Advertisement putting method and device, equipment, storage medium and program product

Abstract

The invention discloses an advertisement putting method and device, equipment, a storage medium and a program product, and relates to the technical field of video processing, and the method comprises the steps: recognizing an advertising board region in a first image of a video source; a target sub-area of the advertising board area is processed, a second image corresponding to the first image is obtained, the target sub-area is an unshielded sub-area in the advertising board area, and the transparency value of the target sub-area is smaller than that of other sub-areas of the advertising board area; determining a target advertisement image corresponding to the target user according to the feature information of the target user; sending the second image and the target advertisement image to a terminal device; thus, the terminal equipment can synthesize and display the second image and the target advertisement image, so that different users can watch advertisements in which the users are interested when watching the video, and the advertisement putting effect is improved.

Description

Technical Field

[0001] The present application relates to the field of video processing technology, and in particular to an advertisement delivery method, apparatus, device, storage medium, and program product. Background Art

[0002] During video playback (such as live broadcasts of sports events), electronic billboards use pre-set advertisements on-site. Therefore, the advertising content that can be displayed in each game is limited. When watching the video, viewers cannot see the expected advertising content, resulting in poor advertising effectiveness. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide an advertisement delivery method, apparatus, device, storage medium and program product, thereby solving the problem of poor advertising delivery effect in the prior art.

[0004] In a first aspect, to achieve the above-mentioned objectives, embodiments of the present application provide an advertisement delivery method, applied to a cloud device, comprising:

[0005] In a first image of the video source, identifying a billboard area;

[0006] Processing a target subregion of the billboard area to obtain a second image corresponding to the first image, wherein the target subregion is an unobstructed subregion of the billboard area, and a transparency value of the target subregion is less than transparency values of other subregions of the billboard area;

[0007] Determining a target advertisement image corresponding to the target user based on characteristic information of the target user;

[0008] The second image and the target advertisement image are sent to a terminal device.

[0009] Optionally, sending the second image and the target advertisement image to a terminal device includes:

[0010] synthesizing the second image with the target advertisement image to obtain a synthesized image;

[0011] The synthesized image is sent to the terminal device.

[0012] Optionally, in the first image of the video source, identifying a billboard area includes:

[0013] Performing target detection in the first image to obtain at least one candidate region;

[0014] Matching a preset billboard template with each candidate area to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area with the highest matching degree with the billboard template among the at least one candidate area;

[0015] Edge detection is performed in the target candidate area to obtain the billboard area.

[0016] Optionally, matching a preset billboard template with each candidate area to obtain a target candidate area includes:

[0017] Extracting features of the target object in the candidate area to obtain feature information of the target object;

[0018] Matching the characteristic information of the target object with the characteristic information of the billboard in the billboard template to obtain a matching degree between the target object and the billboard;

[0019] The candidate area where the target object with the highest matching degree is located is determined as the target candidate area.

[0020] Optionally, processing the target sub-region of the billboard region to obtain a second image corresponding to the first image includes:

[0021] Performing edge detection on the first image to obtain first edge information of the billboard area and second edge information of an obstruction that obstructs the billboard area;

[0022] Segmenting the billboard area and the area corresponding to the obstruction according to the first edge information and the second edge information, and acquiring the target sub-area in the billboard area;

[0023] adjusting the transparency of pixels in the target sub-region, wherein the adjusted transparency value of the target sub-region is smaller than the transparency values of other sub-regions of the billboard region;

[0024] The area corresponding to the obstruction and the processed billboard area are synthesized to obtain the second image.

[0025] Optionally, determining a target advertisement image corresponding to the target user according to characteristic information of the target user includes:

[0026] Acquire a set of candidate advertising images that match the characteristic information of the target user;

[0027] For each candidate advertising image in the set of candidate advertising images, if the legal compliance of the candidate advertising image meets a preset condition, calculating an advertising recommendation value of the candidate advertising image based on the relevance between the video content corresponding to the video source and the candidate advertising image, advertising brand positioning information, and a pre-configured weight;

[0028] The candidate advertising image corresponding to the largest advertising recommendation value is determined as the target advertising image.

[0029] Optionally, in the first image of the video source, identifying a billboard area includes:

[0030] Determining a motion trajectory of the billboard area in the video source based on the billboard area identified in the third image and a target tracking algorithm; wherein the third image is at least one frame of image preceding the first image;

[0031] The billboard area in the first image is identified according to the motion trajectory.

[0032] In a second aspect, to achieve the above-mentioned objectives, embodiments of the present application provide an advertisement delivery method, applied to a terminal device, the method comprising:

[0033] Acquire a second image of the video source and a target advertisement image corresponding to the target user; wherein the transparency value of the unobstructed target sub-region in the billboard area in the second image is less than the transparency values of other sub-regions in the billboard area;

[0034] synthesizing the second image with the target advertisement image to obtain a synthesized image;

[0035] The synthesized image is displayed, wherein all or part of the target advertisement image is displayed on the screen of the terminal device.

[0036] Optionally, obtaining the second image of the video source and the target advertisement image corresponding to the target user includes:

[0037] Receive the second image and the target advertisement image sent by the cloud device.

[0038] Optionally, obtaining the second image of the video source and the target advertisement image corresponding to the target user includes:

[0039] identifying a billboard area in a first image of the video source;

[0040] Processing the target sub-region of the billboard area to obtain a second image corresponding to the first image; wherein the target sub-region is an unobstructed sub-region of the billboard area, and the transparency value of the target sub-region is less than the transparency values of other sub-regions of the billboard area;

[0041] The target advertisement image corresponding to the target user is determined according to the characteristic information of the target user.

[0042] Optionally, identifying a billboard area in the first image of the video source includes:

[0043] Performing target detection in the first image to obtain at least one candidate region;

[0044] Matching a preset billboard template with each candidate area to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area with the highest matching degree with the billboard template among the at least one candidate area;

[0045] Edge detection is performed in the target candidate area to obtain the billboard area.

