KR102310565B1 - Clothing recommendation device and method - Google Patents

Abstract

Provided is a clothing recommendation device, which receives clothing information and location information from a user, collects weather information which matches location information, extracts first color information appearing as a first color area from the clothing information, classifies the first color information according to tone information to create a clothing information group for each color in which the clothing information is grouped, and extracts the clothing information from any one color-specific clothing information group matching the weather information to generate recommended information.

Description

Apparatus and method for recommending clothes {CLOTHING RECOMMENDATION DEVICE AND METHOD}

The present invention relates to an apparatus and method for recommending clothes, and more particularly, to an apparatus and method for recommending clothes suitable for a user.

With the recent development of the Internet, the number of Internet users who purchase goods using e-commerce is increasing. As such, as the Internet shopping mall develops as an important means of commerce, the number of Internet shopping malls handling clothes is also increasing, and the number of products handled in the Internet shopping mall is also increasing.

Currently, clothes are displayed and sold in online shopping malls by simply classifying clothes by brand or type, and providing image information and detailed descriptions (product code, brand name, color, material, manufacturer, product description, etc.) for each costume. ) is exposed on an Internet site such as a Web site, and uniform and passive methods are being used to wait for the Internet user's selection. In addition, a method of exposing clothing information mainly on clothing products purchased by many people is also used, but this refers to the collective purchase record and does not reflect each individual's specialized taste.

SUMMARY OF THE INVENTION The technical problem to be solved by the present invention is to provide a clothing recommendation apparatus and method for analyzing weather information at an arbitrary location and recommending clothes suitable for an arbitrary weather.

One aspect of the present invention, the input unit for receiving an input of clothes information and location information from a user; a collecting unit for collecting weather information matching the location information from an external server; Collecting the clothing information, extracting first color information represented by a first color gamut from the clothing information, and classifying the first color information according to tone information set to classify the first color information into a first color series and a second color series an analysis unit for generating a group of clothes information by color in which the clothes information is grouped; a clothing recommendation unit for generating recommendation information by extracting clothing information from a clothing information group for each color matching the weather information; and an output unit for outputting the recommendation information.

In addition, when the temperature indicated from the wake-up information is lower than a preset reference temperature, the clothes recommendation unit extracts the clothes information from the clothes information group for each color of the first color series to generate the recommended information, When the temperature indicated by the information is higher than the reference temperature, the recommendation information may be generated by extracting the clothing information from the clothing information group for each color of the second color series.

In addition, the input unit further receives destination information from the user, the collecting unit further collects event information matching the destination information, and the analyzing unit, a second color appearing as a second color area from the clothing information. extracting information and classifying the second color information according to the tone information set to classify the third color series, the fourth color series, and the fifth color series to create a clothing information group for each color in which the clothing information is grouped, , the clothing recommendation unit may generate the recommended information by extracting clothing information from a clothing information group for each color matching the event information.

In addition, the clothing recommendation unit, when the temperature indicated by the weather information is higher than a preset reference temperature and the event information indicates an outdoor event, the clothing information group for each color of the second color series and the fourth color series The recommended information may be generated by extracting clothing information that is duplicated in the clothing information group for each color of .

In addition, when a plurality of event information collected according to a plurality of destination information input at different viewpoints appears as a daily event and a night event, the clothing recommendation unit may provide clothing information from the clothing information group for each color of the third color series. extraction, and extracting clothing information representing outerwear from the clothing information group for each color of the fifth color series to generate the recommendation information.

In addition, the collecting unit further collects closed circuit television image information from an external server, the analyzing unit analyzes the closed circuit television image information to generate actual weather information, and the clothes recommendation unit is configured to interact with the actual weather information By comparing the weather information at different points in time, predicted weather information may be generated, and clothing information matching the predicted weather information may be extracted to generate recommended information.

Another aspect of the present invention provides a method for recommending clothes in a clothes recommendation device, the method comprising: receiving, by an input unit, clothes information and location information from a user; collecting, by a collecting unit, weather information matching the location information from an external server; The analysis unit collects the clothes information, extracts first color information appearing in the first color gamut from the clothes information, and classifies the first color information into a first color series and a second color series. creating a clothing information group for each color in which the clothing information is grouped according to the classification; generating, by a clothing recommendation unit, clothing information by extracting clothing information from a clothing information group for each color matching the weather information; and outputting the recommendation information by an output unit.

In addition, the step of generating the recommended information by extracting the clothing information may include extracting clothing information from the clothing information group for each color of the first color series when the temperature indicated from the weather information is lower than a preset reference temperature. The recommendation information may be generated, and when the temperature indicated by the weather information is higher than the reference temperature, the recommendation information may be generated by extracting the clothing information from the clothing information group for each color of the second color series.

In addition, the input unit further receiving destination information from the user; The step of further collecting, by the collection unit, event information matching the destination information, wherein the step of generating the clothing information group by color in which the clothing information is grouped includes a second color area from the clothing information. 2 color information is extracted, and the second color information is classified according to the tone information set to classify the third color series, the fourth color series, and the fifth color series to form a clothing information group by color in which the clothing information is grouped. The step of generating and extracting the clothing information to generate the recommended information may include extracting the clothing information from the clothing information group for each color matching the event information to generate the recommended information.

