Intelligent marketing method and system based on artificial intelligence, and program product

Abstract

The invention discloses an intelligent marketing method and system based on artificial intelligence and a program product, and relates to the technical field of intelligent marketing. Extracting face images in multiple sampling video frames of the monitoring video of the target area in the current time period; determining a hair region in the face image based on the pixel point color features, and determining the gender of a person corresponding to the face image based on the area proportion and distribution features of the hair region; extracting horizontal and vertical gradients of the pixel points to obtain a gradient magnitude image of the face image; performing closed contour extraction after performing morphological processing on the binary gradient magnitude image to obtain a facial contour region in the binary gradient magnitude image; taking an image region corresponding to the facial contour region in the face image as an input of an age recognition model to obtain a character age corresponding to each face image; and generating an advertisement marketing strategy of the next time period based on the character gender and the character age corresponding to each face image. According to the invention, the advertisement reach rate of the delivered marketing advertisement can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent marketing, and particularly relates to an intelligent marketing method, system and program product based on artificial intelligence. Background Art

[0002] Intelligent marketing is a new model that uses technologies such as artificial intelligence, big data, and the Internet of Things to achieve precision, automation, and dynamic optimization of marketing strategies. Its most typical application scenario is advertising marketing in target areas (such as shopping malls, public areas, etc. with a large flow of people).

[0003] Currently, for the intelligent advertising marketing in the target area, the more common method is to pre-statistically analyze the overall flow distribution of the target area, and play advertisements corresponding to the user portraits of the majority according to the overall flow distribution of the target area obtained by statistics, so as to attract as many users as possible and improve the advertisement reach rate. However, in actual situations, the flow distribution of the area is changing dynamically in real time. Using such a method cannot make dynamic adjustments according to the real-time changes in the flow of people, resulting in the advertisement marketing in the target area often not achieving a better reach rate.

[0004] Therefore, how to provide an effective solution to improve the user reach rate in the advertisement marketing process has become an urgent problem to be solved in the existing technology. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent marketing method, system and program product based on artificial intelligence to solve the above problems existing in the prior art.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect, the present invention provides an intelligent marketing method based on artificial intelligence, including: Sampling the monitoring video of the target area in the current time period to obtain multiple sampled video frames; Extracting the face images in all the sampled video frames; Based on the color characteristics of the pixel points in each face image, determining the hair area in each face image, and based on the area ratio and distribution characteristics of the hair area, determining the gender of the person corresponding to each face image; Extracting the horizontal gradient and vertical gradient of each pixel point in each face image through the Sobel operator to obtain the gradient amplitude map corresponding to each face image; Performing binarization processing on the gradient amplitude map corresponding to each face image to obtain the binary gradient amplitude map corresponding to each face image; Performing morphological processing on each binary gradient amplitude map and then extracting the closed contour to obtain the facial contour area in each binary gradient amplitude map; The image area corresponding to the facial contour area in the corresponding binary gradient amplitude map in each face image is used as the input of the pre-trained age recognition model to obtain the age of the person corresponding to each face image; Based on the gender and age of the person corresponding to each face image, an advertising marketing strategy for the target area in the next time period is generated, so that marketing advertisements are placed in the target area in the next time period based on the advertising marketing strategy.

[0007] Based on the above disclosed content, the present invention samples the surveillance video of the target area in the current period to obtain multiple frames of sampled video frames; extracts face images in all sampled video frames; determines the hair area in each face image based on the color characteristics of the pixels in each face image, and determines the gender of the person corresponding to each face image based on the area proportion and distribution characteristics of the hair area; extracts the horizontal gradient and vertical gradient of each pixel in each face image through the Sobel operator to obtain the gradient amplitude map corresponding to each face image; binarizes the gradient amplitude map corresponding to each face image to obtain to the binary gradient amplitude map corresponding to each face image; perform morphological processing on each binary gradient amplitude map and then extract closed contours to obtain the facial contour area in each binary gradient amplitude map; use the image area in each face image corresponding to the facial contour area in the corresponding binary gradient amplitude map as the input of the pre-trained age recognition model to obtain the age of the person corresponding to each face image; based on the gender and age of the person corresponding to each face image, generate an advertising marketing strategy for the target area in the next time period, so as to place marketing advertisements in the target area in the next time period based on the advertising marketing strategy. In this way, the real-time gender and age distribution of the flow of people in the target area can be identified, and based on the real-time gender and age distribution of the flow of people in the target area, a targeted advertising marketing strategy can be generated, so as to attract more user attention and improve the advertising reach of the placed marketing advertisements.

