CN115393617A - Simulated trademark rapid detection method and system based on multi-convolution kernel inspection - Google Patents

Simulated trademark rapid detection method and system based on multi-convolution kernel inspection Download PDF

Info

Publication number
CN115393617A
CN115393617A CN202210952476.XA CN202210952476A CN115393617A CN 115393617 A CN115393617 A CN 115393617A CN 202210952476 A CN202210952476 A CN 202210952476A CN 115393617 A CN115393617 A CN 115393617A
Authority
CN
China
Prior art keywords
trademark image
trademark
image
detected
entropy
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.)
Pending
Application number
CN202210952476.XA
Other languages
Chinese (zh)
Inventor
何肖肖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Huilang Times Technology Co Ltd
Original Assignee
Beijing Huilang Times Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Huilang Times Technology Co Ltd filed Critical Beijing Huilang Times Technology Co Ltd
Priority to CN202210952476.XA priority Critical patent/CN115393617A/en
Publication of CN115393617A publication Critical patent/CN115393617A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/30—Noise filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
    • G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
    • G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/771—Feature selection, e.g. selecting representative features from a multi-dimensional feature space
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/09—Recognition of logos

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a method and a system for quickly detecting an imitated trademark based on multi-convolution kernel inspection, and relates to the technical field of image processing. The method comprises the following steps: and if the entropy difference value is higher than the first preset difference value, the trademark image is a non-imitation trademark image, otherwise, the trademark image to be detected and the trademark image with similar entropy are equally divided to obtain 4 areas. If the areas with similar entropy values are smaller than the preset number, the areas are non-imitation type trademark images, otherwise, smooth convolution kernel processing is utilized, and the Euclidean distance is utilized to calculate the similarity. And (5) carrying out sharpening convolution kernel processing and calculating the similarity by using Euclidean distance. And (4) carrying out denoising convolution kernel processing and calculating the similarity by using Euclidean distance. If the similarity is higher than the preset similarity value, the trademark image is simulated, otherwise, the trademark image is non-simulated. The invention can obviously reduce the consumption of computing resources, avoid huge consumption of computing resources and accurately judge the imitated trademark image and the non-imitated trademark image with lower consumption.

