightness establishing module 300, configured to obtain color channel information of the target image in a plurality of shape feature demarcated regions, calculate a change value of the color channel information of an adjacent region, and perform brightness marking on the shape feature according to a preset brightness setting method and the change value, where the color channel information is used to represent color mixture information and color brightness information.
A retrieval feature reduction module 500, configured to perform resampling processing on the shape features according to the brightness marks, so as to obtain a screening relationship feature, where the resampling processing is used to extract a plurality of shape features whose variation values are greater than a preset standard, so as to establish the screening relationship feature, and the spatial relationship feature includes the screening relationship feature.
The risk detection and determination module 700 is configured to screen a comparison library according to the screening relationship features, obtain a plurality of comparison images that reach a preset coincidence degree, compare and screen the plurality of comparison images according to the spatial relationship features, and generate a risk comparison result, where the risk comparison result is used to represent the possibility of infringement or theft of the target image and the corresponding comparison images.
In the embodiment, an intelligent risk detection system is provided, which is used for retrieving an image and a picture so as to inquire the originality of the picture, under the background of rapid network development at present, phenomena such as material embezzlement, plagiarism and the like become more and more serious, and the system is very important for the benefit safety guarantee of creators; the feature brightness establishing module 300 is used for distinguishing lines of shape features, because on a target image, the number of shape features that can be generated is large, and therefore a primary color and a secondary color need to be distinguished for convenient use and search, where the change judgment of color channel information is adopted, that is, if the color channel information between two adjacent color blocks is very close in value, the brightness is marked as 1 (assuming that the brightness is in the range of 1 to 10, where 10 is the largest change of the color channel information in the image), if the color difference between two adjacent color blocks is very large, the color channel information is marked as 10, at this time, we use the brightness mark as 10 in a detail mode to achieve a better effect, and these lines may be the main structure forming lines of the target image, so that the shape features are reduced by the search feature reducing module 500 (meanwhile, the spatial relationship corresponding to the shapes is also reduced), the bober neon risk judgment module 700 rapidly screens and finally compares the risk values; compared with the prior art, the processing mode can better perform source tracing retrieval on the target image, so that the risk that the target image is embezzled and reused is judged, and particularly for image reuse, the problem that the prior art cannot effectively retrieve, identify and trace the source can be solved.
As another preferred embodiment of the present invention, the present invention further includes a region segmentation module, and the region segmentation module specifically includes:
and the object identification unit is used for identifying a characteristic subject and a background environment of the target image to generate a range mark, and if the characteristic subject is not identified, the step of generating the range mark is not executed, wherein the characteristic subject is used for representing the non-background environment picture content in the target image and can be a person or an animal.
And the object segmentation unit is used for segmenting the target image according to the range mark and respectively acquiring a target main body image and a target environment image, wherein the target main body image and the target environment are both used for establishing spatial relationship characteristics and screening to generate a risk comparison result.
In the embodiment, the region division module is supplemented and is explained with functional division, when in use, the main function is to identify and segment the main body element and the background element in the image, for example, when the target image is a cartoon in a composition such as a novel, a game, etc., the target image usually includes a character and a background, and in the prior art, in the search for risk assessment, the search assessment of the overall characteristics is usually adopted, in fact, unless it is a direct case of embezzlement, in many cases separate alterations of backgrounds and tasks occur, i.e. backgrounds and persons from different images, eventually combined after a simple modification (for example colour matching of the background, a few detailed items; coloration of the garment of the person, etc.), therefore, the existing method for directly retrieving and evaluating is difficult to reliably evaluate, or partial evaluation causes that the evaluation result has no reference.
As shown in fig. 2, as another preferred embodiment of the present invention, the feature brightness establishing module 300 includes a brightness flag unit 301 for performing the brightness setting method, the brightness flag unit 301 including:
the span analysis subunit 3011 is configured to obtain the variation values of a plurality of adjacent areas, obtain a maximum value and a minimum value of the variation values by sorting the magnitude of the variation values, and segment the span areas of the minimum value and the maximum value to obtain brightness levels of the brightness marks.
A marking subunit 3012, configured to perform brightness level determination on the change value of the shape feature at the division of the adjacent area, and perform brightness marking according to the brightness level of the determination result.