[0008] In a possible design, face images in all sampled video frames are extracted, including: Face images in all sampled video frames are extracted based on the YOLO algorithm.

[0009] In a possible design, based on the color characteristics of the pixels in each face image, the hair area in each face image is determined, and based on the area ratio and distribution characteristics of the hair area, the gender of the person corresponding to each face image is determined, including: Convert each face image into HSV format; Based on the hue, saturation and brightness of the pixels in each face image converted into the HSV format, the hair area in each face image is determined; Statistically calculate the proportion of pixel points corresponding to the hair regions in each face image to obtain the area proportion of the hair regions in each face image; Calculate the center point distance between the central pixel point corresponding to the hair region in each face image and the central pixel point of the face image, the maximum distance between the pixel points corresponding to the hair region in the face image and the central pixel point of the face image, and the average distance between the pixel points corresponding to the hair region in the face image and the central pixel point of the face image; Based on the area proportion of the hair regions in each face image, the center point distances corresponding to each face image, the maximum distances corresponding to each face image, and the average distances corresponding to each face image, determine the genders of the persons corresponding to each face image.

[0010] In a possible design, extract the horizontal gradient and vertical gradient of each pixel point in each face image through the Sobel operator to obtain the gradient magnitude map corresponding to each face image, including: Convert each face image into a grayscale image; Calculate the horizontal gradient and vertical gradient of each pixel point in each face image according to the following formula; (1) where Sx represents the horizontal gradient of any pixel point, Sy represents the vertical gradient of any pixel point, a1 represents the grayscale value of the 1st pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a2 represents the grayscale value of the 2nd pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a3 represents the grayscale value of the 3rd pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a4 represents the grayscale value of the 4th pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a5 represents the grayscale value of the 5th pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a6 represents the grayscale value of the 6th pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a7 represents the grayscale value of the 7th pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner, a8 represents the grayscale value of the 8th pixel point in the 9 neighborhood pixel points of any pixel point in the clockwise direction starting from the upper left corner; Based on the horizontal gradient and vertical gradient of each pixel point in each face image, determine the gradient magnitude of each pixel point in each face image to obtain the gradient magnitude map corresponding to each face image.

[0011] In a possible design, perform morphological processing on each binary gradient magnitude map and then extract the closed contours to obtain the facial contour regions in each binary gradient magnitude map, including: After dilation and erosion operations are performed on each binary gradient amplitude map, closed contours are extracted using an edge tracking algorithm to obtain the facial contour area in each binary gradient amplitude map.

[0012] In a possible design, based on the gender and age of the person corresponding to each face image, an advertising marketing strategy for the target area in the next period of time is generated, including: Based on the gender and age of the person corresponding to each face image, determine the gender ratio and age distribution of the person; Generate advertising and marketing strategies for the target area in the next period based on the gender ratio and age distribution of the characters.

