Bulk cargo intelligent checking method based on 3D point cloud segmentation

Abstract

The invention discloses an intelligent bulk cargo checking method based on 3D point cloud segmentation, which realizes intelligence and high efficiency of bulk cargo checking through scientific partitioning, a dual checking mechanism and liquidity index management. Firstly, checking areas are divided according to the size of bulk cargoes, and a full-load threshold value is set to trigger checking. Secondly, the accuracy of bulk cargo information is ensured by comparing the RFID tag with the original purchasing information and combining with 3D point cloud data acquisition. In addition, the bulk cargo and inventory area fluidity index is constructed, the cargo layout is guided, the inventory strategy is flexibly set, and the inventory efficiency and the inventory accuracy are improved. Not only is the inventory flow simplified, but also the validity of the inventory strategy is maintained through a dynamic updating mechanism, and powerful support is provided for warehouse operation.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent inventory counting of bulk cargo, and in particular to an intelligent inventory counting method of bulk cargo based on 3D point cloud segmentation. Background Art

[0002] In the field of logistics and warehousing, inventory counting of bulk goods is an important and tedious task. Traditional bulk inventory counting methods usually rely on manual operations, which are not only time-consuming and labor-intensive, but also prone to errors. With the development of Internet of Things technology, automatic inventory counting systems based on RFID tags have gradually been applied. This system can quickly obtain basic information about bulk goods by scanning RFID tags on shelves, thereby improving inventory counting efficiency. However, traditional RFID-based automatic inventory counting systems still have certain limitations when dealing with complex bulk goods, especially when the bulk goods are irregular in shape and of various sizes. It is often difficult to accurately reflect the actual status of the bulk goods based solely on the information in the RFID tags.

[0003] In recent years, 3D point cloud segmentation technology has made significant progress in the fields of object recognition and shape reconstruction. Through equipment such as depth cameras, three-dimensional point cloud data of bulk goods can be obtained, thereby achieving accurate measurement of the shape and size of bulk goods. Applying 3D point cloud segmentation technology to bulk cargo inventory can make up for the shortcomings of traditional RFID inventory systems in bulk cargo shape and size recognition. By combining RFID tag information and 3D point cloud data, the actual status of bulk goods can be more comprehensively grasped, and the accuracy and efficiency of inventory can be improved. However, there is currently a lack of a bulk cargo intelligent inventory method that can effectively combine the Internet of Things and 3D point cloud segmentation technology on the market.

[0004] The traditional bulk cargo inventory process is cumbersome and inefficient for bulk cargo with complex shapes and varying sizes. In addition, existing RFID-based automatic inventory systems can usually only obtain basic information about bulk cargo, such as name, manufacturer, etc., but cannot accurately reflect the actual shape and size of bulk cargo. As a result, during the inventory process, the system often has difficulty in accurately identifying and recording bulk cargo with irregular shapes or varying sizes. Summary of the invention

[0005] This application provides a method for intelligent inventory counting of bulk cargo based on 3D point cloud segmentation. To address the above problems, this method can achieve accurate measurement and identification of the shape and size of bulk cargo by scientifically dividing the inventory counting area, defining the inventory counting trigger conditions, and combining RFID tag information and 3D point cloud data, thereby achieving accurate inventory counting of bulk cargo. At the same time, this method also has a dynamic update mechanism, which can flexibly adjust the inventory counting strategy according to the actual situation of bulk cargo, thereby improving the pertinence and flexibility of inventory counting.

[0006] The present application provides a bulk cargo intelligent inventory method based on 3D point cloud segmentation, including:

[0007] S101, dividing the shelf into inventory areas of different sizes according to the size of the bulk goods;

[0008] S102, defining conditions for triggering inventory counting according to the full load of the bulk cargo classification area, and when the inventory counting conditions are met, scanning the bulk cargo RFID tags on the shelves to obtain detailed information of the bulk cargo carried in the tags;

[0009] S103, according to the identification code of the bulk cargo, the corresponding original purchase information is retrieved from the purchase information database;

[0010] S104, comparing the retrieved original purchase information with the information in the RFID tag, if all the same fields are consistent, generating an inventory record based on the RFID tag information; if there are inconsistent fields, starting the 3D point cloud data collection process to generate an inventory record;

[0011] S105, traversing the inventory record table stored in the database according to the identification code of the bulk cargo, if the record of the bulk cargo already exists in the inventory record table, replacing it according to a predetermined rule; if the record does not exist in the inventory record table, adding the inventory record of the bulk cargo to the inventory record table.