[0046] Optionally, matching a preset billboard template with each candidate area to obtain a target candidate area includes:

[0047] Extracting features of the target object in the candidate area to obtain feature information of the target object;

[0048] Matching the characteristic information of the target object with the characteristic information of the billboard in the billboard template to obtain a matching degree between the target object and the billboard;

[0049] The candidate area where the target object with the highest matching degree is located is determined as the target candidate area.

[0050] Optionally, processing the target sub-region of the billboard region to obtain the second image corresponding to the first image includes:

[0051] Performing edge detection on the first image to obtain first edge information of the billboard area and second edge information of an obstruction that obstructs the billboard area;

[0052] Segmenting the billboard area and the area corresponding to the obstruction according to the first edge information and the second edge information, and acquiring the target sub-area in the billboard area;

[0053] adjusting the transparency of pixels in the target sub-region, wherein the adjusted transparency value of the target sub-region is smaller than the transparency values of other sub-regions of the billboard region;

[0054] The area corresponding to the obstruction and the processed billboard area are synthesized to obtain the second image.

[0055] Optionally, determining the target advertisement image corresponding to the target user according to characteristic information of the target user includes:

[0056] Acquire a set of candidate advertising images that match the characteristic information of the target user;

[0057] For each candidate advertising image in the set of candidate advertising images, if the legal compliance of the candidate advertising image meets a preset condition, calculating an advertising recommendation value of the candidate advertising image based on the relevance between the video content corresponding to the video source and the candidate advertising image, advertising brand positioning information, and a pre-configured weight;

[0058] The candidate advertising image corresponding to the largest advertising recommendation value is determined as the target advertising image.

[0059] Optionally, identifying a billboard area in the first image of the video source includes:

[0060] Determining a motion trajectory of the billboard area in the video source based on the billboard area identified in the third image and a target tracking algorithm; wherein the third image is at least one frame of image preceding the first image;

[0061] The billboard area in the first image is identified according to the motion trajectory.

[0062] In a third aspect, in order to achieve the above-mentioned objectives, embodiments of the present application provide an advertising delivery device, which is applied to a cloud device, and the device includes:

[0063] an identification module, configured to identify a billboard area in a first image of a video source;

[0064] an image processing module, configured to process a target subregion of the billboard area to obtain a second image corresponding to the first image, wherein the target subregion is an unobstructed subregion of the billboard area, and a transparency value of the target subregion is less than transparency values of other subregions of the billboard area;

[0065] A determination module, configured to determine a target advertisement image corresponding to the target user based on characteristic information of the target user;

[0066] The sending module is configured to send the second image and the target advertisement image to a terminal device.

[0067] In a fourth aspect, in order to achieve the above-mentioned objectives, embodiments of the present application provide an advertising delivery device, applied to a terminal device, comprising:

[0068] an acquisition module, configured to acquire a second image of the video source and a target advertisement image corresponding to the target user; wherein the transparency value of the unobstructed target sub-region in the billboard area in the second image is less than the transparency values of other sub-regions in the billboard area;

[0069] a synthesis module, configured to synthesize the second image with the target advertisement image to obtain a synthesized image;

[0070] The display module is configured to display the synthesized image, wherein all or part of the target advertisement image is displayed on the screen of the terminal device.

[0071] In the fifth aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides an advertising delivery device, including: a transceiver, a processor, a memory, and a computer program stored on the memory and runnable on the processor. When the computer program is executed by the processor, it implements the advertising delivery method as described in the first aspect, or implements the advertising delivery method as described in the second aspect.

[0072] In the sixth aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the advertising delivery method as described in the first aspect, or implements the advertising delivery method as described in the second aspect.

[0073] In the seventh aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a computer program product, including computer instructions, which, when executed by a processor, implement the advertising delivery method as described in the first aspect, or implement the advertising delivery method as described in the second aspect.

[0074] The above technical solution of the present application has at least the following beneficial effects:

[0075] In the advertising delivery method of the embodiment of the present application, first, the cloud device identifies a billboard area in a first image of a video source; second, the cloud device processes a target sub-area of the billboard area to obtain a second image corresponding to the first image, wherein the target sub-area is an unobstructed sub-area of the billboard area, and the transparency value of the target sub-area is less than the transparency values of other sub-areas of the billboard area; third, the cloud device determines a target advertising image corresponding to the target user based on the characteristic information of the target user; finally, the cloud device sends the second image and the target advertising image to the terminal device. In this way, the terminal device can subsequently synthesize the second image and the target advertising image so that when the terminal device plays the video source, the target advertising image is presented to the target user through the target sub-area, thereby ensuring that different users can view advertisements of their interest, thereby improving the effectiveness of advertising delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is one of the flowcharts of the advertising delivery method according to an embodiment of the present application;

[0077] Figure 2 A schematic diagram of a billboard area identified in an embodiment of the present application;

[0078] Figure 3 This is a schematic diagram of a billboard area being blocked in an embodiment of the present application;

[0079] Figure 4 This is a schematic diagram of performing matting on a billboard area in an embodiment of the present application;

[0080] Figure 5 This is a schematic diagram of the updated content in the billboard area in the embodiment of the present application;

[0081] Figure 6 This is a second flow chart of the advertising delivery method according to an embodiment of the present application;

[0082] Figure 7 This is one of the structural diagrams of the advertising delivery device according to an embodiment of the present application;

[0083] Figure 8 This is the second structural diagram of the advertising delivery device according to an embodiment of the present application;

[0084] Figure 9 This is a schematic diagram of the structure of the advertising delivery device according to an embodiment of the present application. DETAILED DESCRIPTION

[0085] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0086] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects and are not used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of this application can be implemented in an order other than those illustrated or described herein. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0087] The following, in conjunction with the accompanying drawings, describes in detail the advertising delivery method, apparatus, device, storage medium, and program product provided by the embodiments of the present application through specific embodiments and their application scenarios.

[0088] The embodiment of the present application provides an advertisement delivery method, which is applied to a cloud device, such as Figure 1 As shown, the method includes:

[0089] Step 101: Identify a billboard area in a first image of a video source.

[0090] In the above step 101, the first image is a frame image in the video source, for example, the first image is a frame image to be played in the video source; taking a football game as an example, Figure 2 As shown, the billboard area is, for example, the area where the fence between the stadium and the audience seats is located.

[0091] Step 102: Process a target sub-region of the billboard area to obtain a second image corresponding to the first image, wherein the target sub-region is an unobstructed sub-region of the billboard area, and the transparency value of the target sub-region is less than the transparency values of other sub-regions of the billboard area.

[0092] In step 102, as a preferred example, the target sub-regions can be set to a transparency that meets preset requirements. This transparency setting is sufficient as long as it ensures that, for example, a target advertising image selected subsequently is presented to the user through these target sub-regions. For example, the target sub-regions can be set to be completely transparent, so that the target advertising image selected subsequently is presented to the user through these target sub-regions. Adjusting the transparency of the target sub-regions can be achieved by setting an alpha channel.

[0093] Step 103: Determine a target advertisement image corresponding to the target user according to the characteristic information of the target user.

[0094] Here, the characteristic information of the target user may include age, gender, occupation, hobbies and consumption habits, etc. The target advertisement image is the advertisement that the target user is most interested in, which is selected based on the characteristic information of the target user.

[0095] Step 104: Send the second image and the target advertisement image to a terminal device.

[0096] Here, by sending the second image and the target advertising image to the terminal device, the terminal device can synthesize the second image and the target advertising image, so as to simultaneously display the target advertising image to the target user when playing the second image, thereby displaying different advertisements to different users and improving the effect of advertising delivery.

[0097] In the advertising delivery method of the embodiment of the present application, first, the cloud device identifies a billboard area in a first image of a video source; second, the cloud device processes a target sub-area of the billboard area to obtain a second image corresponding to the first image, wherein the target sub-area is an unobstructed sub-area of the billboard area, and the transparency value of the target sub-area is less than the transparency values of other sub-areas of the billboard area; third, the cloud device determines a target advertising image corresponding to the target user based on the characteristic information of the target user; finally, the cloud device sends the second image and the target advertising image to the terminal device. In this way, the terminal device can subsequently synthesize the second image and the target advertising image so that when the terminal device plays the video source, the target advertising image is presented to the target user through the target sub-area, thereby ensuring that different users can view advertisements of their interest, thereby improving the effectiveness of advertising delivery.

[0098] As a first optional implementation, step 104 includes:

[0099] The second image is synthesized with the target advertising image to obtain a synthesized image, and the synthesized image is sent to the terminal device. As an example, in this optional implementation, the second image can be superimposed on the target advertising image, wherein at least a portion of the target advertising image should correspond to a target sub-region in the second image, so that the target advertising image is displayed through the target sub-region; that is, when the second image is synthesized with the target advertising image, the target advertising image serves as the background image and the second image serves as the foreground image.

[0100] In the above optional implementation method, the cloud device first synthesizes the second image and the target advertising image, and then sends the synthesized image to the terminal device, so that the terminal device directly displays the synthesized image. In this way, the image processing process of the terminal device is reduced and the power consumption of the terminal device is reduced.

[0101] As a second optional implementation, step 101 includes:

[0102] Target detection is performed in the first image to obtain at least one candidate region.

[0103] In the above steps, specifically, an object detection algorithm (such as YOLO, etc.) may be used to detect an area that may be a billboard (i.e., a candidate area) in the first image; wherein, if the detected candidate area includes a billboard and other objects (such as a player blocking a billboard in a football game or Figure 3 The door frame in the image can be used to distinguish billboards from other objects by extracting the features of different objects (such as color, shape, texture, boundary, etc.).

[0104] A preset billboard template is matched with each candidate area to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area with the highest matching degree with the billboard template among the at least one candidate area.

[0105] In the above steps, by matching the billboard template with each candidate area and determining the candidate area with the highest matching degree as the target candidate area, the position of the billboard can be further determined. Considering that billboards at different angles need to be matched in the video frame, it is necessary to prepare billboard templates at various angles in advance to match each candidate area with billboard templates at different angles to obtain the billboard template with the highest matching degree. In addition, it should be noted that the commonly used template matching algorithms in the above steps include the Normalized Cross-Correlation algorithm (NCC), etc., and when using the template matching algorithm, matching can be performed based on the extracted features.

[0106] Edge detection is performed in the target candidate area to obtain the billboard area.

[0107] In the above steps, edge detection is performed on the candidate target area to detect distinct edge features, which helps to more accurately locate the billboard. In particular, edge detection is performed after template matching to supplement the missing edge information, helping to more accurately locate the target (billboard).

[0108] In the above optional implementation method, the candidate area is first obtained based on target detection and features of objects in the candidate area are extracted. Then, the extracted features are matched with the billboard template to further obtain the position of the billboard. Finally, edge detection is performed to supplement the missing edge information, ultimately achieving a more accurate and complete positioning of the billboard.

[0109] As a specific implementation method, matching a preset billboard template with each candidate area to obtain a target candidate area includes:

[0110] Extracting features of the target object in the candidate area to obtain feature information of the target object;

[0111] Matching the characteristic information of the target object with the characteristic information of the billboard in the billboard template to obtain a matching degree between the target object and the billboard;

[0112] The candidate area where the target object with the highest matching degree is located is determined as the target candidate area.

[0113] That is to say, the above-mentioned specific implementation method includes: for the detected target, extracting the corresponding feature representation, and performing feature matching between the extracted feature representation and the area corresponding to the billboard template (the area where the specific billboard in the billboard template is located) to obtain a matching score or similarity / matching degree.

[0114] As a third optional implementation, step 102 includes:

[0115] Edge detection is performed on the first image to obtain first edge information of the billboard area and second edge information of an obstruction that obstructs the billboard area.

[0116] Here, the above steps can use the Canny edge detection algorithm to perform edge detection on the first image, which helps to extract edge information of the billboard and the obstruction in the first image. In addition, based on the above steps, on the one hand, the detected edges can be further enhanced to reduce noise and enhance edge continuity. Among them, common enhancement methods include: using gradient-based methods or filtering methods; on the other hand, after the edge enhancement process, the edges can also be connected to connect similar edges to form a more complete boundary line. Specifically, an edge connection algorithm such as Hough transform can be used.