In addition, the step of generating the recommended information by extracting the clothing information may include: when a temperature indicated from the weather information is higher than a preset reference temperature and the event information indicates an outdoor event, each color of the second color series The recommendation information may be generated by extracting the clothes information overlapping the clothes information group and the clothes information group for each color of the fourth color series.

In addition, the step of generating the recommended information by extracting the clothing information may include: when a plurality of event information collected according to a plurality of destination information input at different viewpoints appears as a daily event and a nighttime event, the third color series The recommendation information may be generated by extracting clothing information from the clothing information group for each color and extracting clothing information representing outerwear from the clothing information group for each color of the fifth color series.

In addition, the collecting unit further collecting the closed circuit television image information from the external server; The analyzing unit analyzes the closed circuit television image information to generate actual weather information; further comprising, wherein the step of extracting the clothing information and generating the recommended information may include: the actual weather information at a different point in time By comparing the weather information, predicted weather information may be generated, and clothing information matching the predicted weather information may be extracted to generate recommended information.

According to the above-described aspect of the present invention, by providing an apparatus and method for recommending clothes, it is possible to analyze weather information at an arbitrary location and recommend clothes suitable for a certain weather.

1 is a schematic diagram of a clothes recommendation system including a clothes recommendation device according to an embodiment of the present invention.
2 and 3 are control block diagrams of an apparatus for recommending clothes according to an embodiment of the present invention.
4 is a block diagram illustrating a process of generating a clothing information group for each color in the analysis unit of FIG. 2 .
5 is a block diagram illustrating a process of generating recommendation information in the clothes recommendation unit of FIG. 2 .
6 is a block diagram illustrating a process of generating preference information in the analysis unit of FIG. 2 .
7 is a flowchart of a method for recommending clothes according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0023] Reference is made to the accompanying drawings, which show by way of illustration specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different but need not be mutually exclusive. For example, certain shapes, structures, and characteristics described herein with respect to one embodiment may be implemented in other embodiments without departing from the spirit and scope of the invention. In addition, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the present invention. Accordingly, the detailed description set forth below is not intended to be taken in a limiting sense, and the scope of the invention, if properly described, is limited only by the appended claims, along with all scope equivalents to those claimed. Like reference numerals in the drawings refer to the same or similar functions throughout the various aspects.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the drawings.

1 is a schematic diagram of a clothes recommendation system including a clothes recommendation device according to an embodiment of the present invention.

The clothes recommendation system 1 may be provided so that the clothes recommendation device 100 receives arbitrary information from the external server 200 through a wired or wireless network. 200 ) may extract clothes information matching the information transmitted, and the clothes recommendation apparatus 100 may generate recommendation information according to the extracted clothes information.

Here, the external server 200 may be a server device provided in an institution that performs meteorological measurement or forecasting, such as a meteorological agency, a weather forecasting institution, a meteorological measurement institution, etc. In this case, the external server 200 may be any Meteorological information measured at a location or predicted at any point in time at any location may be stored.

In addition, the external server 200 may be a server device connected to a closed circuit television (CCTV: Closed-Circuit Television), in this case, the external server 200 is a closed circuit generated by a closed circuit television installed at an arbitrary location. Circuit television image information may be stored.

In addition, the external server 200 may be a server device provided in any institution, company, or store, and the external server 200 is a blog in which information generated by any institution, company, or store is posted. , a cafe (Cafι), may be a server device provided in organizations and companies that provide services such as web pages.

In this case, the external server 200 may store event information indicating events such as events, lectures, parties, and performances conducted in any institution, company, or store. Accordingly, the event information may include information such as the type, location, start time, end time, dress code, and the like of an event held at any institution, company, or store.

On the other hand, the clothes information may be provided to indicate arbitrary clothes, and for this purpose, the clothes information may include color information indicating the color of the clothes, type information indicating the type of clothes, material information indicating the material of the clothes, etc. .

Here, the color information may include first color information in the first color gamut and second color information in the second color gamut according to a method of representing the color of clothes.

In this case, the first color gamut is CMYK (Cyan Magenta), which is provided so that an arbitrary color appears according to elements of blue (or cyan, cyan), red (or magenta, magenta), yellow (yellow), and black (black). Yellow and Black) region, and the second color region is a Hue, Saturation, Value (HSV) region provided to display an arbitrary color according to elements of hue, saturation, and value. can

In addition, the type information may be provided to indicate the type of clothing such as short sleeves, outerwear, shirt, shorts, sweater, earrings, and ring, and the material information indicates the material of clothing such as cotton, hemp, wool, denim, cashmere, etc. catalog may be prepared.

Accordingly, the recommendation information may be provided to indicate clothes recommended for the user to wear. To this end, the recommended information may be presented as a combination of one or more pieces of clothing information, for example, the recommended information may be presented as a combination of shorts and short sleeves, or the recommended information may be presented as a combination of shorts, a shirt, and a belt. .

In this case, the recommendation information may be generated in a shape in which an arbitrary character provided in three-dimensional graphics or a picture wears clothes according to the clothes information, and the recommendation information is located in a position matching the type of clothes according to the clothes information. It may be generated to represent the shape of the corresponding costume. For example, the recommendation information may be generated such that the short-sleeved top appears at the position and the shorts appear at the position of the bottom.

