Rfid-based fixed asset inventory method and system

Abstract

The present application relates to a RFID-based fixed asset inventory method and system, the present application collects the historical reading data of the current time period, analyzes the historical reading data to obtain the reference inventory period of the current time period. Then analyze the shelf image of multiple time points in the current time period, analyze the activity of RFID tags in each storage unit, shorten the inventory period of the fixed assets with high activity, through high-frequency inventory, the differences can be found and corrected in time, ensure the real-time accuracy of the inventory data of key assets, reduce the risk of information error. For the fixed assets with low activity, the inventory period is extended. The present application concentrates the limited inventory resources (manpower, time) on key fixed assets, avoids wasting resources on low-activity materials, and improves the overall inventory efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a fixed asset inventory method and system based on RFID. BACKGROUND

[0002] Traditional fixed assets adopt manual management form, and there are risks such as that the goods have not been cleaned and verified for many years, and the specific situation cannot be grasped. Business operation has no system support, and the degree of electronicization is low, and it is not connected with the management system. The identification and counting of fixed asset goods completely rely on manual work, and the efficiency of finding and inventory is low; 20,000 fixed assets need two people to spend 10 hours to complete the inventory. Therefore, in the prior art, RFID radio frequency identification tags are generally used to bind fixed assets to facilitate inventory, and when inventory is performed, the radio frequency information in the RFID tag is read to change and record information.

[0003] The existing RFID radio frequency identification tags are divided into active tags and passive tags, and most of the scenes still use passive tags due to cost constraints. Passive tags: rely on the energy of the reader to activate, low cost but need to be close to the reader. Therefore, even if passive RFID tags are used, workers still need to carry reading equipment to read information close to the tags, and the efficiency is also low. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a fixed asset inventory method and system based on RFID to solve the above technical problems.

[0005] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0006] The fixed asset inventory method based on RFID of the present application comprises the following steps:

[0007] Obtain historical reading data of a plurality of RIFD tags at a plurality of time points in a current time period, and obtain shelf images at a plurality of time points in the current time period, wherein the historical reading data includes reading time points and identification information, the RIFD tag is one-to-one bound with a fixed asset, and the fixed asset is placed in a shelf, and the current time period is a time period of a target time length before a current time point;

[0008] Calculate the inventory period of each RFID tag based on the historical reading data at a plurality of time points to obtain the reference inventory period of each RFID tag;

[0009] The RFID tags bound to the fixed assets in each storage unit in the shelf image are determined based on a pre-constructed RFID tag-storage unit mapping table for the current inventory period; the change status of the fixed assets in the storage unit is determined based on the shelf images at any adjacent time points to obtain the change determination result; and the activity level of each RFID tag is calculated based on the change determination result of each storage unit. The RFID tag-storage unit mapping table for the current inventory period is generated based on the inventory results of the previous inventory period.

[0010] The baseline inventory cycle for each RFID tag is adjusted based on the activity level to obtain the adjusted inventory cycle; and the inventory time point for the RFID tag is determined based on the adjusted inventory cycle.

[0011] In one embodiment of this application, the inventory cycle of each RFID tag is calculated based on historical reading data from multiple time points to obtain a baseline inventory cycle for each RFID tag, including:

[0012] Extract the reading time point of each RFID tag from the historical reading data;

[0013] For each RFID tag, calculate the time difference between any two adjacent time points to obtain multiple inventory cycles;

[0014] For each RFID tag, the average value of multiple inventory cycles is calculated to obtain the baseline inventory cycle.

[0015] In one embodiment of this application, the method for constructing the RFID tag-storage unit mapping table for the current inventory period includes:

[0016] Obtain the RFID tag-storage unit correspondence table of the previous inventory period and the change information of the previous judgment period. The change information includes the storage units that have been changed and the identification information of the changed RFID tags.

[0017] Based on the change information of the previous judgment period, the RFID tag-storage unit correspondence table of the previous inventory period is modified to obtain the RFID tag-storage unit correspondence table of the current inventory period.

[0018] In one embodiment of this application, the change status of fixed assets within a storage unit is determined based on a shelf image at any given time point, resulting in a change determination result, including:

[0019] For the shelf image Preprocessing is performed to obtain the preprocessed image. The preprocessing methods include grayscale conversion and high-pass filtering. Indicates a point in time;

[0020] extracting contours in the pre-processed image , screening out closed contours, and extracting size features of the closed contours, wherein the size features include length and width of the minimum circumscribed rectangle, rectangularity, and contour area;

[0021] matching the size features of the closed contours with pre-configured size screening parameters, and taking the closed contours matching the size screening parameters as the overall outer contour of the shelf , wherein the size screening parameters include length range, width range, rectangularity range, and contour area range;

[0022] based on the overall outer contour of the shelf , constructing a mask image, and extracting a shelf region image from the pre-processed image based on the mask image ;

[0023] obtaining a pre-constructed affine transformation matrix, and performing affine transformation on the shelf region image based on the pre-constructed affine transformation matrix to obtain a shelf region front view image ;

[0024] extracting a plurality of storage unit images from the shelf region front view image , and extracting contour distribution features and grayscale distribution features of the plurality of storage unit images, wherein: the subscript i represents the serial number of the storage unit;

[0025] for any two adjacent time points, comparing the contour distribution features and the grayscale distribution features of the storage units with the same position to obtain a change determination result.

[0026] In an embodiment of the present application, a plurality of storage unit images are extracted from the shelf region front view image , and contour distribution features and grayscale distribution features of the plurality of storage unit images are extracted, including:

[0027] aligning a pre-constructed shelf mask with the shelf region front view image , and extracting storage unit images based on the aligned shelf mask and the shelf region front view image , wherein the mathematical expression of the storage unit image is:

[0028]

[0029] For the image of the storage unit Morphological processing is performed to obtain a morphologically processed image;

[0030] Extract the contours from the morphologically processed image; and remove contours with an area smaller than a preset screening threshold to obtain candidate contours of fixed assets.

[0031] The morphologically processed image is meshed to obtain a mesh image; each mesh in the mesh image is calculated. Percentage of outline pixels in candidate outlines of internal fixed assets , ,in, For grid The number of outline pixels in the image. This represents the total number of grid pixels.

[0032] Extract the average gray value of each grid cell The average gray value is then normalized to obtain a normalized gray value. ;

[0033] Contour pixel percentage based on multiple grids Constructing vectorized contour distribution features And based on the normalized gray values ​​of multiple grids. Constructing vectorized grayscale distribution features .

[0034] In one embodiment of this application, the contour distribution features of storage cells with the same location are... and grayscale distribution characteristics The comparison yields the change determination results, including:

[0035] The contour distribution characteristics of each storage cell and grayscale distribution characteristics Combined into feature vectors ;

[0036] Calculate the cosine similarity of the feature vectors of storage units at any two adjacent time points. Wherein, the cosine similarity The mathematical expression is:

[0037]

[0038] When the cosine similarity is greater than or equal to a set similarity threshold, it is determined that the storage unit is at time point [time missing]. Time The time period between the time point where a change exists. The time period between the time point

[0039] In an embodiment of the present application, the activity of each RFID tag is calculated based on the change determination result of each storage unit, including:

[0040] For each storage unit, the number of time points where a change occurs in the current time period is calculated

[0041] The activity is calculated based on the total number of time points in the time period and the number of time points where a change occurs .

[0042] In an embodiment of the present application, the mathematical expression of the adjusted inventory cycle is:

[0043]

[0044] In the formula, represents the reference activity, is the reference inventory cycle of the RFID tag in the storage unit represents the activity of the storage unit

[0045] In an embodiment of the present application, the inventory time point of the RFID tag is determined based on the adjusted inventory cycle, including:

[0046] Based on the adjusted cycle and the latest reading time point, the theoretical inventory time point of each RFID tag is determined;

[0047] Based on the theoretical inventory time point, the density clustering of the plurality of RFID tags is performed to obtain a plurality of clusters; and the average value of the theoretical inventory time points of the plurality of RFID tags in each cluster is calculated to obtain the inventory time point of the RFID tag.

[0048] The present application also provides an RFID-based fixed asset inventory system, including:

[0049] An acquisition module is configured to acquire historical reading data of a plurality of RIFD tags at a plurality of time points in a current time period and acquire shelf images at a plurality of time points in the current time period, wherein the historical reading data includes reading time points and identification information, the RIFD tags are one-to-one bound with fixed assets, the fixed assets are placed in the shelves, and the current time period is a time period of a target time length before a current time point. ​​​​​​

[0050] a period calculation module configured to calculate an inventory period of each RFID tag based on historical reading data of multiple time points to obtain a reference inventory period of each RFID tag;

[0051] an activity calculation module configured to determine a bound RFID tag of a fixed asset in each storage unit in a shelf image based on a pre-constructed RFID tag-storage unit correspondence table of a current inventory period, determine a change state of the fixed asset in the storage unit based on shelf images of any adjacent time points to obtain a change determination result, and calculate an activity of each RFID tag based on the change determination result of each storage unit, wherein the RFID tag-storage unit correspondence table of the current inventory period is generated based on an inventory result of a previous inventory period;

[0052] an inventory management module configured to adjust the reference inventory period of each RFID tag based on the activity to obtain an adjusted inventory period, and perform an inventory based on the adjusted inventory period.