In this embodiment, the brightness setting method is described, and in this process, the brightness level is established based on the span of the maximum value and the minimum value of the variation value in the image, and the number of levels is uniform.
As shown in fig. 3, as another preferred embodiment of the present invention, the image feature extraction module 100 includes:
a shape feature obtaining unit 101, configured to automatically segment the target image according to the color of the target image, and obtain a plurality of image segmentation lines, where the image segmentation lines are the shape features and are used to represent basic graphic lines constituting the target image, and the step of automatically segmenting is only used to segment adjacent color blocks that change discontinuously.
The spatial feature establishing unit 102 is configured to analyze the shape features, randomly select a preset number of shape features from various positions in the target image, analyze a position relationship between the shape features, and generate a spatial relationship feature based on the position relationship and the shape features.
Further, the discontinuous variation of the color blocks depends on the sampling color depth of the shape feature obtaining unit 101 and the sampling resolution of the target image, and the sampling color depth is proportional to the color type that can be identified by the shape feature obtaining unit 101, so when the color blocks are transition colors, the sampling depth is proportional to the number of the generated image dividing lines at the same sampling resolution.
In this embodiment, when performing automatic segmentation to obtain image segmentation lines, that is, shape features, two contents, namely, sampling color depth and sampling resolution are involved, where the sampling color depth determines the type of color that can be identified, and here, as an extreme example, for a complete natural light spectrum strip, when the sampling color depth is infinite, the sampling result is continuous, and when the sampling color depth becomes very low, the sampling result is a strip-shaped light strip, and a segmentation line (that is, a true color) is generated between adjacent color strips, so that the sampling depth can change the number of generated shape features to some extent.
As shown in fig. 4, the present invention further provides an intelligent risk detection method based on image recognition, which includes:
s200, performing feature segmentation processing on the target image, acquiring a plurality of shape features of the target image, and establishing spatial relationship features among the shape features, wherein the shape features are used for representing basic figure line composition elements of the target image.
S400, obtaining color channel information of the target image in a plurality of shape feature demarcating areas, calculating a change value of the color channel information of adjacent areas, and performing brightness marking on the shape features according to a preset brightness setting method and the change value, wherein the color channel information is used for representing color mixing information and color brightness information.
S600, resampling the shape features according to the lightness marks to obtain screening relation features, wherein the resampling process is used for extracting a plurality of shape features of which the variation values are larger than a preset standard to establish the screening relation features, and the spatial relation features comprise the screening relation features.
S800, screening comparison libraries according to the screening relation characteristics to obtain a plurality of comparison images reaching a preset coincidence degree, comparing and screening the plurality of comparison images according to the spatial relation characteristics to generate a risk comparison result, wherein the risk comparison result is used for representing the possibility of infringement or embezzlement of the target image and the corresponding comparison images.
As another preferred embodiment of the present invention, further comprising the steps of:
and identifying a characteristic subject and a background environment of the target image to generate a range mark, and if the characteristic subject is not identified, not executing the step of generating the range mark, wherein the characteristic subject is used for representing the non-background environment picture content in the target image and can be a person or an animal.
And segmenting the target image according to the range mark, and respectively obtaining a target main body image and a target environment image, wherein the target main body image and the target environment are used for establishing spatial relationship characteristics and screening to generate a risk comparison result.
As another preferred embodiment of the present invention, the step of performing brightness marking on the shape feature according to a preset brightness setting method and the variation value specifically includes:
and acquiring the change values of a plurality of adjacent areas, acquiring a maximum value and a minimum value of the change values by sequencing the size of the change values, and segmenting the span areas of the minimum value and the maximum value to acquire brightness marks of a plurality of levels.