On the other hand, the clothes recommendation device 100 may be provided in the shape of a mirror. In this case, the clothes recommendation device 100 reflects the user's appearance through the mirror, but recommends generated by the clothes recommendation device 100 . Information may be provided to be output to one side of the mirror.

Also, the clothes recommendation device 100 may be provided so that the recommendation information generated by the clothes recommendation device 100 is output to the center side of the mirror. By matching the body position, it is effective to know in advance the approximate style according to the recommended information.

In this regard, the clothes recommendation apparatus 100 may be provided with a photographing device such as a camera on one side of the mirror to photograph a user using the clothing recommendation apparatus 100 . The user's location may be detected from the image information generated by the method, and the recommendation information may be output to an area matching the detected location of the user among areas set on the mirror of the clothes recommendation apparatus 100 .

2 and 3 are control block diagrams of an apparatus for recommending clothes according to an embodiment of the present invention.

The clothes recommendation apparatus 100 includes an input unit 110 , a collection unit 120 , an analysis unit 130 , a clothes recommendation unit 140 , an output unit 150 , a storage unit 160 , and a photographing unit 170 . can do.

In addition, the clothes recommendation apparatus 100 may be implemented by more components than the components shown in FIG. 2 , and may be implemented by fewer components than the components shown in FIG. 2 . Alternatively, in the clothes recommendation device 100 , at least two components provided in the clothes recommendation device 100 may be integrated into one component, and one component may perform a complex function. Hereinafter, the above-described components will be described in detail.

The input unit 110 may receive clothes information and location information from a user.

In this case, the input unit 110 may receive a clothes recommendation request command from the user, and in this regard, the clothes recommendation request command may be replaced with location information that the input unit 110 receives from the user. In this case, it may be understood that the clothes recommendation request command is generated when the input unit 110 receives location information from the user.

In this case, the input unit 110 may receive information such as a product name of clothes and a product code from the user. In this case, the collection unit 120 collects clothes information based on the information input from the external server 200 . can do. Accordingly, the storage unit 160 may store the collected clothing information.

Also, the input unit 110 may receive information such as an address as location information.

Meanwhile, the input unit 110 may receive voice information uttered from the user, and in this case, the analysis unit 130 analyzes the input voice information through a technique such as syntax analysis to determine the information input from the user. can do.

The input unit 110 may further receive destination information from the user. In this regard, the input unit 110 may further receive event information from the user.

In this regard, the clothes recommendation request command may be replaced with destination information that the input unit 110 receives from the user. In this case, it may be understood that the clothes recommendation request command is generated when the input unit 110 receives destination information from the user.

Here, the destination information may be location information such as an address indicating an arbitrary location, which may indicate a point the user wants to visit.

In this case, the input unit 110 may receive a plurality of destination information from the user, and in this case, the input unit 110 may further receive the order of the plurality of destination information from the user.

In this case, the order of the plurality of destination information may be set according to the order in which the input unit 110 receives the plurality of destination information from the user.

The collection unit 120 may collect weather information matching the location information from an external server. In this case, the collection unit 120 may collect the weather information at the point in time when the clothes recommendation request command is generated.

Here, the weather information may include information such as weather and temperature of the address according to the location information, and in this case, the collecting unit 120 may collect a plurality of weather information prepared for a preset period range. .

For example, the preset period range may be provided to collect meteorological information from the external server 200 to 23:59 of the corresponding date from the time when the meteorological information is collected.

Meanwhile, the collection unit 120 may further collect closed circuit television image information from the external server 200 . In this case, the collection unit 120 may collect closed circuit television image information at the time when the clothing recommendation request command is generated.

In addition, the collection unit 120 may further collect event information matching the destination information from the external server 200 . In this case, the collection unit 120 may collect event information at the point in time when the clothing recommendation request command is generated.

In this regard, the collection unit 120 may collect event information input from the input unit 110 when event information is input through the input unit 110 .

The analysis unit 130 may collect clothing information and extract first color information represented by the first color area from the clothing information, and the analysis unit 130 may convert the first color information into the first color series and the second color information. By classifying according to the tone information set to be classified into series, it is possible to create a clothing information group for each color in which the clothing information is grouped.

In this regard, the analysis unit 130 includes a clothing information collection module 131 , a first color information extraction module 132 , a first color-specific clothing information classification module 133 , a second color information extraction module 134 , It may include a second color-specific clothing information classification module 135 , a weather information analysis module 136 , and a preference information processing module 137 .

Accordingly, the clothing information collection module 131 may collect clothing information, and the first color information extraction module 132 may extract first color information displayed in the first color region from the clothing information.

In addition, the first color information classification module 133 classifies the first color information according to the tone information set to classify the first color information into the first color series and the second color series to form a clothing information group by color in which the clothing information is grouped. can create

Here, the first color series may mean a warm series, and the second color series may mean a cool series. In this case, the warm series may mean a set of colors set to give a psychologically warm feeling, and the cool series may mean a set of colors set to give a psychologically cold feeling.

Accordingly, the tone information set so that the first color series and the second color series are classified may be determined by a ratio in which a yellow element appears in the first color information.

