KR20240120175A - Customized cosmetic recommendation server and method based on camera - Google Patents

Abstract

A camera-based customized cosmetics recommendation server and its recommendation method are disclosed. The camera-based customized cosmetics recommendation server according to the present invention includes a skin type storage unit that classifies and stores skin types according to items including acne, pores, spots, wrinkles, and dryness; an ingredient effect storage unit that classifies and stores ingredients and effects corresponding to cosmetic products; A facial image receiving unit that receives a facial image from a customer terminal; a skin type search unit that searches for a skin type corresponding to the received facial image among skin types stored in the skin type storage unit; A product image receiving unit that receives product images of cosmetics from a customer terminal; an ingredient effect search unit that searches for ingredients and effects corresponding to the received product image among the ingredients and effects stored in the ingredient effect storage unit; and a product recommendation unit that recommends cosmetic products to the customer terminal based on the skin type searched by the skin type search unit and the ingredients and effects searched by the ingredient effect search unit.

Description

Camera-based customized cosmetics recommendation server and recommendation method {CUSTOMIZED COSMETIC RECOMMENDATION SERVER AND METHOD BASED ON CAMERA}

The present invention relates to a customer-tailored cosmetics recommendation server and its recommendation method. More specifically, the present invention relates to a customer-tailored cosmetics recommendation server and its recommendation method. More specifically, the present invention relates to a customer's skin condition such as acne, pores, spots, wrinkles, dryness, etc., based on a camera, and to the customer's skin condition as determined. It is about a camera-based customized cosmetics recommendation server that recommends optimal cosmetic products to customers and its recommendation method.

Cosmetics are products intended to satisfy the aesthetic needs of users, and have been used historically for a long period of time. In particular, recently, various products have been developed to improve the skin health of users in response to environmental problems such as fine dust and ultraviolet rays, and are attracting attention not only to women but also to men.

Meanwhile, cosmetics contain different ingredients and contents depending on the type of product, and users select and use products that suit them according to their skin conditions such as acne, pores, blemishes, wrinkles, and skin dryness.

However, because users generally select products based solely on their visual condition, they may choose to use cosmetics that do not suit them. As a result, cosmetics that may be effective for some may have adverse effects for others. There is a problem that can be done.

Public Patent Publication No. 10-2022-0123851 (Publication Date: 2022.09.13)

The present invention was created to improve the above-mentioned problems. Based on the camera, the customer's skin condition such as acne, pores, spots, wrinkles, dryness, etc. is determined, and the optimal cosmetics are provided to the customer according to the judged skin condition. The purpose is to provide a camera-based customized cosmetics recommendation server that recommends products and a recommendation method.

In order to achieve the above-described object, the camera-based customized cosmetics recommendation server according to one aspect of the present invention includes a skin type storage unit that classifies and stores skin types according to items including acne, pores, freckles, wrinkles, and dryness. ; an ingredient effect storage unit that classifies and stores ingredients and effects corresponding to cosmetic products; A facial image receiving unit that receives a facial image from a customer terminal; a skin type search unit that searches for a skin type corresponding to the received facial image among skin types stored in the skin type storage unit; A product image receiving unit that receives product images of cosmetics from the customer terminal; an ingredient effect search unit that searches for ingredients and effects corresponding to the received product image among the ingredients and effects stored in the ingredient effect storage unit; and a product recommendation unit that recommends cosmetic products to the customer terminal based on the skin type searched by the skin type search unit and the ingredients and effects searched by the ingredient effect search unit. do.

The cosmetics recommendation server described above includes a review information storage unit that collects and stores review information corresponding to the customer's cosmetic products; and a first review information extraction unit that extracts review information corresponding to the searched skin type from the review information stored in the review information storage unit.

In addition, the cosmetics recommendation server described above includes a second review information extraction unit that extracts review information corresponding to the product video from the review information stored in the review information storage unit; and a review information comparison unit that compares the review information extracted by the first review information extraction unit and the review information extracted by the second review information extraction unit.