[0053] The RFID-based fixed asset inventory method and system has the following advantages: the application collects historical reading data of a current time period, analyzes the historical reading data to obtain a reference inventory period of the current time period, analyzes shelf images of multiple time points in the current time period, analyzes the activity of an RFID tag in each storage unit, shortens the inventory period of a fixed asset with a high activity, discovers and corrects differences in time through high-frequency inventory, ensures real-time and accurate inventory data of key assets, and reduces the risk of information errors. The inventory period of a fixed asset with a low activity is extended. The application concentrates limited inventory resources (manpower and time) on key fixed assets, avoids wasting resources on low-activity assets, and improves overall inventory efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0054] The application will be further described below in combination with the drawings and examples:

[0055] Figure 1 is a use scenario of the RFID-based fixed asset inventory method in an embodiment of the application;

[0056] Figure 2 is a flowchart of the RFID-based fixed asset inventory method in an embodiment of the application;

[0057] Figure 3 is a full-process flowchart of the activity in an embodiment of the application;

[0058] Figure 4is a structural diagram of an RFID-based fixed asset inventory system shown in an embodiment of the present application;

[0059] Figure 5 A structural schematic diagram of a computer system of an electronic device suitable for implementing embodiments of the present application is shown. DETAILED DESCRIPTION

[0060] The advantages and effects of the present application can be easily understood by those skilled in the art from the description of the embodiments of the present application. The present application can also be implemented or applied in other different embodiments, and the details in the description can be modified or changed based on different views and applications without departing from the spirit of the present application. It should be noted that the embodiments below and the features in the embodiments can be combined with each other without conflict.

[0061] It should be noted that the diagrams provided in the following embodiments only schematically illustrate the basic concept of the present application, and only the layers related to the present application are shown in the diagrams, not the number, shape and size ratio of the layers when actually implemented. The actual implementation of each layer can be randomly changed, and the layer layout pattern can also be more complex.

[0062] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application, however, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details.

[0063] Figure 1 is a use scenario diagram of an RFID-based fixed asset inventory method shown in an embodiment of the present application, as shown in Figure 1 Some fixed assets (such as industrial equipment, instruments, etc.) in the present application are placed in the storage units of the shelves 110, and the fixed assets are bound with RFID tags 120. In addition, the scenario also includes a monitoring camera 130, which is inclined towards the shelves 110 and uploads the collected images to a server 140. After analysis, the server 140 sends the analysis results to a terminal 150, and the inventory personnel can perform asset inventory according to the prompts on the terminal 150.

[0064] Figure 2 is a flowchart of an RFID-based fixed asset inventory method shown in an embodiment of the present application, as shown in Figure 2 The RFID-based fixed asset inventory method of the present embodiment can include steps S210 to S250:

[0065] S210, obtain historical reading data of a plurality of RFID tags at a plurality of time points in a current time period, and obtain shelf images at the plurality of time points in the current time period, wherein the historical reading data comprises reading time points and identification information, the RFID tags are bound one-to-one with fixed assets, and the fixed assets are placed on shelves, and the current time period is a time period of a target length before a current time point;

[0066] In the present application, data of the previous week before the current time point is taken as basic data of a reference inventory period.

[0067] In the present application, the shelf images are collected in a timed manner or collected by triggering a change in pressure, and the collection by triggering a change in pressure is more accurate. A pressure sensor is arranged under a storage unit, and the pressure sensor triggers shooting when the pressure data changes.

[0068] S220, calculate an inventory period of each RFID tag based on the historical reading data at the plurality of time points, to obtain a reference inventory period of each RFID tag;

[0069] In the present application, the inventory data of the previous week is taken as basic data, and the basic data is analyzed to obtain a reference inventory period, and the analysis comprises:

[0070] S221, extract a reading time point of each RFID tag from the historical reading data;

[0071] During the operation of the RFID system, reading events (such as time stamps, positions, etc.) of each tag are continuously recorded. The historical reading data is usually stored in a database or a log file, and contains a unique identifier (EPC code) of the tag and a time point of each reading.

[0072] By extracting the reading time point of each tag, time series data of the tag can be constructed. For example, the reading time point of tag A is .

[0073] S222, for each RFID tag, calculate a time difference between any two adjacent time points to obtain a plurality of inventory periods;

[0074] Inventory period ;

[0075] S223, for each RFID tag, calculate an average value of the plurality of inventory periods to obtain a reference inventory period.

[0076] Finally, an average value of the plurality of inventory periods of each RFID tag is calculated, that is, , to obtain the reference inventory period.

[0077] S230, determining the binding RFID tag of the fixed asset in each storage unit in the shelf image based on the pre-constructed RFID tag-storage unit correspondence table of the current inventory cycle; determining the change state of the fixed asset in the storage unit based on the shelf image of any adjacent time point to obtain a change determination result; and calculating the activity of each RFID tag based on the change determination result of each storage unit, wherein the RFID tag-storage unit correspondence table of the current inventory cycle is generated based on the inventory result of the last inventory cycle;

[0078] Since the inventory data of the fixed asset bound by the RFID tag is obtained through the reading data of the RFID tag, the RFID tag-storage unit correspondence table of the current inventory cycle is also needed to determine the position of different RFID tags, so as to obtain the position of the fixed asset. The construction of the RFID tag-storage unit correspondence table of the current inventory cycle adopts a rolling information update, Figure 3 The whole process flow diagram for generating the activity in an embodiment of the present application is shown in FIG. 1, which includes the following steps: Figure 3

[0079] S2301, obtaining the RFID tag-storage unit correspondence table of the last inventory cycle and the change information of the last judgment cycle, wherein the change information includes the storage unit with change and the identification information of the changed RFID tag;

[0080] The RFID tag-storage unit correspondence table records the correspondence between each RFID tag and the storage unit (such as shelf, tray, storage site, etc.). For example, tag A is located in A1 unit of shelf 1, and tag B is located in B3 unit of shelf 2.

[0081] S2302, modifying the RFID tag-storage unit correspondence table of the last inventory cycle based on the change information of the last judgment cycle to obtain the RFID tag-storage unit correspondence table of the current inventory cycle.

[0082] Based on the correspondence table of the last inventory cycle, the change information is combined for incremental update, including:

[0083] Add: add the newly added RFID tag or storage unit to the table.

[0084] Delete: remove the tags or storage units that have been scrapped or failed.

[0085] Modify: adjust the correspondence between the tags and the storage units according to the change information (such as moving the tags).

[0086] ​Secondly, the change state of the fixed assets in the storage unit is determined based on the shelf images at any adjacent time points to obtain a change determination result. The change determination result is based on machine vision processing technology, and specifically includes:

[0087] S2311, preprocessing the shelf image to obtain a preprocessed image , wherein the preprocessing method includes grayscale conversion and high-pass filtering, and wherein represents a time point;

[0088] First, the color image is converted into a grayscale image to reduce the computational complexity and remove color interference. Then, high-pass filtering is used to enhance the high-frequency information (such as edges and contours) in the image and suppress low-frequency background noise to obtain a preprocessed image . The metal frame of the shelf and the boundaries of the goods are more obvious after high-pass filtering, which facilitates subsequent contour extraction.

[0089] S2312, extracting the contour in the preprocessed image , screening out the closed contour, and extracting the size features of the closed contour, wherein the size features include the length and width of the minimum bounding rectangle, the rectangularity, and the contour area;

[0090] Based on edge detection (such as Canny operator) or region growing method, the continuous boundary in the image is identified. By judging the closure of the contour (such as the coincidence of the starting point and the ending point) or using connected component analysis, the broken or open contours are excluded. Then, the length and width of the minimum bounding rectangle of the contour are calculated, and the rectangularity and the contour area are calculated, and the calculation formula of the rectangularity is:

[0091]

[0092] S2313, matching the size features of the closed contour with the preconfigured size screening parameters, and taking the closed contour matching the size screening parameters as the overall outer contour of the shelf , wherein the size screening parameters include length range, width range, rectangularity range, and contour area range;

[0093] The size screening parameters include length range , width range , rectangularity range , and contour area range . When the following conditions are met at the same time, it means matching.

[0094]

[0095]

[0096]

[0097]

[0098] According to the actual size of the shelf (such as length, width, rectangularity, area), a threshold range (such as length range [200, 300] pixels) is set. The size features of each closed contour are compared with the preset parameters one by one, and only the contours that meet all the conditions are retained.

[0099] S2314, based on the overall outer contour of the shelf Construct a mask image, and extract a shelf region image from the pre-processed image based on the mask image ; ;

[0100] Use the shelf outer contour as a mask to set the non-shelf region in the image to zero and only retain the shelf region.

[0101] S2315, obtain a pre-constructed affine transformation matrix, and perform affine transformation on the shelf region image based on the pre-constructed affine transformation matrix to obtain a shelf region front view image ;

[0102] The shelf region image and the preset affine matrix (such as obtained by calibration). The affine transformation matrix (such as rotation, translation, scaling) is used to correct the perspective distortion of the shelf region (such as perspective distortion). Affine transformation eliminates the difference in shooting angle and ensures the accuracy of subsequent storage unit extraction. The mask image reduces background interference and improves processing efficiency.

[0103] S2316, extract a plurality of storage unit images from the shelf region front view image and extract contour distribution features and gray scale distribution features of the plurality of storage unit images, wherein the serial number of the storage unit is represented;

[0104] Based on the grid structure (such as fixed interval storage locations) of the shelf front view image. The mask extraction method is used to extract a plurality of storage unit images, and then the grid features (contour distribution features and gray scale distribution features ) of the fixed assets within the storage unit are extracted, and the specific process is as follows:

[0105] S23161, pre-construct a shelf mask and the shelf region front view image aligning, and based on the aligned shelf mask and the shelf region front view image extracting the storage unit image ;

[0106] Wherein, the shelf mask is precisely aligned with the front view image through affine transformation or feature matching algorithm (such as SIFT, ORB), eliminating the perspective deviation. For example, if the shelf mask is based on the template of the standard shelf design, it needs to be aligned with the current front view image through translation, rotation or scaling, ensuring that the mask covers the shelf area.

[0107] Using the aligned mask as a binary mask, combined with the grid structure of the shelf front view image (such as the pre-set row and column division), the image area of each storage unit is segmented. Among them, the mathematical expression of the storage unit image is:

[0108]

[0109] In the above process, the mask alignment ensures that the shelf area is not offset, avoiding the misplacement of storage units due to perspective differences.

[0110] S23162, morphological processing is performed on the storage unit image to obtain a morphologically processed image;

[0111] Morphological processing includes opening operation (Opening) and closing operation (Opening); opening operation first erodes and then dilates, eliminating small noise and maintaining the outline shape. The closing operation first dilates and then erodes, filling small gaps. Thus, the noise and artifacts in the storage unit image are eliminated, and the accuracy of the outline detection is improved.

[0112] S23163, extracting the outline in the morphologically processed image; and screening out the outline with an area less than a pre-set screening threshold to obtain a fixed asset object candidate outline;

[0113] Based on the morphologically processed image, use edge detection algorithm (such as Canny) or connected component analysis to extract the outline. Then set the threshold according to the minimum size of the fixed asset object (such as the size of the label or the volume of the object). For example, the outline with an area less than 100 pixels may be noise or irrelevant object. Remove the noise or irrelevant object, and only keep the outline features of the fixed asset object. Through area screening, reduce false detection, and ensure that the subsequent analysis focuses on the effective target.

[0114] S23164, gridizing the morphologically processed image to obtain a grid image; calculating the proportion of outline pixel points of the fixed asset object candidate outline in each grid of the grid image , , The number of contour pixels in the grid The total number of grid pixels

[0115] The storage unit image is divided into uniform grids (such as 4x4 or 8x8 grids), and each grid corresponds to a local area. Specifically, the grid is realized by sliding window or fixed row-column division method.

[0116] By grid division, the storage unit is divided into sub-regions, and the local contour distribution pattern (such as item offset, tilt) is captured. Even if the item is partially blocked or offset, the local proportion can still reflect its existence state.

[0117] S23165, extract the average gray value of each grid , and normalize the average gray value to obtain the normalized gray value ;

[0118] The mathematical expression of the normalization process is:

[0119]

[0120] Normalization eliminates the influence of environmental light difference on gray value, ensuring consistent comparison at different time points. The average gray value of the grid reflects the grid filling degree of the storage unit (such as empty gray high, full gray low).

[0121] S23166, based on the contour pixel proportion of multiple grids Construct a vectorized contour distribution feature , and based on the normalized gray value of multiple grids Construct a vectorized gray distribution feature .

[0122] Arrange the contour pixel proportions of all grids in order as a vector, where:

[0123] ; .

[0124] S2317, for any two adjacent time points, compare the contour distribution features and the gray distribution features of the storage units with the same position to obtain the change determination result.

[0125] Since the vectorized contour distribution feature and the gray distribution feature are extracted in the foregoing, a vector similarity calculation method is used for comparison, and the specific process is as follows:

[0126] ​S23171, profile distribution feature of each storage unit and gray scale distribution feature are combined into a feature vector ;

[0127]

[0128] By combining profile and gray scale information, the limitations of a single feature (e.g. profile cannot reflect the material of the item, and gray scale cannot reflect the shape) are compensated. For example, if the profile pixel ratio of a storage unit is high but the gray scale value is low, it may indicate that the item is made of metal; on the contrary, if the gray scale value is high but the profile pixel ratio is low, it may indicate that the item is made of a translucent material. The feature vector combines shape and gray scale information, improving the robustness of change detection. Even if one feature has noise (e.g. light changes affect gray scale), the other feature can still provide complementary information.

[0129] S23172, calculate the cosine similarity of the feature vectors of the storage unit at any two adjacent time points , wherein the cosine similarity is mathematically expressed as:

[0130]

[0131] The cosine similarity ranges from -1 to 1, and the closer the value is to 1, the more similar the directions of the two vectors are (i.e. the more stable the state of the storage unit is).

[0132] S23173, when the cosine similarity is greater than or equal to a set similarity threshold, it is determined that there is no change in the time period between time point and time , otherwise, it is determined that there is a change in the time period between time point and time .

[0133] When the cosine similarity is greater than 0.8, it indicates that the state of the storage unit has not changed significantly (e.g. the item has not been moved or replaced). Otherwise, it indicates that the state of the storage unit has changed significantly (e.g. the item has been moved in / out or the position has been adjusted).

[0134] After analyzing the change state of the fixed assets at multiple time points through machine vision technology, the activity of each fixed asset (bound to the RFID tag and the storage unit) can be further calculated, and the specific process is as follows:

[0135] S2321, for each storage unit, calculate the number of time points that have changed in the current time period ;

[0136] S2322, based on the total number of time points in the time period and the number of time points that change Calculate activity , .

[0137] In this application, the proportion of active time points to all time points is used to represent the activity. The activity range is [0, 1], in addition, the activity calculation can be further optimized by combining weighted time points (such as higher weight for high-frequency monitoring period).

[0138] S240, adjust the reference inventory cycle of each RFID tag based on the activity, to obtain an adjusted inventory cycle; and determine the inventory time point of the RFID tag based on the adjusted inventory cycle.

[0139] Specifically, the adjusted inventory cycle The mathematical expression is:

[0140]

[0141] In the formula, represents the reference activity, is the reference inventory cycle of the RFID tag in the storage unit .

[0142] In the above calculation formula, the proportion calculation is performed by using the pre-constructed reference activity. If it is greater than the reference activity, it means that the activity is high, and the corresponding adjusted inventory cycle increases, and the adjusted inventory cycle shrinks.

[0143] Finally, since the adjustment may cause multiple storage units to have multiple different inventory time points, in order to save the inventory cost, the clustering method is used to unify the inventory time. Specifically, the inventory time point of the RFID tag is determined based on the adjusted inventory cycle, including:

[0144] S241, based on the adjusted cycle and the latest reading time point, determine the theoretical inventory time point of each RFID tag;

[0145] S242, based on the theoretical inventory time point, density clustering is performed on multiple RFID tags to obtain multiple clusters; and calculate the average value of the theoretical inventory time point of the multiple RFID tags in each cluster to obtain the inventory time point of the RFID tag.

[0146] In the above process, the multiple RFID tags are divided into multiple clusters by using a density clustering DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, the inventory time points of the RFID tags in the clusters are similar, and therefore the average value in the cluster is further calculated to obtain a typical inventory time point. The inventory personnel can refer to the inventory time point to perform inventory in the next inventory cycle. The inventory location can be determined with reference to the RFID tag-storage unit correspondence table described above.

[0147] The RFID-based fixed asset inventory method provided in the present application collects historical reading data in a current time period, analyzes the historical reading data to obtain a reference inventory cycle in the current time period, analyzes shelf images at multiple time points in the current time period, analyzes the activity of the RFID tags in each storage unit, shortens the inventory cycle of the fixed assets with high activity, and through high-frequency inventory, can timely find and correct differences, ensure the real-time accuracy of the inventory data of key assets, and reduce the risk of information errors. The inventory cycle of the fixed assets with low activity is extended. The present application concentrates limited inventory resources (manpower and time) on key fixed assets, avoids wasting resources on low-activity materials, and improves the overall inventory efficiency.

[0148] As shown in Figure 4 The present application also provides an RFID-based fixed asset inventory system, which comprises:

[0149] An acquisition module is configured to acquire historical reading data of multiple RIFD tags at multiple time points in a current time period and acquire shelf images at multiple time points in the current time period, wherein the historical reading data comprises reading time points and identification information, the RIFD tags are one-to-one bound to fixed assets, the fixed assets are placed in shelves, and the current time period is a time period of a target length before a current time point;

[0150] A cycle calculation module is configured to calculate an inventory cycle of each RFID tag based on the historical reading data at the multiple time points to obtain a reference inventory cycle of each RFID tag.

[0151] The activity calculation module is configured to determine the bound RFID tag of the fixed asset in each storage unit in the shelf image based on a pre-constructed RFID tag-storage unit correspondence table of a current inventory cycle; determine the change state of the fixed asset in the storage unit based on the shelf image of any adjacent time point to obtain a change determination result; and calculate the activity of each RFID tag based on the change determination result of each storage unit, wherein the RFID tag-storage unit correspondence table of the current inventory cycle is generated based on the inventory result of the last inventory cycle.

[0152] The inventory management module is configured to adjust the reference inventory cycle of each RFID tag based on the activity to obtain an adjusted inventory cycle; and perform inventory based on the adjusted inventory cycle.

[0153] The RFID-based fixed asset inventory system of the present application collects the historical reading data of the current time period, analyzes the historical reading data to obtain the reference inventory cycle of the current time period, analyzes the shelf images of multiple time points in the current time period, analyzes the activity of the RFID tag in each storage unit, shortens the inventory cycle of the fixed asset with high activity, discovers and corrects the difference in time through high-frequency inventory, ensures the real-time accuracy of the inventory data of the key asset, reduces the risk of information error, prolongs the inventory cycle of the fixed asset with low activity, concentrates the limited inventory resources (manpower and time) on the key fixed asset, avoids wasting resources on low-activity assets, and improves the overall inventory efficiency.

[0154] Figure 5 The structure of the computer system of the electronic device suitable for implementing the embodiments of the present application is shown. It should be noted that, Figure 5 The computer system of the electronic device shown is only an example and should not limit the functions and use range of the embodiments of the present application.

[0155] As Figure 5As shown, the computer system includes a central processing unit (CPU) 501 which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 502 or loaded into a random access memory (RAM) 503 from a storage section 508, such as performing the methods in the above-described embodiments. In the RAM 503, various programs and data required for the operation of the system are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0156] Connected to the I / O interface 505 are an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as necessary. A removable recording medium 511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 510 as necessary, so that a computer program read therefrom is installed into the storage section 508 as necessary.

[0157] In particular, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing a computer program for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 509, and / or installed from the removable recording medium 511. When the computer program is executed by the central processing unit (CPU) 501, various functions defined in the system of the present application are performed.

[0158] It should be noted that the computer-readable medium in the embodiments of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may, for example, be an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, the computer-readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer-readable computer programs. Such a propagated data signal can take on various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate or transmit programs for use by or in connection with an instruction execution system, apparatus or device. The computer programs contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination of the above.

[0159] The flowcharts and block diagrams in the drawings illustrate the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. Each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code containing one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders than that shown in the drawings. For example, two blocks that are shown in succession can actually be executed substantially in parallel, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams or flowcharts, and the combination of blocks in the block diagrams or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0160] The units described in the embodiments of the present application can be implemented by software, or by hardware, or by a combination of software and hardware. The units described can also be located in a single processor. In some cases, the names of the units do not constitute a limitation on the units themselves.

[0161] Another aspect of the present application provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor of a computer, and causes the computer to perform the method described above. The computer readable storage medium can be included in the electronic device described in the embodiments above, or can exist separately from the electronic device.

[0162] Another aspect of the present application provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions. The computer instructions are stored in a computer readable storage medium. A processor of a computer reads the computer instructions from the computer readable storage medium. The processor executes the computer instructions, and causes the computer to perform the method described in the embodiments above.

[0163] The above embodiments are merely preferred embodiments of the present application, and the protection scope of the present application is not limited to this. Any equivalent replacement or transformation of the present application made by a person skilled in the art based on the present application is within the protection scope of the present application.

Claims

1. A fixed asset inventory method based on RFID, characterized in that, Including the following steps: The system acquires historical reading data of multiple RIFD tags at multiple time points within the current time period, and acquires shelf images at multiple time points within the current time period. The historical reading data includes reading time points and identification information. The RIFD tags are bound one-to-one with fixed assets, which are placed on shelves. The current time period is the time period of the target duration prior to the current time point. The inventory cycle for each RFID tag is calculated based on historical reading data from multiple time points, thus obtaining the baseline inventory cycle for each RFID tag. Based on a pre-constructed RFID tag-storage unit mapping table for the current inventory period, the bound RFID tags for fixed assets within each storage unit in the shelf image are determined; the change status of fixed assets within the storage unit is determined based on shelf images at any adjacent time points, resulting in a change determination result; and the activity level of each RFID tag is calculated based on the change determination result for each storage unit. The RFID tag-storage unit mapping table for the current inventory period is generated based on the inventory results of the previous inventory period. The change status of fixed assets within the storage unit is determined based on shelf images at any adjacent time points, resulting in a change determination result, including: determining the change status of fixed assets within the storage unit based on the shelf images at any adjacent time points. Preprocessing is performed to obtain the preprocessed image. The preprocessing methods include grayscale conversion and high-pass filtering. Indicates a time point; extracts the preprocessed image. The outline in the data is filtered to identify closed outlines, and the dimensional features of these closed outlines are extracted. These dimensional features include the length and width of the minimum bounding rectangle, the rectangularity, and the outline area. Based on pre-configured dimensional filtering parameters, the dimensional features of the closed outlines are matched, and the closed outlines matching the dimensional filtering parameters are used as the overall outer outline of the shelving unit. The size selection parameters include length range, width range, rectangularity range, and outline area range; based on the overall outer contour of the shelf. Construct a mask image, and extract data from the preprocessed image based on the mask image. Extracting shelf area images Obtain a pre-constructed affine transformation matrix, and apply the pre-constructed affine transformation matrix to the image of the shelf area. Perform an affine transformation to obtain the front view image of the shelving area. Front view image of the shelf area Extract multiple storage cell images and extract the contour distribution features of the multiple storage cell images. and grayscale distribution characteristics ,in, Indicates the sequence number of the storage cell; for any two adjacent time points, the outline distribution characteristics of storage cells with the same location. and grayscale distribution characteristics The changes are compared to obtain the change determination results; based on the change determination results of each storage unit, the activity level of each RFID tag is calculated, including: for each storage unit, calculating the number of time points in the current time period where changes occur. Based on the total number of time points in the time period and the number of time points in time when changes occurred Calculate activity ， ； The baseline inventory cycle for each RFID tag is adjusted based on the activity level to obtain the adjusted inventory cycle; and the inventory time point for the RFID tag is determined based on the adjusted inventory cycle.

2. The RFID-based fixed asset inventory method according to claim 1, characterized in that, The inventory cycle for each RFID tag is calculated based on historical read data from multiple time points, resulting in a baseline inventory cycle for each RFID tag, including: Extract the reading time point of each RFID tag from the historical reading data; For each RFID tag, calculate the time difference between any two adjacent time points to obtain multiple inventory cycles; For each RFID tag, the average value of multiple inventory cycles is calculated to obtain the baseline inventory cycle.

3. The RFID-based fixed asset inventory method according to claim 1, characterized in that, The method for constructing the RFID tag-storage unit mapping table for the current inventory period includes: Obtain the RFID tag-storage unit correspondence table of the previous inventory period and the change information of the previous judgment period. The change information includes the storage units that have been changed and the identification information of the changed RFID tags. Based on the change information of the previous judgment period, the RFID tag-storage unit correspondence table of the previous inventory period is modified to obtain the RFID tag-storage unit correspondence table of the current inventory period.

