Trademark compliance detection method and platform based on multi-dimensional frame selection characteristics of product pictures

Abstract

The invention provides a trademark compliance detection method and platform based on multi-dimensional frame selection characteristics of a product picture, and the method comprises the steps: receiving a to-be-detected image uploaded by a user through terminal equipment, carrying out the recognition and frame selection of a to-be-detected image-text on the to-be-detected image, displaying a selection box and a trademark confidence coefficient for the to-be-detected image-text, carrying out the retrieval of a preset trademark database, and carrying out the detection of the trademark compliance. Obtaining at least one similar trademark similar to the image-text to be detected and the corresponding trademark similarity; sending a multi-dimensional frame selection feature to a server, wherein the multi-dimensional frame selection feature comprises trademark confidence and trademark similarity; and receiving a trademark compliance detection response message from the server, and displaying a trademark compliance detection result for the to-be-detected image-text in each selection box on a trademark detection interface. Therefore, the trademark compliance detection result of each to-be-detected image-text is determined according to the trademark confidence and the trademark similarity, the accuracy and comprehensiveness of trademark identification are improved, and the accuracy of trademark compliance detection is further improved.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to a trademark compliance detection method and platform based on multi-dimensional frame selection features of product images. Background Art

[0002] In the prior art, trademarks are usually identified through YOLO confidence, and then a search is made in the trademark database to determine whether there are similar trademarks based on the identified trademarks. The risk of trademark infringement is determined based on the degree of trademark similarity, thereby detecting whether product images and product trademarks are compliant.

[0003] However, the trademark infringement risk detection results obtained by the above method usually have large errors, and are prone to identifying non-trademark graphics or text as trademarks, and then performing similar trademark searches on "non-trademarks", resulting in a large amount of irrelevant information in the search results, increasing the redundant calculations of the system, and increasing the search cost. Summary of the invention

[0004] The present application provides a trademark compliance detection method and platform based on the multi-dimensional frame selection features of product images. By first obtaining the trademark confidence of the image and text to be detected, and determining the trademark compliance detection result of each image and text to be detected based on the trademark confidence and trademark similarity, the accuracy and comprehensiveness of trademark recognition are improved, thereby improving the accuracy of trademark compliance detection.

[0005] In a first aspect, the present application provides a trademark compliance detection method based on multi-dimensional box selection features of product images, which is applied to a terminal device of a trademark compliance detection system, wherein the trademark compliance detection system includes the terminal device and a server, and the method includes: receiving an image to be detected uploaded by a user, identifying and boxing the image to be detected on the image to be detected; displaying a selection box for the image to be detected and the trademark confidence of the image to be detected on a trademark detection interface, wherein the trademark confidence is used to characterize the probability that the image to be detected is identified as a trademark; searching a preset trademark database to obtain at least one similar trademark similar to the image to be detected and the corresponding trademark similarity; receiving a trigger operation on a trademark compliance detection control on the trademark detection interface, creating a trademark compliance detection request message, and sending the trademark compliance detection request message to the server, wherein the trademark compliance detection request message includes a multi-dimensional box selection feature, wherein the multi-dimensional box selection feature includes the trademark confidence and the trademark similarity; receiving a trademark compliance detection response message from the server, and displaying the trademark compliance detection result for the image to be detected in each selection box on the trademark detection interface.

[0006] In the second aspect, the present application provides a trademark compliance detection platform based on the multi-dimensional frame selection features of product images, characterized in that the trademark compliance detection platform includes a terminal device and a server, the terminal device is used to execute the step instructions executed by the terminal device in the method as described in any one of the first aspects; the server is used to execute the step instructions executed by the server in the method as described in any one of the first aspects.