In the camera-based customized cosmetics recommendation method performed by a customized cosmetics recommendation server according to an aspect of the present invention to achieve the above-described object, items including acne, pores, blemishes, wrinkles, and dryness Classifying and storing skin types according to; Classifying and storing ingredients and effects corresponding to cosmetic products; Receiving a facial image from a customer terminal; Searching for a skin type corresponding to the received facial image among the stored skin types; Receiving a product image of a cosmetic product from the customer terminal; Searching for ingredients and effects corresponding to the received product image among the stored product ingredients; and recommending a cosmetic product to the customer terminal based on the searched skin type and the searched ingredients and effects.

The cosmetic recommendation method described above includes the steps of collecting and storing review information corresponding to a customer's cosmetic product; and extracting review information corresponding to the searched skin type from the stored review information.

In addition, the above-described cosmetic recommendation method may further include a step of extracting review information corresponding to the product image from the stored review information; and a step of comparing the review information corresponding to the skin type and the review information corresponding to the product image.

According to the present invention, the customer's skin condition such as acne, pores, spots, wrinkles, dryness, etc. can be determined based on the camera, and the optimal cosmetic product can be recommended to the customer according to the customer's skin condition.

Figure 1 is a diagram showing a system in which a camera-based customized cosmetics recommendation server is applied according to an embodiment of the present invention.
FIG. 2 is a diagram schematically showing the configuration of the customer-tailored cosmetics recommendation server shown in FIG. 1.
Figure 3 is a diagram showing an example of a facial image captured by a customer terminal.
Figure 4 is a diagram illustrating an example of determining the shape of the face, positions of eyes, nose, mouth, etc. in response to a facial image.
Figure 5 is a diagram illustrating an example of analyzing acne, pores, blemishes, wrinkles, dryness, etc. in response to a facial image.
Figure 6 is a diagram showing an example of receiving a product image from a customer terminal.
Figure 7 is a flowchart showing a camera-based customized cosmetics recommendation method according to an embodiment of the present invention.

Hereinafter, some embodiments of the present invention will be described through exemplary drawings. When assigning reference numerals to components in each drawing, identical components are indicated with the same reference numerals as much as possible, even if they are shown in different drawings. Additionally, when describing embodiments of the present invention, if detailed descriptions of related known configurations or functions are judged to impede understanding of the embodiments of the present invention, the detailed descriptions will be omitted.

Additionally, when describing the components of an embodiment of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are only used to distinguish the component from other components, and the nature, sequence, or order of the component is not limited by the term. When a component is described as being “connected,” “coupled,” or “connected” to another component, that component may be directly connected, coupled, or connected to that other component, but that component and that other component It should be understood that another component may be “connected,” “coupled,” or “connected” between elements.

Figure 1 is a diagram showing a system to which a customized cosmetics recommendation server is applied according to an embodiment of the present invention.

Referring to Figure 1, a camera-based customized cosmetics recommendation server (hereinafter also referred to as 'cosmetics recommendation server') 100 according to an embodiment of the present invention can be connected to the customer terminal 20 through the network 10. there is.

Here, the network 10 is defined as a broad concept that includes not only the Internet but also mobile communication networks such as CDMA (Code Division Multiple Access), WCDMA (Wideband CDMA), LTE (Long Term Evolution), and 5G ( ^5th Generation). .

Additionally, the customer terminal 20 is a terminal capable of network communication and includes not only computers but also mobile communication terminals such as laptops and smart phones.

FIG. 2 is a diagram schematically showing the configuration of the customer-tailored cosmetics recommendation server shown in FIG. 1.

Referring to Figure 2, the camera-based customized cosmetics recommendation server 100 according to an embodiment of the present invention includes a skin type storage unit 102, an ingredient effect storage unit 104, a facial image receiver 106, and a skin type storage unit 102. Search unit 108, product image receiving unit 110, ingredient effect search unit 112, product recommendation unit 114, review information storage unit 116, first review information extraction unit 118, second review information It may include an extraction unit 122 and a review information comparison unit 124.