For example, the first color-specific clothing information classification module 133 selects the corresponding clothing information when the ratio of the appearance of the yellow element among the first color information for any clothing information is within a preset threshold range. It can be classified into a group of a first color series, and the first color-specific clothing information classification module 133 determines that the ratio of the appearance of the yellow element among the first color information for any clothing information is outside a preset threshold range. In this case, the corresponding clothing information may be classified into a group of the second color series.

In this case, according to the tone information, the ratio of the yellow element in the first color information may be set in a range of 30% or more and less than 70%. As such, the tone information is the first color area For each element provided in the CMYK area, it may be set to classify the first color information into a first color series and a second color series according to a ratio in which each element appears from the first color information.

In another embodiment, in the tone information, the ratio of the yellow element among the first color information is within the first range of 40% or more and less than 70%, and the black element is 30% or more and less than 50% of the second color information. When it is within the range, it may be set to be classified into the first color series. In this case, the first color-specific clothing information classification module 133 is configured to, when the ratio of the appearance of the yellow or black element among the first color information for any clothing information is out of a preset threshold range, the corresponding clothing The information may be classified into groups of the second color series.

In addition, the analysis unit 130 may extract second color information that appears as a second color area from the collected clothing information, and the analysis unit 130 converts the second color information into a third color series, a fourth color series and By classifying according to the tone information set to be classified into the fifth color series, a clothing information group for each color in which the clothing information is grouped may be generated.

In this case, the second color information extraction module 134 may extract second color information that appears in the second color area from the clothing information, and the second color information classification module 135 uses the second color information as the third color information. By classifying according to the tone information set to be classified into the color series, the fourth color series, and the fifth color series, a clothing information group for each color in which the clothing information is grouped may be generated.

Here, the third color series may mean a bright series, the fourth color series may mean a light series, and the fifth color series may mean a mute series. In this case, the bright series may mean a set of colors set to give a relatively bright and intense feeling, and the light series may mean a set of colors set to give a relatively bright, soft, and soft feeling. A series may mean a set of colors that are set to give a relatively dark and murky feeling.

Accordingly, the tone information set so that the third color series, the fourth color series, and the fifth color series are classified may be determined by a ratio in which elements of saturation and value appear among the second color information.

For example, the second color-specific clothing information classification module 135 may remove the corresponding clothing information when the ratio of the appearance of the element of saturation among the second color information for any clothing information exceeds a preset threshold. It can be classified into groups of three color series, and the second color information classification module 135 determines that the element of saturation among the second color information for arbitrary clothing information is less than or equal to a preset threshold, and the second color information When the element of lightness exceeds a preset threshold, the corresponding clothing information may be classified into a group of a fourth color series, and the second color-specific clothing information classification module 135 provides information on arbitrary clothing information. When the element of saturation in the second color information is less than or equal to a preset threshold and the element of brightness in the second color information is less than or equal to the preset threshold, the corresponding clothing information may be classified into a group of a fifth color series. .

Accordingly, the second color-specific clothing information classification module 135 is configured to determine if the ratio of the appearance of the element of saturation among the second color information for the arbitrary clothing information exceeds the threshold and the ratio of the appearance of the element of brightness exceeds the threshold. In this case, the corresponding clothing information may be classified into a group of a third color series, and the second color-specific clothing information classification module 135 sets a threshold value for the ratio of the element of saturation among the second color information for arbitrary clothing information. When the ratio exceeds the threshold value and the ratio of the element of brightness is less than or equal to the threshold, the corresponding clothing information may be classified into a group of the third color series.

At this time, in one embodiment, the threshold preset for the tone information may be set to 40%, and in this regard, the threshold preset for the tone information may be set the same for the saturation and the brightness. , or, a threshold set in advance for the tone information may be set differently for the chroma and the brightness.

In addition, the threshold set in advance for the tone information may further include a threshold set for the color element among the second color information.

In this regard, the arbitrary clothing information includes a clothing information group for each color according to at least one of the first color series and the second color series, and at least one of the third color series, the fourth color series, and the fifth color series. It may be set to be classified as being duplicated in the clothing information group for each color according to the series.

Accordingly, the storage unit 160 may store clothes information according to the generated clothes information group for each color.

The analysis unit 130 may analyze the closed circuit television image information to generate actual weather information. In this case, the weather information analysis module 136 may analyze the closed circuit television image information to generate actual weather information.

To this end, the weather information analysis module 136 may recognize an object whose position changes by comparing the closed circuit television image information captured at an adjacent viewpoint, and the weather information analysis module 136 determines the shape of the object whose position changes. It is possible to generate actual weather information according to the

At this time, the weather information analysis module 136 may determine the shape of the object based on the weather learning model provided to determine the shape of the object, for example, the shape of the object is a person wearing an umbrella, a folded umbrella It can mean the person holding it, etc.

To this end, the weather information analysis module 136 may generate a weather learning model by using an artificial intelligence (AI) technique such as machine learning.