[0117] The billboard area and the area corresponding to the obstruction are segmented according to the first edge information and the second edge information, and the target sub-area in the billboard area is obtained.

[0118] Specifically, the above steps can be used to segment the billboard and the obstruction according to the first edge information and the second edge information using certain conditions and rules. Specifically, the judgment and segmentation can be based on geometric information, texture information, color information, etc. Taking the football game as an example, the segmented billboard area (the billboard is cut out) is as follows: Figure 4 The black rectangular area shown in FIG. Further, the segmentation result can be optimized. For example, morphological operations, color similarity, connected component analysis and other methods can be used to remove objects other than the billboard (such as occluders of the billboard) or to refine and merge the billboard.

[0119] The transparency of pixels in the target sub-region is adjusted, wherein the adjusted transparency value of the target sub-region is smaller than the transparency values of other sub-regions of the billboard region.

[0120] The above steps can specifically be to set the alpha channel for the area outside the occluder after the segmentation is completed, by selecting the alpha channel and using software to draw the pixels in the image to define the transparency. Usually, for the area where the cutout is successful (the billboard area in this application), 0 means completely transparent.

[0121] The area corresponding to the occluder and the processed billboard area are synthesized to obtain the second image. In this step, the processed billboard area is used as the background image and the occluder is used as the foreground image for synthesis, wherein the position of the occluder is the same as that in the first image; taking the football game as an example, Figure 5 As shown, the target advertising images are advertisement 10, advertisement 20, and advertisement 30, which are placed behind the billboard area so as to be displayed through the target sub-area.

[0122] The above optional implementation method specifically utilizes an edge detection-based image segmentation method (a segmentation method based on pixel similarity) to handle the problem of occlusion detection and removal. By using an image segmentation algorithm to detect occlusions, the occluded area of the billboard can be detected, and an attempt can be made to restore the occluded content through image restoration or a repair model. Specifically, the matte image (second image) is saved as an image format with an alpha channel and then placed on the target advertising image. This allows the edges of the matte image to smoothly transition to the background.

[0123] As a fourth optional implementation, step 103 includes:

[0124] A set of candidate advertising images matching the characteristic information of the target user is obtained.

[0125] As mentioned above, target user profile information includes the target audience's characteristics, interests, hobbies, and spending habits. Within the audience dimension, you can segment by age, gender, occupation, interests, and hobbies. You can choose from the following 24*X options. 24 is composed of 4 age categories, 2 gender categories, and 3 occupation categories. X is the total number of interests and hobbies.

[0126] Age: Children (0-18 years old), Youth (19-35 years old), Middle-aged (36-60 years old), Elderly (61 years old and above).

[0127] Gender: Male, Female.

[0128] Occupation: primary, secondary and tertiary industries.

[0129] Interests: A hobby or interest that is actively input or synthesized from big data. Active input refers to the user's active input of interests and hobbies through the input method established by the video viewing application (Application, APP).

[0130] In the above steps, the candidate advertising image information is determined based on the characteristic information of the target user. Specifically, the advertisements suitable for the target user are screened out from the overall range of advertisements played (the overall range is equivalent to 24*X choices), that is, the advertising range is limited to 1 / (24*X) of all advertisements. In this way, the target range of the advertisement (the range corresponding to the candidate advertising image set) can be reduced.

[0131] For each candidate advertising image in the candidate advertising image set, when the legal compliance of the candidate advertising image meets the preset conditions, the advertising recommendation value of the candidate advertising image is calculated based on the correlation between the video content corresponding to the video source and the candidate advertising image, the advertising brand positioning information and the pre-configured weight.

[0132] In the above steps, firstly, the legal compliance of the candidate advertising images meets the preset conditions, which means that the advertisement complies with relevant advertising laws and regulations to ensure that the advertising content is legal and compliant. Taking the competition as an example, secondly, the relevance between the video content and the candidate advertising images can be referred to as event relevance. In terms of the event dimension, it can be distinguished according to the event type, event time, venue conditions (advertising location: for example, the corner area that is often given close-ups in football matches), athlete team conditions, etc. Thirdly, advertising brand positioning information means that the advertising content should be consistent with the event positioning (the expected advertising exposure direction set by the advertiser), and the audience appeal index of the advertisement (creativity and highlights, the audience suitable for this advertisement set by the advertising platform).

[0133] The above steps can be expressed by the formula: advertising recommendation value V = [β*B (event relevance) + γ*C (brand positioning)]*D (legal compliance is 0 or 1); where β and γ are pre-configured weights.

[0134] The candidate advertising image corresponding to the largest advertising recommendation value is determined as the target advertising image.

[0135] It should be noted that the ad recommendation value V refers to the selection indicator for all ads within the target audience (target user) defined within the aforementioned ad target range. Once the V values of all candidate ads are calculated, the ad with the current maximum V value is the selected ad. Ad selection is related to event relevance and brand positioning, while legal compliance can veto the ad selection value. Legal compliance is differentiated based on different events, audiences, and countries.

[0136] As a fifth optional implementation, step 101 includes:

[0137] Determining a motion trajectory of the billboard area in the video source based on the billboard area identified in the third image and a target tracking algorithm; wherein the third image is at least one frame of image preceding the first image;

[0138] The billboard area in the first image is identified according to the motion trajectory.

[0139] That is to say, when it is necessary to continuously track the position of a billboard, a target tracking algorithm can be used to track the motion trajectory of the billboard in the video sequence, thereby achieving continuous recognition of the billboard. Compared with the method of separately identifying the billboard area in each frame, the above optional implementation method reduces the complexity of data processing and can save data processing resources.

[0140] Here, it should be noted that the advertising delivery method of the present application needs to complete all steps of processing within the playback time of each frame in a live broadcast scenario. Taking 50 frames per second as an example, all steps of processing need to be completed within 20ms for each frame.

[0141] The embodiment of the present application also provides an advertisement delivery method, which is applied to a terminal device, such as Figure 6 As shown, the method includes:

[0142] Step 601 : Acquire a second image of a video source and a target advertisement image corresponding to a target user; wherein the transparency value of an unobstructed target sub-region in a billboard area in the second image is smaller than the transparency values of other sub-regions in the billboard area.

[0143] In step 601, the terminal device may obtain the second image and the target advertising image based on its own processing, or the terminal device may obtain the second image and the target advertising image from another device (e.g., a cloud device). Furthermore, as a preferred example, the target sub-region is a completely transparent region, i.e., the transparency value of the pixels in the target sub-region is preferably 0.

[0144] Step 602: synthesize the second image and the target advertisement image to obtain a synthesized image.

[0145] The above-mentioned step 602 can specifically be superimposing the second image on the target advertising image, that is, the target advertising image is the background image and the second image is the foreground image; wherein, after synthesis, at least a portion of the target advertising image can be located behind the target sub-region of the second image, so that at least a portion of the target advertising image can be displayed through the target sub-region.

[0146] Step 603: display the synthesized image, wherein all or part of the target advertisement image is displayed on the screen of the terminal device.

[0147] In the advertising delivery method of an embodiment of the present application, first, the terminal device obtains a second image of a video source and a target advertising image corresponding to a target user; wherein the transparency of an unobstructed target sub-area in the billboard area in the second image is greater than the transparency of other sub-areas of the billboard area; secondly, the terminal device synthesizes the second image with the target advertising image to obtain a synthesized image; finally, the synthesized image is displayed, wherein all or part of the target advertising image is displayed on the screen of the terminal device. In this way, when the second image is played, at least a part of the target advertising image that the user is interested in can be displayed through the target sub-area, thereby improving the effect of advertising delivery.

[0148] As a first optional implementation, step 601 includes:

[0149] Receive the second image and the target advertisement image sent by the cloud device.

[0150] In the above optional implementation, the terminal device can receive the second image and the target advertising image from the cloud device. In this way, the terminal device only needs to synthesize the second image and the target advertising image, thereby simplifying the processing process of the terminal device and reducing the power consumption of the terminal device.

[0151] As a second optional implementation, step 601 includes:

[0152] In the first image of the video source, the billboard area is identified; here, the first image is a frame image in the video source, for example, the first image is a frame image to be played in the video source; taking a football game as an example, Figure 2 As shown, the billboard area is, for example, the area where the fence between the stadium and the audience seats is located;

[0153] The target sub-region of the billboard area is processed to obtain a second image corresponding to the first image; wherein the target sub-region is an unobstructed sub-region of the billboard area, and the transparency value of the target sub-region is less than the transparency values of other sub-regions of the billboard area; in this step, as described above, as a preferred example, the target sub-region can be set to a transparency that meets preset requirements, and the transparency setting only needs to ensure that, for example, the target advertising image selected subsequently is presented to the user through these target sub-regions. For example, it can be set to be completely transparent, so that the advertising image selected subsequently can be presented to the user through these target sub-regions; wherein, adjusting the transparency of the target sub-region can be specifically achieved by setting the alpha channel;

[0154] The target advertising image corresponding to the target user is determined based on the characteristic information of the target user. Here, the characteristic information of the target user may include age, gender, occupation, interests and hobbies, and consumption habits. The target advertising image is the advertisement that the target user is most interested in, which is screened based on the characteristic information of the target user.

[0155] In the above optional implementation method, the terminal device can obtain a frame image in the video source, identify the billboard area in the frame image, and set the unobstructed target sub-area in the billboard area as an area with higher transparency, so that the target advertising image determined based on the characteristics of the target user can be displayed through the target sub-area in the future, so that different users can watch the advertisements of their respective interests, thereby improving the advertising delivery effect.

[0156] As a specific implementation, identifying a billboard area in the first image of the video source includes:

[0157] Target detection is performed in the first image to obtain at least one candidate area. In this step, a target detection algorithm (such as YOLO, etc.) can be used to detect an area in the first image that may be a billboard (i.e., a candidate area). If the detected candidate area contains billboards and other objects (such as players blocking billboards in a football game or Figure 3 The door frame in the image can be used to distinguish billboards from other objects by extracting the features of different objects (such as color, shape, texture, boundary, etc.).

[0158] Match the pre-set billboard template with each of the candidate areas to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area with the highest matching degree with the billboard template among the at least one candidate area; in this step, by matching the billboard template with each candidate area and determining the candidate area with the highest matching degree as the target candidate area, the position of the billboard can be further determined. Considering that billboards at different angles need to be matched in the video frame, it is necessary to prepare billboard templates of various angles in advance to match each candidate area with billboard templates at different angles to obtain the billboard template with the highest matching degree. In addition, it should be noted that the commonly used template matching algorithms in the above steps include NCC, etc., and when using the template matching algorithm, matching can be performed based on the extracted features.

[0159] Edge detection is performed on the candidate target region to obtain the billboard region. In this step, edge detection on the candidate target region can detect distinct edge features, which helps to more accurately locate the billboard. Furthermore, performing edge detection after template matching can supplement missing edge information, helping to more accurately locate the target (billboard).

[0160] In the above optional implementation method, the candidate area is first obtained based on target detection and features of objects in the candidate area are extracted. Then, the extracted features are matched with the billboard template to further obtain the position of the billboard. Finally, edge detection is performed to supplement the missing edge information, ultimately achieving a more accurate and complete positioning of the billboard.

[0161] As a more specific implementation, matching a preset billboard template with each candidate area to obtain a target candidate area includes:

[0162] Extracting features of the target object in the candidate area to obtain feature information of the target object;

[0163] Matching the characteristic information of the target object with the characteristic information of the billboard in the billboard template to obtain a matching degree between the target object and the billboard;

[0164] The candidate area where the target object with the highest matching degree is located is determined as the target candidate area.

[0165] That is to say, the above-mentioned specific implementation method includes: for the detected target, extracting the corresponding feature representation, and performing feature matching between the extracted feature representation and the area corresponding to the billboard template (the area where the specific billboard in the billboard template is located) to obtain a matching score or similarity / matching degree.

[0166] As another specific implementation, processing the target sub-region of the billboard region to obtain the second image corresponding to the first image includes:

[0167] Perform edge detection on the first image to obtain first edge information of the billboard area and second edge information of the obstruction blocking the billboard area. In this step, the Canny edge detection algorithm can be used to perform edge detection on the first image, which helps to extract the edge information of the billboard and the obstruction in the first image. In addition, based on the above steps, on the one hand, the detected edges can be further enhanced to reduce noise and enhance edge continuity. Common enhancement methods include: using gradient-based methods or filtering methods; on the other hand, after the edges are enhanced, the edges can also be connected to connect similar edges to form a more complete boundary line. Specifically, an edge connection algorithm such as Hough transform can be used.

[0168] According to the first edge information and the second edge information, the billboard area and the area corresponding to the occluder are segmented, and the target sub-area in the billboard area is obtained; specifically, the above steps can use certain conditions and rules to segment the billboard and the occluder according to the first edge information and the second edge information, wherein the judgment and segmentation can be based on geometric information, texture information, color information, etc., wherein, taking the football game as an example, the segmented billboard area (the billboard is matted) is as follows Figure 4 The black rectangular area shown in FIG. Further, the segmentation result can be optimized. For example, morphological operations, color similarity, connected component analysis and other methods can be used to remove objects other than the billboard (such as occluders of the billboard) or to refine and merge the billboard.

[0169] Adjust the transparency of the pixels in the target sub-region, wherein the adjusted transparency value of the target sub-region is less than the transparency values of other sub-regions of the billboard area. In this step, specifically, after segmentation, an alpha channel is set for the area outside the occluder. By selecting the alpha channel and using software to draw the pixels in the image, the transparency is defined. Generally, for the area that has been successfully cut out, 0 indicates complete transparency.

[0170] The area corresponding to the occluder and the processed billboard area are synthesized to obtain the second image. In this step, the processed billboard area is used as the background image and the occluder is used as the foreground image for synthesis, wherein the position of the occluder is the same as that in the first image. Figure 5 As shown, the target advertising images are advertisement 10, advertisement 20, and advertisement 30, which are placed behind the billboard area so as to be displayed through the target sub-area.

[0171] As another specific implementation, determining the target advertisement image corresponding to the target user according to the characteristic information of the target user includes:

[0172] Obtain a set of candidate advertising images that match the characteristic information of the target user; as mentioned above, the characteristic information of the target user includes the characteristics, interests, hobbies, and consumption habits of the target audience. In terms of the audience dimension, it can be distinguished according to age, gender, occupation, interests, hobbies, etc. You can choose from the following 24*X. 24 is composed of 4 ages * 2 genders * 3 occupations. X is the total number of directions of interests and hobbies. Among them, the specific descriptions of age, gender, occupation, interests and hobbies have been explained in the implementation method on the cloud device side, and will not be repeated here. Among them, this step realizes the screening of some candidate advertising images corresponding to the characteristic information of the target user from the total advertising range, thereby narrowing the scope of the advertisement.

[0173] For each candidate advertising image in the candidate advertising image set, if the legal compliance of the candidate advertising image meets preset conditions, the advertising recommendation value of the candidate advertising image is calculated based on the relevance between the video content corresponding to the video source and the candidate advertising image, the advertising brand positioning information, and the pre-configured weights. As mentioned above, this step can be calculated using the formula: Advertising Recommendation Value V = [β*B(Event Relevance) + γ*C(Brand Positioning)]*D (Legal Compliance is 0 or 1); where β and γ are pre-configured weights. The event relevance is equivalent to the relevance between the video content and the advertising image.

[0174] The candidate advertising image corresponding to the largest advertising recommendation value is determined as the target advertising image; here, the larger the advertising recommendation value, the more closely the corresponding candidate advertising image matches the characteristic information of the target user, and the more interested the target user is in the candidate advertising image. Therefore, in this step, the candidate advertising image with the largest advertising recommendation value is determined as the target advertising image.

[0175] As another specific implementation, identifying a billboard area in the first image of the video source includes:

[0176] Determining a motion trajectory of the billboard area in the video source based on the billboard area identified in the third image and a target tracking algorithm; wherein the third image is at least one frame of image preceding the first image;

[0177] The billboard area in the first image is identified according to the motion trajectory.

[0178] That is to say, when it is necessary to continuously track the position of a billboard, a target tracking algorithm can be used to track the motion trajectory of the billboard in the video sequence, thereby achieving continuous recognition of the billboard. Compared with the method of separately identifying the billboard area in each frame, the above optional implementation method reduces the complexity of data processing and can save data processing resources.

[0179] The advertising delivery method of the embodiment of the present application specifically discloses a method for personalized display of billboards at sports events, including: first, obtaining continuous video frames from a video source, performing target detection and feature extraction on the video frames to obtain the billboard position, using a preset billboard template for matching to further determine the billboard position, and continuously tracking the billboard position, using a target tracking algorithm to track the motion trajectory of the billboard in the video sequence to achieve continuous identification of the billboard position; secondly, using an image segmentation algorithm to detect obstructions at the billboard position, detecting the obscured area of the billboard, and restoring the obscured content through image restoration or a restoration model; thirdly, selecting advertisements based on target audience, event relevance, brand positioning, and legal compliance, and finally pasting the processed video frame image onto the advertising image in the selected sports scene, so that the edge of the matted image can smoothly transition to the background image, and displaying the selected advertising content during video playback.

[0180] The embodiment of the present application also provides an advertisement delivery device, which is applied to a cloud device, such as Figure 7 As shown, the device includes:

[0181] The identification module 701 is configured to identify a billboard area in a first image of a video source;

[0182] an image processing module 702 configured to process a target subregion of the billboard area to obtain a second image corresponding to the first image, wherein the target subregion is an unobstructed subregion of the billboard area and has a transparency value lower than that of other subregions of the billboard area;

[0183] A determination module 703 is configured to determine a target advertisement image corresponding to the target user based on the characteristic information of the target user;

[0184] The sending module 704 is configured to send the second image and the target advertisement image to a terminal device.