The skin type storage unit 102 classifies and stores skin types according to items including acne, pores, blemishes, wrinkles, and dryness. At this time, the skin type storage unit 102 can classify and store skin types in various ways according to the range occupied by each item, such as acne, pores, spots, wrinkles, and dryness, within the scope of the face. For example, the skin type storage unit 102 determines whether the proportion of acne items within the range of the face is 0 to 20%, 20 to 40%, 40 to 60%, 60 to 80%, or 80 to 100%. Depending on your skin type, you can classify it into multiple levels and store them. In this case, the skin type storage unit 102 can classify and store a variety of skin types based on the number of items such as acne, pores, freckles, wrinkles, dryness, etc., and the stage at which each item is classified. For example, as in the above-mentioned example, if the number of each item is 5 and each item is classified into 5 levels, the skin type storage unit 102 can classify and store 5x5x5x5x5=3,125 skin types. You can. However, the number of classified skin types described here is only an example to help understand the invention, and the skin type storage unit 102 can classify and store various numbers of skin types.

The ingredient effect storage unit 104 classifies and stores ingredients and effects corresponding to cosmetic products. At this time, the ingredient effect storage unit 104 stores cosmetic products including the manufacturer, product name, brand, etc. of cosmetics sold in the market through the network 10, the ingredients contained in the cosmetic product, and the corresponding cosmetic products. Big data such as product efficacy and effects can be collected and stored periodically. In this case, the ingredient effect storage unit 104 registers websites corresponding to each shopping mall, cosmetics manufacturer, etc., and can collect cosmetic products, ingredients and effects contained in the cosmetic products, etc. from the registered websites. .

The facial image receiving unit 106 receives facial images from the customer terminal 20 through the network 10. At this time, the customer terminal 20 is installing the application provided from the cosmetics recommendation server 100 according to an embodiment of the present invention, and using the installed application, as shown in FIG. 3, the camera of the customer terminal 20 ( (not shown) can be used to capture facial images.

The skin type search unit 108 searches for the skin type corresponding to the facial image received from the customer terminal 20 among the skin types stored in the skin type storage unit 102. At this time, the skin type search unit 108 analyzes the characteristic points of the facial image received from the customer terminal 20, as shown in FIG. 4, and selects a face corresponding to the customer's facial image based on the analyzed specific point. It is possible to determine the proportions and positions of the face, eyes, nose, mouth, eyebrows, etc. Additionally, the skin type search unit 108 may analyze the spatial ratio of the face based on the determined face shape and the proportions and positions of the eyes, nose, mouth, eyebrows, etc. on the face. Known facial recognition technology can be used to determine the facial shape, facial proportions and positions of eyes, nose, mouth, eyebrows, etc., and spatial ratio on the face from facial images, and detailed descriptions thereof are omitted here. do. In addition, the skin type search unit 108 determines the skin condition for each item such as acne, pores, freckles, wrinkles, and dryness, as shown in FIG. 5 with respect to the facial image received from the customer terminal 20. can do. At this time, the skin type search unit 108 determines the ratio of each item such as acne, pores, blemishes, wrinkles, dryness, etc. in the face corresponding to the facial image, and synthesizes the ratios for each judged item. The skin condition of the facial image can be determined. In addition, when facial images are sequentially received from the customer terminal 20 at temporal intervals, the skin type search unit 108 provides information corresponding to the customer terminal 20 based on the skin condition of the sequentially received facial images. You can analyze the rate of improvement or change in the customer's skin condition.

The product video receiver 110 receives the product video of the cosmetics from the customer terminal 20. At this time, the product image receiving unit 110 can receive product identification information including the barcode and identification code of the cosmetics from the customer terminal 20 as a product image, as shown in FIG. In addition, the product image receiving unit 110 preferably receives a product image of a cosmetic product currently being used by the customer from the customer terminal 20.

The ingredient effect search unit 112 searches for ingredients and effects corresponding to the product image received from the customer terminal 20 among the ingredients and effects stored in the ingredient effect storage unit 104. At this time, the ingredient effect search unit 112 searches for a cosmetic product that matches the product image received from the customer terminal 20 among the various cosmetic products stored in the ingredient effect storage unit 104 and the corresponding classified ingredients and effects. , you can determine the ingredients and effects corresponding to the searched cosmetic product.