[0013] In a second aspect, the present invention provides an intelligent marketing system based on artificial intelligence, comprising: A sampling unit, used to sample the surveillance video of the target area in the current period to obtain multiple sampled video frames; A first extraction unit, used to extract face images from all sampled video frames; A determination unit, configured to determine the hair region in each face image based on the color characteristics of the pixels in each face image, and determine the gender of the person corresponding to each face image based on the area ratio and distribution characteristics of the hair region; The second extraction unit is used to extract the horizontal gradient and vertical gradient of each pixel in each face image by using the Sobel operator to obtain a gradient amplitude map corresponding to each face image; A binarization unit, used for binarizing the gradient amplitude map corresponding to each face image to obtain a binary gradient amplitude map corresponding to each face image; A third extraction unit is used to extract closed contours after morphological processing on each binary gradient amplitude map, so as to obtain a facial contour area in each binary gradient amplitude map; The age recognition unit is used to use the image area corresponding to the facial contour area in the corresponding binary gradient amplitude map in each face image as the input of the pre-trained age recognition model to obtain the age of the person corresponding to each face image; The generating unit is used to generate an advertising marketing strategy for the target area in the next time period based on the gender and age of the person corresponding to each face image, so as to place marketing advertisements in the target area in the next time period based on the advertising marketing strategy.

[0014] In a third aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, executes the artificial intelligence-based intelligent marketing method described in the first aspect or any possible design of the first aspect.

[0015] Fourthly, the present invention provides a computer program product comprising instructions which, when run on a computer, cause the computer to execute the artificial intelligence-based intelligent marketing method as described in the first aspect or any possible design of the first aspect.

[0016] Advantageous effects: The artificial intelligence-based intelligent marketing method, system and program product provided by the present invention can identify the real-time gender and age distribution of the people flow in the target area, and generate targeted advertising marketing strategies based on the real-time gender and age distribution of the people flow in the target area, so as to attract more user attention, improve the advertising reach rate of the placed marketing advertisements, and facilitate practical application and promotion. Description of the drawings

[0017] Figure 1 It is a flowchart of the artificial intelligence-based intelligent marketing method provided by the embodiment of the present application; Figure 2 It is a block diagram schematic of the artificial intelligence-based intelligent marketing system provided by the embodiment of the present application. Detailed implementation manners

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the present invention in combination with the drawings and the descriptions of the embodiments or the prior art. Obviously, the following descriptions of the structures of the drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. It should be noted here that the descriptions of these embodiment modes are used to help understand the present invention, but do not constitute a limitation to the present invention.

[0019] It should be understood that although terms such as first and second may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, the first unit can be called the second unit, and similarly the second unit can be called the first unit, without departing from the scope of the exemplary embodiments of the present invention.

[0020] It should be understood that for the term "and / or" that may appear herein, it is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B may mean: A exists alone, B exists alone, and A and B exist simultaneously; for the term " / and" that may appear herein, it is a description of another association object relationship, indicating that two relationships may exist. For example, A / and B may mean: A exists alone, and A and B exist alone; in addition, for the character " / " that may appear herein, generally it means that the front and rear associated objects are in an "or" relationship.

[0021] In order to improve the user reach rate in the advertising and marketing process, the embodiments of the present application provide an artificial intelligence-based intelligent marketing method, system, and program product. The artificial intelligence-based intelligent marketing method, system, and program product can generate targeted advertising marketing strategies based on the real-time gender and age distribution of the people flow in the target area, and improve the advertising reach rate of the placed marketing advertisements.

[0022] The artificial intelligence-based intelligent marketing method provided by the embodiments of the present application can be applied to advertising placement devices. It can be understood that the execution subject does not constitute a limitation to the embodiments of the present application.

[0023] As Figure 1 shown, it is a flowchart of the artificial intelligence-based intelligent marketing method provided by the first aspect of this embodiment. The artificial intelligence-based intelligent marketing method may but is not limited to include the following steps S101 to S108.

[0024] Step S101. Sample the monitoring video of the target area in the current time period to obtain multiple sampled video frames.

[0025] In one or more embodiments, a video acquisition device may be set in the target area (preferably in the direction of the entrance of the target area), and the monitoring video of the target area may be obtained through the video acquisition device. In the advertising and marketing process, the monitoring video may be divided by time period, and the monitoring video of the target area in the current time period may be sampled to obtain multiple sampled video frames.

[0026] Among them, the time period division interval and the sampling frequency may be set according to the actual situation. For example, a time period may be divided every 3 minutes, and the monitoring video may be sampled every 5 seconds.

[0027] Step S102. Extract the face images in all the sampled video frames.

[0028] In the embodiments of the present application, the face images in all the sampled video frames may be extracted by, but not limited to, the YOLO algorithm, RetinaFace (an advanced single-stage multi-task convolutional neural network face detection model), etc. The extracted face images include the face facial area and the hair area above the face facial area.

[0029] In one or more embodiments, after extracting the face images in all the sampled video frames, the extracted face images may be scaled to a specified size for subsequent analysis.

[0030] Step S103. Based on the color features of the pixel points in each face image, determine the hair area in each face image, and based on the area ratio and distribution characteristics of the hair area, determine the gender of the person corresponding to each face image.

[0031] The HSV format is more in line with human perception of colors and has stronger light robustness. Therefore, when determining the gender of the person corresponding to each face image, each face image can be first converted into the HSV format, and then based on the hue, saturation, and value of the pixel points in each face image converted into the HSV format, the hair region in each face image can be determined.

[0032] Specifically, the value ranges of hue, saturation, and value can be set according to several common typical hair colors such as black, red, gold, brown, etc., and based on the hue, saturation, and value of the pixel points in each face image converted into the HSV format, as well as the value ranges of hue, saturation, and value of various typical hair colors, the hair region in each face image can be determined. For the determined hair region, morphological processing can be used to remove the noise points therein.

[0033] After determining the hair region in each face image, the proportion of the pixel points corresponding to the hair region in each face image to the total pixel points of the face image can be statistically calculated to obtain the area ratio of the hair region in each face image. Then, calculate the center point distance between the center pixel point corresponding to the hair region in each face image and the center pixel point of the face image, the maximum distance between the pixel points corresponding to the hair region in the face image and the center pixel point of the face image, and the average distance between the pixel points corresponding to the hair region in the face image and the center pixel point of the face image.

[0034] Since there are significant differences in hair styles and lengths between men and women in general, there will also be significant differences in the distribution of the hair region in the face images between men and women. Therefore, the center point distance between the center pixel point corresponding to the hair region in the face images of men and women and the center pixel point of the face image, the maximum distance between the pixel points corresponding to the hair region in the face image and the center pixel point of the face image, and the average distance between the pixel points corresponding to the hair region in the face image and the center pixel point of the face image can be statistically calculated in advance through a large number of sample face images. And after calculating the center point distance between the center pixel point corresponding to the hair region in each face image and the center pixel point of the face image, the maximum distance between the pixel points corresponding to the hair region in the face image and the center pixel point of the face image, and the average distance between the pixel points corresponding to the hair region in the face image and the center pixel point of the face image, the gender of the person corresponding to each face image can be directly determined based on the area ratio of the hair region in each face image, the center point distance corresponding to each face image, the maximum distance corresponding to each face image, and the average distance corresponding to each face image.

[0035] Step S104. Extract the horizontal gradient and vertical gradient of each pixel in each face image through the Sobel operator to obtain the gradient magnitude map corresponding to each face image.

[0036] Specifically, each face image can be first converted into a grayscale image, and then the horizontal gradient and vertical gradient of each pixel in each face image are calculated according to the following formula; (1) where Sx represents the horizontal gradient of any pixel, Sy represents the vertical gradient of any pixel, a1 represents the grayscale value of the 1st pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a2 represents the grayscale value of the 2nd pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a3 represents the grayscale value of the 3rd pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a4 represents the grayscale value of the 4th pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a5 represents the grayscale value of the 5th pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a6 represents the grayscale value of the 6th pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a7 represents the grayscale value of the 7th pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner, a8 represents the grayscale value of the 8th pixel in the 9 neighboring pixels of any pixel in the clockwise direction starting from the upper left corner.

[0037] Then, the horizontal gradient Sx and vertical gradient Sy of each pixel in each face image are used to determine the gradient magnitude of each pixel in each face image, and the gradient magnitude map corresponding to each face image is obtained. Among them, the gradient magnitude of a pixel can be expressed as .

[0038] Step S105. Binarize the gradient magnitude map corresponding to each face image to obtain the binary gradient magnitude map corresponding to each face image.

[0039] Step S106. After morphological processing of each binary gradient magnitude map, perform closed contour extraction to obtain the facial contour region in each binary gradient magnitude map.

[0040] Specifically, after performing dilation and erosion operations on each binary gradient magnitude map, closed contour extraction is performed through an edge tracking algorithm (such as the Suzuki algorithm) to obtain the facial contour region in each binary gradient magnitude map.

[0041] Step S107. Using the image area in each face image corresponding to the facial contour area in the corresponding binary gradient amplitude map as the input of the pre-trained age recognition model to obtain the age of the person corresponding to each face image.

[0042] In the embodiment of the present application, an age recognition model for age recognition is pre-trained, and the age recognition model can be obtained by training with the facial contour area of the sample face image as the sample input and the user age (or age group) corresponding to the sample face image as the sample output. The age recognition model can be, but is not limited to, a deep expectation (DEX) model or an SSR-Net (Soft Stagewise Regression Network) model, etc., which is not specifically limited in the embodiment of the present application.

[0043] After obtaining the facial contour area in each binary gradient amplitude map, the image area in each face image corresponding to the facial contour area in the corresponding binary gradient amplitude map can be used as the input of the pre-trained age recognition model to obtain the person's age (or age range) corresponding to each face image.

[0044] Step S108. Generate an advertising marketing strategy for the target area in the next time period based on the gender and age of the person corresponding to each facial image, so as to place marketing advertisements in the target area in the next time period based on the advertising marketing strategy.

[0045] Specifically, based on the gender and age of the person corresponding to each face image, the gender ratio and age distribution of the person can be determined, and then based on the gender ratio and age distribution of the person, an advertising marketing strategy for the target area in the next time period can be generated, so that marketing advertisements can be placed in the target area in the next time period based on the advertising marketing strategy. The advertising marketing strategy includes the advertisement numbers of multiple advertisements placed in the next time period.

[0046] In summary, the artificial intelligence-based intelligent marketing method provided by the present invention can identify the real-time gender and age distribution of the flow of people in the target area, and generate targeted advertising marketing strategies based on the real-time gender and age distribution of the flow of people in the target area, thereby attracting more user attention, improving the advertising reach of the marketing advertisements, and facilitating practical application and promotion.

[0047] See also Figure 2 The second aspect of the embodiment of the present application provides an intelligent marketing system based on artificial intelligence, and the intelligent marketing system based on artificial intelligence includes: A sampling unit, used to sample the surveillance video of the target area in the current period to obtain multiple sampled video frames; A first extraction unit, used to extract face images from all sampled video frames; A determination unit, configured to determine the hair region in each face image based on the color characteristics of the pixels in each face image, and determine the gender of the person corresponding to each face image based on the area ratio and distribution characteristics of the hair region; The second extraction unit is used to extract the horizontal gradient and vertical gradient of each pixel in each face image by using the Sobel operator to obtain a gradient amplitude map corresponding to each face image; A binarization unit, used for binarizing the gradient amplitude map corresponding to each face image to obtain a binary gradient amplitude map corresponding to each face image; A third extraction unit is used to extract closed contours after morphological processing on each binary gradient amplitude map, so as to obtain a facial contour area in each binary gradient amplitude map; The age recognition unit is used to use the image area corresponding to the facial contour area in the corresponding binary gradient amplitude map in each face image as the input of the pre-trained age recognition model to obtain the age of the person corresponding to each face image; The generating unit is used to generate an advertising marketing strategy for the target area in the next time period based on the gender and age of the person corresponding to each face image, so as to place marketing advertisements in the target area in the next time period based on the advertising marketing strategy.

[0048] The working process, working details and technical effects of the artificial intelligence-based intelligent marketing system provided in the second aspect of this embodiment can be found in the first aspect of the embodiment and will not be repeated here.

[0049] A third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor and a transceiver that are communicatively connected in sequence, wherein the memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the artificial intelligence-based intelligent marketing method as described in the first aspect of the embodiment.

[0050] Specifically, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in first-out memory (FIFO), and / or first-in last-out memory (FILO), etc.; the processor is not limited to using a microprocessor of the STM32F105 series, a processor of architectures such as ARM (Advanced RISC Machines), X86, or a processor integrated with an NPU (neural-network processing units); the transceiver may include, but is not limited to, a WiFi (Wireless Fidelity) wireless transceiver, a Bluetooth wireless transceiver, a General Packet Radio Service (GPRS) wireless transceiver, a ZigBee (a low-power local area network protocol based on the IEEE 802.15.4 standard) wireless transceiver, a 3G transceiver, a 4G transceiver, and / or a 5G transceiver, etc.

[0051] In the fourth aspect of this embodiment, a computer-readable storage medium storing instructions for implementing the artificial intelligence-based intelligent marketing method described in the first aspect of the embodiment is provided, that is, instructions are stored on the computer-readable storage medium, and when the instructions run on a computer, they execute the artificial intelligence-based intelligent marketing method described in the first aspect. Among them, the computer-readable storage medium refers to a carrier for storing data, which may include, but is not limited to, floppy disks, optical discs, hard disks, flash memories, USB flash drives, and / or memory sticks, etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0052] In the fifth aspect of this embodiment, a computer program product containing instructions is provided, and when the instructions run on a computer, the computer is made to execute the artificial intelligence-based intelligent marketing method described in the first aspect of the embodiment. Among them, the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0053] It should be understood that specific details are provided in the following description to facilitate a complete understanding of the example embodiments. However, those of ordinary skill in the art should understand that the example embodiments can be implemented without these specific details. For example, a system may be shown in a block diagram to avoid obscuring the example with unnecessary details. In other instances, well-known processes, structures, and technologies may not be shown with unnecessary details to avoid obscuring the example embodiments.

[0054] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent marketing method based on artificial intelligence, characterized in that, include: Sampling the surveillance video of the target area in the current period to obtain multiple sampling video frames; Extracting face images from all sampled video frames; Based on the color characteristics of the pixels in each face image, the hair area in each face image is determined, and based on the area ratio and distribution characteristics of the hair area, the gender of the person corresponding to each face image is determined; The horizontal gradient and vertical gradient of each pixel in each face image are extracted by Sobel operator to obtain the gradient amplitude map corresponding to each face image; Binarizing the gradient amplitude map corresponding to each face image to obtain a binary gradient amplitude map corresponding to each face image; After morphological processing is performed on each binary gradient amplitude map, closed contours are extracted to obtain the facial contour area in each binary gradient amplitude map; The image area corresponding to the facial contour area in the corresponding binary gradient amplitude map in each face image is used as the input of the pre-trained age recognition model to obtain the age of the person corresponding to each face image; Based on the gender and age of the person corresponding to each face image, an advertising marketing strategy for the target area in the next time period is generated, so that marketing advertisements are placed in the target area in the next time period based on the advertising marketing strategy.

2. The intelligent marketing method based on artificial intelligence according to claim 1, wherein, Extract face images from all sampled video frames, including: Face images in all sampled video frames are extracted based on the YOLO algorithm.

3. The intelligent marketing method based on artificial intelligence according to claim 1, wherein Based on the color characteristics of the pixels in each face image, the hair area in each face image is determined, and based on the area ratio and distribution characteristics of the hair area, the gender of the person corresponding to each face image is determined, including: Convert each face image into HSV format; Based on the hue, saturation and brightness of the pixels in each face image converted into the HSV format, the hair area in each face image is determined; Counting the proportion of pixels corresponding to the hair area in each face image, to obtain the area ratio of the hair area in each face image; Calculate the center point distance between the center pixel point corresponding to the hair area in each face image and the center pixel point of the face image, the maximum distance between the pixel point corresponding to the hair area in the face image and the center pixel point of the face image, and the average distance between the pixel point corresponding to the hair area in the face image and the center pixel point of the face image; Based on the area ratio of the hair area in each face image, the center point distance corresponding to each face image, the maximum distance corresponding to each face image and the average distance corresponding to each face image, the gender of the person corresponding to each face image is determined.

