Garment retail method and system based on artificial intelligence

Abstract

The invention discloses a clothing retail method and system based on artificial intelligence, and the method comprises the following steps: scanning the edge feature node of an image where clothing of each size is located, and determining the first incoming quantity of each size of the style according to the season and clothing attribute adaptation degree; analyzing a first influence coefficient of the temperature, holidays and combinations on the clothing sales efficiency, obtaining a corresponding sales attribute according to the change rate of the dehydrated clothing sales, obtaining an estimated sales according to the attribute and a second influence coefficient of an estimated sales day number, and obtaining a smooth or unsalable condition of the clothing according to the proportion of the sales to the inventory; analyzing the maximum sales days of the clothes according to the condition of sales without sale, and when the replenishment period is too long and a stockout phenomenon occurs, selecting the unsalable clothes of an adjacent shop to carry out goods transfer; the stain, component falling and damage conditions of the returned clothes are analyzed, the reuse condition of the returned clothes is obtained, the reused clothes are counted into a stock, and the method has the advantages of improving accuracy and high efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent retail technology, specifically to an artificial intelligence-based method and system for apparel retail. Background Art

[0002] In today's technologically advanced society, online shopping is a major consumption scenario. Clothing is a commodity that is greatly affected by seasons and timeliness. Clothing retail methods are often used to conduct scenario assessments such as online clothing purchase forecasting and replenishment to help shops and manufacturers better create economic benefits and prevent problems such as unsold inventory or waste of funds.

[0003] Current technologies involve scanning flat-laid garments and matching corresponding parameters for ordering, or directly using parameters provided by the manufacturer for labeling. However, differences in machines and production lines among manufacturers lead to misjudgments and incompatibility with target audiences, potentially resulting in large quantities of unsold clothing and economic losses after ordering. Existing technologies use one-way analysis methods such as sales change rates or exposure rates to estimate sales volume and perform replenishment operations. However, clothing sales are easily affected by weather, holidays, and the exposure rate of clothing combinations in stores. Existing technologies are not comprehensive enough, leading to misjudgments of sales. This results in significant discrepancies between ordering and actual sales after seasonal changes or holiday peaks. Furthermore, holidays or sudden clothing bestsellers cannot be predicted or planned for in advance, easily leading to stockouts of popular items, customer loss, and low consumer satisfaction.

[0004] Therefore, it is necessary to design an AI-based apparel retail method and system that can more accurately predict apparel sales inventory and purchase quantities and improve the timeliness of apparel sales. Summary of the Invention

[0005] The purpose of this invention is to provide an artificial intelligence-based method and system for apparel retail, in order to solve the problems mentioned in the background art.

[0006] To address the aforementioned technical problems, the present invention provides the following technical solution: an artificial intelligence-based apparel retail method, the method comprising the following steps:

[0007] Scan the edge feature nodes of the image containing each size of clothing, match the corresponding customer body shape, and determine the first order quantity for each size of the style based on the season and the suitability of the clothing attributes.

[0008] The first influence coefficient of temperature, holidays and combination on clothing sales efficiency is analyzed to obtain the daily sales volume of dehydrated clothing after removing the first influence. The sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combination. The corresponding sales attributes are obtained based on the rate of change of dehydrated clothing sales. The estimated sales volume is obtained based on the second influence coefficient of the attributes and the estimated number of sales days. The clothing best-selling, flat-selling or slow-selling status is obtained based on the proportion of sales volume to inventory.

[0009] Based on the analysis of best-selling and average-selling conditions, determine the maximum number of days the clothing can be sold, obtain the corresponding replenishment quantity, mark the quantity difference of slow-moving items as items to be transferred, and select slow-moving clothing from adjacent stores for transfer when the replenishment cycle is too long and stockouts occur.

[0010] Analyze the stains, missing parts, and damage of returned clothing to determine its reusability, and add the reusable clothing to the inventory.

[0011] According to the above technical solution, the step of scanning the edge feature nodes of the image containing each size of clothing, matching the corresponding customer body shape, and determining the first order quantity for each size of the style based on the suitability of the season and clothing attributes includes:

[0012] Obtain multi-angle visual images of the sample garment, scan the edge feature nodes of the garment, overlap and stitch the same feature nodes to obtain the first 3D model of the garment, obtain the total distance from the same feature node to any remaining feature node, and mark the feature nodes whose total distance exceeds the threshold as contour turning points.

[0013] Based on the material of the clothing fabric, the corresponding elastic coefficient is obtained. The first clothing 3D model is stretched and simulated. The straight-line distance between the contour turning point and the adjacent feature node is required to be within a threshold. Multiple stretched second clothing 3D models are obtained under the elastic coefficient. The second clothing 3D model is illuminated by a laser. The sample clothing 3D model is marked with the average angle of the reflection trajectory within a set threshold. The sample clothing 3D model is curve fitted according to the clothing type. The curve is required to be exactly tangent to the nearest clothing feature node. The parameter girth corresponding to the curve is obtained according to the preset relationship.

[0014] Match the N sizes of the same style of clothing with the discrete circumference parameters of the target group of the clothing, where N is a constant greater than 0. Obtain the proportion of the discrete data of the parameter circumference range corresponding to a single size to the total discrete data and mark it as a proportion coefficient. Retrieve the weather conditions after the clothing is put on the shelves to obtain the correlation coefficient between the estimated clothing put-on time and the weather. Mark the product of the estimated total purchase quantity and the correlation coefficient as the first total purchase quantity. Obtain the first purchase quantity of each size of the style based on the product of the first total purchase quantity and the proportion coefficient.

[0015] According to the above technical solution, the analysis of the first impact coefficient of temperature, holidays, and combinations on clothing sales efficiency yields the daily sales volume of dehydrated clothing after removing the first impact. The sales volume of dehydrated clothing is the sales volume of clothing unaffected by temperature, holidays, and combinations, including:

[0016] Obtain seasonal and meteorological data, determine the suitable temperature range for clothing based on style and thickness, determine the outside temperature range for a single number of days on the shelves, and mark the proportion of overlapping temperatures in the suitable temperature range of clothing as the first temperature influence coefficient α.

[0017] The ratio of total store pageviews during holidays and vacations to the average total pageviews before and after the holidays is denoted as the first holiday impact coefficient β;

[0018] Obtain the number of combinations corresponding to all combination forms including clothing, obtain the number of orders for each combination category, set the corresponding combination weight according to the number of orders for a single combination and sum them up to obtain the first combination influence coefficient γ;

[0019] Calculate the daily sales volume of dehydrated clothing after it is listed using the formula: (α*β*γ)*S p =S r Where α, β, and γ are the coefficients of the first influence of temperature, holiday, and combination on clothing sales, respectively, and S r This represents the actual daily sales volume of clothing.

[0020] According to the above technical solution, the step of obtaining corresponding sales attributes based on the rate of change in dehydrated garment sales, obtaining estimated sales volume based on the attributes and the second influence coefficient of the estimated sales days, and obtaining the garment's best-selling, average-selling, or slow-moving status based on the proportion of sales volume to inventory includes:

[0021] The shelf life is M days, based on the sales volume of dehydrated clothing (S). p Create a line chart to obtain the daily sales change rate. When the sales change rate is positive, mark the sales attribute as a growth period and obtain the location of the historical highest change rate during the growth period. If the historical highest change rate occurs on day M-1, the estimated change rate for the remaining days is expected to first rise to a predetermined threshold and then decrease. Conversely, the estimated change rate for the remaining days is expected to continue to decrease. When the absolute value of the sales change rate fluctuates within a specific threshold, mark the sales attribute as a stable period and calculate the average sales change rate within the stable period. When the sales change rate is negative, mark the sales attribute as a decay period and obtain the decay degree based on the daily change rate value. The estimated sales for the remaining days is expected to continuously decrease based on the decay degree.

[0022] Obtain the first sales day corresponding to the first purchase quantity, mark the difference between the first sales day and the number of days already sold as the second sales day, obtain the sales volume attribute based on the sales volume change rate, and obtain the estimated sales volume for each day in the second sales day based on the corresponding correlation coefficient of the attribute.

[0023] Obtain the second impact coefficient of temperature, holiday and combination on clothing sales during the second sales period, and obtain the estimated second sales volume; obtain the proportion of the estimated second sales volume to the remaining inventory, and mark the clothing as slow-moving when the proportion is less than a set threshold; mark the clothing as flat-moving when the proportion is within the set threshold range; mark the clothing as fast-moving when the proportion exceeds the set threshold.

[0024] According to the above technical solution, the step of analyzing the maximum number of sales days for clothing based on best-selling and average-selling conditions to obtain the corresponding replenishment quantity includes:

[0025] When clothing is in a flat or hot-selling phase, determine the current cycle of the clothing and establish a clothing sales model. When clothing is in a growth phase, determine the date of the last day of the growth phase based on the sales growth rate. Based on the order quantity corresponding to the date, obtain the number of days and sales quantity of the corresponding stable and declining phases.

[0026] When the clothing is in a stable period, when the absolute value of the estimated rate of change in the stable period approaches the set minimum threshold, this is marked as the last day of the stable period. The number of days of the decline period and the sales volume are obtained according to the order quantity corresponding to the date.

[0027] When clothing is in the decline period, the sales change during the decline period is estimated based on the sales change rate. When the sales volume corresponding to the estimated sales change rate during the decline period reaches the lowest threshold, this is marked as the last day of the decline period, and the number of days the decline period lasts is obtained.

[0028] Based on the daily sales changes of clothing, a product sales model is established to estimate the maximum number of days the clothing can be sold, and the total number of clothing orders corresponding to the maximum number of days of sales is determined. When the difference between the number of orders and the inventory is positive, a replenishment request is initiated, and the replenishment quantity is the difference between the total number of clothing orders and the inventory.

[0029] According to the above technical solution, marking the quantity difference of unsold stock as goods to be transferred, and selecting unsold clothing from adjacent stores for transfer when the replenishment cycle is too long and stockouts occur, includes:

[0030] When clothing sales are slow, the difference between the estimated second sales volume and the remaining inventory is marked as the pending transfer difference. The difference between the remaining inventory quantity and the estimated total sales volume within the replenishment cycle is obtained. When the difference is negative and the absolute value accounts for a proportion of the inventory quantity exceeding a threshold, all plans to transfer clothing from the store with the pending transfer difference to the store with the negative difference are obtained. Transfer plans with a arrival time later than the arrival time of the replenishment batch are eliminated, and the transfer plan with the earliest arrival time is selected for transfer.

[0031] According to the above technical solution, the analysis of stains, missing parts, and damage on returned clothing to obtain information on the reusability of the returned clothing, and the inclusion of reusable clothing in inventory, includes:

[0032] Mark returned products that have arrived in inventory, remove the number of returned garments from historical sales, irradiate the 3D model of the returned garments with a laser, compare the laser reflection marks with the 3D model of the standard garments, mark the edge feature nodes where abnormal reflection trajectories are located as abnormal points, and obtain the segmentation image of the abnormal points.

[0033] Identify abnormal pixels in the segmented image, identify the number of abnormal pixels where the stain is located, and mark the garment as reusable when the number of stain pixels is within a threshold, otherwise mark the garment as invalid.

[0034] When a part falls off the garment in the image, the detached part is retrieved. If the detached part is a replaceable part, the garment is marked as reusable; otherwise, the garment is marked as invalid.

[0035] When the distance between feature nodes of clothing in the image exceeds a threshold, the clothing is marked as damaged. The degree of damage is obtained according to the value of the distance exceeding the threshold. When the degree of damage is within the threshold, the clothing is marked as reused; otherwise, the clothing is marked as invalid.

[0036] The proportion of reusable garments to returned garments is marked as the reuse rate, the proportion of returned orders to all orders is marked as the return rate, the quantity of reusable garments before the next replenishment is included in the inventory, the defect rate of the corresponding style of garment is obtained based on the return rate and reuse rate, and the defect rate is included in the replenishment data.

[0037] An artificial intelligence-based apparel retail system, the system comprising:

[0038] The size matching module is used to scan the edge feature nodes of the image where each size of clothing is located, match the corresponding customer body shape, and determine the first purchase quantity of each size of the style based on the season and the suitability of the clothing attributes.

[0039] The Influence Coefficient module is used to analyze the first influence coefficient of temperature, holidays and combinations on clothing sales efficiency, obtain the daily sales volume of dehydrated clothing after removing the first influence, the sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations, obtain the corresponding sales attributes based on the rate of change of dehydrated clothing sales, obtain the estimated sales volume based on the attributes and the second influence coefficient of the estimated number of sales days, and obtain the clothing best-selling, flat-selling or slow-selling status based on the proportion of sales volume to inventory.

[0040] The replenishment and inventory adjustment module is used to analyze the first impact coefficient of temperature, holidays and combinations on clothing sales efficiency, obtain the daily sales volume of dehydrated clothing after removing the first impact, the sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations, obtain the corresponding sales attributes based on the rate of change of dehydrated clothing sales, obtain the estimated sales volume based on the attributes and the second impact coefficient of the estimated number of sales days, and obtain the clothing best-selling, flat-selling or slow-selling status based on the proportion of sales volume to inventory.

[0041] The reuse detection module is used to analyze the stains, missing parts, and damage of returned clothing to determine the reuse status of the returned clothing and to add the reused clothing to the inventory.

[0042] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it causes the electronic device to perform the method described in the first aspect of this application.

[0043] A fourth aspect of this application provides a computer-readable storage medium for storing a computer program that, when run on a computer, causes the computer to perform the method described in the first aspect of this application.

[0044] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: This invention establishes a three-dimensional model of the garment after scanning the sample garment image. Based on the garment's elasticity coefficient, the model is simulated and stretched to calculate the corresponding circumference parameters of the sample garment. These parameters are then matched with the discrete circumference of the target population and seasonal correlation to obtain the corresponding purchase quantity for each size. This more accurately and efficiently completes the calculation of purchase quantity and size, improving the matching degree between purchased garments and target consumers. Furthermore, by analyzing the impact of three parameters on sales—weather, holidays, and the number of items on the shelves—a more comprehensive analysis of sales changes is achieved. This allows for more accurate sales forecasting and determination of best-selling, average-selling, or slow-moving items. Based on these factors, the quantity of the next batch of replenishment is calculated. When garments are slow-moving, they are transferred to adjacent stores experiencing stock shortages, effectively solving the economic losses caused by slow-moving stockpiles and best-selling items being out of stock, and improving customer satisfaction. Attached Figure Description

[0045] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0046] Figure 1 This is a flowchart of the steps of an artificial intelligence-based apparel retail method provided in Embodiment 1 of the present invention;

[0047] Figure 2 This is a schematic diagram of the system modules for clothing retail based on artificial intelligence, provided in Embodiment 2 of the present invention.

[0048] Figure 3 This is a schematic diagram of an electronic device according to an embodiment of this application. DETAILED DESCRIPTION

[0049] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0050] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0052] Please see Figure 1 This invention provides a technical solution: an artificial intelligence-based apparel retail method, comprising:

[0053] Scan the edge feature nodes of the image containing each size of clothing, match the corresponding customer body shape, and determine the first order quantity for each size of the style based on the season and the suitability of the clothing attributes.

[0054] The first influence coefficient of temperature, holidays and combination on clothing sales efficiency is analyzed to obtain the daily sales volume of dehydrated clothing after removing the first influence. The sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combination. The corresponding sales attributes are obtained based on the rate of change of dehydrated clothing sales. The estimated sales volume is obtained based on the second influence coefficient of the attributes and the estimated number of sales days. The clothing best-selling, flat-selling or slow-selling status is obtained based on the proportion of sales volume to inventory.

[0055] Based on the analysis of best-selling and average-selling conditions, determine the maximum number of days the clothing can be sold, obtain the corresponding replenishment quantity, mark the quantity difference of slow-moving items as items to be transferred, and select slow-moving clothing from adjacent stores for transfer when the replenishment cycle is too long and stockouts occur.

[0056] Analyze the stains, missing parts, and damage of returned clothing to determine its reusability, and add the reusable clothing to the inventory.

[0057] This invention monitors carbon particle concentration data in shopping mall areas in real time and issues an alarm upon detecting abnormal data. It also analyzes the intensity of flame combustion to determine the specific location of the fire source, enabling more efficient, rapid, and accurate fire monitoring, alarm, and fire source location identification, thus improving fire early warning efficiency. After a fire occurs in the mall, it identifies the location of trapped personnel and safety exits, monitors the hazard level of flammable materials, and plans evacuation routes, effectively monitoring the fire while efficiently planning evacuation paths. Furthermore, it identifies key features of doors, establishes door models, and analyzes the door models to determine if the paths around the doors are passable, obtaining new evacuation routes and effectively preventing situations where doors cannot be opened, thereby improving evacuation efficiency.

[0058] In some preferred embodiments, the edge feature nodes of the image where each size of clothing is located are scanned to match the corresponding customer body shape. The first purchase quantity of each size of the style is determined according to the season and the suitability of the clothing attributes. The process further includes the following steps: obtaining a multi-angle visual image of the sample clothing, scanning the edge feature nodes of the clothing, overlapping and splicing the same feature nodes to obtain a first clothing three-dimensional model, obtaining the total distance from the same feature node to any remaining feature node, and marking feature nodes whose total distance exceeds a threshold as contour turning points.

[0059] Based on the material of the clothing fabric, the corresponding elastic coefficient is obtained. The first clothing 3D model is stretched and simulated. The straight-line distance between the contour turning point and the adjacent feature node is required to be within the threshold. Multiple stretched second clothing 3D models under the elastic coefficient are obtained. The second clothing 3D model is illuminated by a laser. The average angle of the reflection trajectory within the set threshold is marked as the sample clothing 3D model. The sample clothing 3D model is curve fitted according to the clothing type. The curve is required to be exactly tangent to the feature node closest to the clothing. The parameter girth corresponding to the curve is obtained according to the preset relationship.

[0060] Match the N sizes of the same style of clothing with the discrete circumference parameters of the target group of the clothing, where N is a constant greater than 0. Obtain the proportion of the discrete data of the parameter circumference range corresponding to a single size to the total discrete data and mark it as a proportion coefficient. Retrieve the weather conditions after the clothing is put on the shelves to obtain the correlation coefficient between the estimated clothing put-on time and the weather. Mark the product of the estimated total purchase quantity and the correlation coefficient as the first total purchase quantity. Obtain the first purchase quantity of each size of the style based on the product of the first total purchase quantity and the proportion coefficient.

[0061] In some preferred embodiments, the first influence coefficient of temperature, holidays and combinations on clothing sales efficiency is analyzed to obtain the daily sales volume of dehydrated clothing after removing the first influence. The sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations. The process further includes the following steps: obtaining seasonal and meteorological data, obtaining the suitable temperature range of clothing based on the style and thickness of clothing, obtaining the range of external temperature under a single number of days on the shelves, and obtaining the proportion of overlapping temperatures to the suitable temperature range of clothing, which is marked as the first temperature influence coefficient α.

[0062] The ratio of total store pageviews during holidays and vacations to the average total pageviews before and after the holidays is denoted as the first holiday impact coefficient β;

[0063] Obtain the number of combinations corresponding to all combination forms including clothing, obtain the number of orders for each combination category, set the corresponding combination weight according to the number of orders for a single combination and sum them up to obtain the first combination influence coefficient γ;

[0064] Calculate the daily sales volume of dehydrated clothing after it is listed using the formula: (α*β*γ)*S p =S r Where α, β, and γ are the coefficients of the first influence of temperature, holiday, and combination on clothing sales, respectively, and S r This represents the actual daily sales volume of clothing.

[0065] In some preferred embodiments, the corresponding sales attribute is obtained based on the rate of change in the sales volume of dehydrated garments; the estimated sales volume is obtained based on the attribute and a second influence coefficient of the estimated sales days; and the status of garments as best-selling, average-selling, or slow-moving is determined based on the proportion of sales volume to inventory. This further includes the following steps: the shelf life is M days, and the sales volume of dehydrated garments is S... pCreate a line chart to obtain the daily sales change rate. When the sales change rate is positive, mark the sales attribute as a growth period and obtain the location of the historical highest change rate during the growth period. If the historical highest change rate occurs on day M-1, the estimated change rate for the remaining days is expected to first rise to a predetermined threshold and then decrease. Conversely, the estimated change rate for the remaining days is expected to continue to decrease. When the absolute value of the sales change rate fluctuates within a specific threshold, mark the sales attribute as a stable period and calculate the average sales change rate within the stable period. When the sales change rate is negative, mark the sales attribute as a decay period and obtain the decay degree based on the daily change rate values. The estimated sales for the remaining days is expected to decrease continuously based on the decay degree.

[0066] Obtain the first sales day corresponding to the first purchase quantity, mark the difference between the first sales day and the number of days already sold as the second sales day, obtain the sales volume attribute based on the sales volume change rate, and obtain the estimated sales volume for each day in the second sales day based on the corresponding correlation coefficient of the attribute.

[0067] Obtain the second impact coefficient of temperature, holiday and combination on clothing sales during the second sales period, and obtain the estimated second sales volume; obtain the proportion of the estimated second sales volume to the remaining inventory, and mark the clothing as slow-moving when the proportion is less than a set threshold; mark the clothing as flat-moving when the proportion is within the set threshold range; mark the clothing as fast-moving when the proportion exceeds the set threshold.

[0068] In some preferred embodiments, the maximum number of sales days for clothing is analyzed based on the best-selling and average-selling conditions to obtain the corresponding replenishment quantity. The process further includes the following steps: when clothing is selling at an average or best-selling level, the cycle to which the clothing belongs is determined and a clothing sales model is established. When clothing is in a growth period, the date of the last day of the growth period is determined based on the sales growth rate. The number of days and sales quantity of the corresponding stable period and decline period are obtained based on the order quantity corresponding to the date.

[0069] When the clothing is in a stable period, when the absolute value of the estimated rate of change in the stable period approaches the set minimum threshold, this is marked as the last day of the stable period. The number of days of the decline period and the sales volume are obtained according to the order quantity corresponding to the date.

[0070] When clothing is in the decline period, the sales change during the decline period is estimated based on the sales change rate. When the sales volume corresponding to the estimated sales change rate during the decline period reaches the lowest threshold, this is marked as the last day of the decline period, and the number of days the decline period lasts is obtained.

[0071] Based on the daily sales changes of clothing, a product sales model is established to estimate the maximum number of days the clothing can be sold, and the total number of clothing orders corresponding to the maximum number of days of sales is determined. When the difference between the number of orders and the inventory is positive, a replenishment request is initiated. The replenishment quantity is the difference between the total number of clothing orders and the inventory.

[0072] In some preferred embodiments, the quantity of unsold goods is marked as goods to be transferred. When a stockout occurs due to an excessively long replenishment cycle, unsold clothing from adjacent stores is selected for transfer. The process further includes the following steps: when clothing is unsold, the difference between the estimated second sales volume and the remaining inventory is marked as the difference to be transferred. The difference between the remaining inventory and the estimated total sales volume within the replenishment cycle is obtained. When the difference is negative and the absolute value accounts for a proportion of the inventory exceeding a threshold, all options for transferring clothing from the store with the difference to the store with the negative difference are obtained. Options for transferring goods that arrive later than the replenishment batch arrival time are eliminated, and the option with the earliest arrival time is selected for transfer.

[0073] In some preferred embodiments, the analysis of stains, missing parts, and damage of returned garments is used to obtain information on the reuse of returned garments. The reused garments are then included in the inventory. The process further includes the following steps: marking returned products that have arrived in the inventory, removing the number of returned garments from the historical sales volume, irradiating a 3D model of the returned garments with a laser, comparing the laser reflection marks with those of a standard 3D model of garments, marking the edge feature nodes where abnormal reflection trajectories are located as abnormal points, and obtaining a segmented image of the abnormal points.

[0074] Identify abnormal pixels in the segmented image, identify the number of abnormal pixels where stains are located, and mark the garment as reusable when the number of stain pixels is within a threshold, otherwise mark the garment as invalid.

[0075] When a part falls off a garment in an image, the system retrieves the detached part. If the detached part is replaceable, the garment is marked as reusable; otherwise, it is marked as unusable.

[0076] When the distance between feature nodes of clothing in an image exceeds a threshold, the clothing is marked as damaged. The degree of damage is obtained based on the distance value. When the degree of damage is within the threshold, the clothing is marked as reused; otherwise, the clothing is marked as invalid.

[0077] The proportion of reusable garments to returned garments is marked as the reuse rate, the proportion of returned orders to all orders is marked as the return rate, the quantity of reusable garments before the next replenishment is included in the inventory, and the defect rate of the corresponding garment style is obtained based on the return rate and reuse rate, and the defect rate is included in the replenishment data.

[0078] After being added to inventory, the garments are scored, and various return characteristics such as damage, unraveling, and peeling are categorized. Weights are assigned to each characteristic, and the ratio of return characteristics to total sales is obtained and fed back to the manufacturer. All goods provided by the manufacturer are evaluated, and goods from different manufacturers are categorized and analyzed. The similarity of materials and edge feature nodes of the same type of garments is analyzed, and those with a similarity of more than 80% are categorized. The defect rate of the same type of garment from different manufacturers is compared horizontally, and the garment with the highest profit margin under the same purchase cost is evaluated. Other garments are eliminated, and the garments selected for the next batch of purchases are selected.

[0079] Please see Figure 2 Similar to the embodiments described above, this application also provides an artificial intelligence-based apparel retail system, comprising:

[0080] The size matching module is used to scan the edge feature nodes of the image where each size of clothing is located, match the corresponding customer body shape, and determine the first purchase quantity of each size of the style based on the season and the suitability of the clothing attributes.

[0081] The Influence Coefficient module is used to analyze the first influence coefficient of temperature, holidays and combinations on clothing sales efficiency, obtain the daily sales volume of dehydrated clothing after removing the first influence, the sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations, obtain the corresponding sales attributes based on the rate of change of dehydrated clothing sales, obtain the estimated sales volume based on the attributes and the second influence coefficient of the estimated number of sales days, and obtain the clothing best-selling, flat-selling or slow-selling status based on the proportion of sales volume to inventory.

[0082] The replenishment and inventory adjustment module is used to analyze the first impact coefficient of temperature, holidays and combinations on clothing sales efficiency, obtain the daily sales volume of dehydrated clothing after removing the first impact, the sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations, obtain the corresponding sales attributes based on the rate of change of dehydrated clothing sales, obtain the estimated sales volume based on the attributes and the second impact coefficient of the estimated number of sales days, and obtain the clothing best-selling, flat-selling or slow-selling status based on the proportion of sales volume to inventory.

[0083] The reuse detection module is used to analyze the stains, missing parts, and damage of returned clothing to determine the reuse status of the returned clothing and to add the reused clothing to the inventory.

[0084] It should be noted that although several units or sub-units of the device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units.

[0085] Based on the same inventive concept as the above method embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it enables the electronic device to implement the control method described in the above embodiments.

[0086] In one embodiment, the electronic device may be a server, and in this embodiment, the structure of the electronic device may be as follows: Figure 3 As shown, it includes a memory 2001, a communication module 2003, and one or more processors 2002.

[0087] The memory 2001 is used to store computer programs executed by the processor 2002. The memory 2001 may mainly include a program storage area and a data storage area. The program storage area may store the operating system and programs required to run instant messaging functions, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.

[0088] Memory 2001 may be volatile memory, such as random-access memory (RAM); memory 2001 may also be non-volatile memory, such as read-only memory, flash memory, hard disk drive (HDD), or solid-state drive (SSD); or memory 2001 may be any other medium capable of carrying or storing a desired computer program having the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 2001 may be a combination of the above-mentioned memories.

[0089] Processor 2002 may include one or more central processing units (CPUs) or digital processing units, etc. Processor 2002 is used to implement the above-mentioned audio data processing method when calling computer programs stored in memory 2001.

[0090] The communication module 2003 is used to communicate with terminal devices and other servers.

[0091] This application embodiment does not limit the specific connection medium between the memory 2001, communication module 2003, and processor 2002. This application embodiment... Figure 3 The memory 2001 and the processor 2002 are connected via a bus 2004, which is in... Figure 3The connections between other components are illustrated with arrows and are for illustrative purposes only, not as limiting information. The Bus 2004 can be divided into address bus, data bus, control bus, etc. For ease of description, Figure 3 The text uses only one arrow to describe it, but does not indicate that there is only one bus or one type of bus.

[0092] Based on the same inventive concept as the above-described method embodiments, embodiments of the present invention also provide a computer-readable storage medium for storing a computer program. When the computer program is run on a computer, it enables the electronic device to implement the control method described in the above embodiments. The computer-readable storage medium can be a readable signal medium or a readable storage medium. A readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0093] Based on the same inventive concept as the above-described method embodiments, embodiments of the present invention also provide a computer program product, which includes a computer program that, when run on an electronic device, causes the electronic device to perform the steps of the control methods described above according to various exemplary embodiments of this application. The program product may take the form of any combination of one or more readable media. These computer program commands can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the commands executed by the processor of the computer or other programmable data processing device generate a process for implementing... Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0094] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application. It should be noted that, in this document, relational terms such as "first" and "second" are used merely 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 "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0095] Finally, it should be noted that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An artificial intelligence-based method for apparel retail, characterized in that: The method includes the following steps: Scan the edge feature nodes of the image containing each size of clothing, match the corresponding customer body shape, and determine the first order quantity for each size of the style based on the season and the suitability of the clothing attributes. The first influence coefficient of temperature, holidays and combination on clothing sales efficiency is analyzed to obtain the daily sales volume of dehydrated clothing after removing the first influence. The sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combination. The corresponding sales attributes are obtained based on the rate of change of dehydrated clothing sales. The estimated sales volume is obtained based on the second influence coefficient of the attributes and the estimated number of sales days. The clothing best-selling, flat-selling or slow-selling status is obtained based on the proportion of sales volume to inventory. Based on the analysis of best-selling and average-selling conditions, determine the maximum number of days the clothing can be sold, obtain the corresponding replenishment quantity, mark the quantity difference of slow-moving items as items to be transferred, and select slow-moving clothing from adjacent stores for transfer when the replenishment cycle is too long and stockouts occur. Analyze the stains, missing parts, and damage of returned clothing to determine its reusability, and add the reusable clothing to the inventory.

2. The apparel retail method based on artificial intelligence according to claim 1, characterized in that: The process involves scanning the edge feature nodes of the image containing each size of clothing, matching the corresponding customer body shape, and determining the initial order quantity for each size of the style based on the season and clothing attribute suitability. This includes: Obtain multi-angle visual images of the sample garment, scan the edge feature nodes of the garment, overlap and stitch the same feature nodes to obtain the first 3D model of the garment, obtain the total distance from the same feature node to any remaining feature node, and mark the feature nodes whose total distance exceeds the threshold as contour turning points. Based on the material of the clothing fabric, the corresponding elastic coefficient is obtained. The first clothing 3D model is stretched and simulated. The straight-line distance between the contour turning point and the adjacent feature node is required to be within a threshold. Multiple stretched second clothing 3D models are obtained under the elastic coefficient. The second clothing 3D model is illuminated by a laser. The sample clothing 3D model is marked with the average angle of the reflection trajectory within a set threshold. The sample clothing 3D model is curve fitted according to the clothing type. The curve is required to be exactly tangent to the nearest clothing feature node. The parameter girth corresponding to the curve is obtained according to the preset relationship. Match the N sizes of the same style of clothing with the discrete circumference parameters of the target group of the clothing, where N is a constant greater than 0. Obtain the proportion of the discrete data of the parameter circumference range corresponding to a single size to the total discrete data and mark it as a proportion coefficient. Retrieve the weather conditions after the clothing is put on the shelves to obtain the correlation coefficient between the estimated clothing put-on time and the weather. Mark the product of the estimated total purchase quantity and the correlation coefficient as the first total purchase quantity. Obtain the first purchase quantity of each size of the style based on the product of the first total purchase quantity and the proportion coefficient.

3. The apparel retail method based on artificial intelligence according to claim 1, characterized in that: The analysis of the first impact coefficient of temperature, holidays, and combinations on clothing sales efficiency yields daily dehydrated clothing sales figures after removing the first impact. These dehydrated clothing sales figures are those unaffected by temperature, holidays, and combinations, and include: Obtain seasonal and meteorological data, determine the suitable temperature range for clothing based on style and thickness, determine the outside temperature range for a single number of days on the shelves, and mark the proportion of overlapping temperatures in the suitable temperature range of clothing as the first temperature influence coefficient α. The ratio of total store pageviews during holidays and vacations to the average total pageviews before and after the holidays is denoted as the first holiday impact coefficient β; Obtain the number of combinations corresponding to all combination forms including clothing, obtain the number of orders for each combination category, set the corresponding combination weight according to the number of orders for a single combination and sum them up to obtain the first combination influence coefficient γ; Calculate the daily sales volume of dehydrated clothing after it is listed using the formula: (α*β*γ)*S p =S r Where α, β, and γ are the coefficients of the first influence of temperature, holiday, and combination on clothing sales, respectively, and S r This represents the actual daily sales volume of clothing.

4. The apparel retail method based on artificial intelligence according to claim 1, characterized in that: The process involves obtaining corresponding sales attributes based on the rate of change in dehydrated garment sales, obtaining estimated sales volume based on the attributes and a second influence coefficient based on the estimated number of sales days, and determining whether the garments are best-selling, flat-selling, or slow-moving based on the proportion of sales volume to inventory. This includes: The shelf life is M days, based on the sales volume of dehydrated clothing (S). p Create a line chart to obtain the daily sales change rate. When the sales change rate is positive, mark the sales attribute as a growth period and obtain the location of the historical highest change rate during the growth period. If the historical highest change rate occurs on day M-1, the estimated change rate for the remaining days is expected to first rise to a predetermined threshold and then decrease. Conversely, the estimated change rate for the remaining days is expected to continue to decrease. When the absolute value of the sales change rate fluctuates within a specific threshold, mark the sales attribute as a stable period and calculate the average sales change rate within the stable period. When the sales change rate is negative, mark the sales attribute as a decay period and obtain the decay degree based on the daily change rate value. The estimated sales for the remaining days is expected to continuously decrease based on the decay degree. Obtain the first sales day corresponding to the first purchase quantity, mark the difference between the first sales day and the number of days already sold as the second sales day, obtain the sales volume attribute based on the sales volume change rate, and obtain the estimated sales volume for each day in the second sales day based on the corresponding correlation coefficient of the attribute. Obtain the second impact coefficient of temperature, holiday and combination on clothing sales during the second sales period, and obtain the estimated second sales volume; obtain the proportion of the estimated second sales volume to the remaining inventory, and mark the clothing as slow-moving when the proportion is less than a set threshold; mark the clothing as flat-moving when the proportion is within the set threshold range; mark the clothing as fast-moving when the proportion exceeds the set threshold.

5. The apparel retail method based on artificial intelligence according to claim 1, characterized in that: The process of analyzing the best-selling and average-selling trends of clothing to determine the maximum number of sales days and obtaining the corresponding replenishment quantity includes: When clothing is in a flat or hot-selling phase, determine the current cycle of the clothing and establish a clothing sales model. When clothing is in a growth phase, determine the date of the last day of the growth phase based on the sales growth rate. Based on the order quantity corresponding to the date, obtain the number of days and sales quantity of the corresponding stable and declining phases. When the clothing is in a stable period, when the absolute value of the estimated rate of change in the stable period approaches the set minimum threshold, this is marked as the last day of the stable period. The number of days of the decline period and the sales volume are obtained according to the order quantity corresponding to the date. When clothing is in the decline period, the sales change during the decline period is estimated based on the sales change rate. When the sales volume corresponding to the estimated sales change rate during the decline period reaches the lowest threshold, this is marked as the last day of the decline period, and the number of days the decline period lasts is obtained. Based on the daily sales changes of clothing, a product sales model is established to estimate the maximum number of days the clothing can be sold, and the total number of clothing orders corresponding to the maximum number of days of sales is determined. When the difference between the number of orders and the inventory is positive, a replenishment request is initiated, and the replenishment quantity is the difference between the total number of clothing orders and the inventory.

6. The apparel retail method based on artificial intelligence according to claim 1, characterized in that: The step of marking the unsold quantity as pending replenishment, and selecting unsold clothing from adjacent stores for replenishment when the replenishment cycle is too long and stockouts occur, includes: When clothing sales are slow, the difference between the estimated second sales volume and the remaining inventory is marked as the pending transfer difference. The difference between the remaining inventory quantity and the estimated total sales volume within the replenishment cycle is obtained. When the difference is negative and the absolute value accounts for a proportion of the inventory quantity exceeding a threshold, all plans to transfer clothing from the store with the pending transfer difference to the store with the negative difference are obtained. Transfer plans with a arrival time later than the arrival time of the replenishment batch are eliminated, and the transfer plan with the earliest arrival time is selected for transfer.

7. The apparel retail method based on artificial intelligence according to claim 1, characterized in that: The analysis of stains, missing parts, and damage on returned garments determines their reusability. Reusable garments are then added to inventory, including: Mark returned products that have arrived in inventory, remove the number of returned garments from historical sales, irradiate the 3D model of the returned garments with a laser, compare the laser reflection marks with the 3D model of the standard garments, mark the edge feature nodes where abnormal reflection trajectories are located as abnormal points, and obtain the segmentation image of the abnormal points. Identify abnormal pixels in the segmented image, identify the number of abnormal pixels where the stain is located, and mark the garment as reusable when the number of stain pixels is within a threshold, otherwise mark the garment as invalid. When a part falls off the garment in the image, the detached part is retrieved. If the detached part is a replaceable part, the garment is marked as reusable; otherwise, the garment is marked as invalid. When the distance between feature nodes of clothing in the image exceeds a threshold, the clothing is marked as damaged. The degree of damage is obtained according to the value of the distance exceeding the threshold. When the degree of damage is within the threshold, the clothing is marked as reused; otherwise, the clothing is marked as invalid. The proportion of reusable garments to returned garments is marked as the reuse rate, the proportion of returned orders to all orders is marked as the return rate, the quantity of reusable garments before the next replenishment is included in the inventory, the defect rate of the corresponding style of garment is obtained based on the return rate and reuse rate, and the defect rate is included in the replenishment data.

8. An artificial intelligence-based apparel retail system, characterized in that: The system includes: The size matching module is used to scan the edge feature nodes of the image where each size of clothing is located, match the corresponding customer body shape, and determine the first purchase quantity of each size of the style based on the season and the suitability of the clothing attributes. The Influence Coefficient module is used to analyze the first influence coefficient of temperature, holidays and combinations on clothing sales efficiency, obtain the daily sales volume of dehydrated clothing after removing the first influence, the sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations, obtain the corresponding sales attributes based on the rate of change of dehydrated clothing sales, obtain the estimated sales volume based on the attributes and the second influence coefficient of the estimated number of sales days, and obtain the clothing best-selling, flat-selling or slow-selling status based on the proportion of sales volume to inventory. The replenishment and inventory adjustment module is used to analyze the first impact coefficient of temperature, holidays and combinations on clothing sales efficiency, obtain the daily sales volume of dehydrated clothing after removing the first impact, the sales volume of dehydrated clothing is the sales volume of clothing that is not affected by temperature, holidays and combinations, obtain the corresponding sales attributes based on the rate of change of dehydrated clothing sales, obtain the estimated sales volume based on the attributes and the second impact coefficient of the estimated number of sales days, and obtain the clothing best-selling, flat-selling or slow-selling status based on the proportion of sales volume to inventory. The reuse detection module is used to analyze the stains, missing parts, and damage of returned clothing to determine the reuse status of the returned clothing and to add the reused clothing to the inventory.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the electronic device to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 7.

Images