The product recommendation unit 114 provides cosmetic products to the customer terminal 20 based on the skin type searched by the skin type search unit 108 and the ingredients and effects searched by the ingredient effect search unit 112. I recommend you. At this time, the product recommendation unit 114 analyzes the skin improvement method needed by the customer in response to the skin type searched by the skin type search unit 108, and provides information to the customer terminal 20 based on the analyzed skin improvement method. Can recommend cosmetic products. For example, if it is determined by the skin type search unit 108 that acne accounts for 60 to 80% of the face corresponding to the facial image and freckles accounts for 20 to 40%, the product recommendation unit (114) is the most effective in improving acne and can recommend cosmetic products that contain ingredients that are also effective in improving freckles. In addition, the product recommendation unit 114 selects cosmetic products corresponding to the product image received by the product image receiver 110 among the cosmetic products recommended corresponding to the skin type searched by the skin type search unit 108. Cosmetic products can be recommended based on the ingredients that are lacking. For example, it is determined by the skin type search unit 108 that acne accounts for 60-80% of the face corresponding to the facial image and that freckles account for 20-40%, and the product image receiver 110 ), if the cosmetic product corresponding to the product image received only has the effect of improving acne and does not have the effect of improving freckles, the product recommendation unit 114 provides ingredients that are effective in improving freckles along with the cosmetic product currently used by the customer. It can be recommended to the customer terminal 20 to add and use cosmetic products containing.

Through this, the cosmetics recommendation server 100 according to an embodiment of the present invention recommends the most suitable cosmetic product to the customer, while considering the ingredients of the cosmetic product currently being used by the customer to ensure that the ingredients and effects do not overlap. You will be able to recommend cosmetic products.

The review information storage unit 116 collects and stores review information corresponding to the customer's cosmetic products. At this time, the review information storage unit 114 corresponds to each cosmetic product stored in the ingredient effect storage unit 104, and provides big data of customer review information about the corresponding cosmetic product from various cosmetics sales sites through the network 10. can be collected and stored.

The first review information extraction unit 118 extracts review information corresponding to the skin type searched by the skin type search unit 108 from the review information stored in the review information storage unit 116. At this time, the first review information extraction unit 118 has a skin type that is the same or similar to the skin type searched by the skin type search unit 108 among the review information about various cosmetic products stored in the review information storage unit 116. You can extract customer review information about cosmetic products. In this case, the review information extracted by the first review information extraction unit 118 is a cosmetic product corresponding to the product image received by the product image receiving unit 110, that is, the customer corresponding to the customer terminal 20 is currently using the product. It may include not only review information about the cosmetic product in question, but also review information about various other cosmetic products.

The second review information extraction unit 120 extracts review information corresponding to the product video received by the product video receiving unit 110 from the review information stored in the review information storage unit 116. At this time, the second review information extraction unit 120 selects the same cosmetic product as the cosmetic product in the product video received by the product video receiving unit 110 among the review information about various cosmetic products stored in the review information storage unit 116. Review information can be extracted. In this case, the review information extracted by the second review information extraction unit 120 is not only the review information of customers with the same or similar skin type as the skin type searched by the skin type search unit 108, but also the review information of the skin type search unit 108. It may also include review information of customers with a skin type different from the skin type searched by (108).

The review information comparison unit 122 compares the review information extracted by the first review information extraction unit 118 and the review information extracted by the second review information extraction unit 120. At this time, the review information comparison unit 122 provides review information on various cosmetic products of customers with the same or similar skin type as the skin type searched by the skin type search unit 108, and the second review information extraction unit 120 ) Among the review information extracted, it is desirable to compare the review information of customers with the same or similar skin type to the skin type searched by the skin type search unit 108. That is, the review information comparison unit 122 includes review information of various cosmetic products used by customers with the same or similar skin type as the customer corresponding to the customer terminal 20, and the current use of the customer corresponding to the customer terminal 20. Compare review information from customers who have the same or similar skin type as the customer for the cosmetic product you are using. At this time, the review information comparison unit 122 provides review information about the cosmetics currently used by the customer corresponding to the customer terminal 20, and customers who have the same or similar skin type as the customer corresponding to the customer terminal 20. Among the review information of various cosmetic products used, it is desirable to compare the top review information with a rating greater than a set value and a large number of reviews within the set order.

