Search

CN-121981647-A - Method, device and storage medium for detecting commodity quantity of hook

CN121981647ACN 121981647 ACN121981647 ACN 121981647ACN-121981647-A

Abstract

The application discloses a method, equipment and storage medium for detecting the quantity of commodities of a hook, and belongs to the technical field of commodity management systems. Determining a commodity weight value of a commodity on a hook based on weight sensor data, carrying out time sequence analysis on the commodity weight value, constructing weight time sequence data, identifying a target data segment in the weight time sequence data through a preset sliding time window, identifying waveform characteristics of the target data segment through an event analysis model, determining a weight change event of the commodity weight value based on the waveform characteristics, calculating a weight difference value of the commodity according to the weight change event and the waveform characteristics, calculating the commodity change quantity according to the weight difference value and the commodity unit weight corresponding to the hook, and updating commodity quantity information corresponding to the hook. According to the application, the commodity weight value is determined according to the received weight sensor data, and the change of the commodity number is analyzed according to the commodity weight value, so that the fine sensing of the commodity number change is realized, and the accuracy of inventory management is improved.

Inventors

  • WU LILIN

Assignees

  • 广东瀚墨科技股份有限公司

Dates

Publication Date
20260505
Application Date
20251226

Claims (10)

  1. 1. The method for detecting the commodity quantity of the hooks is characterized by comprising the following steps of: receiving weight sensor data acquired by a weight sensor, and determining a commodity weight value of a commodity on the hook based on the weight sensor data; Carrying out time sequence analysis on the commodity weight value, constructing weight time sequence data, and identifying a target data segment in the weight time sequence data through a preset sliding time window; Identifying waveform characteristics of the target data segment through an event analysis model, and determining a weight change event of the commodity weight value based on the waveform characteristics; Calculating the weight difference of the commodity according to the weight change event and the waveform characteristic, and calculating the commodity change quantity according to the weight difference and the commodity unit weight corresponding to the hook; And updating commodity quantity information corresponding to the hooks according to the commodity change quantity.
  2. 2. The method for detecting the number of commodities in a hook according to claim 1, wherein said step of performing time-series analysis on the commodity weight values, constructing weight time-series data, and identifying a target data segment in the weight time-series data through a preset sliding time window includes: Constructing time-based weight time sequence data of the commodity weight value through the time sequence analysis; extracting a weight value data segment from the weight time sequence data through the sliding time window; Calculating the dispersion of the weight value data segments, and selecting target data segments of the commodity from the weight value data segments according to the dispersion.
  3. 3. The method of claim 2, wherein the step of identifying waveform characteristics of the target data segment by an event analysis model and determining a weight change event of the commodity weight value based on the waveform characteristics comprises: Extracting a multi-dimensional feature vector of the target data segment to form the waveform feature, wherein the multi-dimensional feature vector comprises a rising edge slope, a falling edge slope, a peak value number and/or signal energy distribution; Sequentially carrying out classification decision on the waveform characteristics through hierarchical nodes of a classification decision tree in the event analysis model to form a classification path; determining the event type of the weight change event according to the classification path; the weight change event is formed based on the event type and the multi-dimensional feature vector.
  4. 4. The method of claim 1, wherein the step of identifying waveform characteristics of the target data segment by an event analysis model and determining a weight change event of the commodity weight value based on the waveform characteristics further comprises: determining a starting weight value and an ending weight value of the commodity weight value according to the waveform characteristics; calculating a difference between the initial weight value and the end weight value to obtain the weight difference; Determining a difference confidence interval according to the commodity unit weight corresponding to the hook sensor; and matching the weight difference with the difference confidence interval, and determining the event type of the weight change event according to a matching result.
  5. 5. The method of detecting the number of articles in a hanger according to claim 3 or 4, wherein the step of calculating the weight difference of the articles based on the weight change event and the waveform feature, and calculating the number of article change based on the weight difference and the article unit weight corresponding to the hanger comprises: Acquiring the event type of the weight change event; if the event type is a non-interference event, calculating the weight difference value according to a starting weight value and an ending weight value of the commodity weight value in the waveform characteristic; Acquiring commodity type information associated with the hooks, and acquiring commodity unit weight based on the commodity type information; and calculating the quotient of the weight difference and the unit weight of the commodity to obtain the commodity change quantity.
  6. 6. The method of claim 1, wherein the step of receiving weight sensor data collected by a weight sensor and determining a commodity weight value of the commodity on the hook based on the weight sensor data comprises: Based on sensor time sequence information corresponding to the weight sensor data, identifying a linear increasing trend of the weight sensor data; Determining an environmental impact factor according to the linear growth trend; and removing the environmental impact factor from the weight sensor data to obtain the commodity weight value.
  7. 7. The method for detecting the number of commodities in a hook according to claim 1, wherein after said step of calculating the weight difference of said commodities according to said event type and said waveform characteristics and calculating the number of commodity changes according to said weight difference and the commodity unit weight corresponding to said hook, further comprising: Acquiring a target weight change event of at least one target sensor in a preset time period, and determining the weight of a first unit commodity according to the target weight change event; calculating the weight of a second unit commodity according to the weight difference value and the commodity change quantity; Matching the first unit commodity weight with the second unit commodity weight, and determining the drift amount of the weight sensor according to a matching result; and calibrating the weight sensor based on the drift amount.
  8. 8. The method for detecting the number of commodities in a hook according to claim 1, wherein after said step of calculating the weight difference of said commodities according to said event type and said waveform characteristics and calculating the number of commodity changes according to said weight difference and the commodity unit weight corresponding to said hook, further comprising: Acquiring the weight change event in a preset time interval, and determining time sequence information of the weight change event; Acquiring a time interval sequence and a weight change direction sequence based on the time sequence information; and identifying user behavior information according to the time interval sequence and the weight change direction sequence.
  9. 9. A hook commodity number detection apparatus, characterised in that the apparatus comprises a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the hook commodity number detection method according to any one of claims 1 to 8.
  10. 10. A storage medium, characterized in that the storage medium is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the method for detecting the number of goods of a hanger according to any one of claims 1 to 8.

Description

Method, device and storage medium for detecting commodity quantity of hook Technical Field The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, and a storage medium for detecting the number of commodities in a hook. Background In inventory management and intelligent shelving systems in the retail industry, monitoring of the quantity of items is typically accomplished based on radio frequency identification (RFID, radio Frequency Identification) in order to accurately perceive the change in the quantity of items in real time. Specifically, in the RFID technology, an electronic tag is attached to an article, tag signals are periodically scanned by using readers around the shelf, and the pick-and-place state of the article is determined by identifying the departure and the approach of the tag, so that the change of the inventory quantity is estimated. However, although RFID technology can sense the presence status of goods, it cannot precisely detect the gradual progress of the number of goods. When a customer takes off or hangs back a part of the same goods from the hook, the system can only judge that the goods are operated, but cannot accurately distinguish whether the operation is single-piece picking and placing, multi-piece picking and placing or goods are replaced, so that inventory count deviation is caused, and real-time inventory management accurate to units is difficult to support. The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art. Disclosure of Invention The application mainly aims to provide a method, equipment and storage medium for detecting the quantity of commodities in a hook, and aims to solve the technical problem that a commodity management system is insufficient in counting accuracy of the quantity of the commodities in the hook. In order to achieve the above object, the present application provides a method for detecting the number of articles in a hook, comprising the steps of: receiving weight sensor data acquired by a weight sensor, and determining a commodity weight value of a commodity on the hook based on the weight sensor data; Carrying out time sequence analysis on the commodity weight value, constructing weight time sequence data, and identifying a target data segment in the weight time sequence data through a preset sliding time window; Identifying waveform characteristics of the target data segment through an event analysis model, and determining a weight change event of the commodity weight value based on the waveform characteristics; Calculating the weight difference of the commodity according to the weight change event and the waveform characteristic, and calculating the commodity change quantity according to the weight difference and the commodity unit weight corresponding to the hook; And updating commodity quantity information corresponding to the hooks according to the commodity change quantity. In an embodiment, the step of performing time-series analysis on the commodity weight value to construct weight time-series data, and identifying the target data segment in the weight time-series data through a preset sliding time window includes: Constructing time-based weight time sequence data of the commodity weight value through the time sequence analysis; extracting a weight value data segment from the weight time sequence data through the sliding time window; Calculating the dispersion of the weight value data segments, and selecting target data segments of the commodity from the weight value data segments according to the dispersion. In one embodiment, the step of identifying waveform characteristics of the target data segment by an event analysis model and determining a weight change event of the commodity weight value based on the waveform characteristics comprises: Extracting a multi-dimensional feature vector of the target data segment to form the waveform feature, wherein the multi-dimensional feature vector comprises a rising edge slope, a falling edge slope, a peak value number and/or signal energy distribution; Sequentially carrying out classification decision on the waveform characteristics through hierarchical nodes of a classification decision tree in the event analysis model to form a classification path; determining the event type of the weight change event according to the classification path; the weight change event is formed based on the event type and the multi-dimensional feature vector. In one embodiment, the step of identifying waveform characteristics of the target data segment by an event analysis model and determining a weight change event of the commodity weight value based on the waveform characteristics further comprises: determining a starting weight value and an ending weight value of the commodity weight value acco