[0007] It can be seen that in the embodiment of the present application, the terminal device receives an image to be detected uploaded by a user, identifies and selects the image to be detected on the image to be detected; displays a selection box for the image to be detected and the trademark confidence of the image to be detected on the trademark detection interface, and the trademark confidence is used to characterize the probability that the image to be detected is identified as a trademark; searches a preset trademark database to obtain at least one similar trademark similar to the image to be detected and the corresponding trademark similarity; receives a trigger operation on the trademark compliance detection control on the trademark detection interface, creates a trademark compliance detection request message, and sends a trademark compliance detection request message to the server, the trademark compliance detection request message includes a multi-dimensional selection feature, and the multi-dimensional selection feature includes a trademark confidence and a trademark similarity; receives a trademark compliance detection response message from the server, and displays the trademark compliance detection result for the image to be detected in each selection box on the trademark detection interface. Therefore, in the present application, compared with the prior art method of directly performing trademark similarity search based on the identified image and text information and then determining the risk of trademark infringement, the present application first obtains the trademark confidence of the image and text to be detected, and determines the trademark compliance detection result of each image and text to be detected based on the trademark confidence and trademark similarity, thereby improving the accuracy and comprehensiveness of trademark recognition, and thereby improving the accuracy of trademark compliance detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 A schematic diagram of the structure of an e-commerce platform provided in an embodiment of the present application; Figure 2 A flowchart of a trademark compliance detection method based on multi-dimensional frame selection features of product images provided in an embodiment of the present application; Figure 3 A schematic diagram of an upload interface for an image to be detected provided in an embodiment of the present application; Figure 4 A schematic diagram of image and text recognition of an image to be detected provided in an embodiment of the present application; Figure 5 A diagram showing the trademark compliance test results provided for this application; Figure 6 This is an example of a homologous box-selected image and text provided for this application; Figure 7 For example, the embodiments of the present application provide Figure 6 A display page diagram of a trademark compliance detection structure of an image to be detected is shown; Figure 8 A schematic diagram of the structure of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0011] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0012] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0013] In the embodiments of the present application, "and / or" describes the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent the following three situations: A exists alone; A and B exist at the same time; B exists alone. Among them, A and B can be singular or plural.

[0014] In the embodiment of the present application, the symbol " / " can indicate that the objects associated with each other are in an "or" relationship. In addition, the symbol " / " can also indicate a division sign, that is, performing a division operation. For example, A / B can indicate A divided by B.

[0015] In the embodiments of the present application, "at least one item" or similar expressions refer to any combination of these items, including any combination of single items or plural items, and refer to one or more, and multiple refers to two or more. For example, at least one item of a, b, or c can represent the following seven situations: a, b, c, a and b, a and c, b and c, a, b, and c. Among them, each of a, b, and c can be an element or a set containing one or more elements.

[0016] In the embodiments of the present application, "equal to" can be used in conjunction with greater than, and is applicable to the technical solution adopted when greater than, and can also be used in conjunction with less than, and is applicable to the technical solution adopted when less than. When equal to is used in conjunction with greater than, it is not used in conjunction with less than; when equal to is used in conjunction with less than, it is not used in conjunction with greater than.

[0017] In order to solve the above problems, the present application provides a trademark compliance detection method and platform based on the multi-dimensional frame selection features of product images. By first obtaining the trademark confidence of the image and text to be detected, and determining the trademark compliance detection result of each image and text to be detected based on the trademark confidence and trademark similarity, the accuracy and comprehensiveness of trademark recognition are improved, thereby improving the accuracy of trademark compliance detection.

[0018] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0019] See also Figure 1 and Figure 2 , Figure 1 A schematic diagram of the structure of an e-commerce platform provided in an embodiment of the present application, Figure 2 A flowchart of a trademark compliance detection method based on multi-dimensional frame selection features of product images provided in an embodiment of the present application.

[0020] like Figure 1 As shown, the e-commerce platform 1 includes a terminal device 10 and a server 20, and the terminal device 10 and the server 20 are communicatively connected via wired or wireless means.

[0021] The terminal device 10 may specifically include a front-end device applied to the user side and capable of realizing functions such as data collection and data transmission, and may be a user equipment (UE) such as a mobile phone, a smart phone, a laptop, a digital broadcast receiver, a personal digital assistant (PDA), a tablet computer (PAD), a handheld device, a vehicle-mounted device, a wearable device, a computing device or other processing equipment connected to a wireless modem, a mobile station (MS), a mobile terminal, etc. Alternatively, the terminal device 10 may also be a software application that can be run in the above electronic devices. For example, it may be an APP running on a mobile phone.

[0022] The server 20 may specifically include a server that is applied to one side of the network platform and is responsible for data processing in the background, which can realize functions such as data transmission and data processing. It may be a physical server, or a server cluster or distributed system composed of multiple physical servers. In this embodiment, the number of servers is not specifically limited. Alternatively, it may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0023] The terminal device 10 is Figure 2 The execution body of the trademark compliance detection method and platform based on the multi-dimensional frame selection features of product images shown in the figure includes the following steps S201 to S205: Step S201: receiving an image to be detected uploaded by a user, identifying and selecting the image to be detected on the image to be detected.

[0024] For specific implementation, please refer to Figure 3 , Figure 3 A schematic diagram of an upload interface for an image to be detected provided in an embodiment of the present application, such as Figure 3 As shown, the trademark detection interface 3 includes an image upload area 31 and a product listing information setting area 32. The image upload area 31 is used to upload the image to be detected. The product listing information includes the area to be listed (such as the United States) and the platform to be listed (such as Amazon, etc.). The product listing information setting area 32 is used to set the aforementioned product listing information.

[0025] After the user uploads the image to be detected, the terminal device sends the image to be detected to the server. The server performs recognition processing on the image to be detected, and determines the image to be detected in the image to be detected and the trademark confidence of the image to be detected, wherein the trademark confidence is used to characterize the probability that the image to be detected is identified as a trademark. Specifically, the image to be detected includes graphic elements and text elements, wherein some graphic elements are product images or conventional modified graphics, while another part of the graphic elements may be trademark elements; some text elements may be basic product description information, while another part of the text elements may be trademark elements. In the prior art, it is impossible to accurately distinguish between trademarks or non-trademark elements in graphics and texts. Usually, all graphic elements or text elements are directly regarded as trademark elements, and the trademark similarity with the existing trademark database is determined, and the trademark infringement risk of the image to be detected is directly determined based on the trademark similarity.

[0026] From the above description, it can be seen that directly treating all graphic and text elements as trademarks for trademark similarity identification is likely to lead to large errors in trademark compliance detection.

[0027] Therefore, in this application, we first distinguish between trademark elements and non-trademark elements in graphic and text elements, that is, calculate the trademark confidence of graphic and text elements, and then calculate the trademark infringement risk based on the trademark confidence and trademark similarity, which can improve the accuracy and comprehensiveness of trademark identification, and thus improve the accuracy of trademark compliance detection.

[0028] Specifically, in some embodiments, the image and text to be detected includes a graphic to be detected, and the calculation process of the trademark confidence is as follows: obtaining the area of ​​the framed region corresponding to the selection box of the image and text to be detected, and obtaining the clarity of the at least one similar trademark; for the graphic to be detected, calculating the ratio of the area of ​​the framed region to the area of ​​the image to be detected; and calculating the trademark confidence based on the area ratio, the clarity, and the trademark similarity.

[0029] Among them, after the server identifies all the graphics to be detected, the similarity between the graphics to be detected and the existing registered trademarks is calculated using the image-text similarity method. For the graphics to be detected, the size of the graphic selection area may be large or small, but a selection that is too large or too small may be a wrong selection. Therefore, the area of ​​the selection area is an effective feature for determining whether the graphics to be detected are trademarks. In addition, for non-trademark graphics, there may be no similar trademarks at all in the existing trademark database. Therefore, the acquired trademark may be an abnormal graphic. The clarity of such abnormal graphics is usually very low. Therefore, the clarity of similar trademarks can be an effective feature for determining whether the trademark to be detected is a trademark. In addition, if the similarity is high, the graphics to be detected are more likely to be trademarks. Therefore, trademark similarity can also be determined as an effective feature of whether the trademark to be detected is a trademark.

[0030] Specifically, for the area of ​​the framed region, the ratio of the area of ​​the framed region to the area of ​​the image to be detected is calculated, and the server calculates the trademark confidence according to the area ratio, clarity, and trademark similarity.

[0031] In some embodiments, the trademark confidence is calculated based on the area ratio, the clarity, and the trademark similarity, including: detecting that the area ratio is less than a preset minimum area ratio, or greater than a preset maximum area ratio; and / or detecting that the clarity of at least one similar trademark corresponding to the graphic to be detected is less than a preset clarity, determining that the trademark confidence is a first preset trademark confidence, the first preset trademark confidence is used to characterize that the probability that the graphic to be detected is a trademark is low; detecting that the area ratio is greater than the preset minimum area ratio and less than the preset maximum area ratio, and detecting that the clarity of at least one similar trademark corresponding to the graphic to be detected is greater than the preset clarity, then calculating the trademark confidence according to the following formula: , where T is the trademark confidence, w1, w2, w3 are weight coefficients, r is the area ratio, c is the clarity, s is the trademark similarity, i is the serial number of the graphic to be detected, j is the serial number of the similar trademark, and n is the number of similar trademarks.

[0032] Among them, according to the above embodiments, when the area of ​​the framed area is too large or too small, the possibility that the graphic to be detected is a non-trademark element is relatively high; when the clarity is too low, the possibility that the graphic to be detected is a non-trademark element is relatively high. Therefore, directly based on the area and clarity of the framed area, the graphics to be detected with a framed area that is too large or too small can be excluded first, or the graphics to be detected with a clarity that is too low can be determined to have a trademark confidence level that is a first preset trademark confidence level. The first preset trademark confidence level can be a value less than 0.2 (the trademark confidence level is a value between 0 and 1).

[0033] Among them, when the area of ​​the framed area is normal and the clarity is normal, the trademark confidence of the image to be detected cannot be directly determined. In this case, the trademark confidence is calculated based on the area ratio, clarity, and similarity. Specifically: The server calculates the trademark confidence level according to the following formula: , where T is the trademark confidence, w1, w2, w3 are weight coefficients, r is the area ratio, c is the clarity, s is the trademark similarity, i is the serial number of the graphic to be detected, j is the serial number of the similar trademark, and n is the number of similar trademarks.

[0034] For example, the trademark similarities of the first three similar trademarks retrieved by the to-be-detected graphic 1 are s11=0.8, s12=0.6, s13=0.45, respectively. The clarity of the first three similar trademarks is c11=0.7, c12=0.8, c13=0.4, respectively. The area ratio of the to-be-detected graphic 1 to the to-be-detected image is r1=0.2. Suppose w1=0.3, w2=0.4, w3=0.3, Then I11= 0.3×r1+0.4×c11+0.3×s11=0.58; Then I12 = 0.3×r1+0.4×c12+0.3×s12=0.56; Then I13 = 0.3×r1+0.4×c13+0.3×s13=0.355; Then T1=(0.58+0.56+0.355) / 3≈0.4983.

[0035] It can be seen that in this application, the trademark confidence of the graphic to be detected can be accurately calculated by selecting the area, clarity, and trademark similarity of the region.

[0036] In some embodiments, the image and text to be detected includes text to be detected, and the trademark confidence is calculated as follows: obtaining the text length in the selection box of the image and text to be detected; and obtaining the maximum text length among the at least one similar trademark, and obtaining the clarity of the at least one similar trademark; for the text to be detected, calculating the ratio of the text length to the maximum text length; and calculating the trademark confidence based on the length ratio, the clarity, and the trademark similarity.

[0037] Among them, after the server identifies all the graphics to be detected, the image-text similarity method is used to calculate the similarity between the graphics to be detected and the existing registered trademarks. For the graphics to be detected, if the text length is too long, it may be a wrong selection box. Therefore, the text length can be an effective feature for determining whether the graphics to be detected are trademarks. In addition, for non-trademark graphics, there may be no similar trademarks at all in the existing trademark database. Therefore, the acquired trademark may be an abnormal graphic. The clarity of such abnormal graphics is usually very low. Therefore, the clarity of similar trademarks can be an effective feature for determining whether the trademark to be detected is a trademark. In addition, if the similarity is high, the graphics to be detected are more likely to be trademarks. Therefore, trademark similarity can also be determined as an effective feature of whether the trademark to be detected is a trademark.

[0038] Specifically, for the length of the characters, the maximum length of the characters of the similar trademarks to the characters to be detected is calculated, and the ratio of the length of the characters to be detected to the maximum length is calculated. The server calculates the trademark confidence according to the length ratio, clarity, and trademark similarity.