[0185] Optionally, the sending module 704 includes:

[0186] a first synthesis submodule, configured to synthesize the second image with the target advertisement image to obtain a synthesized image;

[0187] The sending submodule is used to send the synthesized image to the terminal device.

[0188] Optionally, the identification module 701 includes:

[0189] A first detection submodule, configured to perform target detection in the first image and obtain at least one candidate region;

[0190] a matching submodule, configured to match a preset billboard template with each of the candidate areas to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area of the at least one candidate area that has the highest degree of matching with the billboard template;

[0191] The second detection submodule is configured to perform edge detection in the target candidate area to obtain the billboard area.

[0192] Optionally, the matching submodule includes:

[0193] A feature extraction unit, configured to extract features of the target object in the candidate area to obtain feature information of the target object;

[0194] a matching unit, configured to match the characteristic information of the target object with the characteristic information of the billboard in the billboard template to obtain a matching degree between the target object and the billboard;

[0195] The determining unit is configured to determine the candidate area where the target object with the highest matching degree is located as the target candidate area.

[0196] Optionally, the image processing module 702 includes:

[0197] a third detection submodule, configured to perform edge detection on the first image to obtain first edge information of the billboard area and second edge information of an obstruction that obscures the billboard area;

[0198] a segmentation submodule, configured to segment the billboard area and the area corresponding to the obstruction according to the first edge information and the second edge information, and obtain the target sub-area in the billboard area;

[0199] an adjusting submodule, configured to adjust the transparency of pixels in the target subregion, wherein the transparency value of the adjusted target subregion is smaller than the transparency values of other subregions of the billboard region;

[0200] The second synthesis submodule is configured to synthesize the area corresponding to the obstruction and the processed billboard area to obtain the second image.

[0201] Optionally, the determining module 703 includes:

[0202] An acquisition submodule, configured to acquire a set of candidate advertising images that match the characteristic information of the target user;

[0203] a calculation submodule, configured to calculate, for each candidate advertising image in the set of candidate advertising images, an advertising recommendation value of the candidate advertising image based on the relevance between the video content corresponding to the video source and the candidate advertising image, advertising brand positioning information, and pre-configured weights, if the legal compliance of the candidate advertising image meets a preset condition;

[0204] The first determining submodule is configured to determine the candidate advertising image corresponding to the largest advertising recommendation value as the target advertising image.

[0205] Optionally, the identification module 701 includes:

[0206] a second determining submodule, configured to determine a motion trajectory of the billboard area in the video source based on the billboard area identified in a third image and a target tracking algorithm; wherein the third image is at least one frame of image preceding the first image;

[0207] The recognition submodule is configured to recognize the billboard area in the first image according to the motion trajectory.

[0208] It should be noted here that the above-mentioned advertising delivery device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned advertising delivery method embodiment applied to cloud devices, and can achieve the same technical effects. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.

[0209] The embodiment of the present application also provides an advertisement delivery device, which is applied to a terminal device, such as Figure 8 As shown, the device includes:

[0210] An acquisition module 801 is configured to acquire a second image from a video source and a target advertisement image corresponding to a target user; wherein a transparency value of an unobstructed target sub-region in a billboard area in the second image is smaller than a transparency value of other sub-regions in the billboard area;

[0211] A synthesis module 802 is configured to synthesize the second image with the target advertisement image to obtain a synthesized image;

[0212] The display module 803 is configured to display the synthesized image, wherein all or part of the target advertisement image is displayed on the screen of the terminal device.

[0213] Optionally, the acquisition module 801 is specifically configured to receive the second image and the target advertisement image sent by a cloud device.

[0214] Optionally, the acquisition module 801 includes:

[0215] an identification submodule, configured to identify a billboard area in a first image of the video source;

[0216] an image processing submodule, configured to process the target subregion of the billboard area to obtain a second image corresponding to the first image; wherein the target subregion is an unobstructed subregion of the billboard area, and the transparency value of the target subregion is less than the transparency values of other subregions of the billboard area;

[0217] The determination submodule is configured to determine the target advertisement image corresponding to the target user according to the characteristic information of the target user.

[0218] Optionally, the identification submodule includes:

[0219] a first detection unit, configured to perform target detection in the first image to obtain at least one candidate region;

[0220] a matching unit, configured to match a preset billboard template with each of the candidate areas to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area of the at least one candidate area that has the highest degree of matching with the billboard template;

[0221] The second detection unit is configured to perform edge detection in the target candidate area to obtain the billboard area.

[0222] Optionally, the matching unit includes:

[0223] A feature extraction subunit is used to extract features of the target object in the candidate area to obtain feature information of the target object;

[0224] a matching subunit, configured to match the characteristic information of the target object with the characteristic information of the billboard in the billboard template to obtain a matching degree between the target object and the billboard;

[0225] The first determining unit is configured to determine the candidate area where the target object with the highest matching degree is located as the target candidate area.

[0226] Optionally, the image processing submodule includes:

[0227] a third detection unit, configured to perform edge detection on the first image to obtain first edge information of the billboard area and second edge information of an obstruction blocking the billboard area;

[0228] a segmentation unit, configured to segment the billboard area and an area corresponding to the obstruction according to the first edge information and the second edge information, and obtain the target sub-area in the billboard area;

[0229] an adjusting unit, configured to adjust the transparency of pixels in the target sub-region, wherein the transparency value of the target sub-region after adjustment is smaller than the transparency values of other sub-regions of the billboard region;

[0230] A synthesis unit is used to synthesize the area corresponding to the obstruction and the processed billboard area to obtain the second image.