Here, machine learning can be understood as a technique of generating a learning model to classify a plurality of information into one or more groups based on a plurality of pieces of information, and classifying arbitrary information based on the generated learning model, and In the same way, machine learning is supervised learning that creates a learning model to classify arbitrary information according to a plurality of information classified by an administrator, analyzes a plurality of information itself, or performs a clustering process and performs a learning model Unsupervised learning that generates It may include reinforcement learning (Reinforcement Learning) that generates a learning model according to the

Accordingly, the weather information analysis module 136 may generate actual weather information matching the determined shape of the object. For example, the actual weather information may be generated to indicate a raining state when the shape of the object is a person wearing an umbrella, and the actual weather information can be generated when the shape of the object is a person holding a folded umbrella, Rain may be generated to indicate an expected condition.

The clothes recommendation unit 140 may generate recommendation information by extracting clothes information from any one clothes information group for each color matching the weather information.

For example, when the temperature indicated from the weather information is lower than a preset reference temperature, the clothing recommendation unit 140 may extract the clothing information from the clothing information group for each color of the first color series to generate the recommended information. Also, when the temperature indicated from the weather information is higher than a preset reference temperature, the clothing recommendation unit 140 may extract the clothing information from the clothing information group for each color of the second color series to generate the recommended information.

In this case, the preset reference temperature may be set to indicate a temperature at which a person feels cold, for example, the preset reference temperature may be set to 10 degrees.

In this regard, the clothes recommendation unit 140 may generate recommendation information by extracting information about clothes matching the type of weather information. For example, when the weather information indicates rain, the clothes recommendation unit 140 may , by extracting clothing information representing an umbrella or a raincoat, and the like, to generate recommended information.

In addition, when the temperature indicated from the weather information is lower than the preset reference temperature and the weather information indicates that it is raining, the clothing recommendation unit 140 extracts clothing information from the clothing information group for each color of the first color series, Recommendation information may be generated, and in this case, the clothing recommendation unit ( ) may generate the recommendation information by extracting clothing information representing an umbrella or raincoat provided in the clothing information group for each color of the first color series.

In addition, when the temperature indicated from the weather information is higher than the preset reference temperature and the weather information indicates that it is raining, the clothing recommendation unit 140 extracts clothing information from the clothing information group for each color of the second color series, Recommendation information may be generated, and in this case, the clothing recommendation unit ( ) may generate the recommendation information by extracting clothing information representing an umbrella or raincoat provided in the clothing information group for each color of the second color series.

Meanwhile, the clothes recommendation unit 140 may generate recommendation information by extracting clothes information from a clothes information group for each color matching the event information.

For example, when the event information indicates a lecture, the clothing recommendation unit 140 may extract the clothing information from the clothing information group for each color of the fifth color series to generate the recommended information.

In this regard, when a plurality of event information is collected according to a plurality of destination information input at different viewpoints, the clothing recommendation unit 140 extracts clothing information that overlaps with the plurality of event information and provides the recommended information. can create

In addition, when a plurality of event information is collected according to a plurality of destination information input at different viewpoints, the clothing recommendation unit 140 may provide information on clothes that can be worn on different parts of the clothes information group for each color of a different series. Recommendation information can be generated by extracting it.

For example, when a plurality of event information collected according to a plurality of destination information input at different viewpoints appears as a daily event and a nighttime event, the clothes recommendation unit 140 may provide a third color group of clothes information for each color. Recommendation information may be generated by extracting clothing information from the .

In this case, the daily event may mean an event representing daily life such as commuting and work according to destination information indicated by a company, etc., and the night event may mean an event representing a party according to destination information such as a club.

To this end, the clothes recommendation unit 140 may extract clothes information based on a style learning model that is provided to extract one or more pieces of clothes information matching an arbitrary situation.

To this end, the clothes recommendation unit 140 may generate a style learning model by using an artificial intelligence (AI) technique such as machine learning.

Here, machine learning can be understood as a technique of generating a learning model to classify a plurality of information into one or more groups based on a plurality of pieces of information, and classifying arbitrary information based on the generated learning model, and In the same way, machine learning is supervised learning that creates a learning model to classify arbitrary information according to a plurality of information classified by an administrator, analyzes a plurality of information itself, or performs a clustering process and performs a learning model Unsupervised learning that generates It may include reinforcement learning (Reinforcement Learning) that generates a learning model according to the

Meanwhile, the clothes recommendation unit 140 may generate recommended information by extracting clothes information matching the weather information and the event information. In this case, the clothes recommendation unit 140 may generate the recommended information by extracting the clothes information provided redundantly in the clothes information group for each color according to the weather information and the clothes information group for each color according to the event information.

For example, when the temperature indicated by the weather information is higher than a preset reference temperature and the event information indicates an outdoor event, the clothing recommendation unit 140 may include a clothing information group for each color of the second color series and a fourth color. Recommendation information may be generated by extracting clothing information overlapping in the clothing information group for each color of the series.

Here, the outdoor event may mean a seasonal event held outdoors, such as a park.

In addition, when the temperature indicated from the weather information is higher than a preset reference temperature and the event information indicates an indoor event, the clothing recommendation unit 140 is configured to select the clothing information group for each color of the second color series and the fifth color series. Recommendation information may be generated by extracting clothing information overlapping in the clothing information group for each color.

Here, the indoor event may refer to events such as exhibitions and lectures held in exhibition halls and lecture rooms.