[0012] Preferably, the S102 full load condition includes:

[0013] The full load condition of the bulk cargo classification area is defined as the capacity threshold of the shelf in the set area. 80%-85% of the capacity threshold is the full load threshold, which triggers the inventory counting condition.

[0014] Preferably, the step S104 starts a 3D point cloud data collection process, including:

[0015] To use the depth camera Intel RealSense D435 to collect 3D point cloud data of bulk cargo in inconsistent fields, obtain its accurate 3D shape and size data, segment the collected 3D point cloud data, and extract the corresponding 3D model of bulk cargo

[0016] Preferably, the S104 further includes:

[0017] The extracted 3D model features are compared again with the RFID tag information and the original purchase information to see which one has a higher matching degree between the extracted 3D model features and the RFID tag information and the original purchase information, which is used as the basis for generating inventory records.

[0018] Preferably, the S102 includes:

[0019] S201, calculating the bulk cargo liquidity index and the inventory area liquidity index of each inventory area;

[0020] S202, defining the range of bulk cargo liquidity index, and adjusting the regional position of bulk cargo according to the size of the liquidity index;

[0021] S203, defining a threshold range of the liquidity index of the inventory area, and an inventory area exceeding the threshold range triggers an inventory condition.

[0022] Preferably, the step S201 of calculating the bulk cargo liquidity index of each inventory area includes:

[0023] The calculation formula of bulk cargo liquidity index LI is designed as follows:

[0024]

[0025] Among them, coefficient D reflects the market demand for bulk goods, inventory turnover rate I refers to the number of times bulk goods in inventory are turned over within a certain period of time, coefficient E reflects the frequency of bulk goods entering the warehouse, coefficient O reflects the frequency of bulk goods flowing out of the warehouse, and shelf distribution coefficient S takes into account the impact of the position of bulk goods on the shelf on liquidity.

[0026] Preferably, the step S201 of calculating the regional liquidity index includes:

[0027] The calculation formula of regional liquidity index RLI is designed as follows:

[0028]

[0029] Where n represents the number of bulk cargoes in the region, LI_i represents the liquidity index of the i-th bulk cargo, calculated according to the liquidity index formula of a single piece of bulk cargo given previously, and W_i represents the weight of the i-th bulk cargo;

[0030] Wi can be set to 1 / n, that is, the weight of each bulk cargo is equal;

[0031] The weight Wi can be adjusted to reflect the impact of a certain bulk cargo on regional liquidity.

[0032] Preferably, the step S202 defines the bulk cargo liquidity index range, including:

[0033] Define the low liquidity interval LI<1.0, the medium liquidity interval 1.0≤LI<2.0, and the high liquidity interval LI≥2.0; adjust the position of bulk goods in the shelf and the position of the shelf in the classification area according to the threshold size of the flow interval.

[0034] Preferably, the step S203 defines a threshold range of the inventory area liquidity index, including:

[0035] Set at least one threshold value to trigger the inventory condition;

[0036] When more than one threshold is set, the low threshold RLI_low and the high threshold RLI_high define the range of all thresholds that appear;

[0037] Set a basic inventory cycle. When the RLI value of a certain inventory area exceeds the set high threshold RLI_high, the inventory condition is immediately triggered and an emergency inventory is performed on the area.

[0038] When the RLI value is between the low threshold RLI_low and the high threshold RLI_high, an inventory is taken according to the basic inventory cycle;

[0039] When the RLI value is lower than RLI_low, and there is no product about to expire or special customer demand, the inventory counting cycle can be extended by 1.3 times.

[0040] Preferably, the step S201 includes:

[0041] S301, define the liquidity index of bulk cargo within the window time period;

[0042] S302, prioritize the liquidity index of bulk cargo;

[0043] S303, after the inventory counting condition is triggered, the inventory counting starts with the bulk cargo with high priority;

[0044] S304, Regularly update the priority of bulk cargo liquidity index

[0045] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0046] 1. Scientifically partition the shelves according to the size of bulk goods, realizing efficient use of warehouse space and orderly inventory work. This partitioning method not only simplifies the positioning and placement of bulk goods, but also intelligently triggers inventory conditions by setting the full load threshold of the bulk goods classification area, ensuring the timely update and accuracy of inventory information. At the same time, the area number information embedded in the RFID tag facilitates the rapid positioning of bulk goods, further improving inventory efficiency.

[0047] 2. The solution also uses a double-check mechanism to ensure the accuracy of bulk cargo information. During the inventory process, by comparing the RFID tag information with the original purchase information, possible differences can be discovered and corrected in a timely manner. In particular, when the information is inconsistent, the use of 3D point cloud data acquisition technology to obtain accurate three-dimensional data of bulk cargo provides a more reliable basis for inventory. This mechanism not only enhances the credibility of warehouse management, but also brings significant improvement and optimization to enterprise inventory management.

[0048] 3. By comprehensively considering multiple factors such as market demand, inventory turnover rate, frequency of goods in and out, and shelf distribution, a scientific bulk cargo liquidity index and inventory area liquidity index calculation system has been constructed. This system can not only accurately reflect the flow status of goods, but also guide the reasonable layout of goods in the warehouse, thereby effectively improving inventory efficiency and inventory accuracy. At the same time, by setting flexible liquidity index thresholds and corresponding inventory strategies, the solution can trigger inventory conditions according to actual conditions, which not only avoids unnecessary inventory waste, but also ensures the real-time and accuracy of inventory information, greatly improving the overall efficiency of warehouse operations.

[0049] 4. The plan also emphasizes the importance of regularly updating the liquidity index threshold to ensure that the inventory strategy can be adjusted with changes in warehouse operations and market environment to maintain its effectiveness and adaptability. This dynamic management concept not only enhances the flexibility and responsiveness of warehouse management, but also helps enterprises to promptly discover and solve potential problems in inventory management, providing strong support for the sustainable development and optimization of enterprises.

[0050] 5. By scientifically defining the liquidity index of bulk cargo and assigning priorities, accurate measurement and effective management of cargo liquidity are achieved. This helps warehouses prioritize cargo with high liquidity and high management requirements when inventory conditions are triggered, thereby improving the pertinence and efficiency of inventory and reducing unnecessary waste of resources.

[0051] 6. A dynamic update mechanism has been established to regularly update the liquidity index and priority of bulk cargo, ensuring that the inventory strategy is always consistent with the actual flow of goods and management requirements. This flexibility and adaptability enables the warehouse to better respond to market changes and adjust inventory management strategies in a timely manner, thereby improving the overall operation level and competitiveness of the warehouse. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The present invention is a flowchart of a method for intelligent bulk cargo inventory based on 3D point cloud segmentation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0054] It should be noted that the terms “vertical”, “horizontal”, “up”, “down”, “left”, “right” and similar expressions used in this document are only for illustrative purposes and do not represent the only implementation method.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs; the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more related listed items.

[0056] Embodiment 1:

[0057] Figure 1 It is a flow chart of a method for intelligent inventory counting of bulk cargo based on 3D point cloud segmentation according to an embodiment of the present invention.

[0058] like Figure 1 As shown, a bulk cargo intelligent inventory method based on 3D point cloud segmentation includes the following steps:

[0059] S101, dividing the shelf into inventory areas of different sizes according to the size of the bulk goods.

[0060] S102, defining conditions for triggering inventory counting according to the full load condition of the bulk cargo classification area, and when the inventory counting conditions are met, scanning the bulk cargo RFID tags on the shelves to obtain detailed information of the bulk cargo carried in the tags.

[0061] The detailed information includes the name of the bulk cargo, manufacturer, production time, size, shape (for conventional shapes, the description can be simplified), weight, storage time and the area number of the shelf where it is located;

[0062] The detailed information of the bulk cargo is bound to the RFID tag of each bulk cargo in advance; when the bulk cargo is taken out, the detailed information of the bulk cargo is unbound from the RFID tag;

[0063] The full load condition of the bulk cargo classification area is defined as the capacity threshold of the shelf in the set area. 80%-85% of the capacity threshold is the full load threshold, which triggers the inventory counting condition.

[0064] S103, according to the identification code of the bulk cargo, the corresponding original purchase information is retrieved from the purchase information database.

[0065] Among them, the original purchasing information includes the bulk cargo name, identification code, manufacturer, production time, size and weight.

[0066] S104, compare the retrieved original purchase information with the information in the RFID tag. If the contents of all the same fields are consistent, generate an inventory record based on the RFID tag information; if there are inconsistent fields, start the 3D point cloud data collection process to generate an inventory record.

