E-commerce clothing sales value evaluation system based on cloud computing

Abstract

The invention discloses an e-commerce clothing sales value evaluation system based on cloud computing, which comprises a data acquisition module, a sampling module, a quality evaluation module and an output module, and is characterized in that the sampling module is used for acquiring related image features of a target shirt; the quality evaluation module is used for performing feature detection on a target shirt; the data acquisition module, the sampling module, the quality evaluation module and the output module are in communication connection with each other, through the feature overlapping detection module and the pattern overlap ratio detection module, pattern areas of shirts are divided in a sampling detector, slopes of lines are compared, specific parameters of all pattern lines in the shirts are visualized, and the detection accuracy of the shirts is improved. The pattern lines at different positions of the shirt are positioned, marked and disassembled, the line characteristics in the shirt are combed, and the detection difficulty caused by the irregularity of the line characteristics is effectively reduced. The method has the characteristics of high characteristic test efficiency and high accurate monitoring capability.

Description

Technical Field

[0001] The present invention relates to the technical field of clothing production management, and in particular to an e-commerce clothing sales value evaluation system based on cloud computing. Background Art

[0002] In modern clothing design, plaid pattern is an important source of inspiration and design element for fashion designers. With the development and evolution of popular trends, the application form of plaid pattern is constantly updated and has become the production choice of many clothing manufacturers.

[0003] As consumers' requirements for clothing matching increase, the patterns on clothing also affect the selling price and sales rate of clothing. In the prior art, feature detection of different shirt design styles is a detection method that can accurately predict the design value of clothing patterns. It is widely used by e-commerce merchants to evaluate the quality of shirts. However, due to the emergence of more and more clothing styles, simply detecting the color distribution and pattern types of clothing can no longer meet the diversified needs of shirt design evaluation. Therefore, a cloud computing-based e-commerce clothing sales value evaluation system with high design feature testing efficiency and strong accurate monitoring capabilities is very necessary. Summary of the invention

[0004] The purpose of the present invention is to provide an e-commerce clothing sales value evaluation system based on cloud computing to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a cloud computing-based e-commerce clothing sales value evaluation system, comprising a data acquisition module, a sampling module, a quality evaluation module and an output module, wherein the data acquisition module is used to obtain relevant information about shirts sold in shops; the sampling module is used to obtain relevant image features of a target shirt through a detector; the quality evaluation module is used to perform feature detection on the target shirt and obtain relevant parameter information of the shirt pattern; the output module is used to output the detection results of the shirt design scheme, and the data acquisition module, the sampling module, the quality evaluation module and the output module are communicatively connected to each other.

[0006] According to the above technical solution, the data acquisition module includes a shirt information acquisition module and a detector information acquisition module. The shirt information acquisition module is used to obtain relevant information on shirt production; the detector information acquisition module is used to obtain detector information for detecting shirt features.

[0007] According to the above technical solution, the quality assessment module includes a pattern overlap detection module and a feature overlap detection module. The pattern overlap detection module is used to detect the overlap of the patterns set on the shirts for sale; the feature overlap detection module is used to detect whether feature overlap occurs in the target shirt.

[0008] According to the above technical solution, the pattern overlap detection module further includes an arrangement analysis submodule and a line detection submodule. The arrangement analysis submodule is used to analyze the overall arrangement of the patterns on the shirts for sale; the line detection submodule is used to detect the characteristic information presented by the pattern arrangement of the shirts for sale through the slope and concavity of the pattern lines on the shirts for sale.

[0009] According to the above technical solution, the operation method of the sales value evaluation system mainly includes the following steps:

[0010] Step S1: extract the clothing information authorized online by the clothing manufacturer in the shirt purchase and sales testing software connected to the sales value evaluation system, and extract the maximum and minimum information in the clothing size information through data analysis and processing, and obtain the distance range from the left cuff to the right cuff of the target shirt to be tested as [a 1 , a 2 ], the distance from the shirt collar to the bottom button of the shirt is [b 1 , b 2 ];

[0011] Step S2: placing the produced garment on the detection instrument and analyzing the shirt style image captured by the camera;

[0012] Step S3: obtaining the lines of the pattern on the shirt in the detection instrument, calculating the overlap rate of the pattern on the current production clothing, and judging whether the lines of the pattern of the currently selected sampled shirt have feature overlap;

[0013] Step S4: Test different sampled shirts in turn, and output clothing solutions with different clothing but overlapping style features through the output module.

[0014] According to the above technical solution, step S2 further includes:

[0015] Step S21: a detector is set up to detect the pattern on the produced clothing, the robot arm places the detection target shirt on the detector, and uses an automatic ironing machine to remove wrinkles on the target shirt;

[0016] Step S22: In the first image captured by the camera, the system establishes a plane rectangular coordinate system with the center point of the image as the coordinate origin, a straight line through the coordinate origin parallel to the upper limit of the first image as the X-axis, and a Y-axis rotated 90°. The coordinates of the center point of the image are (0, 0), and the length of the X-axis is the distance a from the left cuff to the right cuff of the production shirt. 2 , the length of the Y axis is the distance b from the collar of the shirt to the bottom button of the shirt 2 , the unit length of the coordinate system is 1.

[0017] According to the above technical solution, in step S21, the specific steps of placing the robot arm on the target shirt for detection are:

[0018] Step S211: the system adjusts the direction of the camera set at the highest point on the detector to be perpendicular to the ground, the camera is connected to the loading platform below through a mechanical arm, the direction of the camera is controlled to coincide with the center point of the loading platform set at the bottom of the detector, and the picture taken by the camera is set as the first picture;

[0019] Step S212: The robot arm adjusts the bottom button, the top button, the left cuff and the right cuff of the target shirt to corresponding positions in the coordinate system according to the front view position in the plane rectangular coordinate system in step S2;

[0020] The corresponding position in the coordinate system is: the coordinate of the button with the smallest ordinate among all the identified buttons The coordinate of the button with the largest ordinate among all identified buttons Coordinates of the midpoint of the left cuff Coordinates of the midpoint of the right cuff The button with the smallest ordinate among the buttons corresponds to the bottom button of the detection target shirt, and the button with the largest ordinate among the buttons corresponds to the top button of the detection target shirt. 1 , a 2 ],B∈[b 1 , b 2 ].

[0021] According to the above technical solution, in step S212, the specific method for the robot arm to identify the target shirt is:

[0022] The camera sends the captured image of the target shirt to the sampling module;

[0023] The sampling module preprocesses the image;

[0024] The image is segmented to obtain the target area and the external features of the shirt are extracted;

[0025] The image is input into the recognition model obtained through big data training, the cuff and button features of the shirt in the image are extracted, the location of the features is located and uploaded to the device collaboration system.

[0026] According to the above technical solution, step S3 further includes:

[0027] Step S31: infrared detection of the features of the shirts stacked on the detector loading platform is performed by infrared rays set in the camera, and the initial point coordinates (X 1 , Y 1 ) and the end point coordinates (X 2 , Y 2 ), and get the number of overlaps at the end of each line;

[0028] Step S32: Arrange the coordinate points in descending order according to the number of overlaps, extract the two coordinates with the largest number of overlaps, the two coordinates with the smallest number of overlaps, and S coordinates with the number of overlaps at the midpoint of the sequence, wherein the number of coordinate points to be extracted at the midpoint of the sequence is determined by the total number of coordinate points. When n is an odd number, the number of coordinate points to be extracted is 1; when n is an even number, the number of coordinate points to be extracted is 2;

[0029] Step S33: Obtain the slopes of the (K+1) pattern lines corresponding to each extracted coordinate in turn. Determine the slope difference between the two pattern lines among the slopes of the pattern lines corresponding to the five extracted coordinate points Where L 1 and L 2 They are the corresponding slopes of two lines randomly selected from the pattern lines corresponding to the five coordinate points. When (100%-μ%)<η<(100%+μ%), it is judged that the currently selected pattern lines have feature overlap, where μ is the maximum error limit ratio value of the feature overlap corresponding to the slope difference rate.

[0030] According to the above technical solution, in step S31, the method for identifying the pattern lines arranged on each single-layer shirt mainly includes the following steps:

[0031] Step S311: Cut the image area into k 1 *k 2 There are k small rectangular detection areas. 1 and k 2The length and width of the picture area are divided into corresponding numbers respectively, so that the area of ​​each small detection area is the same, and the n areas with different colors in the small detection area are divided into "first pattern area", "second pattern area" ... "nth pattern area", where the dividing boundary between two adjacent areas is the pattern line, and the pattern line dividing each area is marked as "first X 1 Lines", ... "X G X H Lines", where X 1 , X G and X H are the numbers corresponding to two adjacent divided areas within the monitoring area, X 1 , X G and X H are all integers, min{X 1 , X K , X G}≥1, max{X 1 , X K , X G}≤n;

[0032] Step S312: The initial position of the pattern line in each area is the point (X 1 , Y 1 ) and the end point coordinates (X 2 , Y 2 ) to mark, and obtain the initial position coordinate points (X 11 , Y 11 ), (X 12 , Y 12 )……(X 1n , Y 1n ) and the end point coordinates (X 21 , Y 21 ), (X 22 , Y 22 )……(X 2n , Y 2n ), the number of times each coordinate point overlaps with other initial position coordinate points and end position coordinate points is K, then there are "(K+1)" coordinate points intersecting at this coordinate point.

[0033] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention processes the clothing information of the shirt in advance through the sampling module, and uses the detector to locate and detect the relevant information of the target shirt, determines the length range of the target shirt, and improves the applicability of the clothing feature detection system and the feature detection efficiency; through the feature overlap detection module, the pattern area of ​​the shirt is divided and the slopes of the lines are compared in the sampling detector, and the various visual parameters in the shirt are visualized, which increases the way to identify the technical features of the shirt, greatly improves the detection accuracy, provides a new and powerful basis for e-commerce merchants to choose clothing styles, and effectively avoids the risk of mass production of low-quality clothing that damages corporate sales; at the same time, through the pattern overlap detection module, the pattern lines at different positions of the shirt are located, marked and disassembled, the features formed by all the pattern lines are arranged, the line features in the shirt are sorted out, and the difficulty of detection caused by the irregularity of the line features is effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0035] Figure 1 It is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] See also Figure 1 The present invention provides a technical solution: an e-commerce clothing sales value evaluation system based on cloud computing, comprising:

[0038] The data acquisition module, sampling module, quality assessment module and output module are used to obtain relevant information about shirts sold in shops; the sampling module is used to obtain relevant image features of the target shirt through a detector; the quality assessment module is used to perform feature detection on the target shirt and obtain relevant parameter information of the shirt pattern; the output module is used to output the detection results of the shirt design scheme. The data acquisition module, sampling module, quality assessment module and output module are connected to each other for communication.

[0039] The present invention processes the clothing information of the shirt in advance through a sampling module, and uses a detector to locate and detect relevant information of the target shirt, determines the length range of the target shirt, and improves the applicability of the clothing feature detection system and the feature detection efficiency; through a feature overlap detection module, the pattern area of ​​the shirt is divided and the slopes of the lines are compared in the sampling detector, and various visual parameters in the shirt are visualized, which increases the way to identify the technical features of the shirt, greatly improves the detection accuracy, provides a new and powerful basis for the clothing styles selected by e-commerce merchants, and effectively avoids the risk of mass production of low-quality clothing that damages the sales volume of enterprises; at the same time, through a pattern overlap detection module, the pattern lines at different positions of the shirt are located, marked and disassembled, the features formed by all the pattern lines are arranged, the line features in the shirt are sorted out, and the difficulty of detection caused by the irregularity of the line features is effectively reduced.

[0040] The data acquisition module includes a shirt information acquisition module and a detector information acquisition module. The shirt information acquisition module is used to acquire relevant information of shirt production; the detector information acquisition module is used to acquire detector information for detecting shirt features.

[0041] The quality assessment module includes a pattern overlap detection module and a feature overlap detection module. The pattern overlap detection module is used to detect the overlap of the patterns set on the shirts for sale; the feature overlap detection module is used to detect whether feature overlap occurs in the target shirt.

[0042] The pattern overlap detection module further includes an arrangement analysis submodule and a line detection submodule. The arrangement analysis submodule is used to analyze the overall arrangement of the patterns on the shirts for sale; the line detection submodule is used to detect the characteristic information presented by the pattern arrangement of the shirts for sale through the slope and concavity of the pattern lines on the shirts for sale.

