Method for obtaining storage location information in visual warehouse
By dynamically evaluating the credibility of humidity values and analyzing the humidity change characteristics in the visual warehouse, the inaccuracy problem of traditional warehouse location information acquisition methods is solved, efficient monitoring and control of the warehouse environment is achieved, and the safety of items is ensured. Accuracy of the location information.
Patent Information
- Application Number
- CN202411155146.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-08-22
AI Technical Summary
Traditional warehouse location information acquisition methods are susceptible to environmental interference, label damage, and limited coverage in large-scale and highly dynamic warehouse management, resulting in inaccurate storage conditions and unable to meet the needs of efficient warehouse management.
By obtaining the humidity value sequence according to the preset sampling interval in the visual warehouse, the humidity value credibility at each sampling time is calculated, and the target window and its left and right observation window are constructed. The humidity is analyzed through the confidence correction value and clustering algorithm. The stability and consistency of values, dynamically evaluate the reliability of humidity values, identify abnormal situations, and improve the accuracy and efficiency of environmental monitoring.
It realizes accurate acquisition of warehouse location information, improves the accuracy and responsiveness of environmental monitoring, ensures the stability of the warehouse environment and the safety of items, reduces false alarms and missed reports, and provides a powerful decision-making support tool.
Smart Images

Figure CN118674377B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of electronic digital processing, and more specifically, to a method for acquiring storage location information of a visual warehouse. Background Art
[0002] As the core link of logistics and supply chain management, warehouses undertake multiple functions such as storage, custody, and allocation. Efficient warehouse management can not only reduce inventory costs and improve material circulation efficiency, but also ensure the stability and responsiveness of the supply chain. With the development of information technology, visual warehouse management has gradually become an important means to improve warehouse operation efficiency. Through visualization technology, managers can monitor warehouse status in real time, optimize storage location allocation, and respond to market changes in a timely manner. Traditional methods of obtaining storage location information mainly rely on barcode scanning, RFID technology, etc. These methods have limitations in practical applications, such as being susceptible to environmental interference, label damage, and limited coverage, which makes it difficult to meet large-scale, highly dynamic warehouse management needs.
[0003] The existing Chinese patent application document with publication number CN115271602A discloses a smart fruit and vegetable warehouse management method, device and electronic equipment. The method obtains first fruit and vegetable information, wherein the first fruit and vegetable information includes the name of the fruit and vegetable; according to the preset fruit and vegetable storage conditions, confirms the first storage conditions corresponding to the first fruit and vegetable information, wherein the preset fruit and vegetable storage conditions include temperature and humidity; according to the first storage conditions, determines a first storage area, and stores the first fruit and vegetable in the first storage area.
[0004] However, the temperature and humidity of fruit and vegetable storage are within a range. If the temperature or humidity value fluctuates, the storage conditions of the first storage area will change, resulting in inaccurate storage location information and unsuitable storage. Summary of the invention
[0005] In order to solve the problem of inaccurate storage location information, the present invention proposes a method for obtaining storage location information of a visual warehouse.
[0006] The present invention discloses a method for acquiring storage location information of a visualized warehouse, comprising: acquiring a humidity value sequence in time order according to a preset sampling interval; taking any sampling moment as a target moment, and calculating the credibility of the humidity value corresponding to the target moment; taking a window constructed with the target moment as the center as the target window, and obtaining a left observation window on the left side of the target window and a right observation window on the right side of the target window, wherein the window sizes of the left observation window and the right observation window are consistent with the window size of the target window; and calculating a credibility correction value, wherein the credibility correction value satisfies the relationship:
[0007] , Indicates time The corresponding humidity value is the credibility correction value. Indicates time The credibility of the corresponding humidity value, represents the variance of the first-order difference of the humidity values in the left observation window, represents the variance of the first-order difference of humidity values in the right observation window, represents the mean of the humidity values in the left observation window, represents the mean of the humidity values in the right observation window, Represents the normalization function.
[0008] Traverse the humidity value sequence to obtain the credibility correction value sequence, and cluster the credibility correction value sequence to complete the storage location information.
[0009] The credibility of the humidity value at each sampling moment is calculated, providing a quantitative indicator for the reliability of the data. By constructing the target window and its left and right observation windows, the scheme further analyzes the local variation characteristics of the humidity value, and comprehensively considers the stability and consistency of the humidity value through the calculation of the credibility correction value. Through precise data preprocessing, dynamic credibility evaluation and advanced data analysis technology, the accuracy and efficiency of environmental monitoring are improved.
[0010] Preferably, the credibility includes: setting a target period with the target time as the center, calculating the outlier mean of the humidity value in the target period; calculating the credibility of the humidity value corresponding to the target time, and the credibility satisfies the relationship:
[0011] ,in, Indicates time The credibility of the corresponding humidity value, Indicates time Outliers corresponding to humidity values, Indicates the number of times in the target period when the outlier corresponding to the humidity value is greater than the mean of the outliers. Represents the total number of moments in the target period. In the credibility relation, the moment For the target moment, Represents an exponential function.
[0012] The reliability of the humidity value at each sampling moment can be dynamically evaluated. Using an exponential function to constrain the credibility can make the calculation of the credibility very sensitive to the size and quantity changes of outliers, so that it can quickly respond to abnormal situations in the data and improve the accuracy of anomaly detection.
[0013] Preferably, the target time period is any time period including the target time.
[0014] Preferably, the outlier includes: constructing a first sequence of humidity values corresponding to each moment in the target window; replacing the humidity value at the target moment with the mean of the humidity values at two moments adjacent to the target moment in the first sequence to construct a second sequence; performing curve fitting on the first sequence and the second sequence to calculate the outlier, and the outlier satisfies the relationship:
[0015] ,in, Indicates time Outliers corresponding to humidity values, Indicates the window size of the target window, is the fitting curve of the first sequence, is the fitting curve of the second sequence. In the outlier relationship, the moment The target moment.
[0016] A humidity value sequence based on the target window is constructed, and the mean of the humidity values at adjacent moments is used to smooth the data at the target moment, effectively reducing the impact of random fluctuations on the credibility of the humidity value. The outliers are calculated using the integral method. The scheme not only considers the humidity value changes in the entire target window, but also quantifies the abnormality of the humidity value at each moment by comparing the fitting curve differences between the original and smoothed sequences.
[0017] Preferably, the outlier also includes: constructing a first sequence of humidity values corresponding to each moment in the target window; replacing the humidity value at the target moment with the average of the humidity values at two moments adjacent to the target moment in the first sequence to construct a second sequence; fitting the first sequence to obtain a first curve, and fitting the second sequence to obtain a second curve; calculating the similarity between the first curve and the second curve, and taking the value of the similarity as the outlier.
[0018] Fitting the two series and calculating the similarity of the two fitted curves can identify abnormal fluctuations in humidity values and evaluate the continuity and smoothness of the data, thereby effectively distinguishing normal changes from potential outliers.
[0019] Preferably, the method of completing the storage location information includes: clustering the credibility correction value sequence using a preset clustering algorithm, calculating the distance between each data point and the cluster center, and normalizing the distance, wherein one data point corresponds to one credibility correction value; when the distance is greater than a preset distance threshold, the data point is noise, and the humidity value at the corresponding moment is an abnormal humidity value, and an alarm signal is generated and sent; traversing all data points to complete the denoising of the humidity value sequence.
[0020] The detection accuracy of abnormal environmental humidity data is improved, and by traversing all data points, comprehensive denoising of the entire humidity value sequence is achieved. This not only ensures the accuracy and reliability of environmental monitoring data, but also provides timely feedback for the environmental control system, which helps to quickly respond to and deal with potential environmental problems, thereby ensuring the stability and effectiveness of the environmental monitoring system.
[0021] Preferably, the preset clustering algorithm is density clustering.
[0022] Preferably, the normalization utilizes maximum and minimum value normalization.
[0023] Beneficial effects of the present invention:
[0024] 1. By carefully designing two observation windows, located on both sides of the window corresponding to the target time, the recent change trend of humidity values can be captured, which provides contextual information for understanding the fluctuation of humidity values at the target time. By conducting an in-depth analysis of the variance and mean differences of humidity values in the two observation windows, the consistency and abnormality of humidity changes can be identified, and then the credibility of humidity values at the target time can be corrected in a well-founded manner. This dynamic correction mechanism makes the acquisition of storage location information not only rely on data at a single moment, but also integrates the analysis of time series, increasing the depth and breadth of the evaluation.
[0025] 2. Normalization ensures the comparability of data at different time points or different storage locations, while the introduction of clustering algorithms identifies meaningful data from a large amount of data, which can represent the stability of environmental conditions or potential trends. By setting distance thresholds to identify and process noise, the quality and reliability of data are further improved, and false alarms and missed alarms are reduced, thereby ensuring the accuracy of alarm signals and the timeliness of actions.
[0026] 3. It not only improves the timeliness and accuracy of the acquisition of storage location information, but also enhances the adaptability and responsiveness to environmental changes, providing a powerful decision-making support tool for warehouse managers. The invention can more effectively monitor and control the warehouse environment to ensure the safety and integrity of stored items. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:
[0028] Figure 1 It is a flow chart of a method for acquiring storage location information of a visualized warehouse according to an embodiment of the present invention. DETAILED DESCRIPTION
[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0030] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.
[0031] The present invention provides a method for obtaining storage location information of a visual warehouse. Figure 1 As shown, the method for obtaining storage location information of a visual warehouse includes steps S1 to S4, which are described in detail below.
[0032] S1, obtaining a humidity value sequence in chronological order according to a preset sampling interval.
[0033] In one embodiment, a humidity sensor is used to collect humidity data of a specific storage location. The specific storage location is generally a warehouse that requires humidity monitoring, such as a warehouse storing fresh food, medicine, and electronic products.
[0034] For warehouses storing perishable foods such as fruits, vegetables, meat and dairy products, humidity control is essential to prevent food spoilage and extend shelf life. Pharmaceuticals and biological products have strict requirements for storage conditions, and too high or too low humidity may affect the stability and effectiveness of drugs. Electronic components and equipment are very sensitive to humidity, and improper humidity conditions may cause equipment damage or performance degradation.
[0035] Exemplarily, the preset sampling interval is set to 10 seconds in the present invention, and can be specifically set by those skilled in the art. The humidity value at each sampling moment is obtained according to the preset sampling interval, and a humidity value sequence is constructed.
[0036] S2, calculate the reliability of the humidity value corresponding to the target time.
[0037] It should be noted that since there are workers entering and exiting the warehouse, moving items, cleaning, etc., this will cause the humidity sensor to be affected by environmental factors and produce noise. The credibility of the humidity value at each moment is the probability that the humidity value is normal data (humidity value not affected by noise), and the credibility of the humidity value is negatively correlated with the probability that the humidity value is a noise point.
[0038] In one embodiment, any sampling moment is used as the target moment, and the target period is set around the target moment. It should be noted that in a warehouse where sensitive items are stored, the flow of personnel is relatively small, and the flow of personnel will be concentrated in a period of time in order to minimize the impact of the flow of personnel on the environment in the warehouse. The target period is to better distinguish whether the fluctuation of humidity values in the period is caused by the flow of personnel or by a fault. Exemplarily, the target period is set to 5 minutes.
[0039] The window constructed with the target time as the center is used as the target window, and the humidity values corresponding to each time in the target window are constructed as the first sequence; the humidity value of the target time is replaced by the mean of the humidity values of the two times adjacent to the target time in the first sequence to construct the second sequence.
[0040] For example, the window size of the target window is set to 9, and the humidity values contained in the first sequence are , is the humidity value at the target time, then the humidity values contained in the second sequence are .
[0041] Perform curve fitting on the first and second sequences and calculate outliers. The outliers satisfy the relationship:
[0042] ,in, Indicates time Outliers corresponding to humidity values, Indicates the window size of the target window, is the fitting curve of the first sequence, is the fitting curve of the second sequence. In the outlier relationship, the moment The target moment.
[0043] In the domain 1 to The fitting curve of the first sequence between The area enclosed by the axis, In the domain 1 to The fitting curve of the second sequence between Since there are different results of curve fitting, the difference in area can reflect the change of the second sequence compared with the first sequence by using the integral form.
[0044] By multiplying the integral It is to balance the influence of window length on the integral result. For example, if the goal is to obtain the end of the sampling time for the start of the sampling time, that is, the window at the target time is incomplete, then the integral result when the window is incomplete will be smaller than the integral result when the window is complete. After averaging, the difference in the integral result caused by reducing the window size can be avoided.
[0045] Calculate the mean of the outliers of the humidity value in the target period, and calculate the credibility of the humidity value corresponding to the target time. The credibility satisfies the relationship:
[0046] ,in, Indicates time The credibility of the corresponding humidity value, Indicates time Outliers corresponding to humidity values, Indicates the number of times in the target period when the outlier corresponding to the humidity value is greater than the mean of the outliers. Represents the total number of moments in the target period. In the credibility relation, the moment For the target moment, Represents an exponential function.
[0047] In one embodiment, the calculation of the outlier also includes: constructing the humidity value corresponding to each moment in the target window into a first sequence; replacing the humidity value at the target moment with the mean of the humidity values at two moments adjacent to the target moment in the first sequence to construct a second sequence; fitting the first sequence to obtain a first curve, and fitting the second sequence to obtain a second curve; calculating the similarity between the first curve and the second curve, and using the value of the similarity as the outlier. Exemplarily, the similarity includes DTW (dynamic time warping), cosine similarity, and Pearson correlation coefficient, etc., and the present invention adopts DTW.
[0048] In one embodiment, the target period can also be any period including the target moment. In this way, the target period includes different times, and the corresponding humidity values will also have different situations. There are many possibilities, and the accuracy is lower than the target period constructed around the target moment.
[0049] S3, calculating the credibility correction value.
[0050] It should be noted that the humidity in warehouses storing sensitive items often needs to be maintained within a set range. When the humidity exceeds the range, the dehumidifier or humidifier needs to be started immediately to ensure that the humidity is within the set range. The start of the dehumidifier or humidifier will cause the collected data to change immediately and then stabilize. That is, when the warehouse controls the humidity, there will be points where the data changes rapidly. This change may lead to an increase in outlier calculations, so the credibility of the humidity value needs to be corrected.
[0051] In one embodiment, a left observation window on the left side of the target window and a right observation window on the right side of the target window are obtained, wherein the window sizes of the left observation window and the right observation window are consistent with the window size of the target window.
[0052] Calculate the credibility correction value, the credibility correction value satisfies the relationship:
[0053] , Indicates time The corresponding humidity value is the credibility correction value. Indicates time The credibility of the corresponding humidity value, represents the variance of the first-order difference of the humidity values in the left observation window, represents the variance of the first-order difference of humidity values in the right observation window, represents the mean of the humidity values in the left observation window, represents the mean of the humidity values in the right observation window, Represents the normalization function.
[0054] The variance of the first-order difference is to first make adjacent differences in the data in the window to obtain a difference sequence, and then calculate the variance of the difference sequence. For example, the window size is set to 5, and the humidity values contained in the window are 50, 46, 54, 60, and 55. After calculating the first-order difference, the difference sequence is 4, 8, 6, and 5, and the calculated variance is 2.1875.
[0055] It is the reciprocal of the mean, the main influencing factor of the credibility correction value, and measures the average change of humidity values in two observation windows.
[0056] It should be noted that if the window sizes of the two observation windows cannot be complete, that is, the size of the left observation window is 7, while the size of the right observation window can only be 5, conversely, the size of the right observation window can only be 7, while the size of the left observation window can only be 5, then Set to 0 to ignore humidity changes.
[0057] S4, traverse the humidity value sequence to obtain the credibility correction value sequence, cluster the credibility correction value sequence to complete the acquisition of storage location information.
[0058] In one embodiment, a preset clustering algorithm is used to cluster the credibility correction value sequence, the distance between each data point and the cluster center is calculated, and the distance is normalized, wherein one data point corresponds to one credibility correction value; when the distance is greater than a preset distance threshold, the data point is noise, and the humidity value at the corresponding moment is an abnormal humidity value, and an alarm signal is generated and sent. Exemplarily, the preset distance threshold is set to 0.72.
[0059] Traverse all data points to complete the denoising of the humidity value sequence.
[0060] Exemplarily, the preset clustering algorithm is density clustering, and the normalization uses maximum and minimum value normalization.
[0061] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.
[0062] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for obtaining storage location information of a visual warehouse, characterized in that: include: Obtain a sequence of humidity values in chronological order according to a preset sampling interval; Take any sampling time as the target time and calculate the credibility of the humidity value corresponding to the target time, including: The target time period is set with the target time as the center, and the outlier mean of the humidity value in the target time period is calculated. The outliers include: Construct the humidity value corresponding to each moment in the target window into a first sequence; The humidity value at the target time is replaced by the average of the humidity values at two times adjacent to the target time in the first sequence to construct a second sequence; Perform curve fitting on the first and second sequences and calculate outliers. The outliers satisfy the relationship: ,in, Indicates time Outliers corresponding to humidity values, Indicates the window size of the target window, is the fitting curve of the first sequence, is the fitting curve of the second sequence. In the outlier relationship, the moment For the target moment; Calculate the reliability of the humidity value corresponding to the target time, and the reliability satisfies the relationship: ,in, Indicates time The credibility of the corresponding humidity value, Indicates the number of times in the target period when the outlier corresponding to the humidity value is greater than the mean of the outliers. Represents the total number of moments in the target period. In the credibility relation, the moment For the target moment, represents the exponential function; The window constructed with the target time as the center is used as the target window, and a left observation window on the left side of the target window and a right observation window on the right side of the target window are obtained, wherein the window sizes of the left observation window and the right observation window are consistent with the window size of the target window; Calculate the credibility correction value, the credibility correction value satisfies the relationship: , Indicates time The corresponding humidity value is the credibility correction value. represents the variance of the first-order difference of the humidity values in the left observation window, represents the variance of the first-order difference of humidity values in the right observation window, represents the mean of the humidity values in the left observation window, represents the mean of the humidity values in the right observation window, represents the normalization function; The humidity value sequence is traversed to obtain the credibility correction value sequence, and the credibility correction value sequence is clustered to complete the acquisition of storage location information.
2. A method for obtaining storage location information of a visual warehouse, characterized in that: include: Obtain a sequence of humidity values in chronological order according to a preset sampling interval; Take any sampling time as the target time and calculate the credibility of the humidity value corresponding to the target time, including: The target time period is set with the target time as the center, and the outlier mean of the humidity value in the target time period is calculated. The outliers include: Construct the humidity value corresponding to each moment in the target window into a first sequence; The humidity value at the target time is replaced by the average of the humidity values at two times adjacent to the target time in the first sequence to construct a second sequence; The first sequence is fitted to obtain a first curve, and the second sequence is fitted to obtain a second curve; Calculate the similarity between the first curve and the second curve, and use the value of the similarity as an outlier; Calculate the reliability of the humidity value corresponding to the target time, and the reliability satisfies the relationship: ,in, Indicates time The credibility of the corresponding humidity value, Indicates time Outliers corresponding to humidity values, Indicates the number of times in the target period when the outlier corresponding to the humidity value is greater than the mean of the outliers. Represents the total number of moments in the target period. In the credibility relation, the moment For the target moment, represents the exponential function; The window constructed with the target time as the center is used as the target window, and a left observation window on the left side of the target window and a right observation window on the right side of the target window are obtained, wherein the window sizes of the left observation window and the right observation window are consistent with the window size of the target window; Calculate the credibility correction value, the credibility correction value satisfies the relationship: , Indicates time The corresponding humidity value is the credibility correction value. represents the variance of the first-order difference of the humidity values in the left observation window, represents the variance of the first-order difference of humidity values in the right observation window, represents the mean of the humidity values in the left observation window, represents the mean of the humidity values in the right observation window, represents the normalization function; The humidity value sequence is traversed to obtain the credibility correction value sequence, and the credibility correction value sequence is clustered to complete the acquisition of storage location information.
3. The method for obtaining storage location information of a visual warehouse according to claim 1 or 2, characterized in that: The target time period is any time period including the target time.
4. The method for obtaining storage location information of a visual warehouse according to claim 1 or 2, characterized in that: The completed storage location information includes: Clustering the credibility correction value sequence using a preset clustering algorithm, calculating the distance between each data point and the cluster center, and normalizing the distance, wherein one data point corresponds to one credibility correction value; When the distance is greater than the preset distance threshold, the data point is noise, and the humidity value at the corresponding moment is an abnormal humidity value, an alarm signal is generated and sent; Traverse all data points to complete the denoising of the humidity value sequence.
5. The method for obtaining storage location information of a visual warehouse according to claim 4, characterized in that: The preset clustering algorithm is density clustering.
6. The method for obtaining storage location information of a visual warehouse according to claim 4, characterized in that: The normalization uses maximum and minimum value normalization.
