KR102425057B1 - System for online shopping using neural networks - Google Patents

Abstract

An online shopping system according to one embodiment of the present invention includes: a user terminal running an Internet web browser; a first server accessible through the Internet web browser; a second server connected to the first server; and a third server connected to the first server and the second server. According to embodiments, clothes purchased by a user may be repaired to suit the body size of the user.

Description

Online shopping system using a neural network {SYSTEM FOR ONLINE SHOPPING USING NEURAL NETWORKS}

Embodiments of the present invention relate to a technology for providing clothes that fit a user's body during online shopping.

With the generalization of the Internet and the development of information technology (IT), people's consumption patterns are also changing from offline to online. For example, in the past, there was no way to purchase groceries other than purchasing groceries in a traditional market, but recently it has become possible to purchase groceries through an online shopping mall, and the size of the market is also increasing. In addition to groceries, various household items, clothing, and furniture can also be purchased online, and there are almost no items that cannot be purchased online.

Among items that can be purchased online, in the case of clothing, the size notation method is different for each brand, and even if the notation method is the same, the size may be different depending on the characteristics of the brand and product. Therefore, it may be difficult for consumers to determine whether a corresponding product fits their body only with the size listed in the shopping mall. If the size does not fit after purchasing the product, the consumer has to return the product, and the seller has to resend the product, which can be cumbersome.

In addition, consumers are often unaware of their body size accurately. For example, the number of consumers who are aware of their leg length, arm length, shoulder width, etc. is extremely small, and even if they are aware of their body size, in the case of clothing, they can know the exact size only by wearing the product.

The technical problems to be achieved in the embodiments are not limited to those mentioned above, and other technical problems not mentioned may be considered by those of ordinary skill in the art from various embodiments to be described below. can

An online shopping system according to an embodiment of the present invention includes a user terminal running an Internet web browser, a first server accessible through the Internet web browser, a second server connected to the first server, and a third server connected to the first server and the second server, wherein the first server includes a user ID, a password corresponding to the ID, the user's name, the user's address, and the user's nickname storing at least one of the above, and if the ID and password input through the user terminal match the stored ID and password, the user terminal is allowed to access, and a screen including a plurality of clothes photos is output, and the plurality of When any one of the clothes photos of When an image including the entire body is received through the user terminal, and the price of the selected clothing is paid, the second server includes the first user data, the second user data, and the third user data. receive a first uniform resource locator (URL) from the first server, obtain the third user data from the first URL, receive the image from the first server, Squared Difference; SSD) extracts the first object by connecting the section equal to or greater than a specified value, divides the first object into pixel units, and includes in the first area of the first object based on the angle at which the image is captured The gap between pixels to be used is enlarged, pixels around the enlarged gap are merged and inserted into the gap, and the gap between pixels included in the second area of the first object is reduced or the number of pixels included in the second area is reduced. at least a portion of the first Calculating the size of the object, calculating a repair completion date based on the number of repair details received within a specified period and the third user data, and transmitting the calculated repair completion date to the user terminal, and the third server receives a second uniform resource locator (URL) including the user's address from the first server, obtains the user's address from the second URL, and determines the calculated repair completion date from the second server Receive, calculate a delivery date based on the user's address and the calculated repair completion date, and transmit the calculated delivery date to the user terminal.

When the repair of the selected clothing is completed, the second server according to an embodiment of the present invention may transmit a repair result to the user terminal.

The user terminal according to an embodiment of the present invention may transmit user feedback on the repair result to the second server.

The third server according to an embodiment of the present invention may transmit the location of the vehicle that delivers the selected clothing to the user terminal.

The third server according to an embodiment of the present invention may transmit the delivery time to the user terminal based on the location of the vehicle and the traffic condition.

The second server according to an embodiment of the present invention

[Equation 1]

는 상기 이미지에 포함되는 제2 픽셀의 명도, Y는 상기 제1 픽셀의 채도,

는 상기 제2 픽셀의 채도, Z는 상기 제1 픽셀의 색상,

는 상기 제2 픽셀의 색상을 의미하고, 상기 [수학식 1]에 따라 상기 이미지의 영상 변화량을 산출할 수 있다.X is the brightness of the first pixel included in the image,

is the brightness of the second pixel included in the image, Y is the saturation of the first pixel,

is the saturation of the second pixel, Z is the color of the first pixel,

denotes the color of the second pixel, and the image change amount of the image may be calculated according to Equation 1 above.