The product recommendation unit 114 recommends the other cosmetic product to the customer terminal 20 when the review information of another cosmetic product is higher than the set rating compared to the cosmetic product currently used by the customer corresponding to the customer terminal 20. can do.

Through this, the cosmetics recommendation server 100 according to an embodiment of the present invention recommends cosmetic products in consideration of various customer review information about various cosmetic products in addition to the cosmetic products currently used by the customer, thereby providing optimal customized cosmetics to the customer. can be recommended.

Meanwhile, when a cosmetic product different from the cosmetic product the customer is currently using is recommended through the product recommendation unit 114, the customer terminal 20 takes a product video after using the recommended cosmetic product for a certain period of time and recommends the cosmetic product. It can be transmitted to the server 100. In this case, the cosmetics recommendation server 100 uses a deep learning algorithm to perform the same process as described above, improving skin condition during the period the customer used the cosmetic product recommended by the customer terminal 20, The customer terminal 20 learns changes in review information about recommended cosmetic products, and through this, the customer terminal 20 can perform verification through feedback after using the recommended cosmetic product for a certain period of time. In addition, the cosmetics recommendation server 100 compares the customer's skin condition for the previous product video with the customer's skin condition for the product video of the recommended cosmetic product, and compares the customer's skin condition with the recommended cosmetic product for the previous product video. You can also analyze the improved improvement rate and perform verification of the recommended product based on the analyzed improvement rate.

Figure 7 is a flowchart showing a camera-based customized cosmetics recommendation method according to an embodiment of the present invention. The camera-based customized cosmetics recommendation method according to an embodiment of the present invention can be performed by the cosmetics recommendation server 100 shown in FIG. 1.

Referring to Figures 1 to 7, the cosmetics recommendation server 100 classifies and stores skin types according to items including acne, pores, blemishes, wrinkles, and dryness (S101). At this time, the cosmetics recommendation server 100 can classify and store skin types in various ways according to the range occupied by each item such as acne, pores, spots, wrinkles, dryness, etc. within the scope of the face. For example, the cosmetics recommendation server 100 determines whether the proportion of acne items within the face range is 0 to 20%, 20 to 40%, 40 to 60%, 60 to 80%, or 80 to 100%. Skin types can be classified into multiple levels and stored.

The cosmetics recommendation server 100 classifies and stores the ingredients and effects corresponding to the cosmetic products (S103). At this time, the cosmetics recommendation server 100 provides cosmetic products including the manufacturer, product name, and brand of the cosmetics sold in the market through the network 10, the ingredients contained in the cosmetic product, and the product. Big data such as efficacy and effects can be collected and stored.

The cosmetics recommendation server 100 collects and stores review information corresponding to the customer's cosmetics product (S105). At this time, the cosmetics recommendation server 100 may collect and store big data of customers' review information about the corresponding cosmetic product from various cosmetics sales sites through the network 10 in response to each stored cosmetic product.

The cosmetics recommendation server 100 receives a facial image from the customer terminal 20 through the network 10 (S107). At this time, the customer terminal 20 is installing the application provided from the cosmetics recommendation server 100 according to an embodiment of the present invention, and facial images are captured through a camera (not shown) of the customer terminal 20 using the installed application. can be filmed.

The cosmetics recommendation server 100 searches for the skin type corresponding to the facial image received from the customer terminal 20 among the stored skin types (S109). At this time, the cosmetics recommendation server 100 analyzes feature points for the facial image received from the customer terminal 20, and based on the analyzed specific points, the facial shape, eyes, nose, mouth, and face corresponding to the customer's facial image. The position of eyebrows, etc. can be determined. A method of determining the shape of the face and the positions of eyes, nose, mouth, eyebrows, etc. from a facial image can use known facial recognition technology, and detailed description thereof will be omitted here.