[0231] Optionally, the determining submodule includes:

[0232] an acquiring unit, configured to acquire a set of candidate advertising images matching the characteristic information of the target user;

[0233] a calculation unit configured to calculate, for each candidate advertising image in the set of candidate advertising images, an advertising recommendation value of the candidate advertising image based on a relevance between the video content corresponding to the video source and the candidate advertising image, advertising brand positioning information, and a pre-configured weight, if the legal compliance of the candidate advertising image satisfies a preset condition;

[0234] The second determining unit is configured to determine the candidate advertising image corresponding to the largest advertising recommendation value as the target advertising image.

[0235] Optionally, the identification submodule includes:

[0236] a third determining unit, configured to determine a motion trajectory of the billboard area in the video source based on the billboard area identified in a third image and a target tracking algorithm; wherein the third image is at least one frame of image preceding the first image;

[0237] A recognition unit is configured to recognize the billboard area in the first image according to the motion trajectory.

[0238] It should be noted here that the above-mentioned advertising delivery device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned advertising delivery method embodiment applied to the terminal device, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0239] An embodiment of the present application also provides an advertising delivery device, including a transceiver 910, a processor 900, a memory 920, and a program stored on the memory 920 and executable on the processor 900; wherein, when the processor 900 executes the program, the advertising delivery method applied to a cloud device or to a terminal device as described above is implemented.

[0240] The transceiver 910 is configured to receive and send data under the control of the processor 900 .

[0241] Among them, Figure 9 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically various circuits of one or more processors represented by processor 900 and memory represented by memory 920, which are linked together. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are all well known in the art and, therefore, will not be described further herein. The bus interface provides an interface. The transceiver 910 may be a plurality of components, i.e., a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. For different devices, the processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 may store data used by the processor 900 when performing operations.

[0242] The present application also provides a readable storage medium having a program stored thereon. When the program is executed by a processor, the various processes of the above-described advertising delivery method embodiment applied to a cloud device or applied to a terminal device are implemented, and the same technical effects are achieved. To avoid repetition, the description is omitted here. The readable storage medium may be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0243] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course, by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), and includes a number of instructions for executing the methods described in each embodiment of the present application.

[0244] Therefore, an embodiment of the present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the advertising delivery method applied to a cloud device or to a terminal device as described above, and can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0245] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0246] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An advertisement delivery method, characterized in that: Applied to a cloud device, the method includes: In a first image of the video source, identifying a billboard area; Processing a target subregion of the billboard area to obtain a second image corresponding to the first image, wherein the target subregion is an unobstructed subregion of the billboard area, and a transparency value of the target subregion is less than transparency values of other subregions of the billboard area; Determining a target advertisement image corresponding to the target user based on characteristic information of the target user; The second image and the target advertisement image are sent to a terminal device.

2. The method according to claim 1, characterized in that In a first image of a video source, a billboard area is identified, including: Performing target detection in the first image to obtain at least one candidate region; Matching a preset billboard template with each candidate area to obtain a target candidate area, wherein the target candidate area includes the billboard area, and the target candidate area is the area with the highest matching degree with the billboard template among the at least one candidate area; Edge detection is performed in the target candidate area to obtain the billboard area.

3. The method according to claim 1, characterized in that Processing a target sub-region of the billboard region to obtain a second image corresponding to the first image includes: Performing edge detection on the first image to obtain first edge information of the billboard area and second edge information of an obstruction that obstructs the billboard area; Segmenting the billboard area and the area corresponding to the obstruction according to the first edge information and the second edge information, and acquiring the target sub-area in the billboard area; adjusting the transparency of pixels in the target sub-region, wherein the adjusted transparency value of the target sub-region is smaller than the transparency values of other sub-regions of the billboard region; The area corresponding to the obstruction and the processed billboard area are synthesized to obtain the second image.

4. The method according to claim 1, wherein Determining a target advertisement image corresponding to the target user based on characteristic information of the target user includes: Acquire a set of candidate advertising images that match the characteristic information of the target user; For each candidate advertising image in the set of candidate advertising images, if the legal compliance of the candidate advertising image meets a preset condition, calculating an advertising recommendation value of the candidate advertising image based on the relevance between the video content corresponding to the video source and the candidate advertising image, advertising brand positioning information, and a pre-configured weight; The candidate advertising image corresponding to the largest advertising recommendation value is determined as the target advertising image.

5. An advertisement delivery method, characterized in that: Applied to a terminal device, the method includes: Acquire a second image of the video source and a target advertisement image corresponding to the target user; wherein the transparency value of the unobstructed target sub-region in the billboard area in the second image is less than the transparency values of other sub-regions in the billboard area; synthesizing the second image with the target advertisement image to obtain a synthesized image; The synthesized image is displayed, wherein all or part of the target advertisement image is displayed on the screen of the terminal device.

6. The method according to claim 5, characterized in that Acquiring the second image of the video source and the target advertisement image corresponding to the target user includes: identifying a billboard area in a first image of the video source; Processing the target sub-region of the billboard area to obtain a second image corresponding to the first image; wherein the target sub-region is an unobstructed sub-region of the billboard area, and the transparency value of the target sub-region is less than the transparency values of other sub-regions of the billboard area; The target advertisement image corresponding to the target user is determined according to the characteristic information of the target user.

7. The method according to claim 6, characterized in that Processing the target sub-region of the billboard region to obtain the second image corresponding to the first image includes: Performing edge detection on the first image to obtain first edge information of the billboard area and second edge information of an obstruction that obstructs the billboard area; Segmenting the billboard area and the area corresponding to the obstruction according to the first edge information and the second edge information, and acquiring the target sub-area in the billboard area; adjusting the transparency of pixels in the target sub-region, wherein the adjusted transparency value of the target sub-region is smaller than the transparency values of other sub-regions of the billboard region; The area corresponding to the obstruction and the processed billboard area are synthesized to obtain the second image.

8. An advertising device, characterized in that: include: A transceiver, a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the advertising delivery method according to any one of claims 1 to 4, or implements the advertising delivery method according to any one of claims 5 to 7.

9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the advertising delivery method according to any one of claims 1 to 4, or implements the advertising delivery method according to any one of claims 5 to 7.

10. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the advertising delivery method according to any one of claims 1 to 4, or implement the advertising delivery method according to any one of claims 5 to 7.

Images