The online shopping system according to an embodiment of the present invention further includes an online shopping model, wherein the online shopping model includes an input layer, one or more hidden layers, and an output layer, and training data configured to extract the first object. is input to the input layer of the online shopping model and passes through the one or more hidden layers and output layers to output an output vector, the output vector is input to a loss function layer connected to the output layer, and the loss function layer is The loss value is output using a loss function that compares the output vector with the correct vector for each training data, and the parameter of the online shopping model is learned in a direction in which the loss value becomes smaller,

[Equation]

The loss function follows the above equation, where N is the number of the plurality of training data, n is a natural number identifying the training data, k is a natural number identifying the value of the nth training data, nk is the nth It may mean the k-th value of the training data, t may mean correct answer data, y may mean the output vector, and E may mean a loss value.

According to embodiments, the user's body size may be accurately measured.

According to embodiments, clothes purchased by the user may be repaired to fit the user's body size.

According to embodiments, a repair completion date and a delivery date of the clothing purchased by the user may be notified to the user.

Effects obtainable from the embodiments are not limited to the effects mentioned above, and other effects not mentioned are clearly derived and understood by those of ordinary skill in the art based on the following detailed description. can be

BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are included as part of the detailed description to aid understanding of the embodiments, provide various embodiments and, together with the detailed description, explain technical features of the various embodiments.
1 illustrates an online shopping system according to an exemplary embodiment.
2 illustrates a process of acquiring first user data to third user data from a first URL according to an exemplary embodiment.
3 illustrates an image including the entire body of a user according to an exemplary embodiment.
4 illustrates a process of extracting a first object from an image according to an exemplary embodiment.
5 illustrates a process in which a part of an image is enlarged and another part of an image is reduced according to an exemplary embodiment.
6 illustrates a repair result transmitted to a user terminal according to an embodiment.
7 illustrates a process of obtaining a user's address from a second URL according to an exemplary embodiment.
8 illustrates a location of a delivery vehicle transmitted to a user terminal according to an exemplary embodiment.

The following embodiments combine elements and features of the embodiments in a predetermined form. Each component or feature may be considered optional unless explicitly stated otherwise. Each component or feature may be implemented in a form that is not combined with other components or features. In addition, various embodiments may be configured by combining some components and/or features. The order of operations described in various embodiments may be changed. Some features or features of one embodiment may be included in another embodiment, or may be replaced with corresponding features or features of another embodiment.

In the description of the drawings, procedures or steps that may obscure the gist of various embodiments are not described, and procedures or steps that can be understood at the level of those of ordinary skill in the art are also not described. did.

Throughout the specification, when a part is said to "comprising or including" a certain component, it does not exclude other components unless otherwise stated, meaning that other components may be further included. do. In addition, terms such as "... unit", "... group", and "module" described in the specification mean a unit that processes at least one function or operation, which is hardware or software or a combination of hardware and software. can be implemented as Also, "a or an", "one", "the" and like related terms are used herein in the context of describing various embodiments (especially in the context of the claims that follow). Unless otherwise indicated or clearly contradicted by context, it may be used in a sense including both the singular and the plural.

Hereinafter, embodiments according to various embodiments will be described in detail with reference to the accompanying drawings. DETAILED DESCRIPTION The detailed description set forth below in conjunction with the appended drawings is intended to describe exemplary embodiments of various embodiments, and is not intended to represent the only embodiments.

In addition, specific terms used in various embodiments are provided to help the understanding of various embodiments, and the use of these specific terms may be changed to other forms without departing from the technical spirit of various embodiments. .

1 illustrates an online shopping system 100 according to an embodiment.

Referring to FIG. 1 , the online shopping system 100 may include a user terminal 110 , a first server 120 , a second server 130 , and a third server 140 .

The user terminal 110 may be an electronic device that can access the first server 120 through an Internet web browser. For example, the user terminal 110 may include at least one of a smart phone, a tablet PC, a laptop computer, and a desktop computer.