4. The intelligent marketing method based on artificial intelligence according to claim 1, characterized in that The horizontal gradient and vertical gradient of each pixel in each face image are extracted by the Sobel operator to obtain the gradient amplitude map corresponding to each face image, including: Convert each face image into a grayscale image; The horizontal gradient and vertical gradient of each pixel in each face image are calculated according to the following formula; （1） Wherein, Sx represents the horizontal gradient of any pixel point, Sy represents the vertical gradient of any pixel point, a1 represents the gray value of the 1st pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a2 represents the gray value of the 2nd pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a3 represents the gray value of the 3rd pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a4 represents the gray value of the 4th pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a5 represents the gray value of the 5th pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a6 represents the gray value of the 6th pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a7 represents the gray value of the 7th pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point, a8 represents the gray value of the 8th pixel point in the clockwise direction starting from the upper left corner among the 9 neighboring pixel points of any pixel point; Based on the horizontal gradient and vertical gradient of each pixel point in each face image, determine the gradient magnitude of each pixel point in each face image, and obtain the gradient magnitude map corresponding to each face image.

5. The intelligent marketing method based on artificial intelligence according to claim 1, characterized in that, After performing morphological processing on each binary gradient magnitude map, perform closed contour extraction to obtain the facial contour region in each binary gradient magnitude map, including: After performing dilation and erosion operations on each binary gradient magnitude map, perform closed contour extraction through an edge tracking algorithm to obtain the facial contour region in each binary gradient magnitude map.

6. The intelligent marketing method based on artificial intelligence according to claim 1, wherein Based on the gender and age of the person corresponding to each face image, generate an advertising marketing strategy for the target area in the next time period, including: Based on the gender and age of the person corresponding to each face image, determine the gender ratio and age distribution of the people. Based on the gender ratio and age distribution of the people, generate an advertising marketing strategy for the target area in the next time period.

7. An intelligent marketing system based on artificial intelligence, characterized in that, Including: A sampling unit for sampling the surveillance video of the target area in the current time period to obtain multiple sampled video frames; A first extraction unit for extracting the face images in all the sampled video frames; A determination unit for determining the hair region in each face image based on the color characteristics of the pixel points in each face image, and determining the gender of the person corresponding to each face image based on the area ratio and distribution characteristics of the hair region; A second extraction unit for extracting the horizontal gradient and vertical gradient of each pixel point in each face image through a Sobel operator to obtain the gradient magnitude map corresponding to each face image; A binarization unit for binarizing the gradient magnitude map corresponding to each face image to obtain the binary gradient magnitude map corresponding to each face image; A third extraction unit for performing morphological processing on each binary gradient magnitude map and then performing closed contour extraction to obtain the facial contour region in each binary gradient magnitude map; An age recognition unit, configured to use the image regions corresponding to the facial contour regions in the corresponding binary gradient magnitude maps in each face image as the input of a pre-trained age recognition model, so as to obtain the ages of the persons corresponding to each face image; A generation unit, configured to generate an advertising marketing strategy for the target region in the next time period based on the gender and age of the person corresponding to each face image, so as to deliver marketing advertisements to the target region in the next time period based on the advertising marketing strategy.

8. A computer-readable storage medium, characterized in that, Instructions are stored on the computer-readable storage medium, and when the instructions are run on a computer, the artificial intelligence-based intelligent marketing method according to any one of claims 1 to 6 is executed.

9. A computer program product, comprising a computer program or instructions, characterized in that, The computer program or the instructions, when executed by a computer, implement the artificial intelligence-based intelligent marketing method according to any one of claims 1 to 6.

Images