[0039] In some embodiments, the trademark confidence is calculated based on the length ratio, the clarity, and the trademark similarity, including: detecting that the length ratio is greater than a preset length ratio, and / or detecting that the clarity of at least one similar trademark corresponding to the text to be detected is less than a preset clarity, determining that the trademark confidence is a first preset trademark confidence, the first preset trademark confidence is used to characterize that the probability that the text to be detected is a trademark is low; detecting that the length ratio is less than the preset length ratio, and detecting that the clarity of at least one similar trademark corresponding to the text to be detected is greater than the preset clarity, then calculating the trademark confidence according to the following formula: , where T is the trademark confidence, w1, w2, w3 are weight coefficients, l is the length ratio, c is the clarity, s is the trademark similarity, i is the serial number of the text to be detected, j is the serial number of the similar trademark, and n is the number of similar trademarks.

[0040] Among them, according to the above embodiments, it can be known that when the length of text is too long, the possibility that the graphic to be detected is a non-trademark element is relatively high; when the clarity is too low, the possibility that the graphic to be detected is a non-trademark element is relatively high. Therefore, directly based on the text length and clarity, the graphics to be detected with too long text length can be excluded first, or the graphics to be detected with too low clarity can be determined to have a trademark confidence level that is a first preset trademark confidence level. The first preset trademark confidence level can be a value less than 0.2 (the trademark confidence level is a value between 0 and 1).

[0041] Among them, when the length and clarity of the text are normal, it is impossible to directly determine the trademark confidence of the graphic to be detected. In this case, the trademark confidence is calculated based on the length ratio, clarity, and similarity. Specifically: The server calculates the trademark confidence level according to the following formula: Among them, T is the trademark confidence, w1, w2, w3 are weight coefficients, l is the length ratio, c is the clarity, s is the trademark similarity, i is the serial number of the text to be detected, j is the serial number of the similar trademark, and n is the number of similar trademarks.

[0042] It can be seen that in this application, the trademark confidence of the graphic to be detected can be accurately calculated through text length, clarity, and trademark similarity.

[0043] Step S202: displaying a selection box for the image and text to be detected and the trademark confidence level of the image and text to be detected on the trademark detection interface.

[0044] The trademark confidence is used to represent the probability that the image and text to be detected is recognized as a trademark.

[0045] For specific implementation, please refer to Figure 4 , Figure 4 is a schematic diagram of graphic and text recognition for the image to be detected provided by an embodiment of the present application. The server recognizes the graphic 41 to be detected in the image 4 to be detected, and recognizes the text to be detected, including "Cha Yan Guan Se" t1, "Bi Yun Jian" t2, and "Modern China Tea Shop" t3, and displays the selection boxes for each text and graphic to be detected, as well as the trademark confidence levels of each text and graphic to be detected.

[0046] Step S203: Retrieve the preset trademark database to obtain at least one similar trademark similar to the text and graphic to be detected and the corresponding trademark similarity.

[0047] Among them, the server can convert the text and graphic to be detected into a vector, and use the method of vector similarity to calculate the similarity between the text and graphic to be detected and the trademarks in the preset trademark database. This method is an existing technology, or other appropriate methods can be used for calculation. The above methods are all existing technologies, and will not be elaborated in this application.

[0048] Step S204: Receive a trigger operation on the trademark compliance detection control on the trademark detection interface, create a trademark compliance detection request message, and send the trademark compliance detection request message to the server.

[0049] Among them, the trademark compliance detection request message includes multi-dimensional box selection features, and the multi-dimensional box selection features include the trademark confidence level and the trademark similarity.

[0050] Among them, see Figure 4 , a trademark compliance detection control is also displayed on the trademark detection interface. After the user clicks the trademark compliance detection control, the server calculates the trademark compliance detection results of each text and graphic to be detected according to the trademark confidence level and the trademark similarity.

[0051] Step S205: Receive a trademark compliance detection response message from the server, and display the trademark compliance detection results for each text and graphic to be detected within each selection box on the trademark detection interface.

[0052] In some embodiments, the creation process of the trademark compliance detection results includes the following steps: calculating the trademark infringement risk probability of the text and graphic to be detected according to the trademark confidence level and the trademark similarity; when it is detected that the trademark infringement risk probability is greater than a first preset threshold, determining that the text and graphic to be detected is a high-infringement-risk text, and determining that the trademark compliance detection result of the high-infringement-risk text is that there is an infringement risk; when it is detected that the trademark infringement risk probability is less than or equal to the first preset threshold, determining that the text and graphic to be detected is a low-infringement-risk text, and determining that the trademark compliance detection result of the low-infringement-risk text is that there is no infringement risk.

[0053] Among them, the server calculates the trademark infringement risk probability based on the trademark confidence level and trademark similarity. This trademark infringement risk probability directly reflects which graphics or texts in the current image to be detected have infringement risks. Specifically, when the trademark infringement risk probability is greater than the first preset threshold, it indicates that the current text and image to be detected have a relatively high infringement risk, and the text and image to be detected are determined as high-infringement-risk text and image. When the trademark infringement risk probability is less than or equal to the first preset threshold, it indicates that the current text and image to be detected have no infringement risk or have a relatively low infringement risk. When the first preset threshold is relatively low, it is directly regarded as having no infringement risk, and the text and image to be detected are determined as low-infringement-risk text and image. Moreover, the trademark compliance detection result of the low-infringement-risk text and image is determined to be no infringement risk.

[0054] In some embodiments, displaying the trademark compliance detection result of the text and image to be detected in each selection box on the trademark detection interface includes: displaying the selection box of the text and image to be detected and the trademark infringement risk probability on the trademark detection interface, and displaying the similar trademarks of the high-infringement-risk text and image on the trademark detection interface; and highlighting the high-infringement-risk text and image.

[0055] For specific implementation, please refer to Figure 5 , Figure 5 which is the display page diagram of the trademark compliance detection result provided by this application. As Figure 5 shown, in this embodiment, it is determined that "Cha Yan Guan Se" t1 is a high-infringement-risk text and image, the to-be-detected graphic 41 is a high-infringement-risk text and image, "Bi Yun Jian" t2 and "Modern China Tea Shop" t3 are low-infringement-risk text and images. Then, on the trademark detection interface, the selection boxes of "Cha Yan Guan Se" t1, the to-be-detected graphic 41, "Bi Yun Jian" t2, and "Modern China Tea Shop" t3 and their corresponding trademark infringement risk probabilities are displayed, and the two high-infringement-risk text and images, namely "Cha Yan Guan Se" t1 and the to-be-detected graphic 41, are highlighted. In addition, on the trademark detection interface 3, the similar trademarks similar to the text and image to be detected and their corresponding similarity degrees are also displayed.

[0056] In some embodiments, before calculating the trademark infringement risk probability of the text and image to be detected according to the trademark confidence level and the trademark similarity, the method further includes: obtaining the graphic contours of the multiple to-be-detected graphics; determining whether there are at least two to-be-detected graphics with the same graphic contour among the multiple to-be-detected graphics; if so, determining the at least two to-be-detected graphics as homologous selected graphics, and setting the trademark confidence levels of the at least two to-be-detected graphics to the maximum value of the trademark confidence levels of the at least two to-be-detected graphics; if not, determining that all the multiple to-be-detected graphics are non-homologous selected graphics.

[0057] Among them, since the server may select the same image frame multiple times in the process of identifying the image to be detected, forming multiple homologous frames, for example, see Figure 6 , Figure 6 The example pictures of the same source frame selection pictures and texts provided for this application are as follows: Figure 6 As shown, when performing image recognition on the image to be detected 6, the server performs three frame selections on the image to be detected 61, forming three image and text objects to be detected. The area of ​​the three selection frames is different, but the content of the frame selections is the same. At this time, since the image and text content is the same, when calculating similar trademarks, the trademark similarities of the three images to be detected and the similar trademarks may be roughly the same, but due to the different frame selection areas, there are large differences in calculating the trademark confidence.

[0058] In order to avoid the errors caused by the above-mentioned ore, the server needs to determine whether there are homologous box-selected images and texts in the identified images and texts to be detected before calculating the trademark infringement risk probability. If so, the trademark confidence of the image and text to be detected is determined to be the highest among the multiple homologous box-selected images and texts. If not, the calculation is performed according to the normal calculation process.

[0059] In some embodiments, the display of the selection box and the trademark infringement risk probability of the image and text to be detected on the trademark detection interface includes: for at least two of the homologous box selection graphics, displaying a selection box and the maximum trademark infringement risk probability in the homologous box selection graphics on the trademark detection interface; for the non-homologous box selection graphics, displaying the selection box and the trademark infringement risk probability of the non-homologous box selection graphics on the trademark detection interface.