According to an embodiment, the user may select clothes to purchase through the user terminal 110 . After selecting clothes, the user may input first user data related to the color of the clothes, second user data related to the size of the clothes, and third user data related to the repair plan of the clothes through the user terminal 110 .

According to an embodiment, the user may transmit an image (eg, 300 of FIG. 3 ) including the user's entire body to the first server 120 through the user terminal 110 . The image 300 may be used to measure the user's body size, and all or part of purchased clothing may be repaired to correspond to the measured body size.

The first server 120 may be an online shopping mall server through which a user can purchase clothes. The first server 120 may store at least one of the user's ID, the password corresponding to the ID, the user's name, the user's address, and the user's nickname. The first server 120 may allow access to the user terminal 110 when the ID and password input through the user terminal 110 match the stored ID and password.

According to an embodiment, the first server 120 may be connected to the second server 130 and/or the third server 140 . The first server 120 may transmit data received from the user terminal 110 to the second server 130 and/or the third server 140 . For example, the first server 120 may store the clothes purchased by the user, first user data related to the color of the clothes, second user data related to the size of the clothes, third user data related to the repair plan of the clothes, and the entire body of the user. The image 300 including , may be transmitted to the second server 130 and/or the third server 140 .

The second server 130 may be a server of the repair company 220 that repairs clothes by receiving a repair request from an online shopping mall. For example, when the user purchases the coat 210 on the online shopping mall, the coat 210 may be delivered to the repair company 220 and then the coat 210 may be repaired according to the user's request.

According to an embodiment, the second server 130 may extract the first object from the image 300 including the entire body of the user. The first object may mean an outline of a user excluding the background in the image 300 . For example, the second server 130 may extract the first object by connecting a section in the image 300 in which a sum of squared difference (SSD) is greater than or equal to a specified value. The image change amount may mean a change degree of brightness, saturation, color, etc. of the image 300 . That is, between the background and the first object, there is a section in which brightness, saturation, color, etc. change rapidly. By connecting the sections, the first object can be extracted.

When the first object is extracted, the second server 130 may calculate the size of the first object by enlarging a portion of the region based on the angle at which the image 300 is captured and reducing the remaining portion. As an example, when the image 300 is captured near the user's lower body, the lower body may appear larger than the user's actual body size, and the upper body may appear smaller. In this case, the user's body size can be accurately measured by reducing the lower body portion and enlarging the upper body portion. Conversely, when the image 300 is captured near the user's upper body, the upper body may appear larger than the user's actual body size, and the lower body may appear smaller. In this case, the user's body size can be accurately measured by reducing the upper body portion and enlarging the lower body portion. According to an embodiment of the present invention, the user's body size can be accurately measured by enlarging a partial area and reducing the remaining partial area based on the angle at which the image 300 is captured.

The size of the first object (or the body size of the user) may be used to repair clothes purchased by the user, and the repaired clothes may be delivered to the user. For example, if the user's arm length measured in the image 300 is 60 cm and the clothing purchased by the user is a sweater, the repair company 220 may repair the arm portion of the sweater to 60 cm. As another example, if the waist width of the user measured in the image 300 is 80 cm and the clothes purchased by the user are pants, the class company may repair the waist portion of the pants to 80 cm. According to an embodiment of the present invention, the user's body size can be accurately measured by extracting the user's contour from the image 300 and then expanding and/or reducing the user's body size, and the clothes purchased by the user are repaired to fit the user's body size. can do.

The third server 140 may be a server of the delivery company 230 that delivers the clothing purchased by the user or the repaired clothing to the user's address 240 . When there is no repair request from the user, the clothes purchased by the user may be delivered to the user's address without going through the repair company 220 . Conversely, when there is a user's request for repair, the clothing purchased by the user may be repaired by the repair company 220 and then delivered to the user's address.

According to an embodiment, the third server 140 may receive the user's address from the first server 120 and receive the repair completion date from the second server 130 . The third server 140 may calculate a delivery date based on the user's address and repair completion date. According to an embodiment, the third server 140 may transmit the real-time location of the delivery vehicle to the user terminal 110, and the user may recognize the delivery date and delivery time of the purchased clothing.

The user terminal 110 , the first server 120 , the second server 130 , and the third server 140 described in FIG. 1 are exemplary, and various embodiments described in this document are the same as those described in FIG. 1 . not limited Hereinafter, the contents described with reference to FIG. 1 may be applied to components having the same reference numerals as those of the components described with reference to FIG. 1 .

2 illustrates a process of obtaining first user data 211 to third user data 213 from a first URL 120U1 according to an exemplary embodiment.

Referring to FIG. 2 , when the user terminal 110 is connected, the first server 120 may output a screen including a plurality of clothes photos. When the user selects one of the plurality of clothes photos, the first server 120 generates first user data 211 related to the color of the selected clothes and second user data 212 related to the size of the selected clothes. , and third user data 213 related to the repair plan of the selected clothing. For example, the first user data 211 may be white, yellow, or red, and the second user data 212 may be large, medium, or small. The third user data 213 may be length repair, waist reduction, and sleeve length repair.

The second server 130 may receive the first URL 120U1 including the first user data 211 , the second user data 212 , and the third user data 213 from the first server 120 . can The second server 130 may extract third data from the first URL 120U1 and may calculate a repair completion date based on the third data. For example, in the case of length repair, it does not take a long time to repair, but in the case of a waist reduction, the required time increases, so the date of completion of the repair may also be delayed.

According to an embodiment, the second server 130 may calculate a repair completion date based on the number of repair details received within a specified period and the third user data 213 . The calculated repair completion date may be transmitted to the user terminal 110 , and the user may check the repair completion date. For example, since the number of repair details received within one week is less than 10, and the required time is not long in the case of millet repair, the second server 130 may transmit a date within the near future to the user terminal 110 .

3 illustrates an image 300 including the entire body of a user according to an exemplary embodiment. 3 is a diagram for explaining a process in which the second server 130 accurately calculates the user's body size.

Referring to FIG. 3 , the second server 130 may extract the first object from the image 300 . The first object may mean an outline of a user excluding the background in the image 300 . When the first object is extracted, the user may distinguish an area to be enlarged and an area to be reduced based on the angle at which the image 300 is captured.

For example, as shown in FIG. 3 , in the case of a picture taken near a user's lower body, the lower body may be larger than the actual user's body size, and the upper body may be taken smaller than the actual user's body size. Accordingly, the second server 130 may reduce the lower body area and enlarge the upper body area among the first objects. When the lower body area is reduced and the upper body area is enlarged, the second server 130 may acquire the same contour as the actual user's body proportions. Then, the second server 130 may acquire the length, size, width, etc. of each body part (eg, arms, legs, etc.) based on the actual user's body proportions.

Conversely, in the case of a picture taken near the user's upper body, the upper body may be larger than the actual user's body size, and the lower body may be taken smaller than the actual user's body size. Accordingly, the second server 130 may reduce the upper body area and enlarge the lower body area of the first object. When the upper body area is reduced and the lower body area is enlarged, the second server 130 may acquire the same contour as the actual user's body proportions. Then, the second server 130 may acquire the length, size, width, etc. of each body part (eg, arms, legs, etc.) based on the actual user's body proportions.

According to an embodiment, the second server 130 may determine the angle at which the image 300 is captured based on the positions of the camera and the smart phone included in the image 300 . For example, when the camera is located in the upper region of the image 300 , it may be determined that the image 300 is captured near the user's upper body. Conversely, when the camera is located in the lower region of the image 300 , it may be determined that the image 300 is photographed near the user's lower body.

4 illustrates a process of extracting a first object from an image 300 according to an exemplary embodiment. FIG. 4 is an enlarged view of the first area 310 shown in FIG. 3 , and is a view for explaining in detail a process of extracting a first object from the image 300 including the entire body of the user.

Referring to FIG. 4 , the second server 130 may obtain a sum of squared difference (SSD) of the image 300 . The image change amount may mean a change degree of brightness, saturation, color, etc. of the image 300 . When the image change amount is obtained, the second server 130 may extract the first object by connecting the section 310L in which the image change amount is greater than or equal to a specified value.

For example, as shown in FIG. 4 , the brightness, saturation, and color of the area a 310A and the area b 310B of the image 300 are very different, so that the image change amount may be greater than or equal to a specified value. The second server 130 may connect a section 310L having an image change amount greater than or equal to a specified value, and the connected section 310L may form a first object.

According to an embodiment, the second server 130 may change the designated value based on the image change amount distribution of the image 300 . For example, when there is little difference in brightness, saturation, and color between the first object and the background, the second server 130 may decrease the specified value. When the specified value decreases, the second server 130 may extract the first object even if the difference between the first object and the background is small.

According to an embodiment, the second server 130 may calculate the image change amount based on the following [Equation 1].

[Equation 1]

는 제2 픽셀의 명도, Y는 제1 픽셀의 채도,

는 제2 픽셀의 채도, Z는 제1 픽셀의 색상,

는 제2 픽셀의 색상을 의미할 수 있다. 제2 픽셀은 제1 픽셀의 상, 하, 좌, 우측에 위치한 픽셀들 중 어느 하나일 수 있다. 즉, 제2 서버(130)는 제1 픽셀의 명도, 채도, 색상 값을 획득한 후 제2 픽셀의 명도, 채도, 색상 각각의 차를 제곱하여 병합함으로써 영상 변화량을 산출할 수 있다. 상기 [수학식 1]에 의해 산출된 영상 변화량이 지정된 값 이상이면 제2 서버(130)는 제1 픽셀과 제2 픽셀 사이를 제1 객체의 일부로 지정할 수 있으며, 이미지(300)에 포함된 모든 픽셀에 대하여 상기 [수학식 1]을 적용함으로써 제1 객체를 추출할 수 있다.In [Equation 1], X is the brightness of the first pixel,

is the brightness of the second pixel, Y is the saturation of the first pixel,

is the saturation of the second pixel, Z is the color of the first pixel,

may mean the color of the second pixel. The second pixel may be any one of pixels located above, below, left, and right of the first pixel. That is, the second server 130 may calculate the image change amount by squaring and merging the differences between the brightness, saturation, and hue of the second pixel after obtaining the brightness, saturation, and color values of the first pixel. When the amount of image change calculated by Equation 1 is greater than or equal to a specified value, the second server 130 may designate a space between the first pixel and the second pixel as a part of the first object, and all The first object may be extracted by applying Equation 1 to the pixel.

5 illustrates a process in which a part of the image 300 is enlarged and another part of the image 300 is reduced according to an exemplary embodiment. 5 is a diagram for explaining in detail a method for accurately measuring a user's body size. FIG. 5( a ) may mean a third area 330 in FIG. 3 , and FIG. 5( b ) may mean a second area 320 in FIG. 3 .

Referring to FIGS. 5A and 5B , the second server 130 may divide the second server 130 into a plurality of pixels in the second area 320 and the third area 330 . After the second area 320 and the third area 330 are divided into a plurality of pixels, the second server 130 removes some of the pixels included in the second area 320 or an interval between the pixels. By reducing , the size of the second region 320 may be reduced. For example, as shown in FIG. 5B , the second server 130 includes an a-th pixel 521 , a b-th pixel 522 , a c-th pixel 523 , a d-th pixel 524 , and an e-th pixel 521 . The second region 320 may be reduced by removing at least one of the pixels 525 or by overlapping pixels.

According to an embodiment, the second server 130 enlarges a gap between pixels included in the third region 330 , merges pixels around the enlarged gap, and inserts it into the gap in the third region 330 . size can be enlarged. For example, the second server 130 may enlarge an interval between the first pixel 511 , the second pixel 512 , the third pixel 513 , and the fourth pixel 514 . The second server 130 may calculate the average of the brightness, saturation, and color values of the first pixel 511 , the second pixel 512 , the third pixel 513 , and the fourth pixel 514 , A fifth pixel 515 having the calculated average value may be inserted between the first pixel 511 , the second pixel 512 , the third pixel 513 , and the fourth pixel 514 . According to an embodiment of the present invention, when a part of the image 300 is enlarged, a natural image may be obtained by inserting pixels having an average of the brightness, saturation, and color values of the pixels into the enlarged space.

According to an embodiment, when merging neighboring pixels, the third server 140 may assign weights in the order of hue, brightness, and saturation. For example, the human eye may react sensitively in the order of hue, brightness, and saturation, and in particular, if the colors are different in the relationship between neighboring pixels, the human eye can easily recognize the corresponding part. Accordingly, the third server 140 may determine the hue, brightness, and saturation of the fifth pixel 515 based on Equation 2 to Equation 4 below.

[Equation 2]

[Equation 3]

[Equation 4]

In [Equation 2] to [Equation 4], H1 is the color of the first pixel 511, H2 is the color of the second pixel 512, H3 is the color of the third pixel 513, H4 is the fourth pixel A color of 514, H5 is the color of the fifth pixel 515, α is a constant, V1 is the brightness of the first pixel 511, V2 is the brightness of the second pixel 512, V3 is the third pixel 513 ), V4 is the brightness of the fourth pixel 514, V5 is the brightness of the fifth pixel 515, β is a constant, S1 is the saturation of the first pixel 511, S2 is the brightness of the second pixel 512 Saturation, S3 is the saturation of the third pixel 513 , S4 is the saturation of the fourth pixel 514 , S5 is the saturation of the fifth pixel 515 , and γ is a constant. Here, the size may be increased in the order of α, β, and γ. According to an embodiment of the present invention, a portion of the image 300 may be naturally enlarged by assigning weights to elements sensitive to the human eye.

When the second area 320 and the third area 330 are reduced and enlarged, respectively, the second server 130 may calculate the size of the first object. The second server 130 may calculate the size of a part of the body included in the first object or may calculate the size of the entire first object. According to an embodiment, the second server 130 may calculate the size of the first object by using the number of pixels included in the first object. For example, the length and width of the arm may be calculated based on the number of pixels included in the arm of the first object. The calculated body size may be used when clothing is repaired, and accordingly, clothing may be repaired to a size suitable for the user's body even without a separate user input.

After the image is converted through the above method, object identification may be performed through the converted result.

6 shows a repair result transmitted to the user terminal 110 according to an embodiment.

Referring to FIG. 6 , when the repair is completed, the second server 130 may transmit the repair result to the user terminal 110 . The repair result may be a photograph, and dimensions may be displayed in the repair result. For example, when the repaired clothing is a top, arm length, shoulder width, chest circumference, etc. may be displayed together on the photo. As another example, when the repaired clothing is a bottom, leg length, waist circumference, etc. may be displayed together on the photo.

According to an embodiment, the user terminal 110 may transmit feedback on the repair result to the second server 130 . The feedback may be transmitted to the second server 130 via the first server 120 , or may be transmitted from the user terminal 110 to the second server 130 .

7 illustrates a process of obtaining a user's address from the second URL 120U2 according to an exemplary embodiment.

Referring to FIG. 7 , the first server 120 may output a screen including the user's ID, password, and user's address 410 in response to a user request. The third server 140 may receive the second URL 120U2 including the user's ID, password, and user's address 410 from the first server 120 . The third server 140 may extract the user's address 410 from the second URL 120U2 and may calculate a delivery completion date based on the user's address 410 . For example, when the user's address 410 is Gyeonggi-do, the delivery time is not long, but when the user's address 410 is on Jeju Island, the delivery time is longer, so the delivery completion date may also be delayed.

According to an embodiment, the third server 140 may calculate the delivery completion date based on the repair completion date and the user's address 410 . The calculated delivery completion date may be transmitted to the user terminal 110 , and the user may check the delivery completion date. For example, the third server 140 may calculate the delivery completion date by adding up the time required for delivery to the repair completion date.

8 illustrates a location of a delivery vehicle transmitted to the user terminal 110 according to an exemplary embodiment.

Referring to FIG. 8 , the third server 140 may transmit the real-time delivery status of the clothes purchased by the user to the user terminal 110 . That is, the third server 140 may calculate the delivery completion date by adding up the time required for delivery to the repair completion date, and then transmit the real-time delivery status to the user terminal 110 . Through the real-time delivery status, the user can accurately recognize the date and time when the purchased clothing will be delivered.