Description

Simulated trademark rapid detection method and system based on multi-convolution kernel inspection
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a system for quickly detecting an imitated trademark based on multi-convolution kernel inspection.
Background
Trademarks are playing an increasingly important role as important components of intellectual property rights of enterprises. However, in today's society, it is not uncommon for trademarks to be emulated. The condition that the trademark is simulated misleads consumers, causes huge economic loss for known enterprises and brings many adverse effects to society and countries.
In order to solve the above problems, many researchers have conducted more intensive research and proposed some more reasonable simulation type trademark detection methods. However, as the number of enterprises increases sharply, the number of trademark images becomes more and more huge, and many imitation type trademark detection methods consume huge computing resources and cannot guarantee computing time. Therefore, how to establish a low-consumption imitation type trademark rapid detection method has very important value and significance in ensuring the precision and simultaneously accelerating the detection speed.
Disclosure of Invention
The invention aims to provide a method and a system for quickly detecting an imitated trademark based on multi-convolution kernel inspection, which are used for solving the problems that detection of the imitated trademark in the prior art consumes huge computing resources and cannot guarantee computing time.
The embodiment of the invention is realized by the following steps:
in a first aspect, an embodiment of the present application provides a method for quickly detecting an emulated trademark based on multiple convolution kernel tests, which includes the following steps:
step S110: carrying out entropy calculation on the trademark image to be detected and all trademark images in a preset image database;
step S120: judging the trademark image to be detected as a non-imitation trademark image if the entropy difference values of all trademark images in the trademark image to be detected and a preset image database are higher than a first preset difference value, and taking the trademark image as a similar entropy trademark image and carrying out next judgment if the entropy difference value of at least one trademark image in the trademark image to be detected and the preset image database is not higher than the first preset difference value;
step S130: respectively carrying out quartering on the trademark image to be detected and the similar entropy trademark image to obtain 4 areas corresponding to the trademark image and the similar entropy trademark image, and calculating the entropy values of the 4 areas corresponding to the trademark image and the similar entropy trademark image;
step S140: regarding 4 areas of the trademark image to be detected and 4 areas of the similar entropy trademark image, regarding the areas with the entropy difference value lower than a second preset difference value as the areas with the similar entropy, if the number of the areas with the similar entropy is less than a preset number, judging that the trademark image to be detected is a non-imitation trademark image, and if the number of the areas with the similar entropy is not less than the preset number, performing next judgment;
step S150: processing the trademark image to be detected and the similar entropy trademark image by utilizing smooth convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy;
step S160: processing the trademark image to be detected and the similar entropy trademark image by utilizing sharpening convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy trademark image;
step S170: processing the trademark image to be detected and the similar entropy trademark image by using denoising convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by using Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy trademark image;
step S180: in the execution of steps S150 to S170, if the similarity between the trademark image to be detected and the near-entropy trademark image is higher than the preset similarity value, the trademark image to be detected is determined as the imitation trademark image, otherwise, if the similarity between the trademark image to be detected and the near-entropy trademark image is not higher than the preset similarity value, the trademark image to be detected is determined as the non-imitation trademark image.
In some embodiments of the present invention, the step S110 includes:
respectively calculating the gray value of each pixel point of all trademark images in the trademark image to be detected and a preset image database;
calculating the probability of any gray value appearing in the corresponding image based on the gray value of each pixel point;
using formulas
Figure 347485DEST_PATH_IMAGE001
Calculating an entropy value of the corresponding image, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in the corresponding image.
In some embodiments of the present invention, the step S130 includes:
and (3) carrying out one-to-one correspondence on the 4 areas of the trademark image to be detected and the 4 areas of the trademark image with the similar entropy value.
In some embodiments of the present invention, the step S130 includes:
calculating the gray value of each pixel point in each region, and calculating the probability of any gray value appearing in the region;
using formulas
Figure 999046DEST_PATH_IMAGE002
Calculating entropy of the region, wherein H is entropy, i is gray value, and p is i Is the probability that the grey value i appears in this region.
In some embodiments of the present invention, before the step S110, the method further includes:
a plurality of trademark images are acquired in advance and stored in a preset image database.
In some embodiments of the present invention, after the step S180, the method further includes:
and if the trademark image to be detected is a non-imitation trademark image, storing the trademark image to be detected into a preset image database.
In some embodiments of the present invention, before the step S110, the method further includes:
and acquiring a trademark image to be detected.
In a second aspect, an embodiment of the present application provides an emulated trademark rapid detection system based on multiple convolution kernel tests, which includes:
the image entropy calculation module is used for calculating the entropy of the trademark image to be detected and all trademark images in the preset image database;
the image entropy comparison module is used for judging the trademark image to be detected as a non-imitation trademark image if the entropy difference values of all the trademark images in the trademark image to be detected and the preset image database are higher than a first preset difference value, and taking the trademark image as a similar entropy trademark image and carrying out next judgment if the entropy difference value of at least one trademark image in the trademark image to be detected and the preset image database is not higher than the first preset difference value;
the image halving module is used for respectively quartering the trademark image to be detected and the similar entropy trademark image to obtain 4 areas corresponding to the trademark image to be detected and the similar entropy value, and calculating the entropy values of the 4 areas corresponding to the trademark image to be detected and the similar entropy value;
the region entropy comparison module is used for regarding 4 regions of the trademark image to be detected and 4 regions of the similar entropy trademark image, regarding the regions with the entropy difference value lower than a second preset difference value as the regions with the similar entropy, if the number of the regions with the similar entropy is less than the preset number, judging that the trademark image to be detected is a non-imitation trademark image, and if the number of the regions with the similar entropy is not less than the preset number, performing the next judgment;
the smooth convolution kernel processing module is used for processing the trademark image to be detected and the trademark image with the similar entropy value by utilizing smooth convolution kernel to respectively obtain filtered results of the trademark image to be detected and the trademark image with the similar entropy value, and calculating the similarity of the trademark image to be detected and the trademark image with the similar entropy value by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the trademark image with the similar entropy value;