[0060] Among them, when there is a general selection box selection graphic, when displaying the trademark compliance detection results on the trademark detection interface, only one selection box needs to be displayed, such as Figure 7 shown.

[0061] It can be seen that in the embodiment of the present application, the terminal device receives an image to be detected uploaded by a user, identifies and selects the image to be detected on the image to be detected; displays a selection box for the image to be detected and the trademark confidence of the image to be detected on the trademark detection interface, and the trademark confidence is used to characterize the probability that the image to be detected is identified as a trademark; searches a preset trademark database to obtain at least one similar trademark similar to the image to be detected and the corresponding trademark similarity; receives a trigger operation on the trademark compliance detection control on the trademark detection interface, creates a trademark compliance detection request message, and sends a trademark compliance detection request message to the server, the trademark compliance detection request message includes a multi-dimensional selection feature, and the multi-dimensional selection feature includes a trademark confidence and a trademark similarity; receives a trademark compliance detection response message from the server, and displays the trademark compliance detection result for the image to be detected in each selection box on the trademark detection interface. Therefore, in the present application, compared with the prior art method of directly performing trademark similarity search based on the identified image and text information and then determining the risk of trademark infringement, the present application first obtains the trademark confidence of the image and text to be detected, and determines the trademark compliance detection result of each image and text to be detected based on the trademark confidence and trademark similarity, thereby improving the accuracy and comprehensiveness of trademark recognition, and thereby improving the accuracy of trademark compliance detection.

[0062] It should be noted that the specific implementation process of this embodiment can refer to the specific implementation process described in the above method embodiment, which will not be described here.

[0063] With the above Figure 2 For details on the embodiments shown in the drawings, please refer to Figure 8 , Figure 8 A schematic diagram of the structure of a server provided in an embodiment of the present application, such as Figure 8 As shown, the server 20 includes a processor 81, a memory 83, a communication interface 82, and one or more programs 831. The one or more programs 831 are stored in the memory 83 and are configured to be executed by the processor 81. The programs include methods for executing the methods described in the above embodiments.

[0064] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method described in the above method embodiments, and the above computer includes an electronic device.

[0065] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0066] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0067] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0068] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0069] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0070] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the above-mentioned methods in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0071] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable memory, which may include a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0072] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A trademark compliance detection method based on multi-dimensional frame selection features of product images, characterized in that: A terminal device applied to a trademark compliance detection system, wherein the trademark compliance detection system comprises the terminal device and a server, and the method comprises: Receive the image to be detected uploaded by the user, identify and select the image to be detected on the image to be detected; Displaying a selection box for the image and text to be detected and the trademark confidence of the image and text to be detected on the trademark detection interface, wherein the trademark confidence is used to represent the probability that the image and text to be detected is recognized as a trademark; Searching a preset trademark database to obtain at least one similar trademark similar to the image and text to be detected and the corresponding trademark similarity; receiving a triggering operation on a trademark compliance detection control on the trademark detection interface, creating a trademark compliance detection request message, and sending the trademark compliance detection request message to the server, wherein the trademark compliance detection request message includes a multi-dimensional box selection feature, and the multi-dimensional box selection feature includes the trademark confidence and the trademark similarity; A trademark compliance detection response message is received from the server, and the trademark compliance detection result for the image and text to be detected in each selection box is displayed on the trademark detection interface.

2. The method according to claim 1, characterized in that The process of creating the trademark compliance test results includes the following steps: Calculating the trademark infringement risk probability of the image and text to be detected according to the trademark confidence and the trademark similarity; When it is detected that the trademark infringement risk probability is greater than a first preset threshold, the image to be detected is determined to be a high infringement risk image, and the trademark compliance detection result of the high infringement risk image is determined to be an infringement risk; When it is detected that the trademark infringement risk probability is less than or equal to the first preset threshold, the image and text to be detected is determined to be a low infringement risk image and text, and the trademark compliance detection result of the low infringement risk image and text is determined to be no infringement risk.

3. The method according to claim 2, characterized in that The trademark detection interface displays the trademark compliance detection result for the image and text to be detected in each selection box, including: Displaying the selection box of the image and text to be detected and the probability of trademark infringement risk on the trademark detection interface, and displaying similar trademarks of the image and text with high infringement risk on the trademark detection interface; and highlighting the image and text with high infringement risk.