An online shopping system according to an embodiment of the present invention includes a user terminal running an Internet web browser, a first server accessible through the Internet web browser, a second server connected to the first server, and a third server connected to the first server and the second server, wherein the first server includes a user ID, a password corresponding to the ID, the user's name, the user's address, and the user's nickname storing at least one of the above, and if the ID and password input through the user terminal match the stored ID and password, the user terminal is allowed to access, and a screen including a plurality of clothes photos is output, and the plurality of When any one of the clothes photos of When an image including the entire body is received through the user terminal, and the price of the selected clothing is paid, the second server includes the first user data, the second user data, and the third user data. receive a first uniform resource locator (URL) from the first server, obtain the third user data from the first URL, receive the image from the first server, Squared Difference; SSD) extracts the first object by connecting the section equal to or greater than a specified value, divides the first object into pixel units, and includes in the first area of the first object based on the angle at which the image is captured The gap between pixels to be used is enlarged, pixels around the enlarged gap are merged and inserted into the gap, and the gap between pixels included in the second area of the first object is reduced or the number of pixels included in the second area is reduced. at least a portion of the first Calculating the size of the object, calculating a repair completion date based on the number of repair details received within a specified period and the third user data, and transmitting the calculated repair completion date to the user terminal, and the third server receives a second uniform resource locator (URL) including the user's address from the first server, obtains the user's address from the second URL, and determines the calculated repair completion date from the second server Receive, calculate a delivery date based on the user's address and the calculated repair completion date, and transmit the calculated delivery date to the user terminal.

When the repair of the selected clothing is completed, the second server according to an embodiment of the present invention may transmit a repair result to the user terminal.

The user terminal according to an embodiment of the present invention may transmit user feedback on the repair result to the second server.

The third server according to an embodiment of the present invention may transmit the location of the vehicle that delivers the selected clothing to the user terminal.

The third server according to an embodiment of the present invention may transmit the delivery time to the user terminal based on the location of the vehicle and the traffic condition.

The second server according to an embodiment of the present invention

[Equation 1]

는 상기 이미지에 포함되는 제2 픽셀의 명도, Y는 상기 제1 픽셀의 채도,

는 상기 제2 픽셀의 채도, Z는 상기 제1 픽셀의 색상,

는 상기 제2 픽셀의 색상을 의미하고, 상기 [수학식 1]에 따라 상기 이미지의 영상 변화량을 산출할 수 있다.X is the brightness of the first pixel included in the image,

is the brightness of the second pixel included in the image, Y is the saturation of the first pixel,

is the saturation of the second pixel, Z is the color of the first pixel,

denotes the color of the second pixel, and the image change amount of the image may be calculated according to Equation 1 above.

The online shopping system according to an embodiment of the present invention further includes an online shopping model, wherein the online shopping model includes an input layer, one or more hidden layers, and an output layer, and training data configured to extract the first object. is input to the input layer of the online shopping model and passes through the one or more hidden layers and output layers to output an output vector, the output vector is input to a loss function layer connected to the output layer, and the loss function layer is The loss value is output using a loss function that compares the output vector with the correct vector for each training data, and the parameter of the online shopping model is learned in a direction in which the loss value becomes smaller,

[Equation]

The loss function follows the above equation, where N is the number of the plurality of training data, n is a natural number identifying the training data, k is a natural number identifying the value of the nth training data, nk is the nth It may mean the k-th value of the training data, t may mean correct answer data, y may mean the output vector, and E may mean a loss value.

The embodiments described above may be implemented by a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the apparatus, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

The software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with reference to the limited drawings, those skilled in the art may apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (6)