[0043] In a preferred embodiment, the operation method of the sales value evaluation system mainly includes the following steps:

[0044] Step S1: extract the clothing information authorized online by the clothing manufacturer in the shirt purchase and sales testing software connected to the sales value evaluation system, and extract the maximum and minimum information in the clothing size information through data analysis and processing, and obtain the distance range from the left cuff to the right cuff of the target shirt to be tested as [a 1 , a 2 ], the distance from the shirt collar to the bottom button of the shirt is [b 1 , b 2 ];

[0045] Step S2: placing the produced garment on the detection instrument and analyzing the shirt style image captured by the camera;

[0046] Step S3: obtaining the lines of the pattern on the shirt in the detection instrument, calculating the overlap rate of the pattern on the current production clothing, and judging whether the lines of the pattern of the currently selected sampled shirt have feature overlap;

[0047] Step S4: Test different sampled shirts in turn, and output clothing solutions with different clothing but overlapping style features through the output module.

[0048] Step S2 further comprises:

[0049] Step S21: a detector is set up to detect the pattern on the produced clothing, the robot arm places the detection target shirt on the detector, and uses an automatic ironing machine to remove wrinkles on the target shirt;

[0050] Step S22: In the first image captured by the camera, the system establishes a plane rectangular coordinate system with the center point of the image as the coordinate origin, a straight line through the coordinate origin parallel to the upper limit of the first image as the X-axis, and a Y-axis rotated 90°. The coordinates of the center point of the image are (0, 0), and the length of the X-axis is the distance a from the left cuff to the right cuff of the production shirt. 2 , the length of the Y axis is the distance b from the collar of the shirt to the bottom button of the shirt 2 , the unit length of the coordinate system is 1.

[0051] In step S21, the specific steps of placing the robot arm on the target shirt for inspection are:

[0052] Step S211: the system adjusts the direction of the camera set at the highest point on the detector to be perpendicular to the ground, the camera is connected to the loading platform below through a mechanical arm, the direction of the camera is controlled to coincide with the center point of the loading platform set at the bottom of the detector, and the picture taken by the camera is set as the first picture;

[0053] Step S212: The robot arm adjusts the bottom button, the top button, the left cuff and the right cuff of the target shirt to corresponding positions in the coordinate system according to the front view position in the plane rectangular coordinate system in step S2;

[0054] The corresponding position in the coordinate system is: the coordinate of the button with the smallest ordinate among all the identified buttons The coordinate of the button with the largest ordinate among all identified buttons Coordinates of the midpoint of the left cuff Coordinates of the midpoint of the right cuff The button with the smallest ordinate among the buttons corresponds to the bottom button of the detection target shirt, and the button with the largest ordinate among the buttons corresponds to the top button of the detection target shirt. 1 , a 2],B∈[b 1 , b 2 ].

[0055] In shirts produced by manufacturers, there are often problems that shirts of the same style but different sizes are different in size, and shirts of the same size but different styles are different in size, which makes it difficult for the same detection system to be applied to target clothing of multiple styles when testing the characteristics of shirts. This technical solution processes the clothing information of the shirts in advance and determines the length range of the target shirts. This improves the applicability of the clothing feature detection system while speeding up the feature detection efficiency.

[0056] In step S212 of this embodiment, the specific method for the robot arm to identify the target shirt is:

[0057] The camera sends the captured image of the target shirt to the sampling module;

[0058] The sampling module preprocesses the image;

[0059] The image is segmented to obtain the target area and the external features of the shirt are extracted;

[0060] The image is input into the recognition model obtained through big data training, the cuff and button features of the shirt in the image are extracted, the location of the features is located and uploaded to the device collaboration system.

[0061] In this embodiment, step S3 further includes:

[0062] Step S31: infrared detection of the features of the shirts stacked on the detector loading platform is performed by infrared rays set in the camera, and the initial point coordinates (X 1 , Y 1 ) and the end point coordinates (X 2 , Y 2 ), and get the number of overlaps at the end of each line;

[0063] Step S32: Arrange the coordinate points in descending order according to the number of overlaps, extract the two coordinates with the largest number of overlaps, the two coordinates with the smallest number of overlaps, and S coordinates with the number of overlaps at the midpoint of the sequence, wherein the number of coordinate points to be extracted at the midpoint of the sequence is determined by the total number of coordinate points. When n is an odd number, the number of coordinate points to be extracted is 1; when n is an even number, the number of coordinate points to be extracted is 2;

[0064] Step S33: Obtain the slopes of the (K+1) pattern lines corresponding to each extracted coordinate in turn. Determine the slope difference between the two pattern lines among the slopes of the pattern lines corresponding to the five extracted coordinate points Where L 1and L 2 They are the corresponding slopes of two lines randomly selected from the pattern lines corresponding to the five coordinate points. When (100%-μ%)<η<(100%+μ%), it is judged that the currently selected pattern lines have feature overlap, where μ is the maximum error limit ratio value of the feature overlap corresponding to the slope difference rate.

[0065] The value of a feature is directly related to its occurrence rate and repetition rate. As far as the value of a feature is concerned, the higher the occurrence rate and the greater the repetition, the lower the value of the feature; the lower the occurrence rate and the smaller the repetition, the higher the value of the feature. In other words, if the feature is formed with strong uniqueness and contingency, its value is higher; if the feature is formed with regularity and universality, its value is lower.

[0066] This technical solution solves the problem that the identification method is too single, resulting in a single analysis channel and serious limitation of the analysis effect, because the shirt pattern design technology is judged by identifying the tone modulation of different shirts; it also solves the problem that the patterns and lines of the shirt pattern design are difficult to quantify, resulting in a lack of screening ability for unsold shirts and surplus production capacity. By dividing the pattern area of ​​the shirt in the sampling detector and comparing the slopes of the lines, the various visual parameters in the shirt are visualized, which increases the way to identify the technical characteristics of the shirt and greatly improves the detection accuracy.

[0067] By testing the slope of the patterns on the inside of the shirts, the company's assessment accuracy of the production quality of shirt patterns has been further improved, providing a new and powerful basis for e-commerce merchants to choose clothing styles, and effectively avoiding the risk of mass production of low-quality clothing that damages the sales of e-commerce merchants.

[0068] In step S31 of this embodiment, the method for identifying the pattern lines arranged on each single-layer shirt mainly includes the following steps:

[0069] Step S311: Cut the image area into k 1 *k 2 There are k small rectangular detection areas. 1 and k 2 The length and width of the picture area are divided into corresponding numbers respectively, so that the area of ​​each small detection area is the same, and the n areas with different colors in the small detection area are divided into "first pattern area", "second pattern area" ... "nth pattern area", where the dividing boundary between two adjacent areas is the pattern line, and the pattern line dividing each area is marked as "first X 1 Lines", ... "X G X H Lines", where X 1 , X Gand X H are the numbers corresponding to two adjacent divided areas within the monitoring area, X 1 , X G and X H are all integers, min{X 1 , X K , X G}≥1, max{X 1 , X K , X G}≤n;

[0070] Step S312: The initial position of the pattern line in each area is the point (X 1 , Y 1 ) and the end point coordinates (X 2 , Y 2 ) to mark, and obtain the initial position coordinate points (X 11 , Y 11 ), (X 12 , Y 12 )……(X 1n , Y 1n ) and the end point coordinates (X 21 , Y 21 ), (X 22 , Y 22 )……(X 2n , Y 2n ), the number of times each coordinate point overlaps with other initial position coordinate points and end position coordinate points is K, then there are "(K+1)" coordinate points intersecting at this coordinate point.

[0071] This technical solution solves the problem that when detecting the pattern arrangement on the target shirt, the shirts in different position division areas have different shapes and sizes, different pattern extension lengths, and even some pattern lines are distorted, which makes it difficult to detect the pattern features of the shirts as expected. By locating, marking and disassembling the pattern lines at different positions of the shirt, the line features in the shirt are sorted out, and the features corresponding to the specific parameters of all the pattern lines are arranged and combined, which effectively reduces the difficulty of detection caused by the irregularity of the line features.