[0067] The 3D point cloud data collection process specifically uses the depth camera Intel RealSense D435 to collect 3D point cloud data of bulk cargo with inconsistent fields, obtains its precise 3D shape and size data, segments the collected 3D point cloud data, and extracts the 3D model of the corresponding bulk cargo;

[0068] The extracted 3D model features are compared again with the RFID tag information and the original purchase information to see which one has a higher matching degree between the extracted 3D model features and the RFID tag information and the original purchase information, which is used as the basis for generating inventory records.

[0069] It should be noted that 3D point cloud data collection is an existing technology and will not be described in detail in this article.

[0070] S105, traversing the inventory record table stored in the database according to the identification code of the bulk cargo, if the record of the bulk cargo already exists in the inventory record table, replacing it according to a predetermined rule; if the record does not exist in the inventory record table, adding the inventory record of the bulk cargo to the inventory record table.

[0071] For example, a warehouse needs to conduct regular inventory of stored bulk goods. According to the size of the bulk goods, the shelves are divided into three inventory areas: small, medium, and large, and obvious signs are set. The full load threshold is set to 80%, that is, when the bulk goods in a certain area reach 80% of the capacity of the area, the inventory condition is triggered to ensure that all bulk goods have been bound to RFID tags and contain detailed bulk goods information;

[0072] The system detects that the bulk cargo volume in the medium inventory area has reached 80%, triggering the inventory condition. The system uses an RFID scanner to scan the bulk cargo RFID tags on the shelves in the medium inventory area to obtain the bulk cargo detailed information on the tags.

[0073] According to the identification code of the bulk cargo, the corresponding original procurement information is retrieved from the procurement information database, and the retrieved original procurement information is compared with the information in the RFID tag. It is found that the weight information of a certain bulk cargo is inconsistent. The Intel Real Sense D435 depth camera is used to collect 3D point cloud data of the bulk cargo, and the collected data is pre-processed and segmented to extract the 3D model of the bulk cargo. The extracted 3D model features are compared again with the RFID tag information and the original procurement information, and it is found that the 3D model features have a higher matching degree with the original procurement information;

[0074] Find the corresponding record in the inventory record sheet according to the identification code of the bulk cargo. Since the record of the bulk cargo already exists, but the weight information is inconsistent, the new information (i.e. the weight in the original purchase information) is used to replace the old information. Update the inventory record sheet to ensure the accuracy of the information.

[0075] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0076] 1. Scientifically partition the shelves according to the size of bulk goods, realizing efficient use of warehouse space and orderly inventory work. This partitioning method not only simplifies the positioning and placement of bulk goods, but also intelligently triggers inventory conditions by setting the full load threshold of the bulk goods classification area, ensuring the timely update and accuracy of inventory information. At the same time, the area number information embedded in the RFID tag facilitates the rapid positioning of bulk goods, further improving inventory efficiency.

[0077] 2. The solution also uses a double-check mechanism to ensure the accuracy of bulk cargo information. During the inventory process, by comparing the RFID tag information with the original purchase information, possible differences can be discovered and corrected in a timely manner. In particular, when the information is inconsistent, the use of 3D point cloud data acquisition technology to obtain accurate three-dimensional data of bulk cargo provides a more reliable basis for inventory. This mechanism not only enhances the credibility of warehouse management, but also brings significant improvement and optimization to enterprise inventory management.

[0078] Embodiment 2:

[0079] The above-mentioned embodiment 1 divides the shelves into partitions by bulk cargo size, and sets a full load threshold to automatically trigger an inventory count. The bulk cargo information is double-checked using RFID tags and original purchase information, and 3D point cloud data collection and confirmation is used when there is a discrepancy, thereby improving the accuracy and efficiency of the inventory count. In order to further improve the efficiency and accuracy of the inventory count and the efficiency of bulk cargo storage and retrieval, the embodiment 1 is further improved, and a liquidity index threshold is set for each partition, and the inventory count condition is triggered according to the threshold of the liquidity index.

[0080] Now, further improvements are made on the basis of step S102 in the first embodiment, specifically:

[0081] S201, calculating the bulk cargo liquidity index of each inventory area and the inventory area liquidity index.

[0082] Among them, the calculation formula of bulk cargo liquidity index LI is designed as follows:

[0083]

[0084] The coefficient D reflects the market demand for bulk goods. It is determined by analyzing historical sales data, market trends, seasonal changes and other factors. For example, it can be expressed as the ratio of the average monthly sales volume of a certain bulk product to the average monthly sales volume of all bulk products in the past year;

[0085] Inventory turnover rate I refers to the number of times that bulk inventory is turned over within a certain period (such as a month or a quarter). The calculation formula is: Inventory turnover rate = outbound volume / average inventory volume. The average inventory volume can be calculated by (beginning inventory volume + ending inventory volume) / 2.

[0086] The coefficient E reflects the frequency of bulk cargo entering the warehouse, which can be determined by counting the ratio of the number of bulk cargo entering the warehouse to the total number of warehouse entries within a certain period of time;

[0087] The coefficient O reflects the frequency of bulk cargo outflow from the warehouse, which can be determined by counting the ratio of the number of bulk cargo outflows to the total number of warehouse outflows within a certain period of time.

[0088] The shelf distribution coefficient S takes into account the impact of the location of bulk goods on the shelf on liquidity. It can be set according to the distance of the bulk goods from the entrance, exit or channel. The closer the distance, the smaller the coefficient. For example, a piecewise function can be used to represent that the coefficient of the location close to the entrance, exit or channel is 0.5, the center location is 1.0, and the far distance location is 1.5.

[0089] The calculation formula of regional liquidity index RLI is designed as follows:

[0090]

[0091] n represents the number of bulk cargoes in the region, LI_i represents the liquidity index of the ith bulk cargo, calculated according to the single bulk cargo liquidity index formula given above, and W_i represents the weight of the ith bulk cargo, which can be determined based on the value, importance, sales volume, etc. of the bulk cargoes. If all bulk cargoes have the same importance, W_i can be simply set to 1 / n, that is, each bulk cargo has an equal weight. If some bulk cargoes have a greater impact on regional liquidity, the weight W_i can be adjusted to reflect this.

[0092] S202, defining a range of bulk cargo liquidity indexes, and adjusting the regional position of the bulk cargoes according to the size of the liquidity indexes.

[0093] Specifically, the calculated bulk cargo liquidity index LI threshold can be divided into several intervals, each of which corresponds to a different liquidity level;

[0094] In this embodiment, the following liquidity intervals are defined:

[0095] The low liquidity interval is LI<1.0, the medium liquidity interval is 1.0≤LI<2.0, and the high liquidity interval is LI≥2.0; bulk cargo within the high liquidity interval threshold should be placed in areas close to entrances and exits or near passages for easy inventory.

[0096] S203, defining a threshold range of the liquidity index of the inventory area, and an inventory area exceeding the threshold range triggers an inventory condition.

[0097] Specifically, based on historical analysis, we understand the operational efficiency and inventory counting frequency of the warehouse under different RLI values ​​and the business needs. Under different RLI levels, we can understand the inventory counting frequency and inventory accuracy requirements that warehouse management can accept, and set one or more RLI thresholds to trigger inventory counting conditions.

[0098] In this embodiment, two thresholds can be set: a low threshold RLI_low=1.5, which is used to indicate that the regional liquidity is moderate and regular inventory counting is required to ensure inventory accuracy; a high threshold RLI_high=2.5, which is used to indicate that the regional liquidity is high and frequent inventory counting is required to prevent out-of-stock or excess inventory;

[0099] When the RLI value of a certain inventory area exceeds the set high threshold RLI_high, the inventory condition is triggered immediately and an emergency inventory is performed on the area; when the RLI value is between the low threshold RLI_low and the high threshold RLI_high, the inventory is performed according to the predetermined plan or cycle; if the RLI value is lower than RLI_low and there are no other special circumstances (such as bulk goods are about to expire, special customer needs, etc.), the inventory cycle can be extended.

[0100] For example, assume the following RLI thresholds are set:

[0101] RLI_low=1.5, RLI_high=2.5;

[0102] When the RLI of a certain inventory area is greater than 2.5, the inventory condition is immediately triggered and an emergency inventory is performed.

[0103] When 1.5≤RLI≤2.5, inventory is conducted once a week.

[0104] When RLI < 1.5, inventory is taken monthly.

[0105] S204, regularly updating the liquidity index threshold.

[0106] It should be noted that the bulk liquidity index LI interval and the regional liquidity index RLI threshold range need to be adjusted regularly.

[0107] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0108] 1. By comprehensively considering multiple factors such as market demand, inventory turnover rate, bulk cargo in-and-out frequency, and shelf distribution, a scientific bulk cargo liquidity index and inventory area liquidity index calculation system has been constructed. This system can not only accurately reflect the flow status of bulk cargo, but also guide the reasonable layout of bulk cargo in the warehouse, thereby effectively improving inventory efficiency and inventory accuracy. At the same time, by setting flexible liquidity index thresholds and corresponding inventory strategies, the solution can trigger inventory conditions according to actual conditions, which not only avoids unnecessary inventory waste, but also ensures the real-time and accuracy of inventory information, greatly improving the overall efficiency of warehouse operations.

[0109] 2. The plan also emphasizes the importance of regularly updating the liquidity index threshold to ensure that the inventory strategy can be adjusted with changes in warehouse operations and market environment to maintain its effectiveness and adaptability. This dynamic management concept not only enhances the flexibility and responsiveness of warehouse management, but also helps enterprises to promptly discover and solve potential problems in inventory management, providing strong support for the sustainable development and optimization of enterprises.

[0110] Embodiment three:

[0111] The above-mentioned second embodiment optimizes the bulk cargo layout and inventory strategy by comprehensively considering the liquidity factors of bulk cargo, thereby improving warehouse operation efficiency and inventory accuracy. At the same time, the threshold is flexibly set and the management strategy is dynamically adjusted to ensure efficient and adaptable inventory management and comprehensively optimize warehouse management. In order to further improve the efficiency of bulk cargo inventory, the second embodiment is further improved. According to the liquidity index of the bulk cargo being counted within a certain window period, the liquidity index is prioritized to achieve efficient and rapid inventory of bulk cargo.

[0112] Now, further improvements are made on the basis of step S201 in the second embodiment, specifically:

[0113] S301, define the liquidity index of bulk cargo within the window time period.

[0114] Among them, for each type of bulk cargo counted, within a certain window period (a window period is defined as 15 days)

[0115] S302, prioritizing the liquidity index of bulk cargo.

[0116] S303, after the inventory counting condition is triggered, the inventory counting starts with the bulk cargo with high priority.

[0117] Specifically, after the inventory condition is triggered, instead of counting all bulk cargoes at the same time, the bulk cargoes with the highest priority are counted first according to the previously determined priorities. The bulk cargoes are counted in order of priority to ensure that the bulk cargoes with high liquidity and high management requirements are processed first.

[0118] S304, regularly updating the priority of bulk cargo liquidity index.

[0119] Among them, the basic cycle is set at 15 days.

[0120] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0121] 1. By scientifically defining the liquidity index of bulk cargo and assigning priorities, accurate measurement and effective management of cargo liquidity are achieved. This helps warehouses prioritize cargo with high liquidity and high management requirements when inventory conditions are triggered, thereby improving the pertinence and efficiency of inventory counting and reducing unnecessary waste of resources.

[0122] 2. A dynamic update mechanism has been established to regularly update the liquidity index and priority of bulk cargo, ensuring that the inventory strategy is always consistent with the actual flow of goods and management requirements. This flexibility and adaptability enables the warehouse to better respond to market changes and adjust inventory management strategies in a timely manner, thereby improving the overall operation level and competitiveness of the warehouse.

[0123] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent inventory counting of bulk cargo based on 3D point cloud segmentation, characterized in that: include: S101, dividing the shelf into inventory areas of different sizes according to the size of the bulk goods; S102, defining conditions for triggering inventory counting according to the full load of the bulk cargo classification area, and when the inventory counting conditions are met, scanning the bulk cargo RFID tags on the shelves to obtain detailed information of the bulk cargo carried in the tags; S103, according to the identification code of the bulk cargo, the corresponding original purchase information is retrieved from the purchase information database; S104, comparing the retrieved original purchase information with the information in the RFID tag, and if the contents of all identical fields are consistent, generating an inventory record based on the RFID tag information; If there are inconsistent fields, the 3D point cloud data collection process is started to generate inventory records; S105, traversing the inventory record table stored in the database according to the identification code of the bulk cargo, and if the record of the bulk cargo already exists in the inventory record table, performing replacement processing according to a predetermined rule; If there is no record in the inventory record sheet, add a new inventory record for the bulk cargo in the inventory record sheet.

2. A bulk cargo intelligent inventory method based on 3D point cloud segmentation as claimed in claim 1, characterized in that: The S102, full load condition, includes: The full load condition of the bulk cargo classification area is defined as the capacity threshold of the shelf in the set area. 80%-85% of the capacity threshold is the full load threshold, which triggers the inventory counting condition.

3. A bulk cargo intelligent inventory method based on 3D point cloud segmentation as claimed in claim 1, characterized in that: The step S104, starting the 3D point cloud data acquisition process, includes: In order to use the depth camera Intel RealSense D435 to collect 3D point cloud data of bulk cargo in inconsistent fields, obtain its precise 3D shape and size data, segment the collected 3D point cloud data, and extract the 3D model of the corresponding bulk cargo.

4. A bulk cargo intelligent inventory method based on 3D point cloud segmentation as claimed in claim 1, characterized in that: The S104 further includes: The extracted 3D model features are compared again with the RFID tag information and the original purchase information to see which one has a higher matching degree between the extracted 3D model features and the RFID tag information and the original purchase information, which is used as the basis for generating inventory records.

5. A bulk cargo intelligent inventory method based on 3D point cloud segmentation as claimed in claim 1, characterized in that: The S102 includes: S201, calculating the bulk cargo liquidity index and the inventory area liquidity index of each inventory area; S202, defining the range of bulk cargo liquidity index, and adjusting the regional position of bulk cargo according to the size of the liquidity index; S203, defining a threshold range of the liquidity index of the inventory area, and an inventory area exceeding the threshold range triggers an inventory condition.

6. A method for intelligent inventory counting of bulk cargo based on 3D point cloud segmentation as claimed in claim 5, characterized in that: The step S201, calculating the bulk cargo liquidity index of each inventory area, includes: The calculation formula of bulk cargo liquidity index LI is designed as follows: Among them, coefficient D reflects the market demand for bulk goods, inventory turnover rate I refers to the number of times bulk goods in inventory are turned over within a certain period of time, coefficient E reflects the frequency of bulk goods entering the warehouse, coefficient O reflects the frequency of bulk goods flowing out of the warehouse, and shelf distribution coefficient S takes into account the impact of the position of bulk goods on the shelf on liquidity.

7. A bulk cargo intelligent inventory method based on 3D point cloud segmentation as claimed in claim 5, characterized in that: The step S201, calculating the regional liquidity index, includes: The calculation formula of regional liquidity index RLI is designed as follows: Where n represents the number of bulk cargoes in the region, LI_i represents the liquidity index of the i-th bulk cargo, calculated according to the liquidity index formula of a single piece of bulk cargo given previously, and W_i represents the weight of the i-th bulk cargo; Wi can be set to 1 / n, that is, the weight of each bulk cargo is equal; The impact of a certain bulk cargo on regional liquidity can be reflected by adjusting the weight W_i.

8. A method for intelligent bulk cargo inventory based on 3D point cloud segmentation as claimed in claim 5, characterized in that: S202, defining the range of bulk cargo liquidity index, includes: Define the low liquidity interval LI<1.0, the medium liquidity interval 1.0≤LI<2.0, and the high liquidity interval LI≥2.0; adjust the position of bulk goods in the shelf and the position of the shelf in the classification area according to the threshold size of the flow interval.

9. A method for intelligent bulk cargo inventory based on 3D point cloud segmentation as claimed in claim 5, characterized in that: The step S203, defining a threshold range of the liquidity index of the inventory area, includes: Set at least one threshold value to trigger the inventory condition; When more than one threshold is set, the low threshold RLI_low and the high threshold RLI_high define the range of all thresholds that appear; Set a basic inventory cycle. When the RLI value of a certain inventory area exceeds the set high threshold RLI_high, the inventory condition is immediately triggered and an emergency inventory is performed on the area. When the RLI value is between the low threshold RLI_low and the high threshold RLI_high, the inventory is taken according to the basic inventory cycle; when the RLI value is lower than RLI_low and there is no goods about to expire or special customer needs, the inventory cycle can be extended by 1.3 times.

10. The method for intelligent bulk cargo inventory based on 3D point cloud segmentation according to claim 5, characterized in that: The S201 includes: S301, defining the liquidity index of bulk cargo within the window period; S302, prioritize the liquidity index of bulk cargo; S303, after the inventory counting condition is triggered, the inventory counting starts with the bulk cargo with high priority; S304, regularly updating the priority of bulk cargo liquidity index.

Images