the sharpening convolution kernel processing module is used for utilizing sharpening convolution to check the trademark image to be detected and the trademark image with the similar entropy value to process the trademark image to be detected and the trademark image with the similar entropy value, respectively obtaining filtered results of the trademark image and the trademark image with the similar entropy value, and calculating the similarity of the trademark image and the trademark image with the Euclidean distance based on the filtered results of the trademark image and the trademark image with the similar entropy value;
the de-noising convolution kernel processing module is used for processing the trademark image to be detected and the trademark image with the similar entropy value by utilizing de-noising convolution kernel to respectively obtain filtered results of the trademark image to be detected and the trademark image with the similar entropy value, and calculating the similarity of the trademark image to be detected and the trademark image with the similar entropy value by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the trademark image with the similar entropy value;
and the similarity judging module is used for judging the trademark image to be detected as an imitated trademark image if the similarity between the trademark image to be detected and the trademark image with the similar entropy is higher than a preset similarity value in the process of executing the smooth convolution kernel processing module to the denoising convolution kernel processing module, otherwise, judging the trademark image to be detected as a non-imitated trademark image if the similarity between the trademark image to be detected and the trademark image with the similar entropy is not higher than the preset similarity value.
In some embodiments of the invention, the image entropy calculation module includes:
the image gray value calculating unit is used for calculating the gray value of each pixel point of all the trademark images in the trademark image to be detected and the preset image database respectively;
the image gray value probability calculating unit is used for calculating the probability of any gray value appearing in the corresponding image based on the gray value of each pixel point;
an image entropy calculation unit for using a formula
Figure 870794DEST_PATH_IMAGE001
Calculating an entropy value of the corresponding image, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in the corresponding image.
In some embodiments of the invention, the image equally dividing module comprises:
and the region corresponding unit is used for carrying out one-to-one correspondence on the 4 regions of the trademark image to be detected and the 4 regions of the trademark image with the similar entropy value.
In some embodiments of the invention, the image equally dividing module comprises:
the regional gray value calculation unit is used for calculating the gray value of each pixel point in each region and calculating the probability of any gray value appearing in the region;
a region entropy calculation unit for using a formula
Figure 424DEST_PATH_IMAGE002
Entropy of the regionCalculating values, wherein H is an entropy value, i is a gray value, p i Is the probability that the grey value i appears in this region.
In some embodiments of the present invention, the simulated trademark rapid detection system based on multiple convolution kernel inspection further includes:
and the trademark image pre-storing module is used for pre-acquiring a plurality of trademark images and storing the trademark images into a preset image database.
In some embodiments of the present invention, the simulated trademark rapid detection system based on multiple convolution kernel inspection further includes:
and the to-be-detected trademark image storage module is used for storing the to-be-detected trademark image into a preset image database if the to-be-detected trademark image is a non-imitation trademark image.
In some embodiments of the present invention, the above simulated trademark rapid detection system based on multi-convolution kernel inspection further includes:
and the trademark image acquisition module to be detected is used for acquiring the trademark image to be detected.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory for storing one or more programs; a processor. The program or programs, when executed by a processor, implement the method of any of the first aspects as described above.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method according to any one of the first aspect described above.
Compared with the prior art, the embodiment of the invention has at least the following advantages or beneficial effects:
the invention provides a method and a system for quickly detecting an imitated trademark based on multi-convolution kernel inspection, which comprises the following steps: step S110: and carrying out entropy calculation on the trademark image to be detected and all trademark images in a preset image database. Step S120: and if the entropy difference values of all trademark images in the trademark image to be detected and the preset image database are not higher than the first preset difference value, taking the trademark image as a similar entropy trademark image, and performing next judgment. Step S130: and respectively carrying out quartering on the trademark image to be detected and the similar entropy trademark image to obtain 4 areas corresponding to the trademark image and the similar entropy trademark image, and calculating the entropy values of the 4 areas corresponding to the trademark image and the similar entropy trademark image. Step S140: regarding 4 areas of the trademark image to be detected and 4 areas of the similar entropy trademark image, regarding the area with the entropy difference value lower than the second preset difference value as the area with the similar entropy value, if the number of the areas with the similar entropy value is smaller than the preset number, judging that the trademark image to be detected is the non-imitation trademark image, and if the number of the areas with the similar entropy value is not smaller than the preset number, performing the next judgment. Step S150: and (3) utilizing smooth convolution to check the trademark image to be detected and the similar entropy trademark image for processing, respectively obtaining filtered results of the trademark image to be detected and the similar entropy trademark image to be detected, and utilizing Euclidean distance to calculate similarity of the trademark image to be detected and the similar entropy based on the filtered results of the trademark image to be detected and the similar entropy. Step S160: and (3) processing the trademark image to be detected and the similar entropy trademark image by utilizing a sharpening convolution kernel to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy. Step S170: and (3) utilizing denoising convolution to check the trademark image to be detected and the similar entropy trademark image for processing, respectively obtaining filtered results of the trademark image to be detected and the similar entropy trademark image, and utilizing Euclidean distance to calculate similarity of the trademark image to be detected and the similar entropy trademark image based on the filtered results of the trademark image to be detected and the similar entropy. Step S180: in the step S150 to the step S170, if the similarity between the trademark image to be detected and the near entropy trademark image is higher than the preset similarity value, the trademark image to be detected is determined as an imitation trademark image, otherwise, if the similarity between the trademark image to be detected and the near entropy trademark image is not higher than the preset similarity value, the trademark image to be detected is determined as a non-imitation trademark image.
The method and the system firstly compare the entropy values of all trademark images in a trademark image to be detected and a preset image database to judge whether the preset image database has the trademark image with the entropy value close to that of the trademark image to be detected, if not, the non-imitation trademark image can be quickly identified, and if so, all the trademark images with the entropy value close to that of the trademark image to be detected are used as the trademark images with the close entropy value to obtain all the trademark images with the close entropy value. And then, for any similar entropy trademark image, dividing the trademark image to be detected and the similar entropy trademark image into 4 regions in a quartering mode respectively, comparing the entropy values of the two corresponding regions to obtain the number of the regions with similar entropy values, judging that the trademark image to be detected is a non-imitation trademark image if the number of the regions with similar entropy values is smaller than a preset number, and processing the trademark image to be detected and the similar entropy trademark image by using different convolutions if the number of the regions with similar entropy values is not smaller than the preset number. And after filtering based on different convolution kernels, judging that the trademark image to be detected and the trademark image with the similar entropy have higher similarity, and judging the trademark image to be detected as the imitation trademark image. And if the trademark image to be detected and the trademark image with the similar entropy value do not have higher similarity after filtering based on different convolution kernels, judging the trademark image to be detected as a non-imitation trademark image. The method and the system adopt entropy calculation, convolution kernel processing and similarity calculation, which can remarkably reduce the consumption of calculation resources and accelerate the calculation speed, thereby avoiding huge consumption of calculation resources, accelerating the detection speed and distinguishing the imitated trademark image and the non-imitated trademark image with lower consumption and higher precision.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of an emulated trademark rapid detection method based on multiple convolution kernel tests according to an embodiment of the present invention;
fig. 2 is a flowchart of a step S110 according to an embodiment of the present invention;
fig. 3 is a flowchart of a step S130 according to an embodiment of the present invention;
fig. 4 is a block diagram of an emulated trademark rapid detection system based on multiple convolution kernel tests according to an embodiment of the present invention;
fig. 5 is a schematic structural block diagram of an electronic device according to an embodiment of the present invention.
An icon: 100-a simulated trademark rapid detection system based on multi-convolution kernel inspection; 110-an image entropy calculation module; 120-image entropy comparison module; 130-image aliquoting module; 140-region entropy comparison module; 150-smooth convolution kernel processing module; 160-sharpening the convolution kernel processing module; 170-denoising convolution kernel processing module; 180-similarity determination module; 101-a memory; 102-a processor; 103-communication interface.
Detailed Description
In order to make the objects, 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 with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not construed as indicating or implying relative importance.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Furthermore, if the terms "comprise," "include," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, the presence of an element identified by the phrases "comprising a component of' 8230; \8230;" does not exclude the presence of additional like elements in any process, method, article, or apparatus that comprises the element.
In the description of the present application, it should be noted that if the terms "upper", "lower", "inner", "outer", etc. are used to indicate an orientation or positional relationship based on that shown in the drawings or that the application product is usually placed in use, the description is merely for convenience and simplicity, and it is not intended to indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore should not be construed as limiting the present application.
In the description of the present application, it should also be noted that, unless explicitly stated or limited otherwise, the terms "disposed" and "connected" should be interpreted broadly, e.g., as being fixed or detachable or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art.
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the individual features of the embodiments can be combined with one another without conflict.
Examples
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for rapidly detecting an emulated trademark based on multiple convolution kernel tests according to an embodiment of the present application. The embodiment of the application provides a simulation type trademark rapid detection method based on multi-convolution kernel inspection, which comprises the following steps:
step S110: carrying out entropy calculation on the trademark image to be detected and all trademark images in a preset image database;
specifically, entropy calculation is performed on the trademark image to be detected and all trademark images respectively, so that entropy values corresponding to the trademark image to be detected and each trademark image are obtained. And sequentially comparing the entropy values of the trademark images to be detected with the entropy values of the trademark images one by one to obtain an entropy value difference value.
In the implementation process, entropy values of all trademark images in a preset image database can be calculated in advance, and then the trademark image to be detected is obtained. Therefore, after the trademark image to be detected is obtained, only the entropy value of the trademark image to be detected needs to be calculated, and then the comparison is carried out with the entropy values of all the trademark images, so that the huge calculation resource consumption is avoided, and the calculation time is effectively reduced.
Step S120: judging the trademark image to be detected as a non-imitation trademark image if the entropy difference values of all trademark images in the trademark image to be detected and a preset image database are higher than a first preset difference value, and taking the trademark image as a similar entropy trademark image and carrying out next judgment if the entropy difference value of at least one trademark image in the trademark image to be detected and the preset image database is not higher than the first preset difference value;
specifically, by comparing the entropy difference value of any one of the trademark images in the trademark image to be detected and the preset image database with the first preset difference value, whether the entropy values of the trademark image to be detected and the trademark image are similar or not can be judged, and if the entropy values of all the trademark images in the trademark image to be detected and the preset image database are not similar, the trademark image to be detected is directly judged to be the non-imitation trademark image. And if the trademark image with the entropy value close to that of the trademark image to be detected exists, taking all the trademark images with the entropy values close to that of the trademark image to be detected as the trademark images with the close entropy values, so as to obtain all the trademark images with the close entropy values.
Step S130: respectively carrying out quartering on the trademark image to be detected and the similar entropy trademark image to obtain 4 areas corresponding to the trademark image and the similar entropy trademark image, and calculating the entropy values of the 4 areas corresponding to the trademark image and the similar entropy trademark image;
for example, for any similar entropy trademark image, the trademark image to be detected and the similar entropy trademark image are divided into 4 regions by quartering, and the 4 regions of the trademark image to be detected and the 4 regions of the similar entropy trademark image correspond to each other one by one, for example, the upper left region of the trademark image to be detected corresponds to the upper left region of the similar entropy trademark image.
Step S140: regarding 4 areas of the trademark image to be detected and 4 areas of the similar entropy trademark image, regarding the areas with the entropy difference value lower than a second preset difference value as the areas with the similar entropy, if the number of the areas with the similar entropy is less than a preset number, judging that the trademark image to be detected is a non-imitation trademark image, and if the number of the areas with the similar entropy is not less than the preset number, performing next judgment;
illustratively, the preset number may be 3. And comparing the entropy values of the two corresponding regions to obtain the number of the regions with similar entropy values, directly judging that the trademark image to be detected is a non-imitation trademark image if the number of the regions with similar entropy values is less than 3, and continuing to judge in the next step if the number of the regions with similar entropy values is not less than 3. Therefore, by comparing the entropy values of the regions corresponding to the trademark image to be detected and the trademark image with the similar entropy values, whether the trademark image to be detected is a non-imitation trademark image or not can be judged with low consumption and high precision.
Step S150: processing the trademark image to be detected and the similar entropy trademark image by utilizing smooth convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy;
illustratively, the size of the smoothing convolution kernel may be 3 x 3.
Step S160: processing the trademark image to be detected and the similar entropy trademark image by utilizing sharpening convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy trademark image;
illustratively, the size of the sharpening convolution kernel may be 3 x 3.
Step S170: processing the trademark image to be detected and the similar entropy trademark image by using denoising convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by using Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy trademark image;
illustratively, the size of the de-noised convolution kernel may be 3 x 3.
Step S180: in the execution of steps S150 to S170, if the similarity between the trademark image to be detected and the near-entropy trademark image is higher than the preset similarity value, the trademark image to be detected is determined as the imitation trademark image, otherwise, if the similarity between the trademark image to be detected and the near-entropy trademark image is not higher than the preset similarity value, the trademark image to be detected is determined as the non-imitation trademark image.
Specifically, the method comprises the steps of comparing entropy values of all trademark images in a trademark image to be detected and a preset image database to judge whether the trademark image with the entropy value close to that of the trademark image to be detected exists in the preset image database or not, if not, rapidly identifying a non-imitation type trademark image, and if so, taking all the trademark images with the entropy values close to that of the trademark image to be detected as the trademark images with the close entropy values to obtain all the trademark images with the close entropy values. And then, for any similar entropy trademark image, dividing the trademark image to be detected and the similar entropy trademark image into 4 regions in a quartering mode respectively, comparing the entropy values of the two corresponding regions to obtain the number of the regions with similar entropy values, judging that the trademark image to be detected is a non-imitation trademark image if the number of the regions with similar entropy values is smaller than a preset number, and processing the trademark image to be detected and the similar entropy trademark image by using different convolutions if the number of the regions with similar entropy values is not smaller than the preset number. And after filtering based on different convolution kernels, judging that the trademark image to be detected and the trademark image with the similar entropy have higher similarity, and judging the trademark image to be detected as the imitation trademark image. And if the trademark image to be detected and the trademark image with the similar entropy value do not have higher similarity after filtering based on different convolution kernels, judging the trademark image to be detected as a non-imitation trademark image. The method adopts entropy calculation, convolution kernel processing and similarity calculation, which can remarkably reduce the consumption of calculation resources and accelerate the calculation speed, thereby avoiding huge consumption of calculation resources, accelerating the detection speed and distinguishing the imitated trademark image and the non-imitated trademark image with low consumption and high precision.
Referring to fig. 2, fig. 2 is a flowchart of a step S110 according to an embodiment of the present invention. In some embodiments of this embodiment, the step S110 includes:
respectively calculating the gray value of each pixel point of all trademark images in the trademark image to be detected and a preset image database;
calculating the probability of any gray value appearing in the corresponding image based on the gray value of each pixel point;
using a formula
Figure 549217DEST_PATH_IMAGE001
Calculating an entropy value of the corresponding image, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in the corresponding image.
Illustratively, for any trademark image to be detected or any trademark image in a preset image database, firstly, the gray value of each pixel point in the image is acquired, and the probability of each gray value appearing in the image is calculated, so that a formula is utilized
Figure 106100DEST_PATH_IMAGE001
An entropy value of the image is obtained.
In some embodiments of this embodiment, the step S130 includes:
and (3) carrying out one-to-one correspondence on the 4 areas of the trademark image to be detected and the 4 areas of the trademark image with the similar entropy value.
Illustratively, the upper left area of the trademark image to be detected corresponds to the upper left area of the trademark image with similar entropy, the lower left area of the trademark image to be detected corresponds to the lower left area of the trademark image with similar entropy, the upper right area of the trademark image to be detected corresponds to the upper right area of the trademark image with similar entropy, and the lower right area of the trademark image to be detected corresponds to the lower right area of the trademark image with similar entropy.
Referring to fig. 3, fig. 3 is a flowchart illustrating a step S130 according to an embodiment of the present invention. In some embodiments of the present embodiment, the step S130 includes:
calculating the gray value of each pixel point in each region, and calculating the probability of any gray value appearing in the region;
using a formula
Figure 232188DEST_PATH_IMAGE002
Calculating the entropy value of the region, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in this region.
Exemplarily, for any one of the regions obtained by quartering, the gray value of each pixel point in the region is obtained first, and the probability of each gray value occurring in the region is calculated, so that a formula is utilized
Figure 165509DEST_PATH_IMAGE002
An entropy value for the region is obtained.
In some embodiments of this embodiment, before the step S110, the method further includes:
a plurality of trademark images are acquired in advance and stored in a preset image database.
In some embodiments of this embodiment, after step S180, the method further includes:
and if the trademark image to be detected is a non-imitation trademark image, storing the trademark image to be detected into a preset image database. Thereby enriching the data of the brand images in the preset image database.
In some embodiments of this embodiment, before the step S110, the method further includes:
and acquiring a trademark image to be detected.
Referring to fig. 4, fig. 4 is a block diagram illustrating an exemplary trademark rapid detection system 100 based on multi-convolution kernel inspection according to an embodiment of the present invention. The embodiment of the present application provides an emulation formula trade mark rapid detection system 100 based on many convolution kernel tests, it includes:
the image entropy calculation module 110 is configured to perform entropy calculation on the trademark image to be detected and all trademark images in the preset image database;
the image entropy comparison module 120 is configured to determine the trademark image to be detected as a non-imitation trademark image if the entropy difference values of the trademark image to be detected and all trademark images in the preset image database are higher than a first preset difference value, and determine the trademark image as a similar entropy trademark image if the entropy difference value of at least one trademark image in the trademark image to be detected and the preset image database is not higher than the first preset difference value;
the image halving module 130 is configured to perform quartering on the trademark image to be detected and the trademark image with the similar entropy respectively to obtain 4 regions corresponding to the trademark image and the trademark image with the similar entropy, and calculate entropy values of the 4 regions corresponding to the trademark image and the trademark image with the similar entropy;
the region entropy comparison module 140 is configured to regard, for 4 regions of the to-be-detected trademark image and 4 regions of the similar entropy trademark image, the region whose entropy difference is lower than the second preset difference as a region with similar entropy, determine that the to-be-detected trademark image is a non-imitation trademark image if the number of the regions with similar entropy is smaller than the preset number, and perform the next determination if the number of the regions with similar entropy is not smaller than the preset number;
the smooth convolution kernel processing module 150 is used for processing the trademark image to be detected and the trademark image with the similar entropy value by utilizing smooth convolution kernel to respectively obtain filtered results of the trademark image to be detected and the trademark image with the similar entropy value, and calculating the similarity of the trademark image to be detected and the trademark image with the similar entropy value by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the trademark image with the similar entropy value;
the sharpening convolution kernel processing module 160 is configured to utilize a sharpening convolution kernel to check the trademark image to be detected and the trademark image with the similar entropy value, to obtain filtered results of the trademark image to be detected and the trademark image with the similar entropy value respectively, and to utilize an euclidean distance to calculate similarity between the two filtered results;
the denoising convolution kernel processing module 170 is configured to utilize denoising convolution kernel to process the trademark image to be detected and the trademark image with the similar entropy, respectively obtain filtered results of the two, and calculate similarity between the two by using the euclidean distance based on the filtered results of the two;
and the similarity determination module 180 is configured to determine the trademark image to be detected as the mimic trademark image if the similarity between the trademark image to be detected and the trademark image with the similar entropy is higher than a preset similarity value in the processes from the smoothing convolution kernel processing module 150 to the denoising convolution kernel processing module 170, and otherwise determine the trademark image to be detected as the non-mimic trademark image if the similarity between the trademark image to be detected and the trademark image with the similar entropy is not higher than the preset similarity value.
Specifically, the system firstly compares the entropy values of all trademark images in a trademark image to be detected and a preset image database to judge whether the preset image database has the trademark image close to the entropy value of the trademark image to be detected, if not, the non-imitation trademark image can be quickly identified, and if so, all the trademark images with the entropy values close to the entropy value of the trademark image to be detected are used as the trademark images with the close entropy value to obtain all the trademark images with the close entropy value. And then, for any similar entropy trademark image, quartering the trademark image to be detected and the similar entropy trademark image into 4 areas respectively, comparing the entropy values of the two corresponding areas to obtain the number of the areas with similar entropy values, judging that the trademark image to be detected is a non-imitation trademark image if the number of the areas with similar entropy values is less than a preset number, and processing the trademark image to be detected and the similar entropy trademark image by using different convolutions if the number of the areas with similar entropy values is not less than the preset number. And after filtering based on different convolution kernels, judging that the trademark image to be detected and the trademark image with the similar entropy have higher similarity, and judging the trademark image to be detected as the imitation trademark image. And if the trademark image to be detected and the trademark image with the similar entropy value do not have higher similarity after filtering based on different convolution kernels, judging the trademark image to be detected as a non-imitation trademark image. The system adopts entropy calculation, convolution kernel processing and similarity calculation, which can remarkably reduce the consumption of calculation resources and accelerate the calculation speed, thereby avoiding huge consumption of calculation resources, accelerating the detection speed and distinguishing the imitated trademark image and the non-imitated trademark image with low consumption and high precision.
In some implementations of this embodiment, the image entropy calculation module 110 includes:
the image gray value calculating unit is used for respectively calculating the gray value of each pixel point of all trademark images in the trademark image to be detected and the preset image database;
the image gray value probability calculating unit is used for calculating the probability of any gray value appearing in the corresponding image based on the gray value of each pixel point;
an image entropy calculation unit for using a formula
Figure 303229DEST_PATH_IMAGE001
Calculating an entropy value of the corresponding image, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in the corresponding image.
Illustratively, for any trademark image to be detected or any trademark image in a preset image database, firstly, the gray value of each pixel point in the image is acquired, and the probability of each gray value appearing in the image is calculated, so that a formula is utilized
Figure 155647DEST_PATH_IMAGE001
An entropy value of the image is obtained.
In some embodiments of this embodiment, the image equally dividing module 130 includes:
and the region corresponding unit is used for carrying out one-to-one correspondence on the 4 regions of the trademark image to be detected and the 4 regions of the trademark image with the similar entropy value.
Illustratively, the upper left area of the trademark image to be detected corresponds to the upper left area of the trademark image with similar entropy, the lower left area of the trademark image to be detected corresponds to the lower left area of the trademark image with similar entropy, the upper right area of the trademark image to be detected corresponds to the upper right area of the trademark image with similar entropy, and the lower right area of the trademark image to be detected corresponds to the lower right area of the trademark image with similar entropy.
In some embodiments of this embodiment, the image equally dividing module 130 includes:
the regional gray value calculation unit is used for calculating the gray value of each pixel point in each region and calculating the probability of any gray value appearing in the region;
a region entropy calculation unit for using a formula
Figure 909977DEST_PATH_IMAGE002
Calculating the entropy value of the region, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in this region.
For any one area obtained by quartering, exemplarily, a gray value of each pixel point in the area is obtained first, and the probability of each gray value appearing in the area is calculated, so that a formula is utilized
Figure 115830DEST_PATH_IMAGE002
An entropy value for the region is obtained.
In some embodiments of this embodiment, the above-mentioned simulated trademark rapid detection system 100 based on multi-convolution kernel inspection further includes:
and the trademark image pre-storing module is used for acquiring a plurality of trademark images in advance and storing the trademark images into a preset image database.
In some embodiments of this embodiment, the simulated trademark rapid detection system 100 based on multiple convolution kernel inspection further includes:
and the to-be-detected trademark image storage module is used for storing the to-be-detected trademark image into a preset image database if the to-be-detected trademark image is a non-imitation trademark image. Thereby enriching the data of the brand images in the preset image database.
In some embodiments of this embodiment, the above-mentioned simulated trademark rapid detection system 100 based on multi-convolution kernel inspection further includes:
and the trademark image acquisition module is used for acquiring the trademark image to be detected.
Referring to fig. 5, fig. 5 is a schematic structural block diagram of an electronic device according to an embodiment of the present disclosure. The electronic device comprises a memory 101, a processor 102 and a communication interface 103, wherein the memory 101, the processor 102 and the communication interface 103 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 101 may be used to store software programs and modules, such as program instructions/modules corresponding to the simulated trademark rapid detection system 100 based on multi-convolution kernel inspection, and the processor 102 executes the software programs and modules stored in the memory 101 to perform various functional applications and data processing. The communication interface 103 may be used for communicating signaling or data with other node devices.
The Memory 101 may be, but not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Read Only Memory (EPROM), an electrically Erasable Read Only Memory (EEPROM), and the like.
The processor 102 may be an integrated circuit chip having signal processing capabilities. The Processor 102 may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
It will be appreciated that the configuration shown in fig. 5 is merely illustrative and that the electronic device may include more or fewer components than shown in fig. 5 or have a different configuration than shown in fig. 5. The components shown in fig. 5 may be implemented in hardware, software, or a combination thereof.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus embodiments described above are merely illustrative and, for example, the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Claims (10)