In the online shopping system,
A user terminal running an Internet web browser;
A first server accessible through the Internet web browser,
a second server connected to the first server, and
A third server connected to the first server and the second server,
The first server is:
storing at least one of a user's ID, a password corresponding to the ID, the user's name, the user's address, and the user's nickname,
If the ID and password input through the user terminal match the stored ID and password, the access of the user terminal is allowed,
Outputs a screen including a plurality of clothing photos,
When any one of the plurality of clothes photos is selected, first user data related to the color of the selected clothes, second user data related to the size of the clothes, third user data related to the repair plan of the clothes, and the Receive an image including the entire body of the user through the user terminal,
The second server is:
When the price of the selected clothing is paid, receiving a first uniform resource locator (URL) including the first user data, the second user data, and the third user data from the first server,
obtaining the third user data from the first URL;
receiving the image from the first server;
Extracting a first object by connecting a section in which the sum of squared difference (SSD) of the image is greater than or equal to a specified value,
dividing the first object into pixel units,
Based on the angle at which the image was captured, it is determined in which area the image was taken from among the upper region close to the horizontal position of the user's upper body and the lower region close to the horizontal position of the lower body included in the image,
When the image is taken from the upper region, the gap between pixels included in the first region close to the lower region of the first object is enlarged, and pixels around the enlarged gap are merged and inserted into the gap;
calculating the size of the first object by reducing a gap between pixels included in a second region close to the upper region of the first object or removing at least a portion of pixels included in the second region;
calculating a repair completion date based on the number of repair details received within a specified period and the third user data;
Transmitting the calculated repair completion date to the user terminal,
The third server is:
Receiving a second uniform resource locator (URL) including the user's address from the first server,
obtaining the address of the user from the second URL;
receiving the calculated repair completion date from the second server;
Calculating a delivery date based on the user's address and the calculated repair completion date,
Transmitting the calculated delivery date to the user terminal,
The online shopping system further comprises an online shopping model,
The online shopping model includes an input layer, one or more hidden layers and an output layer,
The training data configured to extract the first object is input to the input layer of the online shopping model and passes through the one or more hidden layers and output layers to output an output vector, wherein the output vector is a loss function connected to the output layer input to the layer, and the loss function layer outputs a loss value using a loss function that compares the output vector with a correct vector for each training data, and the parameter of the online shopping model is a direction in which the loss value decreases is learned with
[Equation]

The loss function follows the above equation,
In the above formula, N is the number of the plurality of learning data, n is a natural number identifying the learning data, k is a natural number identifying the value of the n-th learning data, nk is the k-th value of the n-th learning data, t means correct answer data, y means the output vector, E means loss value,
The second server calculates a sum of squared difference (SSD) of the image by using Equation 1,
[Equation 1]

In [Equation 1], X is the brightness of the sixth pixel included in the image,

is the brightness of the 7th pixel included in the image, Y is the saturation of the 6th pixel,

is the saturation of the 7th pixel, Z is the color of the 6th pixel,

may mean the color of the seventh pixel. The second pixel is any one of pixels located above, below, left, and right of the sixth pixel,
When the gap between pixels included in the first region is enlarged, the third server assigns weights to hue, brightness, and saturation according to the user's sensitivity to color, brightness, and saturation, Equation 2, Equation 2 According to Equations 3 and 4, the pixels around the enlarged gap are merged and inserted into the gap,
[Equation 2]

[Equation 3]

[Equation 4]

In [Equation 2] to [Equation 4], H1 is the color of the first pixel 511, H2 is the color of the second pixel 512, H3 is the color of the third pixel 513, H4 is the fourth pixel A color of 514, H5 is the color of the fifth pixel 515, α is a weight for the color, V1 is the brightness of the first pixel 511, V2 is the brightness of the second pixel 512, V3 is the third The brightness of the pixel 513, V4 is the brightness of the fourth pixel 514, V5 is the brightness of the fifth pixel 515, β is the weight for the brightness, S1 is the saturation of the first pixel 511, S2 is the first pixel 2 is the saturation of the pixel 512, S3 is the saturation of the third pixel 513, S4 is the saturation of the fourth pixel 514, S5 is the saturation of the fifth pixel 515, γ is the weight for the saturation ,
, online shopping system. The method of claim 1,
and the second server transmits a repair result to the user terminal when the repair of the selected clothing is completed. 3. The method of claim 2,
The user terminal transmits a user feedback on the repair result to the second server. The method of claim 1,
and the third server transmits a location of a vehicle that delivers the selected clothing to the user terminal. 5. The method of claim 4,
The third server transmits the delivery time to the user terminal based on the location of the vehicle and the traffic condition. delete

Images