Meanwhile, the clothes recommendation unit 140 may generate predicted weather information by comparing the actual weather information with the weather information at different points in time, and the clothes recommendation unit 140 extracts clothes information matching the predicted weather information. Thus, the recommendation information can be generated.

At this time, it can be understood that the predicted weather information is generated by predicting the weather according to the transition of the weather information at different points in time with the actual weather information, for example, the predicted weather information indicates that the weather information represents clear weather, When the actual weather information appears to be currently raining, it may be generated to indicate the raining weather.

In addition, the predicted weather information may be generated to indicate clear weather when the weather information indicates that it is raining, the actual weather information indicates that it is clear, and the weather information after the current time indicates clear weather.

Accordingly, the clothes recommendation unit 140 may generate recommended information by extracting clothes information matching the expected weather information.

The output unit 150 may output recommendation information. To this end, the output unit 150 may output the recommended information using a display device such as a monitor, and in this case, the output unit 150 may be an arbitrary character provided as a three-dimensional graphic or picture wears clothes. The recommendation information may be output in the shape of the corresponding clothes, or the output unit 150 may output the recommendation information to indicate the shape of the clothes at a position matching the type of clothes according to the clothes information.

Meanwhile, the photographing unit 170 may generate image information by photographing the user's clothes. In this case, the photographing unit 170 may be provided to photograph a pre-set area, and accordingly, image information may photograph a user located in the pre-set area.

Accordingly, the storage unit 160 may store the image information generated by the photographing unit 170 , and in this case, the storage unit 160 stores weather information at a time point at which the image information is captured and at a time point at which the image information is captured. and information such as event information may be further stored.

In this case, the input unit 110 may receive a selection command for any recommended information at the time the image information is generated. In this case, the storage unit 160 may store the image information by matching it with the recommended information selected according to the selection command.

Accordingly, the analysis unit 130 may generate preference information according to the number of image information matching any recommendation information, and in this case, the preference information may be generated to indicate clothes preferred by the user.

Here, the preference information processing module 137 may generate preference information according to the number of image information matching any recommendation information, and in this case, the preference information may be generated to indicate clothes preferred by the user.

In this case, the preference information processing module 137 may generate purchase information based on a purchase learning model that is provided to indicate a pattern of clothing information appearing from the preference information.

Here, the purchase information may be generated to indicate the color information of the clothing information group for each color identical to the clothing information displayed from the preference information, but may indicate clothing information different from the clothing information stored in the storage unit 160 .

In this case, the purchase information may be generated to further include a shopping mall, a website, a brand, and the like, from which arbitrary clothing information can be purchased.

To this end, the preference information processing module 137 may generate a purchase learning model by using an artificial intelligence (AI) technique such as reinforcement learning.

In this case, in the process of generating the recommendation information, the clothes recommendation unit 140 may generate the recommendation information by preferentially extracting the clothes information preferred by the user according to the preference information.

4 is a block diagram illustrating a process of generating a clothing information group for each color in the analysis unit of FIG. 2 .

Referring to FIG. 4 , the input unit 110 may receive clothes information and location information from a user. In this case, the input unit 110 may receive information such as a product name of clothes and a product code from the user. In this case, the collection unit 120 collects clothes information based on the information input from the external server 200 . can do.

The analysis unit 130 may collect clothing information and extract first color information represented by the first color area from the clothing information, and the analysis unit 130 may convert the first color information into the first color series and the second color information. By classifying according to the tone information set to be classified into series, it is possible to create a clothing information group for each color in which the clothing information is grouped.

In addition, the analysis unit 130 may extract second color information that appears as a second color area from the collected clothing information, and the analysis unit 130 converts the second color information into a third color series, a fourth color series and By classifying according to the tone information set to be classified into the fifth color series, a clothing information group for each color in which the clothing information is grouped may be generated.

Accordingly, the storage unit 160 may store clothes information according to the generated clothes information group for each color.

5 is a block diagram illustrating a process of generating recommendation information in the clothes recommendation unit of FIG. 2 .

Referring to FIG. 5 , the input unit 110 may receive clothes information and location information from the user, and the input unit 110 may further receive destination information from the user.

Accordingly, the collecting unit 120 may collect weather information matching the location information from the external server, and the collecting unit 120 may further collect event information matching the destination information from the external server 200 . .

The clothes recommendation unit 140 may generate recommendation information by extracting clothes information from any one clothes information group for each color matching the weather information.

For example, when the temperature indicated from the weather information is lower than the preset reference temperature, the clothing recommendation unit 140 may extract the clothing information from the clothing information group for each color of the first color series to generate the recommended information. Also, when the temperature indicated by the weather information is higher than a preset reference temperature, the clothing recommendation unit 140 may extract the clothing information from the clothing information group for each color of the second color series to generate the recommended information.

Meanwhile, the clothes recommendation unit 140 may generate recommendation information by extracting clothes information from a clothes information group for each color matching the event information.

For example, when the event information indicates a lecture, the clothing recommendation unit 140 may extract the clothing information from the clothing information group for each color of the fifth color series to generate the recommended information.

Accordingly, the output unit 150 may output the recommendation information.

6 is a block diagram illustrating a process of generating preference information in the analysis unit of FIG. 2 .