And judging the brightness level of the change value of the shape feature at the division position of the adjacent region, and marking brightness according to the brightness level of the judgment result.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by a computer program, which may be stored in a non-volatile computer readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (8)
1. An intelligent risk detection system based on image recognition, comprising:
the spatial feature extraction module is used for carrying out feature segmentation processing on a target image, acquiring a plurality of shape features of the target image and establishing spatial relationship features among the shape features, wherein the shape features are used for representing basic graphic line composition elements of the target image;
the characteristic brightness establishing module is used for acquiring color channel information of the target image in a plurality of shape characteristic demarcating areas, calculating a change value of the color channel information of adjacent areas, and performing brightness marking on the shape characteristics according to a preset brightness setting method and the change value, wherein the color channel information is used for representing color mixing information and color brightness information;
the retrieval feature reduction module is used for performing resampling processing on the shape features according to the brightness marks to obtain screening relation features, the resampling processing is used for extracting a plurality of shape features of which the variation values are larger than a preset standard to establish the screening relation features, and the spatial relation features comprise the screening relation features;
and the risk detection judging module is used for screening the comparison library according to the screening relation characteristics, acquiring a plurality of comparison images reaching the preset contact ratio, comparing and screening the plurality of comparison images according to the spatial relation characteristics, and generating a risk comparison result, wherein the risk comparison result is used for representing the possibility of infringement or embezzlement of the target image and the corresponding comparison images.
2. The intelligent risk detection system based on image recognition according to claim 1, further comprising a region segmentation module, wherein the region segmentation module specifically comprises:
an object recognition unit, configured to perform recognition of a feature subject and a background environment on the target image and generate a range mark, and if the feature subject is not recognized, the step of generating the range mark is not performed, where the feature subject is used to represent non-background environment picture content in the target image and may be a person or an animal;
and the object segmentation unit is used for segmenting the target image according to the range mark and respectively acquiring a target main body image and a target environment image, wherein the target main body image and the target environment are both used for establishing spatial relationship characteristics and screening to generate a risk comparison result.
3. The intelligent risk detection system based on image recognition according to claim 2, wherein the feature brightness establishing module comprises a brightness marking unit for performing the brightness setting method, the brightness marking unit comprising:
the span analysis subunit is used for acquiring the change values of a plurality of adjacent areas, acquiring a maximum value and a minimum value of the change values by sorting the size of the change values, and segmenting the span areas of the minimum value and the maximum value to acquire brightness levels of a plurality of brightness marks;
and the marking subunit is used for judging the brightness level of the change value of the shape characteristic at the division position of the adjacent area and marking the brightness according to the brightness level of the judgment result.
4. The intelligent risk detection system based on image recognition according to claim 1, wherein the image feature extraction module comprises:
a shape feature obtaining unit, configured to automatically segment the target image according to a color of the target image, and obtain a plurality of image segmentation lines, where the image segmentation lines are the shape features and are used to represent basic graphic lines constituting the target image, and the step of automatically segmenting is only used to segment adjacent color blocks that change discontinuously;
and the spatial feature establishing unit is used for analyzing the shape features, randomly selecting a preset number of shape features from various positions in the target image, analyzing the position relationship among the shape features, and generating spatial relationship features based on the position relationship and the shape features.
5. The intelligent risk detection system based on image recognition of claim 4, wherein the discontinuous variation of the color blocks depends on the sampling color depth of the shape feature obtaining unit and the sampling resolution of the target image, the sampling color depth is proportional to the color types that can be recognized by the shape feature obtaining unit, so when the color blocks are the gradual change colors of continuous transition, the sampling depth is proportional to the number of the generated image dividing lines at the same sampling resolution.
6. An intelligent risk detection method based on image recognition is characterized by comprising the following steps:
performing feature segmentation processing on a target image to obtain a plurality of shape features of the target image, and establishing spatial relationship features among the shape features, wherein the shape features are used for representing basic figure line composition elements of the target image;
acquiring color channel information of the target image in a plurality of shape feature demarcating areas, calculating a change value of the color channel information of adjacent areas, and performing brightness marking on the shape features according to a preset brightness setting method and the change value, wherein the color channel information is used for representing color mixing information and color brightness information;
resampling the shape features according to the lightness marks to obtain screening relation features, wherein the resampling process is used for extracting a plurality of shape features of which the variation values are larger than a preset standard to establish the screening relation features, and the spatial relation features comprise the screening relation features;
and screening a comparison library according to the screening relation characteristics to obtain a plurality of comparison images reaching a preset contact ratio, comparing and screening the plurality of comparison images according to the spatial relation characteristics to generate a risk comparison result, wherein the risk comparison result is used for representing the possibility of infringement or embezzlement of the target image and the corresponding comparison images.
7. The intelligent risk detection method based on image recognition according to claim 6, further comprising the steps of:
identifying a characteristic subject and a background environment of the target image to generate a range mark, and if the characteristic subject is not identified, not executing the step of generating the range mark, wherein the characteristic subject is used for representing the non-background environment picture content in the target image and can be a person or an animal;
and segmenting the target image according to the range mark, and respectively obtaining a target main body image and a target environment image, wherein the target main body image and the target environment are used for establishing spatial relationship characteristics and screening to generate a risk comparison result.
8. The intelligent risk detection method based on image recognition according to claim 7, wherein the step of lightness marking the shape feature according to a preset lightness setting method and the variation value specifically comprises:
acquiring the change values of a plurality of adjacent areas, acquiring a maximum value and a minimum value of the change values by sorting the change values, and segmenting a span area of the minimum value and the maximum value to acquire brightness marks of a plurality of levels;
and judging the brightness level of the change value of the shape feature at the division position of the adjacent region, and marking the brightness according to the brightness level of the judgment result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210696467.9A CN114782711B (en) | 2022-06-20 | 2022-06-20 | Intelligent risk detection method and system based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210696467.9A CN114782711B (en) | 2022-06-20 | 2022-06-20 | Intelligent risk detection method and system based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114782711A true CN114782711A (en) | 2022-07-22 |
| CN114782711B CN114782711B (en) | 2022-09-16 |
Family
ID=82420591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210696467.9A Active CN114782711B (en) | 2022-06-20 | 2022-06-20 | Intelligent risk detection method and system based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114782711B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116030115A (en) * | 2023-03-22 | 2023-04-28 | 四川航天职业技术学院(四川航天高级技工学校) | Visual design image analysis method and system applied to AI |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110243434A1 (en) * | 2008-11-11 | 2011-10-06 | Panasonic Corporation | Feature value extracting device, object identification device, and feature value extracting method |
| CN103824081A (en) * | 2014-02-24 | 2014-05-28 | 北京工业大学 | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition |
| CN106446925A (en) * | 2016-07-07 | 2017-02-22 | 哈尔滨工程大学 | Dolphin identity recognition method based on image processing |
| CN106485700A (en) * | 2016-09-23 | 2017-03-08 | 电子科技大学 | A kind of automatic testing method of the renal cells based on convolutional neural networks |
| US20200364860A1 (en) * | 2019-05-16 | 2020-11-19 | Retrace Labs | Artificial Intelligence Architecture For Identification Of Periodontal Features |
| CN112232303A (en) * | 2020-11-16 | 2021-01-15 | 内蒙古自治区农牧业科学院 | Grassland road information extraction method based on high-resolution remote sensing image |
| CN112884740A (en) * | 2021-02-20 | 2021-06-01 | 联想(北京)有限公司 | Image detection method and device, electronic equipment and storage medium |
| CN112991238A (en) * | 2021-02-22 | 2021-06-18 | 上海市第四人民医院 | Texture and color mixing type food image segmentation method, system, medium and terminal |
| CN112990792A (en) * | 2021-05-11 | 2021-06-18 | 北京智源人工智能研究院 | Method and device for automatically detecting infringement risk and electronic equipment |
| CN112989098A (en) * | 2021-05-08 | 2021-06-18 | 北京智源人工智能研究院 | Automatic retrieval method and device for image infringement entity and electronic equipment |
| CN113240682A (en) * | 2021-05-31 | 2021-08-10 | 华中科技大学 | Overturn-preventing construction driving map generation method and system for crawler crane |
| CN114638825A (en) * | 2022-05-12 | 2022-06-17 | 深圳市联志光电科技有限公司 | Defect detection method and system based on image segmentation |
-
2022
- 2022-06-20 CN CN202210696467.9A patent/CN114782711B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110243434A1 (en) * | 2008-11-11 | 2011-10-06 | Panasonic Corporation | Feature value extracting device, object identification device, and feature value extracting method |
| CN103824081A (en) * | 2014-02-24 | 2014-05-28 | 北京工业大学 | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition |
| CN106446925A (en) * | 2016-07-07 | 2017-02-22 | 哈尔滨工程大学 | Dolphin identity recognition method based on image processing |
| CN106485700A (en) * | 2016-09-23 | 2017-03-08 | 电子科技大学 | A kind of automatic testing method of the renal cells based on convolutional neural networks |
| US20200364860A1 (en) * | 2019-05-16 | 2020-11-19 | Retrace Labs | Artificial Intelligence Architecture For Identification Of Periodontal Features |
| CN112232303A (en) * | 2020-11-16 | 2021-01-15 | 内蒙古自治区农牧业科学院 | Grassland road information extraction method based on high-resolution remote sensing image |
| CN112884740A (en) * | 2021-02-20 | 2021-06-01 | 联想(北京)有限公司 | Image detection method and device, electronic equipment and storage medium |
| CN112991238A (en) * | 2021-02-22 | 2021-06-18 | 上海市第四人民医院 | Texture and color mixing type food image segmentation method, system, medium and terminal |
| CN112989098A (en) * | 2021-05-08 | 2021-06-18 | 北京智源人工智能研究院 | Automatic retrieval method and device for image infringement entity and electronic equipment |
| CN112990792A (en) * | 2021-05-11 | 2021-06-18 | 北京智源人工智能研究院 | Method and device for automatically detecting infringement risk and electronic equipment |
| CN113240682A (en) * | 2021-05-31 | 2021-08-10 | 华中科技大学 | Overturn-preventing construction driving map generation method and system for crawler crane |
| CN114638825A (en) * | 2022-05-12 | 2022-06-17 | 深圳市联志光电科技有限公司 | Defect detection method and system based on image segmentation |
Non-Patent Citations (4)
| Title |
|---|
| FENG WU等: "《Design of a Computer-Based Legal Information Retrieval System》", 《COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE》 * |
| 张财广: "《光学卫星遥感图像舰船目标检测综述》", 《电波科学学报》 * |
| 罗妙辉: "《基于图像内容检索技术的纺织品图像侵权检测》", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
| 高秀东 等: "《高职电子商务专业核心课程教学研究与实践》", 《当代教育实践与教学研究》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116030115A (en) * | 2023-03-22 | 2023-04-28 | 四川航天职业技术学院(四川航天高级技工学校) | Visual design image analysis method and system applied to AI |
| CN116030115B (en) * | 2023-03-22 | 2023-06-02 | 四川航天职业技术学院(四川航天高级技工学校) | Visual design image analysis method and system applied to AI |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114782711B (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110930353B (en) | Method and device for detecting state of hole site protection door, computer equipment and storage medium | |
| CN111583180B (en) | Image tampering identification method and device, computer equipment and storage medium | |
| JP6882874B6 (en) | Ground change interpretation support device, ground change interpretation support method, and program | |
| CN111753692A (en) | Target object extraction method, product detection method, device, computer and medium | |
| CN114782711B (en) | Intelligent risk detection method and system based on image recognition | |
| CN111260645B (en) | Tampered image detection method and system based on block classification deep learning | |
| Jeong et al. | Neural network-based text location for news video indexing | |
| CN110751013A (en) | Scene recognition method, device and computer-readable storage medium | |
| CN111797830A (en) | Rapid red seal detection method, system and device for bill image | |
| CN115019310B (en) | Image-text identification method and equipment | |
| CN116091503A (en) | Method, device, equipment and medium for discriminating panel foreign matter defects | |
| CN112861861B (en) | Method and device for recognizing nixie tube text and electronic equipment | |
| CN111402185A (en) | Image detection method and device | |
| Lu et al. | An Efficient Detection Approach of Content Aware Image Resizing. | |
| CN111046878B (en) | Data processing method and device, computer storage medium and computer | |
| CN115393748A (en) | Method for detecting infringement trademark based on Logo recognition | |
| CN112884740A (en) | Image detection method and device, electronic equipment and storage medium | |
| CN112203053A (en) | Intelligent supervision method and system for subway constructor behaviors | |
| Gao et al. | True color distributions of scene text and background | |
| US20080187229A1 (en) | Method For Recognizing and Indexing Digital Media | |
| CN116092099B (en) | Multi-target administrative law enforcement document information integrity recognition detection method and system | |
| CN115273061B (en) | Image content level extraction method and system based on principal component analysis | |
| CN110674830B (en) | Image privacy identification method and device, computer equipment and storage medium | |
| Chen | Bank Card Number Identification Program Based on Template Matching | |
| KR20240011567A (en) | Method and device for processing building exteriror image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |