CN111681234A - Method, system and equipment for detecting standard of trial product placed on store shelf - Google Patents
Method, system and equipment for detecting standard of trial product placed on store shelf Download PDFInfo
- Publication number
- CN111681234A CN111681234A CN202010529064.6A CN202010529064A CN111681234A CN 111681234 A CN111681234 A CN 111681234A CN 202010529064 A CN202010529064 A CN 202010529064A CN 111681234 A CN111681234 A CN 111681234A
- Authority
- CN
- China
- Prior art keywords
- shelf
- trial
- trial products
- detection model
- target detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention relates to a method, a system and equipment for detecting the specification of trial products placed on a store shelf, wherein a monitoring camera is adopted to obtain an image of the trial products placed on the shelf, the obtained image is trained through deep learning to obtain a segmentation model, pictures of a trial product placement area in the segmentation model are trained to obtain a target detection model, the image of the shelf to be detected with the specification is input into the target detection model to be subjected to specification detection and analysis, evaluation results of the number, display flatness, interval and plumpness of the trial products are output, a store manager carries out supervision and correction on the trial products on the shelf according to the result output by the target detection model, and the detection efficiency is improved. The method realizes the standard inspection of the trial products placed on the automatic detection goods shelf, improves the detection efficiency, and solves the technical problems that the prior standard inspection of the trial products in stores adopts manpower, the efficiency is low, and the replacement of the trial products by workers in time is inconvenient.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method, a system and equipment for detecting the specification of trial products placed on a store shelf.
Background
In a retail store, a layer of display board is usually arranged above a shelf and used for displaying the trial of goods on the shelf, so that customers can experience the goods, the knowledge of the customers on the goods is promoted, and the goods are better sold.
The shop usually has certain regulation to putting of using the dress, for example pleasing to the eye, evenly puts the product on probation, product quantity on probation etc.. Management of the placements of the use garments is now typically handled by the area manager. Firstly, the retail stores are large in quantity, the regulations of the trial garments are checked manually, the work is complicated, and unified and standardized management is not facilitated; secondly, the number of people in the peak period of the store is too large, which is not beneficial to the responsible person to replace or maintain the trial products in time.
Disclosure of Invention
The embodiment of the invention provides a method, a system and equipment for detecting the specification of trial products placed on a store shelf, which are used for solving the technical problems that the prior trial product specification inspection of stores is manual, the efficiency is low, and workers are inconvenient to replace the trial products in time.
In order to achieve the above object, the embodiments of the present invention provide the following technical solutions:
a method for detecting the specification of trial products placed on a store shelf comprises the following steps:
acquiring images on a trial shelf of a store in real time, and establishing an image database of a trial placement area;
deep learning training is carried out on all images in the image database to establish a segmentation model;
selecting pictures of the trial assembly and placement area from all the images by adopting the segmentation model, and establishing a target detection model by adopting deep learning training on all the selected pictures;
the method comprises the steps of obtaining a goods shelf image with standard trial products to be detected, inputting the goods shelf image into a target detection model, analyzing the number, display flatness, intervals and plumpness of the trial products in the goods shelf image by the target detection model, and outputting a result whether the trial products placed on a goods shelf are standard or not.
Preferably, the specific steps of the target detection model detecting the specifications of the trial products of the shelf image include:
calculating the number of the trial products in the goods shelf image, and if the number is smaller than a specified number threshold value, outputting the information of the shortage of the trial products on the goods shelf by the target detection model;
fitting the calculation points on the same plane of all the trial products in the goods shelf image into a fitting straight line, calculating and accumulating the distances between the calculation points of all the trial products and the fitting straight line to obtain a fitting distance value, and if the fitting distance value is larger than a set error threshold value, outputting the information of uneven placement of the trial products on the goods shelf by the target detection model;
calculating the interval distance between every two trial products in the goods shelf image, and if the interval distance is larger than a set distance threshold or a proportion threshold, outputting information that the trial products on the goods shelf do not accord with the placing interval by the target detection model;
and calculating the distance between the center of all the trial products in the goods shelf image and the goods shelf area boundary, and if the distance is greater than a set distance threshold value, outputting space information for placing the trial products on the goods shelf by the target detection model.
Preferably, a least square method is adopted to fit a fitting straight line of the trial product in the target detection model.
Preferably, the target detection model further comprises an output module which is communicated with a mobile terminal of a store manager, and the output module is used for outputting result information for detecting that the trial products placed on the shelf are not normal.
Preferably, a 360-degree ball machine is adopted to acquire images on the trial shelf of the store in real time.
The embodiment of the invention also provides a system for detecting the specification of the trial products placed on the store shelf, which comprises an image data acquisition unit, a segmentation model establishment unit, a detection model establishment unit and a detection unit;
the image data acquisition unit is used for acquiring images on a trial shelf of a store in real time and establishing an image database of a trial placement area;
the segmentation model establishing unit is used for carrying out deep learning training on all images in the image database to establish a segmentation model;
the detection model establishing unit is used for selecting pictures of the trial assembly and placement area in all the images by adopting the segmentation model and establishing a target detection model by adopting deep learning training on all the selected pictures;
the detection unit is used for acquiring a goods shelf image with standard trial products to be detected and inputting the goods shelf image into the target detection model, and the target detection model analyzes the number, display flatness, intervals and plumpness of the trial products in the goods shelf image and outputs a result of whether the trial products placed on the goods shelf are standard or not.
Preferably, the detection unit includes a first calculation and discrimination subunit, a second calculation and discrimination subunit, a third calculation and discrimination subunit, and a fourth calculation and discrimination subunit;
the first calculation judging subunit is configured to calculate the number of the trial products in the shelf image, and if the number is smaller than a specified number threshold, the target detection model outputs information that the trial products on the shelf are out of stock;
the second calculation and judgment subunit is used for fitting a fitting straight line according to the calculation points on the same plane of all the trial products in the shelf image, calculating and accumulating the distances between the calculation points of all the trial products and the fitting straight line to obtain a fitting distance value, and if the fitting distance value is greater than a set error threshold value, the target detection model outputs the information that the trial products on the shelf are placed unevenly;
the third calculation and judgment subunit is configured to calculate an interval distance between every two trial products in the shelf image, and if the interval distance is greater than a set distance threshold or a set proportion threshold, the target detection model outputs information that the trial products on the shelf do not conform to the placement interval;
and the fourth calculation and judgment subunit is used for calculating the distance between the centers of all the trial products in the shelf image and the boundary of the shelf area, and if the distance is greater than a set distance threshold, the target detection model outputs the space information for placing the trial products on the shelf.
Preferably, the second calculation and judgment subunit is further configured to fit a fitting straight line of the trial product by using a least square method.
Preferably, the system for detecting the specification of the trial products placed on the store shelf further comprises an output unit, and the output unit is used for outputting result information for detecting the irregularity of the trial products placed on the shelf.
The invention also provides a device for detecting the specification of the trial products placed on the store shelf, which comprises a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is used for executing the specification detection method for the trial products placed on the store shelf according to the instructions in the program codes.
According to the technical scheme, the embodiment of the invention has the following advantages:
1. the method for detecting the specification of the trial products on the store shelf adopts a monitoring camera to acquire the images of the trial products on the shelf, trains the acquired images through deep learning to obtain a segmentation model and trains the images of the regions of the trial products in the segmentation model to obtain a target detection model, inputs the shelf images to be detected to be specified into the target detection model to perform specification detection analysis, outputs the number of the trial products, the display flatness, the interval and the plumpness evaluation result, and a store manager supervises and modifies the trial products on the shelf according to the result output by the target detection model, thereby improving the detection efficiency. The method realizes the standard inspection of the trial products placed on the automatic detection goods shelf, improves the detection efficiency, and solves the technical problems that the prior standard inspection of the trial products in stores adopts manpower, the efficiency is low, and the replacement of the trial products by workers in time is inconvenient.
2. The system for detecting the specification of the trial products on the store shelf adopts the image data acquisition unit to acquire the images of the trial products on the shelf, the acquired images are trained through the deep learning of the segmentation model establishment unit to obtain the segmentation model, the pictures of the trial product areas in the segmentation model, which are put in the segmentation model, are trained by the detection model establishment unit to obtain the target detection model, the shelf images to be detected for specification are input into the target detection model of the detection unit to be subjected to the specification detection analysis, the number of the trial products is output, the display flatness, the interval and the plumpness evaluation result are output, a store manager supervises and modifies the trial products on the shelf according to the result output by the target detection model, and the detection efficiency is improved. The system method realizes the standard inspection of the trial products placed on the automatic detection goods shelf, improves the detection efficiency, and solves the technical problems that the prior standard inspection of the trial products in stores is manual, the efficiency is low, and the workers are not convenient to replace the trial products in time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a flowchart illustrating steps of a method for detecting a specification of a product to be placed on a store shelf according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a shelf for a specification detection method of trial products placed on a store shelf according to an embodiment of the invention.
Fig. 3 is a block diagram of a system for detecting specification of products for trial placement on a store shelf according to an embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the application provides a method, a system and equipment for detecting the standard of trial products placed on a store shelf, which are used for solving the technical problems that the conventional standard inspection of the trial products in stores is manual, the efficiency is low, and workers are not convenient to replace the trial products in time.
The first embodiment is as follows:
fig. 1 is a flowchart illustrating steps of a method for detecting a specification of a product to be placed on a store shelf according to an embodiment of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a method for detecting a specification of a sample product placed on a store shelf, including the following steps:
s1, acquiring images on a trial shelf of a store in real time, and establishing an image database of a trial placement area;
s2, carrying out deep learning training on all images in an image database to establish a segmentation model;
s3, selecting pictures in the trial assembling and placing area from all the images by adopting a segmentation model, and establishing a target detection model by adopting deep learning training on all the selected pictures;
s4, acquiring a goods shelf image with standard to-be-tested products and inputting the goods shelf image into a target detection model, analyzing the number, display flatness, intervals and plumpness of the to-be-tested products in the goods shelf image by the target detection model, and outputting a result whether the to-be-tested products placed on the goods shelf are standard or not.
In S1 in the embodiment of the present invention, a 360 ° ball machine or a high-definition wide-angle camera is used to acquire an image on a trial shelf of a store in real time. Specifically, a ball machine or a wide-angle camera for acquiring shelf pictures is installed above a store shelf (namely, on the ceiling of the store), images of the store shelf are captured in real time by rotating the ball machine or the wide-angle camera at different angles, the captured images are stored in a memory or on a server, and an image database of a shelf trial installation area is gradually established.
It should be noted that the dome camera is called as a dome camera, is a representative of the development of modern television monitoring, integrates multiple functions of a color integrated camera, a pan-tilt, a decoder, a protective cover and the like, is convenient to install, simple to use and powerful in function, and is widely applied to monitoring in an open area.
In S2 in the embodiment of the present invention, data of all images is obtained in an image database as segmentation model training samples, partial images are selected as samples of an example segmentation model, a trial product area on a shelf is labeled, and a segmentation model is established by deep learning training. As shown in fig. 2, specifically, the area a in fig. 2 is a target area, the other areas are non-target areas, the target area is a position where a trial product is placed, and the target area and the non-target areas in the image can be segmented and distinguished by a deep learning segmentation model. If several target areas exist in the image, several target areas can be divided.
It should be noted that deep learning is to learn the intrinsic rules and the expression levels of sample data, and the information obtained in these learning processes is helpful to the interpretation of data such as text, image and sound. The final aim of the method is to enable a machine to have the analysis and learning capability like a human, to recognize data such as characters, images and sounds, and the like, and the deep learning is also a complex machine learning algorithm, and the effect in the aspect of voice and image recognition is far superior to that of the prior related technology.
In S3 in the embodiment of the present invention, in S2, the obtained image is mainly distinguished into a target area and a non-target area, but in an actual situation, products placed in the target area are not only trial-use products, but also other products may be placed, and in S3, a deep learning training is performed on all images containing the target area in the segmentation model to obtain a target detection model, so that the accuracy of specification detection of the products placed on the shelf with the trial-use products is improved. Specifically, the target area picture obtained by dividing the trial product placement area in S2 is selected and stored separately, so that the picture only includes the area for placing the trial product, and then the selected picture is used as a sample of the target detection model, and deep learning training is performed on the selected picture according to the number, display flatness, interval and fullness of the trial product, so as to obtain the target detection model.
It should be noted that, all trial placement areas (i.e., target areas) in the image are segmented by using the segmentation model trained in S2 and stored as independent pictures, and only the trial placement areas are included in the pictures, and then the segmented pictures are used as samples of the target detection model.
In the embodiment of the invention, S4, the shelf image of the trial product specification to be detected is obtained by adopting the same technical means as S1, the shelf image is input to the target detection model for analysis, and the result of checking the shelf trial product placement specification is output. The output result is also transmitted to the mobile terminal of the store manager to inform the store manager, so that the store manager arranges the shop clerk to correct the trial products on the goods shelf and supervise the store manager, the efficiency is greatly improved, and the store patrol cost of the store manager is reduced.
According to the method for detecting the specification of the trial products on the shop shelf, the monitoring camera is adopted to obtain the image of the trial product on the shelf, the obtained image is trained through deep learning to obtain the segmentation model, the picture of the area where the trial products are placed in the segmentation model is trained to obtain the target detection model, the shelf image to be detected with the specification is input into the target detection model to be subjected to specification detection and analysis, the evaluation results of the number, the display flatness, the interval and the plumpness of the trial products are output, and a shop manager supervises and modifies the trial products on the shelf according to the result output by the target detection model, so that the detection efficiency is improved. The method realizes the standard inspection of the trial products placed on the automatic detection goods shelf, improves the detection efficiency, and solves the technical problems that the prior standard inspection of the trial products in stores adopts manpower, the efficiency is low, and the replacement of the trial products by workers in time is inconvenient.
In one embodiment of the present invention, the specific steps of the target detection model detecting the specifications of the trial products of the shelf image include:
calculating the number of the trial products in the goods shelf image, and if the number is smaller than a specified number threshold value, outputting the information of the shortage of the trial products on the goods shelf by the target detection model;
fitting the calculation points on the same plane of all the trial products in the goods shelf image into a fitting straight line, calculating and accumulating the distances between the calculation points of all the trial products and the fitting straight line to obtain a fitting distance value, and if the fitting distance value is larger than a set error threshold value, outputting the information of uneven placement of the trial products on the goods shelf by a target detection model;
calculating the spacing interval between every two trial products in the goods shelf image, and if the spacing interval is larger than a set spacing threshold or a proportion threshold, outputting information that the trial products on the goods shelf do not accord with the placing interval by the target detection model;
and calculating the distance between the center of all the trial products in the goods shelf image and the goods shelf area boundary, and if the distance is greater than a set distance threshold, outputting space information for placing the trial products on the goods shelf by the target detection model.
It should be noted that the fact that the target detection model outputs the space information for placing the trial products on the goods shelf means that the number of the trial products placed in the target area on the goods shelf is not enough, and the space for placing the trial products is available.
In the embodiment of the invention, a least square method is adopted to fit the fitting straight line of the trial product in the target detection model.
The trial products are projected on a plane to form four sides of a detection frame of the trial products, the central point of the bottom end face of the trial products placed on a shelf is taken as a calculation point, if the calculation point of the first trial product is a, the calculation point of the second trial product is b, the calculation point of the third trial product is c, the calculation point of the fourth trial product is d, a least square method is adopted to fit all the calculation points of the trial products into a straight line and mark the straight line as a fitting straight line, the distance between each calculation point of the trial products and the fitting straight line is calculated to obtain a fitting distance, the fitting distances between all the trial products and the fitting straight line are added to obtain a fitting distance value, and if the fitting distance value exceeds a set error threshold value, information of uneven trial placement is output.
The least square method is explained by using the minimum of the square sum of the residual errors of the dependent variable values and the actual values on the fitting straight line of the linear regression fitting as an optimization target, so as to determine the coefficients needed to be found. For example: given a set of data X (X1, X2.,. times, Xn) and Y (Y1, Y2.,. times, Yn), it is found by looking at scattergrams that there is a strong linear relationship between X, Y, and finding a suitable linear coefficient by linear fitting best reflects the correlation between X, Y, as known from "https:// www.jianshu.com/p/d28dfbcb 0007" linear regression fitting based on least squares "published by website," the fitted straight line is calculated by the least squares method according to the calculated points of all trial products.
In the embodiment of the invention, the specific calculation of the spacing distance between every two trial products in the shelf image is as follows: the central point of the bottom end surface of the shelf on which the trial products are placed is used as a calculation point, if the calculation point of the first trial product is a, the calculation point of the second trial product is b, and the distance between the first trial product and the second trial product is labThe width of the bottom end face of the first trial product is waThe width of the bottom end face of the second trial product is wbThen the first and second trial products are spaced apart by a distance lab-1/2(wa+wb)。
In the embodiment of the present invention, calculating the distance between the center of all the trial products in the shelf image and the boundary of the shelf area specifically includes: the central point of the bottom end surface of the shelf on which the trial products are placed is taken as a calculation point, if the calculation point of the trial products is taken as e, the boundaries of the shelf area are four edges which enclose the shelf area and are respectively marked as a left frame left, a right frame right, an upper frame up and a lower frame down, as shown in fig. 2. And if the distance between the calculation point e and any boundary of the shelf area is greater than a set distance threshold value, the target detection model outputs the space information for placing the trial products on the shelf.
In an embodiment of the invention, the target detection model further comprises an output module which is communicated with a mobile terminal of a store manager, and the output module is used for outputting result information for detecting that the trial products placed on the shelves are not normal.
It should be noted that the output result information generates a work order report, and the work order report is transmitted to the mobile terminal of the store manager to inform the store manager, so that the store manager arranges the store staff to correct the trial products on the shelf and supervise the store manager, the efficiency is greatly improved, and the store patrol cost of the store manager is also reduced.
Example two:
fig. 3 is a block diagram of a system for detecting specification of products for trial placement on a store shelf according to an embodiment of the present invention.
As shown in fig. 3, an embodiment of the present invention further provides a system for detecting specifications of trial products placed on a store shelf, including an image data obtaining unit 10, a segmentation model establishing unit 20, a detection model establishing unit 30, and a detection unit 40;
the image data acquisition unit 10 is used for acquiring images on a trial shelf of a store in real time and establishing an image database of a trial placement area;
a segmentation model establishing unit 20, configured to perform deep learning training on all images in the image database to establish a segmentation model;
the detection model establishing unit 30 is used for selecting the pictures of the trial assembly and placement areas in all the images by adopting the segmentation model, and establishing a target detection model by adopting deep learning training on all the selected pictures;
and the detection unit 40 is used for acquiring a shelf image to be detected with the standard trial products and inputting the shelf image into the target detection model, and the target detection model analyzes the number, display flatness, interval and plumpness of the trial products in the shelf image and outputs a result of whether the trial products placed on the shelf are standard or not.
In the embodiment of the present invention, the detecting unit 40 includes a first calculation and discrimination sub-unit 41, a second calculation and discrimination sub-unit 42, a third calculation and discrimination sub-unit 43, and a fourth calculation and discrimination sub-unit 44;
a first calculation and judgment subunit 41, configured to calculate the number of trial products in the shelf image, and if the number is smaller than a specified number threshold, the target detection model outputs information that the trial products on the shelf are out of stock;
the second calculation and judgment subunit 42 is configured to fit the calculation points on the same plane of all the trial products in the shelf image into a fitting straight line, calculate and accumulate distances between the calculation points of all the trial products and the fitting straight line to obtain a fitting distance value, and if the fitting distance value is greater than a set error threshold, the target detection model outputs information that the trial products on the shelf are not placed smoothly;
a third calculating and judging subunit 43, configured to calculate an interval between two products to be tried in the shelf image, and if the interval is greater than a set interval threshold or a ratio threshold, the target detection model outputs information that the products to be tried on the shelf do not conform to the placement interval;
and the fourth calculation and judgment subunit 44 is configured to calculate distances between centers of all the trial products in the shelf image and the boundary of the shelf area, and if the distance is greater than a set distance threshold, the target detection model outputs space information for placing the trial products on the shelf.
It should be noted that the units in the system in the second embodiment are arranged corresponding to the method in the first embodiment, and the content of the steps corresponding to the units has been described in detail in the first embodiment, which is not described in this embodiment.
The system for detecting the specification of the trial products on the shop shelf adopts the image data acquisition unit to acquire the images of the trial products on the shop shelf, trains the acquired images through the deep learning of the segmentation model establishment unit to obtain the segmentation model, trains the pictures of the regions where the trial products are placed in the segmentation model to obtain the target detection model, inputs the shelf images to be detected into the target detection model of the detection unit for specification detection and analysis, outputs the evaluation results of the number, display flatness, interval and plumpness of the trial products, and a shop manager supervises and modifies the trial products on the shop shelf according to the results output by the target detection model, thereby improving the detection efficiency. This system has realized putting on the automated inspection goods shelves and has adorned product standard inspection on probation, has improved detection efficiency, has solved current dress product standard inspection on probation to the store and has adopted the manual work, and the technical problem of dress product on probation that the staff of just being not convenient for in time changes the inefficiency.
In one embodiment of the present invention, the system for detecting the specification of the products on the shelves of the store for placing the trial products further comprises an output unit 50, wherein the output unit 50 is configured to output result information for detecting the irregularity of the products on the shelves for placing the trial products.
It should be noted that the output result information generates a work order report, and the work order report is transmitted to the mobile terminal of the store manager to inform the store manager, so that the store manager arranges the store staff to correct the trial products on the shelf and supervise the store manager, the efficiency is greatly improved, and the store patrol cost of the store manager is also reduced.
Example three:
the embodiment of the invention provides a specification detection device for trial products placed on a store shelf, which comprises a processor and a memory, wherein the processor is used for processing the specification detection device;
a memory for storing the program code and transmitting the program code to the processor;
and the processor is used for executing the specification detection method for the trial products placed on the store shelf according to the instructions in the program codes.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system and method may be implemented in other ways. For example, the above-described system embodiments are merely illustrative, and for example, the division of the units is only one logical functional division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, systems or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. The method for detecting the specification of the trial products placed on the store shelf is characterized by comprising the following steps of:
acquiring images on a trial shelf of a store in real time, and establishing an image database of a trial placement area;
deep learning training is carried out on all images in the image database to establish a segmentation model;
selecting pictures of the trial assembly and placement area from all the images by adopting the segmentation model, and establishing a target detection model by adopting deep learning training on all the selected pictures;
the method comprises the steps of obtaining a goods shelf image with standard trial products to be detected, inputting the goods shelf image into a target detection model, analyzing the number, display flatness, intervals and plumpness of the trial products in the goods shelf image by the target detection model, and outputting a result whether the trial products placed on a goods shelf are standard or not.
2. The method for detecting the specification of the trial products placed on the store shelf according to claim 1, wherein the step of detecting the specification of the trial products of the shelf image by the target detection model comprises the following steps:
calculating the number of the trial products in the goods shelf image, and if the number is smaller than a specified number threshold value, outputting the information of the shortage of the trial products on the goods shelf by the target detection model;
fitting the calculation points on the same plane of all the trial products in the goods shelf image into a fitting straight line, calculating and accumulating the distances between the calculation points of all the trial products and the fitting straight line to obtain a fitting distance value, and if the fitting distance value is larger than a set error threshold value, outputting the information of uneven placement of the trial products on the goods shelf by the target detection model;
calculating the interval distance between every two trial products in the goods shelf image, and if the interval distance is larger than a set distance threshold or a proportion threshold, outputting information that the trial products on the goods shelf do not accord with the placing interval by the target detection model;
and calculating the distance between the center of all the trial products in the goods shelf image and the goods shelf area boundary, and if the distance is greater than a set distance threshold value, outputting space information for placing the trial products on the goods shelf by the target detection model.
3. The method for detecting the specification of the trial products placed on the store shelf according to claim 2, wherein a least square method is used for fitting a straight line of the trial products in the target detection model.
4. The method for detecting the specification of the trial products placed on the store shelf according to claim 1, wherein the target detection model further comprises an output module which is in communication with a mobile terminal of a store manager, and the output module is used for outputting result information for detecting the irregularity of the trial products placed on the shelf.
5. The method for detecting the specification of the trial products placed on the store shelf according to claim 1, wherein a 360-degree ball machine is used to obtain the images of the trial shelves in real time.
6. A specification detection system for trial products placed on a store shelf is characterized by comprising an image data acquisition unit, a segmentation model establishment unit, a detection model establishment unit and a detection unit;
the image data acquisition unit is used for acquiring images on a trial shelf of a store in real time and establishing an image database of a trial placement area;
the segmentation model establishing unit is used for carrying out deep learning training on all images in the image database to establish a segmentation model;
the detection model establishing unit is used for selecting pictures of the trial assembly and placement area in all the images by adopting the segmentation model and establishing a target detection model by adopting deep learning training on all the selected pictures;
the detection unit is used for acquiring a goods shelf image with standard trial products to be detected and inputting the goods shelf image into the target detection model, and the target detection model analyzes the number, display flatness, intervals and plumpness of the trial products in the goods shelf image and outputs a result of whether the trial products placed on the goods shelf are standard or not.
7. The system for detecting the specification of the product on the store shelf on which the trial products are placed according to claim 6, wherein the detection unit comprises a first calculation and judgment subunit, a second calculation and judgment subunit, a third calculation and judgment subunit and a fourth calculation and judgment subunit;
the first calculation judging subunit is configured to calculate the number of the trial products in the shelf image, and if the number is smaller than a specified number threshold, the target detection model outputs information that the trial products on the shelf are out of stock;
the second calculation and judgment subunit is used for fitting a fitting straight line according to the calculation points on the same plane of all the trial products in the shelf image, calculating and accumulating the distances between the calculation points of all the trial products and the fitting straight line to obtain a fitting distance value, and if the fitting distance value is greater than a set error threshold value, the target detection model outputs the information that the trial products on the shelf are placed unevenly;
the third calculation and judgment subunit is configured to calculate an interval distance between every two trial products in the shelf image, and if the interval distance is greater than a set distance threshold or a set proportion threshold, the target detection model outputs information that the trial products on the shelf do not conform to the placement interval;
and the fourth calculation and judgment subunit is used for calculating the distance between the centers of all the trial products in the shelf image and the boundary of the shelf area, and if the distance is greater than a set distance threshold, the target detection model outputs the space information for placing the trial products on the shelf.
8. The system for detecting the specification of the placement of the trial products on the store shelf as claimed in claim 7, wherein the second calculating and judging subunit is further configured to fit the fitting straight line of the trial products by a least square method.
9. The system for detecting the specification of a trial product placed on a store shelf as claimed in claim 6, further comprising an output unit for outputting result information for detecting that the trial product placed on the shelf is not normal.
10. A device for detecting the specification of a trial product placed on a store shelf is characterized by comprising a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is used for executing the specification detection method for placing the trial products on the store shelf according to any one of claims 1 to 5 according to the instructions in the program code.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210602992.XA CN115049598A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
CN202010529064.6A CN111681234A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010529064.6A CN111681234A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210602992.XA Division CN115049598A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111681234A true CN111681234A (en) | 2020-09-18 |
Family
ID=72454618
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210602992.XA Pending CN115049598A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
CN202010529064.6A Pending CN111681234A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210602992.XA Pending CN115049598A (en) | 2020-06-11 | 2020-06-11 | Method, system and equipment for detecting standard of trial product placed on store shelf |
Country Status (1)
Country | Link |
---|---|
CN (2) | CN115049598A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112836578A (en) * | 2020-12-31 | 2021-05-25 | 广西慧云信息技术有限公司 | Shelf out-of-stock detection method based on apparent characteristics |
CN113762862A (en) * | 2020-12-28 | 2021-12-07 | 京东城市(北京)数字科技有限公司 | Cargo abnormity monitoring method and device, electronic equipment and storage medium |
CN114511820A (en) * | 2022-04-14 | 2022-05-17 | 美宜佳控股有限公司 | Goods shelf commodity detection method and device, computer equipment and storage medium |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117934791A (en) * | 2022-10-11 | 2024-04-26 | 中国科学院沈阳计算技术研究所有限公司 | Target detection method and device for workshop inspection robot |
CN116773546B (en) * | 2023-06-20 | 2024-03-22 | 上海感图网络科技有限公司 | Copper plating plate stacking defect detection method, copper plating plate stacking defect detection device and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108549870A (en) * | 2018-04-16 | 2018-09-18 | 图麟信息科技(深圳)有限公司 | A kind of method and device that article display is differentiated |
CN108898141A (en) * | 2018-06-14 | 2018-11-27 | 中山大学 | Commodity shelf segmentation and layer information statistical method based on image detection |
CN109117824A (en) * | 2018-08-31 | 2019-01-01 | 北京旷视科技有限公司 | Management method, device, electronic equipment and the storage medium of commodity |
CN109522933A (en) * | 2018-10-19 | 2019-03-26 | 北京陌上花科技有限公司 | Commodity detection method and device for shelf |
CN110705371A (en) * | 2019-09-09 | 2020-01-17 | 上海零眸智能科技有限公司 | Refrigerator structure and commodity arrangement detection method thereof |
CN111145430A (en) * | 2019-12-27 | 2020-05-12 | 北京每日优鲜电子商务有限公司 | Method and device for detecting commodity placing state and computer storage medium |
-
2020
- 2020-06-11 CN CN202210602992.XA patent/CN115049598A/en active Pending
- 2020-06-11 CN CN202010529064.6A patent/CN111681234A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108549870A (en) * | 2018-04-16 | 2018-09-18 | 图麟信息科技(深圳)有限公司 | A kind of method and device that article display is differentiated |
CN108898141A (en) * | 2018-06-14 | 2018-11-27 | 中山大学 | Commodity shelf segmentation and layer information statistical method based on image detection |
CN109117824A (en) * | 2018-08-31 | 2019-01-01 | 北京旷视科技有限公司 | Management method, device, electronic equipment and the storage medium of commodity |
CN109522933A (en) * | 2018-10-19 | 2019-03-26 | 北京陌上花科技有限公司 | Commodity detection method and device for shelf |
CN110705371A (en) * | 2019-09-09 | 2020-01-17 | 上海零眸智能科技有限公司 | Refrigerator structure and commodity arrangement detection method thereof |
CN111145430A (en) * | 2019-12-27 | 2020-05-12 | 北京每日优鲜电子商务有限公司 | Method and device for detecting commodity placing state and computer storage medium |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113762862A (en) * | 2020-12-28 | 2021-12-07 | 京东城市(北京)数字科技有限公司 | Cargo abnormity monitoring method and device, electronic equipment and storage medium |
CN112836578A (en) * | 2020-12-31 | 2021-05-25 | 广西慧云信息技术有限公司 | Shelf out-of-stock detection method based on apparent characteristics |
CN112836578B (en) * | 2020-12-31 | 2022-09-23 | 广西慧云信息技术有限公司 | Shelf out-of-stock detection method based on apparent characteristics |
CN114511820A (en) * | 2022-04-14 | 2022-05-17 | 美宜佳控股有限公司 | Goods shelf commodity detection method and device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN115049598A (en) | 2022-09-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111681234A (en) | Method, system and equipment for detecting standard of trial product placed on store shelf | |
US11403839B2 (en) | Commodity detection terminal, commodity detection method, system, computer device, and computer readable medium | |
CN103632158B (en) | Forest fire prevention monitor method and forest fire prevention monitor system | |
CN111191576A (en) | Personnel behavior target detection model construction method, intelligent analysis method and system | |
US20210174299A1 (en) | Method, system and device for collecting sales information of commodities in cabinet, and storage medium | |
CN111401269B (en) | Commodity hot spot detection method, device and equipment based on monitoring video | |
CN112115906A (en) | Open dish identification method based on deep learning target detection and metric learning | |
CN111414948B (en) | Target object detection method and related device | |
JP2017049983A (en) | Analytics system and method | |
CN101316371B (en) | Flame detecting method and device | |
CN111951258A (en) | Goods shelf out-of-stock early warning analysis system and method based on edge calculation | |
CN111310733A (en) | Method, device and equipment for detecting personnel entering and exiting based on monitoring video | |
CN109086696B (en) | Abnormal behavior detection method and device, electronic equipment and storage medium | |
CN112153373A (en) | Fault identification method and device for bright kitchen range equipment and storage medium | |
CN111415339A (en) | Image defect detection method for complex texture industrial product | |
CN110619308A (en) | Aisle sundry detection method, device, system and equipment | |
TW202201275A (en) | Device and method for scoring hand work motion and storage medium | |
CN111444806B (en) | Commodity touch information clustering method, device and equipment based on monitoring video | |
CN111563492B (en) | Fall detection method, fall detection device and storage device | |
CN111597889B (en) | Method, device and system for detecting target movement in video | |
KR100543706B1 (en) | Vision-based humanbeing detection method and apparatus | |
CN110378299B (en) | Clothing identification system under indoor lighting condition | |
CN111325731A (en) | Installation detection method and device of remote control device | |
CN112418025A (en) | Weight detection method and device based on deep learning | |
CN112153320A (en) | Method and device for measuring size of article, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20201130 Address after: Room 011, first floor, no.2429, Xingang East Road, Haizhu District, Guangzhou City, Guangdong Province (office only) Applicant after: CHUANGYOU digital technology (Guangdong) Co.,Ltd. Address before: Room 205, 2f office, No.156, nanshanzui Road, Hengqin, Zhuhai City, Guangdong Province Applicant before: MINISO (HENGQIN) ENTERPRISE MANAGEMENT Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200918 |
|
RJ01 | Rejection of invention patent application after publication |