[0072] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0073] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An e-commerce clothing sales value evaluation system based on cloud computing, characterized by: It includes a data acquisition module, a sampling module, a quality assessment module and an output module. The data acquisition module is used to obtain relevant information about shirts sold in shops; the sampling module is used to obtain relevant image features of target shirts through a detector; the quality assessment module is used to perform feature detection on the target shirt and obtain relevant parameter information of the shirt pattern; the output module is used to output the detection result of the shirt design scheme, and the data acquisition module, the sampling module, the quality assessment module and the output module are connected to each other in communication; The data acquisition module includes a shirt information acquisition module and a detector information acquisition module, wherein the shirt information acquisition module is used to acquire relevant information of shirt production; and the detector information acquisition module is used to acquire detector information for detecting shirt features; The quality assessment module includes a pattern overlap detection module and a feature overlap detection module. The pattern overlap detection module is used to detect the overlap of the patterns set on the shirts for sale; the feature overlap detection module is used to detect whether feature overlap occurs in the target shirts. The pattern overlap detection module further includes an arrangement analysis submodule and a line detection submodule, wherein the arrangement analysis submodule is used to analyze the overall arrangement of patterns on the shirts for sale; The line detection submodule is used to detect the characteristic information presented by the pattern arrangement of the shirt through the slope and concave-convex degree of the pattern lines on the selling shirt; The operation method of the sales value evaluation system mainly includes the following steps: Step S1: extracting clothing information authorized online by clothing manufacturers in the shirt purchase and sales testing software connected to the sales value evaluation system, and extracting the maximum and minimum information in the clothing size information through data analysis and processing, and obtaining the distance range from the left cuff to the right cuff of the target shirt to be tested as [a1, a2], and the distance range from the shirt collar to the bottom button of the shirt as [b1, b2]; Step S2: placing the produced garment on the detection instrument and analyzing the shirt style image captured by the camera; Step S3: obtaining the lines of the pattern on the shirt in the detection instrument, calculating the overlap rate of the pattern on the current production clothing, and judging whether the lines of the pattern of the currently selected sampled shirt have feature overlap; Step S4: testing different sampled shirts in turn, and outputting clothing solutions with different clothing but overlapping style features through an output module; It is characterized in that: the step S3 further comprises: Step S31: infrared detection of the features of the shirts stacked on the detector loading platform is performed by infrared rays set in the camera, and the initial point coordinates (X1, Y1) and the end point coordinates (X2, Y2) of the pattern lines arranged on each single-layer shirt in the coordinate system are identified by image feature analysis technology, and the number of overlaps at the end of each line is obtained; Step S32: Arrange the coordinate points in descending order according to the number of overlaps, extract the two coordinates with the largest number of overlaps, the two coordinates with the smallest number of overlaps, and S coordinates with the number of overlaps at the midpoint of the sequence, wherein the number of coordinate points to be extracted at the midpoint of the sequence is determined by the total number of coordinate points. When n is an odd number, the number of coordinate points to be extracted is 1; when n is an even number, the number of coordinate points to be extracted is 2; Step S33: Obtain the slopes of the (K+1) pattern lines corresponding to each extracted coordinate in turn. Determine the slope difference between the two pattern lines among the slopes of the pattern lines corresponding to the five extracted coordinate points Wherein L1 and L2 are the corresponding slopes of two lines randomly selected from the pattern lines corresponding to the five coordinate points. When (100%-μ%)<η<(100%+μ%), it is judged that the currently selected pattern lines have feature overlap, where μ is the maximum error limit ratio value of the feature overlap corresponding to the slope difference rate.

2. The cloud computing-based e-commerce clothing sales value evaluation system according to claim 1 is characterized in that: In step S31, the method for identifying the pattern lines arranged on each single-layer shirt mainly includes the following steps: Step S311: In each shirt inner pattern, the screen area is cut into k1*k2 rectangular small detection areas, where k1 and k2 are the corresponding numbers of the length and width of the screen area, respectively, so that the area of ​​each small detection area is the same, and n areas with different colors in the small detection area are respectively divided into "first pattern area", "second pattern area" ... "nth pattern area", where the dividing boundary between two adjacent areas is the pattern line, and the pattern lines dividing each area are marked as "first x1 line", ... "nth pattern area". G X H Lines", where X1, X G and X H are the numbers corresponding to two adjacent divided areas within the monitoring area, X1, X G and X H are all integers, min{X1, X K , X G }≥1, max{X1, X K , X G }≤n; Step S312: Mark the initial position of the pattern line in the coordinate system (X1, Y1) and the end point coordinate (X2, Y2) in the pattern line divided into each area, and obtain the initial position coordinate point (X1, Y1) of the pattern line in the small detection area in turn. 11 , Y 11 ), (X 12 , Y 12 )……(X 1n , Y 1n ) and the end point coordinates (X 21 , Y 21 ), (X 22 , Y 22 )……(X 2n , Y 2n ), the number of times each coordinate point overlaps with other initial position coordinate points and end position coordinate points is K, then there are "(K+1)" coordinate points intersecting at this coordinate point.

Images