Referring to FIG. 6 , the photographing unit 170 may generate image information by photographing a user's clothes. Accordingly, the storage unit 160 may store the image information generated by the photographing unit 170 , and in this case, the storage unit 160 stores weather information at a time point at which the image information is captured and at a time point at which the image information is captured. and information such as event information may be further stored.

In this case, the input unit 110 may receive a selection command for any recommended information at the time the image information is generated. In this case, the storage unit 160 may store the image information by matching it with the recommended information selected according to the selection command.

Accordingly, the analysis unit 130 may generate preference information according to the number of image information matching any recommendation information, and in this case, the preference information may be generated to indicate clothes preferred by the user.

In this case, the analysis unit 130 may generate purchase information based on a purchase learning model that is provided to indicate a pattern of clothing information appearing from the preference information.

7 is a flowchart of a method for recommending clothes according to an embodiment of the present invention.

Since the clothes recommendation method according to an embodiment of the present invention proceeds in substantially the same configuration as the clothes recommendation device 100 shown in FIG. 1 , the same reference numerals are used for the same components as the clothes recommendation device 100 of FIG. 1 . , and repeated descriptions will be omitted.

The clothes recommendation method includes the steps of receiving input of clothes information and location information (600), collecting weather information matching the location information (610), creating a group of clothes information by color in which the clothes information is grouped (620), It may include extracting the clothing information to generate recommended information ( 630 ) and outputting the recommended information ( 640 ).

The step 600 of receiving clothes information and location information may be a step in which the input unit 110 receives clothes information and location information from a user.

The step 610 of collecting the weather information matching the location information may be a step in which the collecting unit 120 collects the weather information matching the location information from the external server 200 .

In the step 620 of generating the clothing information group by color in which the clothing information is grouped, the analysis unit 130 collects the clothing information, extracts the first color information appearing in the first color area from the clothing information, and the first color This may be a step of classifying information according to tone information set to classify the information into the first color series and the second color series to generate a clothing information group for each color in which the clothing information is grouped.

The step 630 of extracting the clothes information to generate the recommended information may be a step in which the clothes recommendation unit 140 extracts the clothes information from the clothes information group for each color matching the weather information to generate the recommended information. .

The step 640 of outputting the recommendation information may be a step in which the output unit 150 outputs the recommendation information.

Although the above has been described with reference to the embodiments, those skilled in the art will understand that various modifications and changes can be made to the present invention without departing from the spirit and scope of the present invention as set forth in the following claims. will be able

1: Clothing recommendation system
100: clothing recommendation device
200: external server

Claims (12)