4. The RFID-based fixed asset inventory method according to claim 1, characterized in that, Front view of the shelving area Extract multiple storage cell images and extract the contour distribution features of the multiple storage cell images. And, including: Pre-built shelf mask Front view image of the shelving area Alignment is performed, and the alignment is based on the aligned shelf mask. Front view of the shelving area Extract storage unit image Among them, the storage unit image The mathematical expression is: For the image of the storage unit Morphological processing is performed to obtain a morphologically processed image; Extract the contours from the morphologically processed image; and remove contours with an area smaller than a preset screening threshold to obtain candidate contours of fixed assets. The morphologically processed image is meshed to obtain a mesh image; each mesh in the mesh image is calculated. Percentage of outline pixels in candidate outlines of internal fixed assets , ,in, For grid The number of outline pixels in the image. This represents the total number of grid pixels. Extract the average gray value of each grid cell The average gray value is then normalized to obtain a normalized gray value. ; Contour pixel percentage based on multiple grids Constructing vectorized contour distribution features And based on the normalized gray values ​​of multiple grids. Constructing vectorized grayscale distribution features .

5. The RFID-based fixed asset inventory method according to claim 4, characterized in that, The contour distribution characteristics of storage cells in the same location and grayscale distribution characteristics The comparison yields the change determination results, including: The contour distribution characteristics of each storage cell and grayscale distribution characteristics Combined into feature vectors ; Calculate the cosine similarity of the feature vectors of storage units at any two adjacent time points. Wherein, the cosine similarity The mathematical expression is: When the cosine similarity is greater than or equal to a set similarity threshold, it is determined that the storage unit is at time point [time missing]. Time If the time interval remains unchanged, otherwise, determine that the storage unit is at the specified time point. Time The time periods between them vary.

6. The RFID-based fixed asset inventory method according to claim 1, characterized in that, The adjusted inventory cycle The mathematical expression is: In the formula, Indicates the baseline activity level. For storage units The baseline inventory cycle for internal RFID tags. Represents storage unit Activity level.

7. The RFID-based fixed asset inventory method according to claim 1, characterized in that, The inventory time point for RFID tags is determined based on the adjusted inventory cycle, including: Based on the adjusted cycle and the latest reading time, the theoretical inventory time for each RFID tag is determined; Based on the theoretical inventory time points, multiple RFID tags are density-clustered to obtain multiple clusters; and the average of the theoretical inventory time points of multiple RFID tags in each cluster is calculated to obtain the inventory time points of the RFID tags.

8. A fixed asset inventory system based on RFID, characterized in that, include: The acquisition module is used to acquire historical reading data of multiple RIFD tags at multiple time points within the current time period, and to acquire shelf images at multiple time points within the current time period. The historical reading data includes reading time points and identification information. The RIFD tags are bound one-to-one with fixed assets, which are placed on shelves. The current time period is the time period of the target duration before the current time point. The cycle calculation module is used to calculate the inventory cycle of each RFID tag based on historical reading data from multiple time points, and obtain the baseline inventory cycle of each RFID tag. The activity calculation module is used to determine the bound RFID tags of fixed assets in each storage unit in the shelf image based on a pre-built RFID tag-storage unit mapping table for the current inventory period; to determine the change status of fixed assets in the storage unit based on shelf images at any adjacent time points, and obtain change determination results; and to calculate the activity level of each RFID tag based on the change determination results of each storage unit. The RFID tag-storage unit mapping table for the current inventory period is generated based on the inventory results of the previous inventory period. The determination of the change status of fixed assets in the storage unit based on shelf images at any adjacent time points, and obtaining change determination results, includes: determining the activity level of each RFID tag in the shelf image. Preprocessing is performed to obtain the preprocessed image. The preprocessing methods include grayscale conversion and high-pass filtering. Indicates a time point; extracts the preprocessed image. The outline in the data is filtered to identify closed outlines, and the dimensional features of these closed outlines are extracted. These dimensional features include the length and width of the minimum bounding rectangle, the rectangularity, and the outline area. Based on pre-configured dimensional filtering parameters, the dimensional features of the closed outlines are matched, and the closed outlines matching the dimensional filtering parameters are used as the overall outer outline of the shelving unit. The size selection parameters include length range, width range, rectangularity range, and outline area range; based on the overall outer contour of the shelf. Construct a mask image, and extract data from the preprocessed image based on the mask image. Extracting shelf area images Obtain a pre-constructed affine transformation matrix, and apply the pre-constructed affine transformation matrix to the image of the shelf area. Perform an affine transformation to obtain the front view image of the shelving area. Front view image of the shelf area Extract multiple storage cell images and extract the contour distribution features of the multiple storage cell images. and grayscale distribution characteristics ,in, Indicates the sequence number of the storage cell; for any two adjacent time points, the outline distribution characteristics of storage cells with the same location. and grayscale distribution characteristics The changes are compared to obtain the change determination results; based on the change determination results of each storage unit, the activity level of each RFID tag is calculated, including: for each storage unit, calculating the number of time points in the current time period where changes occur. Based on the total number of time points in the time period and the number of time points in time when changes occurred Calculate activity , ; The inventory management module is used to adjust the baseline inventory cycle of each RFID tag based on the activity level to obtain the adjusted inventory cycle; and to perform inventory based on the adjusted inventory cycle.

Images