4. The method according to claim 3, characterized in that The image to be detected includes a graphic to be detected, and the calculation process of the trademark confidence is as follows: Obtaining the area of ​​the framed region corresponding to the frame of the image and text to be detected, and obtaining the clarity of the at least one similar trademark; For the graphic to be detected, calculating the ratio of the area of ​​the framed region to the area of ​​the image to be detected; The trademark confidence is calculated according to the area ratio, the clarity, and the trademark similarity.

5. The method according to claim 4, characterized in that The calculating the trademark confidence according to the area ratio, the clarity, and the trademark similarity includes: It is detected that the area ratio is less than a preset minimum area ratio, or greater than a preset maximum area ratio; and / or it is detected that the clarity of at least one similar trademark corresponding to the to-be-detected graphic is less than a preset clarity, and the trademark confidence is determined to be a first preset trademark confidence, and the first preset trademark confidence is used to indicate that the probability that the to-be-detected graphic is a trademark is low; If it is detected that the area ratio is greater than the preset minimum area ratio and less than the preset maximum area ratio, and it is detected that the clarity of at least one similar trademark corresponding to the to-be-detected graphic is greater than the preset clarity, the trademark confidence is calculated according to the following formula: Among them, T is the trademark confidence, w1, w2, w3 are weight coefficients, r is the area ratio, c is the clarity, s is the trademark similarity, i is the serial number of the graphic to be detected, j is the serial number of the similar trademark, and n is the number of similar trademarks.

6. The method according to claim 3, characterized in that The image and text to be detected include the text to be detected, and the calculation process of the trademark confidence is as follows: Obtaining the length of the text in the selection box of the image to be detected; and, obtaining the maximum length of the text in the at least one similar trademark, and, obtaining the clarity of the at least one similar trademark; For the text to be detected, calculating the ratio of the length of the text to the maximum length of the text; The trademark confidence is calculated according to the length ratio, the clarity, and the trademark similarity.

7. The method according to claim 6, characterized in that The calculating the trademark confidence according to the length ratio, the clarity, and the trademark similarity includes: When it is detected that the ratio of the lengths is greater than a preset length ratio, and / or it is detected that the clarity of at least one similar trademark corresponding to the text to be detected is less than a preset clarity, the trademark confidence is determined to be a first preset trademark confidence, and the first preset trademark confidence is used to indicate that the probability that the text to be detected is a trademark is low; If it is detected that the ratio of the lengths is less than the preset length ratio, and it is detected that the clarity of at least one similar trademark corresponding to the text to be detected is greater than the preset clarity, the trademark confidence is calculated according to the following formula: Among them, T is the trademark confidence, w1, w2, w3 are weight coefficients, l is the length ratio, c is the clarity, s is the trademark similarity, i is the serial number of the text to be detected, j is the serial number of the similar trademark, and n is the number of similar trademarks.

8. The method according to claim 5, characterized in that Before calculating the trademark infringement risk probability of the to-be-detected image and text according to the trademark confidence and the trademark similarity, the method further includes: Acquire the graphic contours of the multiple graphics to be detected; Determine whether there are at least two graphics to be detected with the same contour among the multiple graphics to be detected; If so, determining that the at least two graphics to be detected are homologous frame selection graphics, and setting the trademark confidence of the at least two graphics to be detected to the maximum value of the trademark confidence of the at least two graphics to be detected; If not, it is determined that the multiple graphics to be detected are all non-homologous frame selection graphics.

9. The method according to claim 8, characterized in that The displaying of the selection box of the image and text to be detected and the trademark infringement risk probability on the trademark detection interface includes: For at least two of the same source framed selection graphics, displaying a selection box and the maximum trademark infringement risk probability in the same source framed selection graphics on the trademark detection interface; For the non-homologous framed selection graphic, the selection box of the non-homologous framed selection graphic and the trademark infringement risk probability are displayed on the trademark detection interface.

10. A trademark compliance detection platform based on multi-dimensional frame selection features of product images, characterized in that: The trademark compliance detection platform includes terminal equipment and a server. The terminal device is used to execute the step instructions executed by the terminal device in the method according to any one of claims 1 to 9; The server is used to execute the step instructions executed by the server in the method according to any one of claims 1 to 9.

Images