an input unit for receiving clothes information and location information from a user and receiving destination information from the user;
a collecting unit for collecting weather information matching the location information from an external server and collecting event information matching the destination information;
Collecting the clothes information, extracting first color information from the clothes information according to a first color gamut that is one of preset color models, and extracting the first color information according to preset tone information in the first color gamut an analysis unit for classifying the clothing information into a first color series or a second color series and grouping the clothing information into clothing information groups for each color;
a clothing recommendation unit for generating recommended information by extracting clothing information from a clothing information group for each color matching the weather information and the event information;
a photographing unit provided on one side of the mirror to photograph the clothes of the user to generate image information;
a storage unit for storing the image information generated by the photographing unit; and
An output unit for outputting the recommendation information to an area matching the location of the user detected from the image information;
The input unit receives a selection command for the recommendation information from the user,
The storage unit stores the image information by matching the recommended information selected according to the user's selection command,
The collecting unit further collects closed circuit television image information from an external server,
The analysis unit recognizes an object whose position changes from the closed circuit television image information, generates actual weather information according to the shape of the object whose position changes, and uses machine learning to determine the shape of the object, and Generates preference information indicating clothes preferred by the user according to the number of image information matching the recommendation information, and generates purchase information based on a purchase learning model provided to indicate a pattern of clothing information appearing from the preference information, , wherein the purchase information represents color information of a clothing information group for each color identical to the clothing information displayed from the preference information, and is generated to indicate clothing information different from the clothing information included in the image information stored in the storage unit, The information is generated to include shopping malls, websites and brands where the clothing information can be purchased,
The clothing recommendation section,
When a plurality of event information is collected according to a plurality of destination information input at different viewpoints, a plurality of different color series matching each event information is extracted, and a plurality of different colors according to each color series are extracted. A plurality of pieces of clothing information are extracted from the clothing information group to generate recommended information, and information on clothes that can be worn on different parts is extracted from each color clothing information group,
The actual weather information is compared with the weather information at different time points to generate predicted weather information, and extracting clothing information matching the predicted weather information to generate recommended information, wherein the user prefers according to the preference information. A clothing recommendation device that preferentially extracts clothing information.
According to claim 1, wherein the clothes recommendation unit,
When the temperature indicated from the weather information is lower than a preset reference temperature, the recommended information is generated by extracting the clothing information from the clothing information group for each color of the first color series, and the temperature indicated from the weather information is the reference temperature. When the temperature is higher than the temperature, the clothing recommendation apparatus generates the recommendation information by extracting the clothing information from the clothing information group for each color of the second color series.
According to claim 1,
The input unit further receives destination information from the user,
The collection unit further collects event information matching the destination information,
The analysis unit may be configured to extract second color information from the clothing information according to a second color gamut that is different from the first color gamut among preset color models, and to extract the second color information from the second color gamut according to preset tone information in the second color gamut. Classifying the second color information into a third color series, a fourth color series, and a fifth color series, and grouping the clothing information into a clothing information group for each color, wherein the tone information preset in the second color region is in the second color region It is set to classify the second color information according to the ratio in which any element in
The clothing recommendation section,
A clothing recommendation device for generating the recommended information by extracting clothing information from a clothing information group for each color matching the event information.
According to claim 3, wherein the clothes recommendation unit,
When the temperature indicated from the weather information is higher than the preset reference temperature and the event information indicates an outdoor event, the second color series clothes information group and the fourth color series clothes information group A clothing recommendation apparatus for generating the recommended information by extracting overlapping clothing information.
According to claim 3, wherein the clothes recommendation unit,
When a plurality of event information collected according to a plurality of destination information input at different viewpoints appears as a daily event and a night event, the clothes information is extracted from the clothes information group for each color of the third color series, and the fifth color A clothing recommendation device for generating the recommendation information by extracting clothing information representing outerwear from a clothing information group for each color of a series.
delete In the clothes recommendation method in the clothes recommendation device,
receiving, by an input unit, clothing information and location information from a user, and receiving destination information from the user;
collecting, by a collecting unit, weather information matching the location information from an external server, and collecting event information matching the destination information;
The analysis unit collects the clothes information, extracts first color information from the clothes information according to a first color gamut that is one of preset color models, and the first color information according to preset tone information in the first color gamut classifying the information into a first color series or a second color series and grouping the clothing information into clothing information groups for each color;
generating, by a clothing recommendation unit, clothing information by extracting clothing information from a clothing information group for each color matching the weather information and the event information;
generating, by a photographing unit, image information of the user's clothes;
storing, by a storage unit, image information generated from the photographing unit;
outputting, by an output unit, the recommendation information to an area matching the location of the user detected from the image information;
receiving, by the input unit, a selection command for the recommendation information from the user at a time point when the image information is generated;
further collecting, by the collecting unit, closed circuit television image information from an external server;
The analysis unit recognizes the object whose position changes from the closed circuit television image information, and generates actual weather information according to the shape of the object whose position changes, but determines the shape of the object using machine learning, and the recommendation Generates preference information indicating clothes preferred by the user according to the number of image information matching the information, and generates purchase information based on a purchase learning model provided to indicate a pattern of clothes information appearing from the preference information, The purchase information is generated to indicate color information of a clothing information group for each color identical to the clothing information displayed from the preference information, and indicates clothing information different from the clothing information included in the image information stored in the storage unit, and the purchase information generating the clothing information to include purchaseable shopping malls, websites, and brands; further comprising,
The step of generating the recommended information by extracting the clothing information,
When a plurality of event information is collected according to a plurality of destination information input at different viewpoints, a plurality of different color series matching each event information is extracted, and a plurality of different colors according to each color series are extracted. A plurality of pieces of clothing information are extracted from the clothing information group to generate recommended information, and information on clothes that can be worn on different parts is extracted from each color clothing information group,
The actual weather information is compared with the weather information at different time points to generate predicted weather information, and clothes information matching the predicted weather information is extracted, but clothes information preferred by the user is prioritized according to the preference information. A clothing recommendation method that extracts and generates recommended information.
The method of claim 7, wherein the step of extracting the clothing information to generate the recommended information,
When the temperature indicated from the weather information is lower than a preset reference temperature, the recommended information is generated by extracting the clothing information from the clothing information group for each color of the first color series, and the temperature indicated from the weather information is the reference temperature. When the temperature is higher than the temperature, the method for recommending clothes is to extract the clothes information from the clothes information group for each color of the second color series to generate the recommendation information.
8. The method of claim 7,
receiving, by the input unit, further input of destination information from the user;
Further comprising; further collecting, by the collecting unit, event information matching the destination information;
The generating of the clothing information group by color in which the clothing information is grouped may include: extracting second color information from the clothing information according to a second color region that is different from the first color region among preset color models; Classifying the second color information into a third color series, a fourth color series, and a fifth color series according to preset tone information in the second color area to group the clothes information into clothes information groups for each color, The preset tone information in the color gamut is set to classify the second color information according to a ratio of an arbitrary element in the second color gamut,
The step of generating the recommended information by extracting the clothing information,
A clothing recommendation method for generating the recommended information by extracting clothing information from a clothing information group for each color matching the event information.
The method of claim 9, wherein the step of extracting the clothing information to generate the recommended information,
When the temperature indicated from the weather information is higher than the preset reference temperature and the event information indicates an outdoor event, the second color series clothes information group and the fourth color series clothes information group A clothing recommendation method for generating the recommended information by extracting overlapping clothing information.
The method of claim 9, wherein the step of extracting the clothing information to generate the recommended information,
When a plurality of event information collected according to a plurality of destination information input at different viewpoints appears as a daily event and a night event, the clothes information is extracted from the clothes information group for each color of the third color series, and the fifth color A clothing recommendation method for generating the recommended information by extracting clothing information representing outerwear from a clothing information group for each color of a series.
delete

Images