1. A simulation type trademark rapid detection method based on multi-convolution kernel inspection is characterized by comprising the following steps:
step S110: carrying out entropy calculation on the trademark image to be detected and all trademark images in a preset image database;
step S120: judging the trademark image to be detected as a non-imitation trademark image if the entropy difference values of the trademark image to be detected and all trademark images in the preset image database are higher than a first preset difference value, and taking the trademark image as a similar entropy trademark image and carrying out next judgment if the entropy difference value of the trademark image to be detected and at least one trademark image in the preset image database is not higher than the first preset difference value;
step S130: quartering the trademark image to be detected and the similar entropy trademark image respectively to obtain 4 areas corresponding to the trademark image to be detected and the similar entropy trademark image, and calculating the entropy values of the 4 areas corresponding to the trademark image to be detected and the similar entropy trademark image;
step S140: regarding the 4 regions of the trademark image to be detected and the 4 regions of the similar entropy trademark image, regarding the regions with the entropy difference value lower than a second preset difference value as regions with similar entropy values, if the number of the regions with similar entropy values is less than a preset number, judging that the trademark image to be detected is a non-imitation trademark image, and if the number of the regions with similar entropy values is not less than the preset number, performing next judgment;
step S150: processing the trademark image to be detected and the similar entropy trademark image by utilizing smooth convolution check to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy;
step S160: utilizing sharpening convolution to check the trademark image to be detected and the similar entropy trademark image for processing, respectively obtaining filtered results of the trademark image to be detected and the similar entropy trademark image, and utilizing Euclidean distance to calculate similarity of the trademark image to be detected and the similar entropy trademark image based on the filtered results of the trademark image to be detected and the similar entropy;
step S170: processing the trademark image to be detected and the close entropy trademark image by using denoising convolution kernel to respectively obtain filtered results of the trademark image to be detected and the close entropy trademark image, and calculating the similarity of the two by using Euclidean distance based on the filtered results of the trademark image to be detected and the close entropy trademark image;
step S180: in the step S150 to the step S170, if the similarity between the trademark image to be detected and the near entropy trademark image is higher than a preset similarity value, the trademark image to be detected is determined as an imitation trademark image, otherwise, if the similarity between the trademark image to be detected and the near entropy trademark image is not higher than the preset similarity value, the trademark image to be detected is determined as a non-imitation trademark image.
2. The method for rapidly detecting the imitated trademark based on the multi-convolution kernel inspection as claimed in claim 1, wherein the step S110 comprises:
respectively calculating the gray value of each pixel point of all trademark images in the trademark image to be detected and a preset image database;
calculating the probability of any gray value appearing in the corresponding image based on the gray value of each pixel point;
using formulas
Figure 390268DEST_PATH_IMAGE001
Calculating an entropy value of the corresponding image, wherein H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in the corresponding image.
3. The method for rapidly detecting the imitated trademark based on the multi-convolution kernel inspection as claimed in claim 1, wherein the step S130 comprises:
and (3) carrying out one-to-one correspondence on the 4 areas of the trademark image to be detected and the 4 areas of the trademark image with the similar entropy value.
4. The method for rapidly detecting the imitated trademark based on the multi-convolution kernel inspection as claimed in claim 1, wherein the step S130 comprises:
calculating the gray value of each pixel point in each region, and calculating the probability of any gray value appearing in the region;
using formulas
Figure 554533DEST_PATH_IMAGE002
The entropy value of the region is calculated,where H is the entropy value, i is the gray value, p i Is the probability that the grey value i appears in this region.
5. The method for rapidly detecting the imitated trademark based on the multi-convolution kernel inspection as claimed in claim 1, wherein before the step S110, the method further comprises:
a plurality of trademark images are acquired in advance and stored in a preset image database.
6. The method for rapidly detecting the imitated trademark based on the multi-convolution kernel inspection as claimed in claim 1, wherein after the step S180, the method further comprises:
and if the trademark image to be detected is a non-imitation trademark image, storing the trademark image to be detected into a preset image database.
7. The method for rapidly detecting the imitated trademark based on the multi-convolution kernel test of claim 1, wherein before the step S110, the method further comprises:
and acquiring a trademark image to be detected.
8. A simulation type trademark rapid detection system based on multi-convolution kernel inspection is characterized by comprising:
the image entropy calculation module is used for calculating the entropy of the trademark image to be detected and all trademark images in the preset image database;
the image entropy comparison module is used for judging the trademark image to be detected as a non-imitation trademark image if the difference values of the entropy values of all the trademark images in the trademark image to be detected and the preset image database are higher than a first preset difference value, and taking the trademark image as a similar entropy trademark image and carrying out next judgment if the difference value of the entropy values of at least one trademark image in the trademark image to be detected and the preset image database is not higher than the first preset difference value;
the image halving module is used for respectively quartering the trademark image to be detected and the similar entropy trademark image to obtain 4 areas corresponding to the trademark image to be detected and the similar entropy trademark image, and calculating the entropy values of the 4 areas corresponding to the trademark image to be detected and the similar entropy trademark image;
the region entropy comparison module is used for regarding the 4 regions of the trademark image to be detected and the 4 regions of the similar entropy trademark image, regarding the regions with the entropy difference value lower than a second preset difference value as regions with similar entropy, if the number of the regions with similar entropy is less than a preset number, judging that the trademark image to be detected is a non-imitation trademark image, and if the number of the regions with similar entropy is not less than the preset number, performing the next judgment;
the smooth convolution kernel processing module is used for processing the trademark image to be detected and the similar entropy trademark image by utilizing a smooth convolution kernel, respectively obtaining filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by utilizing the Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy;
the sharpening convolution kernel processing module is used for utilizing sharpening convolution to check the trademark image to be detected and the similar entropy value trademark image for processing, respectively obtaining filtered results of the trademark image to be detected and the similar entropy value trademark image, and utilizing Euclidean distance to calculate similarity of the trademark image to be detected and the similar entropy value trademark image based on the filtered results of the trademark image to be detected and the similar entropy value trademark image;
the de-noising convolution kernel processing module is used for processing the trademark image to be detected and the similar entropy trademark image by using de-noising convolution kernel to respectively obtain filtered results of the trademark image to be detected and the similar entropy trademark image, and calculating the similarity of the trademark image to be detected and the similar entropy trademark image by using Euclidean distance based on the filtered results of the trademark image to be detected and the similar entropy trademark image;
and the similarity judging module is used for judging the trademark image to be detected as a simulated trademark image if the similarity between the trademark image to be detected and the similar entropy trademark image is higher than a preset similarity value in the process of executing the smooth convolution kernel processing module to the denoising convolution kernel processing module, otherwise, judging the trademark image to be detected as a non-simulated trademark image if the similarity between the trademark image to be detected and the similar entropy trademark image is not higher than the preset similarity value.
9. An electronic device, comprising:
a memory for storing one or more programs;
a processor;
the one or more programs, when executed by the processor, implement the method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.
CN202210952476.XA 2022-08-09 2022-08-09 Simulated trademark rapid detection method and system based on multi-convolution kernel inspection Pending CN115393617A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210952476.XA CN115393617A (en) 2022-08-09 2022-08-09 Simulated trademark rapid detection method and system based on multi-convolution kernel inspection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210952476.XA CN115393617A (en) 2022-08-09 2022-08-09 Simulated trademark rapid detection method and system based on multi-convolution kernel inspection

Publications (1)

Publication Number Publication Date
CN115393617A true CN115393617A (en) 2022-11-25

Family

ID=84118110

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210952476.XA Pending CN115393617A (en) 2022-08-09 2022-08-09 Simulated trademark rapid detection method and system based on multi-convolution kernel inspection

Country Status (1)

Country Link
CN (1) CN115393617A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116705250A (en) * 2023-06-07 2023-09-05 北京海上升科技有限公司 Low-consumption optimization and intelligent storage method and system for medical image big data

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116705250A (en) * 2023-06-07 2023-09-05 北京海上升科技有限公司 Low-consumption optimization and intelligent storage method and system for medical image big data
CN116705250B (en) * 2023-06-07 2024-06-14 宁波全网云医疗科技股份有限公司 Low-consumption optimization and intelligent storage method and system for medical image big data

Similar Documents

Publication Publication Date Title
CN114241484B (en) Social network-oriented image big data accurate retrieval method and system
CN110705718A (en) Model interpretation method and device based on cooperative game and electronic equipment
CN112085056B (en) Target detection model generation method, device, equipment and storage medium
CN115100450B (en) Intelligent traffic brand automobile big data detection method and system based on artificial intelligence
CN110909712A (en) Moving object detection method and device, electronic equipment and storage medium
CN112580668B (en) Background fraud detection method and device and electronic equipment
CN107578438A (en) Circle recognition methods, device and electronic equipment
CN112613543A (en) Enhanced policy verification method and device, electronic equipment and storage medium
CN109213862A (en) Object identification method and device, computer readable storage medium
CN116432604A (en) Data verification method and device and electronic equipment
CN108288274B (en) Mold detection method and device and electronic equipment
CN116562991B (en) Commodity big data information identification method and system for meta-space electronic commerce platform
CN111680680B (en) Target code positioning method and device, electronic equipment and storage medium
CN115393617A (en) Simulated trademark rapid detection method and system based on multi-convolution kernel inspection
CN114596441A (en) Massive image denoising method and system for live webcast social big data
WO2023029491A1 (en) Panel array short circuit detection method and apparatus, electronic device, and storage medium
CN113723467A (en) Sample collection method, device and equipment for defect detection
CN113191277B (en) Table image area identification method and system based on entropy verification
CN109389110A (en) A kind of area determination method and device
CN110689230B (en) Regional poverty degree determining method, electronic device and storage medium
CN112329810B (en) Image recognition model training method and device based on significance detection
CN116188917A (en) Defect data generation model training method, defect data generation method and device
CN111382052A (en) Code quality evaluation method and device and electronic equipment
CN112861874B (en) Expert field denoising method and system based on multi-filter denoising result
CN111950644A (en) Model training sample selection method and device and computer equipment

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