In addition, the cosmetics recommendation server 100 can determine the skin condition for each item such as acne, pores, blemishes, wrinkles, and dryness with respect to the facial image received from the customer terminal 20. At this time, the cosmetics recommendation server 100 determines the proportion of each item such as acne, pores, freckles, wrinkles, dryness, etc. in the face corresponding to the facial image, and synthesizes the proportions of each judged item to determine the proportion of each item. The skin condition can be judged on facial images.

The cosmetics recommendation server 100 extracts review information corresponding to the searched skin type from the review information (S111). At this time, the cosmetic recommendation server 100 may extract review information about cosmetic products of customers who have the same or similar skin type as the searched skin type from among the stored review information about various cosmetic products. In this case, the extracted review information may include not only review information about the cosmetic product currently being used by the customer corresponding to the customer terminal 20, but also review information about various other cosmetic products.

The cosmetics recommendation server 100 receives the product video of the cosmetics from the customer terminal 20 (S113). At this time, the cosmetics recommendation server 100 may receive product identification information including the barcode and identification code of the cosmetics from the customer terminal 20 in the form of a product image. In addition, the cosmetics recommendation server 100 preferably receives a product image of the cosmetic product currently being used by the customer from the customer terminal 20.

The cosmetics recommendation server 100 searches for ingredients and effects corresponding to the product image received from the customer terminal 20 among the stored ingredients and effects (S115). At this time, the cosmetics recommendation server 100 searches for a cosmetic product that matches the product image received from the customer terminal 20 among the various stored cosmetic products and the ingredients and effects classified correspondingly, and the ingredients corresponding to the searched cosmetic product. and effectiveness can be determined.

The cosmetics recommendation server 100 extracts review information corresponding to the received product video from the stored review information (S117). At this time, the cosmetics recommendation server 100 may extract review information about the same cosmetic product as the cosmetic product in the received product video from among the review information about various stored cosmetic products. In this case, the extracted review information may include not only review information of customers with the same or similar skin type as the searched skin type, but also review information of customers with a skin type different from the searched skin type.

The cosmetics recommendation server 100 compares the primarily extracted review information and the secondarily extracted review information (S119). At this time, the cosmetics recommendation server 100 provides review information on various cosmetic products from customers with the same or similar skin type as the searched skin type, and skin types that are the same or similar to the searched skin type among the secondarily extracted review information. It is desirable to compare review information of customers with . In other words, the cosmetics recommendation server 100 provides review information on various cosmetic products used by customers with the same or similar skin type as the customer corresponding to the customer terminal 20, and information currently used by the customer corresponding to the customer terminal 20. Compare review information from customers with the same or similar skin type for existing cosmetic products.

The cosmetics recommendation server 100 recommends cosmetic products to the customer terminal 20 based on the searched skin type and the searched ingredients and effects (S121). At this time, the cosmetics recommendation server 100 may analyze the skin improvement method needed by the customer in response to the searched skin type, and recommend a cosmetic product to the customer terminal 20 based on the analyzed skin improvement method. For example, if it is determined that acne accounts for 60 to 80% of the face corresponding to the facial image and freckles accounts for 20 to 40%, the cosmetics recommendation server 100 determines whether the effect of improving acne is effective. It is the highest, and cosmetic products containing ingredients that have the effect of improving freckles can be recommended.

In addition, the cosmetics recommendation server 100 recommends cosmetics products based on ingredients lacking in the cosmetics product corresponding to the product video received from the customer terminal 20 among the cosmetics products recommended in response to the searched skin type. can do. For example, it is judged that acne accounts for 60 to 80% of the face and blemishes account for 20 to 40% of the face corresponding to the facial image, and cosmetics corresponding to the product image received from the customer terminal 20 If the product only has the effect of improving acne and does not have the effect of improving freckles, the cosmetics recommendation server 100 recommends that the customer use a cosmetic product containing ingredients effective in improving freckles along with the cosmetic product the customer is currently using. It can be recommended to the terminal 20.

Through this, the cosmetics recommendation server 100 according to an embodiment of the present invention recommends the most suitable cosmetic product to the customer, while considering the ingredients of the cosmetic product currently being used by the customer to ensure that the ingredients and effects do not overlap. You will be able to recommend cosmetic products.

In addition, if the review information of another cosmetic product is more positive than the set value compared to the cosmetic product currently used by the customer corresponding to the customer terminal 20, the cosmetics recommendation server 100 recommends the other cosmetic product to the customer terminal. (20) can be recommended.

Through this, the cosmetics recommendation server 100 according to an embodiment of the present invention recommends cosmetic products in consideration of various customer review information about various cosmetic products in addition to the cosmetic products currently used by the customer, thereby providing optimal customized cosmetics to the customer. can be recommended.

Meanwhile, when a cosmetic product different from the cosmetic product the customer is currently using is recommended through the product recommendation unit 114, the customer terminal 20 takes a product video after using the recommended cosmetic product for a certain period of time and recommends the cosmetic product. It can be transmitted to the server 100. In this case, the cosmetics recommendation server 100 uses a deep learning algorithm to perform the same process as described above, improving skin condition during the period the customer used the cosmetic product recommended by the customer terminal 20, The customer terminal 20 learns changes in review information for recommended cosmetic products, and through this, the customer terminal 20 can perform verification through feedback after using the recommended cosmetic product for a certain period of time ( S123). In addition, the cosmetics recommendation server 100 compares the customer's skin condition for the previous product video with the customer's skin condition for the product video of the recommended cosmetic product, and compares the customer's skin condition with the recommended cosmetic product for the previous product video. You can also analyze the improved improvement rate and perform verification of the recommended product based on the analyzed improvement rate.

Although embodiments according to the present invention have been described above, they are merely illustrative, and those skilled in the art will understand that various modifications and equivalent scope of embodiments are possible therefrom. Accordingly, the scope of protection of the present invention should be determined not only by the following claims but also by their equivalents.

Claims (6)

a skin type storage unit that classifies and stores skin types according to items including acne, pores, freckles, wrinkles, and dryness;
an ingredient effect storage unit that classifies and stores ingredients and effects corresponding to cosmetic products;
A facial image receiving unit that receives a facial image from a customer terminal;
a skin type search unit that searches for a skin type corresponding to the received facial image among skin types stored in the skin type storage unit;
A product image receiving unit that receives product images of cosmetics from the customer terminal;
an ingredient effect search unit that searches for ingredients and effects corresponding to the received product image among the ingredients and effects stored in the ingredient effect storage unit; and
a product recommendation unit that recommends cosmetic products to the customer terminal based on the skin type searched by the skin type search unit and the ingredients and effects searched by the ingredient effect search unit;
A camera-based customized cosmetics recommendation server comprising: According to paragraph 1,
a review information storage unit that collects and stores review information corresponding to the customer's cosmetic products; and
a first review information extraction unit that extracts review information corresponding to the searched skin type from the review information stored in the review information storage unit;
A camera-based customized cosmetics recommendation server further comprising: According to paragraph 2,
a second review information extraction unit that extracts review information corresponding to the product video from the review information stored in the review information storage unit; and
a review information comparison unit that compares the review information extracted by the first review information extraction unit and the review information extracted by the second review information extraction unit;
A camera-based customized cosmetics recommendation server further comprising: In the camera-based method of providing customized cosmetics information performed by a customer-tailored cosmetics information provision server,
Classifying and storing skin types according to items including acne, pores, blemishes, wrinkles, and dryness;
Classifying and storing ingredients and effects corresponding to cosmetic products;
Receiving a facial image from a customer terminal;
Searching for a skin type corresponding to the received facial image among the stored skin types;
Receiving a product image of a cosmetic product from the customer terminal;
Searching for ingredients and effects corresponding to the received product image among the stored product ingredients; and
Recommending a cosmetic product to the customer terminal based on the searched skin type and the searched ingredients and effects;
A customer-tailored cosmetics recommendation method comprising: According to paragraph 4,
Collecting and storing review information corresponding to the customer's cosmetic products; and
Extracting review information corresponding to the searched skin type from the stored review information;
A customer-tailored cosmetics recommendation method further comprising: According to clause 5,
extracting review information corresponding to the product video from the stored review information; and
Comparing review information corresponding to the skin type and review information corresponding to the product video;
A customer-tailored cosmetics recommendation method further comprising: