Consigned luggage data intelligent management method and system, product and medium

Abstract

The invention discloses an intelligent management method and system for consigned luggage data, a product and a medium. The method comprises the steps that when no-label luggage is detected, basic feature data of the luggage is obtained through scanning and compared with a sports equipment feature database, an equipment feature matching probability value is calculated, and a high-probability candidate equipment set is screened out; acquiring a flight consignment equipment list in a preset time period, and extracting an intersection of the candidate equipment set and the consignment list to obtain candidate consignment equipment and a corresponding flight; calculating a number matching probability of a flight corresponding to each candidate device, and calculating a final matching probability by combining the device feature matching probability and the number matching probability with a weight coefficient; and when the maximum matching probability exceeds a preset threshold value, associating the label-free luggage with the corresponding flight and equipment type, and adding an electronic label. By implementing the technical scheme provided by the invention, the identification efficiency of the label-free luggage is improved during a large-scale sports event.

Description

Technical Field

[0001] The present application relates to the field of baggage tracking systems, and in particular to a method, system, product and medium for intelligent management of checked baggage data. Background Art

[0002] The transportation and management of specialized equipment, a crucial component of sporting events, places higher demands on airport baggage handling systems. Especially during concentrated sporting events, the safe and efficient transportation of large quantities of specialized sports equipment directly impacts the smooth running of the events, making intelligent airport baggage management a critical component of sporting event security.

[0003] In related technologies, airport baggage management systems primarily process baggage through an intelligent sorting system. This system uses image recognition technology to obtain basic feature information of baggage and compares it with a pre-established baggage database.

[0004] However, in related technologies, when multiple sports teams need to transit through the same transfer airport to different competition zones, a large amount of sports equipment is centrally handled at the transfer airport. These sports equipment from different teams enter the baggage sorting system simultaneously. Due to the excessive size of these equipment, labels can easily become detached or damaged during transportation, making it impossible for the system to identify the specific flights to which these equipment belong. In this case, matching only a single piece of equipment feature makes it difficult to accurately determine which flight each untagged piece of equipment should be sorted, which can easily lead to misclassification. Inefficient baggage identification can also delay event transit times. Summary of the Invention

[0005] The present application provides a method, system, product, and medium for intelligent management of checked baggage data, which are used to improve the efficiency of identifying untagged baggage during large-scale sporting events.

[0006] In a first aspect of the present application, a method for intelligent management of checked baggage data is provided, the method comprising: When an untagged baggage item for which no valid tag is detected occurs, the baggage's three dimensions and outline data are scanned. Based on the baggage's three dimensions and outline data, the baggage is compared with a preset sports equipment feature database to calculate the probability of matching the untagged baggage's features with various types of sports equipment. Equipment types whose equipment feature matching probability exceeds a preset first feature matching probability threshold are selected to obtain a candidate equipment set. The checked equipment list for all flights within a preset time period based on the current time is obtained. The intersection of the candidate equipment set and the equipment types in the checked equipment list for each flight is extracted to obtain a candidate checked equipment set and candidate flights corresponding to the candidate checked equipment set. For each equipment type in the candidate checked equipment set, the equipment quantity matching probability for all candidate flights is calculated. Based on all equipment quantity matching probabilities and equipment feature matching probabilities, the probabilities are combined pairwise and combined with preset weight coefficients to obtain a final matching probability set. The final matching probability set is sorted to obtain the maximum matching probability value. When the maximum matching probability value exceeds the preset matching probability threshold, the untagged baggage item is associated with the flight and equipment type corresponding to the maximum matching probability value and an electronic tag is added.

[0007] In the above-described embodiment, untagged baggage is 3D scanned to obtain size and contour data, which is then matched against a pre-set database of sports equipment features to screen out possible equipment types. Subsequently, the checked equipment lists for recent flights are queried, and the correlation between the baggage and the flight is comprehensively assessed by calculating the intersection of equipment types and quantity matching probabilities, combined with the equipment feature matching probabilities. This combined physical feature recognition and flight information verification is suitable for handling large quantities of specialized equipment during sporting events. When athletes check in specialized equipment with detached tags, dual verification of equipment features and flight information allows the baggage to be quickly and accurately linked to the correct flight, significantly improving the efficiency of untagged baggage identification during major sporting events.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, before comparing the luggage's three-dimensional dimensions and the luggage's outline data with a preset sports equipment feature database, the method further includes: Calculate the matching degrees between the three-dimensional size parameters and the external contour data of the luggage and the complete form parameters and each separated form parameter of each sports equipment in the sports equipment feature database respectively, to obtain the complete form matching degree and multiple separated form matching degrees; when the maximum value among the multiple separated form matching degrees of the candidate separated equipment is higher than the complete form matching degree, mark the unlabeled luggage as a separated part, and obtain the set of candidate separated equipment types; according to each candidate separated equipment in the set of candidate separated equipment types, calculate the three-dimensional size parameters and the external contour data of the separated part of the luggage, and calculate the remaining separated shape parameters and the remaining separated three-dimensional size parameters of the remaining separated parts; according to the remaining separated shape parameters and the remaining separated three-dimensional size parameters, calculate the complementary matching degrees for all the unlabeled luggage within the preset detection time period; when there is unlabeled luggage with a complementary matching degree exceeding the preset matching threshold, mark it as a candidate complementary part; take the candidate complementary part with the highest complementary matching degree value among all the candidate complementary parts as the complementary part; according to the candidate separated equipment type corresponding to the complementary part, splice and reconstruct the three-dimensional size parameters and the external contour data of the separated part and the candidate complementary part to obtain the reconstructed complete form parameters and the complete three-dimensional size parameters.

[0009] In the above embodiment, the separated parts are identified through feature matching analysis, then the complementary parts are searched within the preset time window, and finally the complete equipment form is restored through the reconstruction algorithm, realizing the intelligent processing of the disassembled equipment. By establishing the correlation between the separated parts, not only the recognition accuracy of the disassembled equipment is improved, but also in the case of partial label damage, other related parts can be reversely located through the identified parts, ensuring the integrity tracking and accurate delivery of the separated equipment, and enhancing the adaptability of the luggage sorting system to special-shaped equipment.

[0010] Combined with some embodiments of the first aspect, in some embodiments, it is characterized in that calculating the complementary matching degrees for all the unlabeled luggage within the preset detection time period according to the remaining separated shape parameters and the remaining separated three-dimensional size parameters specifically includes: Obtain the equipment weight parameter and the equipment material parameter of each candidate equipment in the set of candidate equipment types; according to the equipment weight parameter, the equipment material parameter, the separated shape parameter and the separated three-dimensional size parameter of the separated part, calculate the separated weight parameter and the separated material feature parameter of the separated part; perform distribution statistics on the weight parameters and the material feature parameters of all the unlabeled luggage within the detection time period to obtain the weight distribution interval and the material feature distribution interval; according to the remaining separated shape parameters and the remaining separated three-dimensional size parameters, calculate the complementary matching degrees for the unlabeled luggage within the distribution intervals corresponding to the expected weight parameters and the material feature parameters.

[0011] In the above embodiments, by analyzing the weight distribution law and material characteristics of standard equipment, and combining the shape and size parameters of the separation components, the weight and material characteristics of the components to be matched are predicted. It is possible to statistically analyze the weight and material characteristics of all unlabeled luggage within the time window, establish a distribution range, and thus locate potential complementary components among a large number of luggage. This improves the accuracy and efficiency of the matching of separation components. At the same time, by establishing a distribution model of weight and material, the recognition ability for different types of complementary components is enhanced, providing a more accurate and efficient matching solution for the handling of special luggage.

[0012] In combination with some embodiments of the first aspect, in some embodiments, after scanning the three-dimensional size parameters and the external contour data of the unlabeled luggage, it further includes: Obtain the grid point cloud data on the surface of the unlabeled luggage and the pressure distribution map of the contact surface with the conveyor belt, and reconstruct a surface model; when it is determined that there is a break or missing part in the surface model or calculate the force values of each area in the pressure distribution map, and when the force value of any area exceeds the preset upper threshold or is lower than the preset lower threshold, the unlabeled luggage is determined to be damaged luggage; determine the force concentration area according to the pressure distribution map or the surface model, generate a buffer support plan for strengthening the force concentration area, and output a buffer operation instruction corresponding to the buffer support plan.

[0013] In the above embodiments, surface features are obtained through high-precision scanning, and at the same time, force data is collected using a pressure sensor array. The surface reconstruction algorithm is used to fuse these data into a complete three-dimensional model. When a surface break or abnormal pressure is found, the damaged situation can be identified in time, and a corresponding buffer support plan can be generated based on the force analysis, which can effectively prevent secondary damage during transportation. By combining surface feature analysis with pressure distribution monitoring, the accuracy of luggage damage recognition and the pertinence of protective measures are significantly improved, providing a reliable guarantee for the safe transportation of luggage.

[0014] In combination with some embodiments of the first aspect, in some embodiments, to determine the force concentration area according to the pressure distribution map or the surface model and generate a buffer support plan for strengthening the force concentration area, it specifically includes: Calculate the ratio of the force value to the force area of the force concentration area to obtain the pressure value per unit area, and select a matching buffer material type from the preset corresponding table of buffer material pressure grades according to the pressure value per unit area; based on the force area and the buffer material type, calculate the required size and quantity of the buffer material.

[0015] In the above embodiments, by calculating the pressure value per unit area of the stress concentration area and combining the preset corresponding relationship between the pressure levels of the buffer materials, the selection and dosage calculation of the buffer materials are realized. A buffer protection plan can be formulated according to the specific stress characteristics of the equipment. This method of material selection and matching based on calculation improves the pertinence and effectiveness of buffer protection. At the same time, by establishing the corresponding relationship between the stress characteristics and the material properties, the system's ability to respond to different types of damage situations is enhanced, providing a more scientific protection plan for the safe transportation of checked luggage.

[0016] In combination with some embodiments of the first aspect, in some embodiments, when there is an unlabeled luggage for which no valid label is detected, after scanning the three-dimensional size parameters and the external contour data of the unlabeled luggage, it further includes: Obtaining the surface temperature distribution data of the unlabeled luggage; dividing the surface temperature distribution data into multiple temperature regions according to a preset temperature difference range; counting the area proportion of each temperature region; when there is a special temperature region in the unlabeled luggage, marking the unlabeled luggage as a luggage to be processed preferentially.

[0017] In the above embodiments, by obtaining the surface temperature distribution data of the luggage, identifying and statistically analyzing the area of the temperature abnormal region, the timely discovery and preferential processing of the temperature abnormal luggage are realized. By analyzing the temperature distribution and statistically analyzing the region, the luggage that needs special treatment can be identified, improving the timeliness of the treatment of special luggage. At the same time, by establishing an identification mechanism for temperature abnormality, the protection ability for temperature-sensitive luggage is enhanced, providing a more efficient treatment plan for the safe transportation of checked luggage.

[0018] In combination with some embodiments of the first aspect, in some embodiments, scanning the three-dimensional size parameters and the external contour data of the unlabeled luggage specifically includes: Detecting the reflectance of the surface of the unlabeled luggage; when the detected reflectance exceeds the preset reflectance threshold, adjusting the scanning light intensity of the scanning device to obtain the scanning data of the unlabeled luggage; performing reflectance compensation processing on the scanning data to obtain the three-dimensional size parameters and the external contour data of the luggage.

[0019] In the above embodiments, by detecting the reflectance of the luggage surface in real time, dynamically adjusting the scanning light intensity, and performing reflectance compensation processing on the obtained data, the precise scanning of the high-reflectance luggage is realized. Through the self-adaptive adjustment of the light intensity and the data compensation processing, clear and complete luggage feature data can be obtained, improving the scanning quality and data accuracy of the luggage with special surfaces.

[0020] Second aspect, embodiments of the present application provide an intelligent management system for checked baggage data. The intelligent management system for checked baggage data includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the intelligent management system for checked baggage data to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] Third aspect, embodiments of the present application provide a computer program product containing instructions. When the computer program product runs on the intelligent management system for checked baggage data, it causes the intelligent management system for checked baggage data to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] Fourth aspect, embodiments of the present application provide a computer-readable storage medium including instructions. When the instructions run on the intelligent management system for checked baggage data, it causes the intelligent management system for checked baggage data to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] It can be understood that the intelligent management system for checked baggage data provided in the second aspect above, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the intelligent management method for checked baggage data provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, and will not be elaborated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By collecting and analyzing the basic characteristics of the baggage, and combining flight data and checked baggage list information, the present application realizes the association between equipment and flights. Especially during large-scale sports events, when multiple participating teams check a large number of professional equipment at the same time, and some equipment labels are damaged or fallen off, it is possible to identify the ownership of unlabeled baggage through equipment feature matching and quantity distribution analysis. On the one hand, the accuracy of equipment type judgment is ensured through the matching of the feature database, and on the other hand, the quantity association of the flight checked baggage list provides an additional verification basis to improve the identification efficiency of unlabeled baggage during large-scale sports events.

[0025] 2. This application identifies separated components through feature matching analysis, then searches for complementary components within a preset time window, and finally restores the complete equipment form through a reconstruction algorithm, realizing the intelligent processing of disassembled equipment. By establishing the correlation between separated components, not only the recognition accuracy of disassembled equipment is improved, but also when some labels are damaged, other related components can be inversely located through the identified components, ensuring the integrity tracking and accurate delivery of separated equipment, and enhancing the adaptability of the baggage sorting system to equipment in special forms.

[0026] 3. This application reconstructs the luggage surface model through grid point cloud data to identify structural anomalies; then through pressure distribution analysis, evaluates the impact degree of damage on the luggage structure; finally, based on the force characteristics, formulates an accurate buffer support plan. This method is applicable when checked luggage is damaged. It can identify risk points through structural analysis and mechanical modeling and generate corresponding protection plans. It improves the safety handling ability of damaged luggage, and at the same time, by formulating a targeted buffer support plan, enhances the protection effect on damaged luggage, providing an accurate and reliable guarantee plan for the safe transportation of checked luggage. Brief Description of the Drawings

[0027] Figure 1 is a flowchart of the intelligent management method for checked luggage data in an embodiment of this application; Figure 2 is another flowchart of the intelligent management method for checked luggage data in an embodiment of this application; Figure 3 is an exemplary hardware structure diagram of the intelligent management system for checked luggage data in an embodiment of this application. Detailed Embodiments

[0028] The terms used in the following embodiments of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in the specification and appended claims of this application, the singular forms "a", "an", "the", "above", "said", "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as implying or indicating relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of this application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0030] In the related art, the airport baggage management system mainly relies on an intelligent sorting system to process baggage. This system obtains the basic characteristic information of the baggage through a scanning device, and performs characteristic matching with a preset baggage database to achieve automatic classification of the baggage. However, during large-scale sports events, when multiple representative teams need to transfer through the same transit airport to different competition areas, a large number of professional sports equipment will be centrally processed at the transit airport. Due to the oversized volume of this sports equipment from different representative teams, the labels are prone to falling off or being damaged during transportation. At this time, it is difficult to accurately determine which flight each unlabeled piece of equipment should be sorted to only by matching a single equipment characteristic, which is likely to cause misclassification, reduce the baggage recognition efficiency, and may lead to delays in event transfers.

[0031] In the embodiment of the present application, an intelligent management method for checked baggage data is proposed. This method not only considers the physical characteristics of the equipment, but also combines the flight checked baggage list information to establish a dual verification mechanism. First, the candidate equipment type is determined through characteristic matching, and then the flight checked baggage list within a preset time period is obtained, and the intersection of the two is extracted to obtain the candidate checked baggage and the corresponding flight. By calculating the weighted combination of the equipment characteristic matching probability and the quantity matching probability, the unlabeled equipment is associated with the correct flight, improving the recognition efficiency of unlabeled baggage during large-scale sports events.

[0032] Figure 1 is a schematic flowchart of using the intelligent management method for checked baggage data in the embodiment of the present application, including the following steps: S101. When unlabeled baggage with no valid label detected appears, scan the three-dimensional size parameters and the outer contour data of the unlabeled baggage.

[0033] Specifically, the baggage is detected for labels through label reading devices arranged on both sides of the conveyor belt. When the label reading device fails to detect valid electronic label information within the preset label reading time, the three-dimensional scanning device is activated. This three-dimensional scanning device uses the multi-point laser ranging technology, emits lasers to the surface of the baggage through laser emitters at multiple different angles, and receives the reflected signals through the receiver. According to the time difference between the emission time and the reception time of the laser signal, combined with the emission angle of the laser, the spatial coordinate information of each point on the surface of the baggage is calculated using the triangulation ranging principle.

[0034] Secondly, the obtained spatial coordinate information is processed. First, a three-dimensional point cloud model of the baggage is constructed through the spatial coordinate set, and then the minimum circumscribed rectangle algorithm is used to calculate the length, width, and height parameters of the smallest cuboid including the point cloud model, so as to obtain the three-dimensional size parameters of the baggage. At the same time, the point cloud data is subjected to surface reconstruction, and the discrete point cloud data is converted into a continuous surface model using the grid processing method, so as to obtain the outer contour data of the baggage.

[0035] In some embodiments, in the case of stacked luggage, a multi-level scanning mode is enabled. The boundaries of different pieces of luggage are distinguished through an image segmentation algorithm, and scanning and data processing are performed separately.

[0036] In other embodiments, when unlabeled luggage is detected, the mesh point cloud data and pressure distribution map of the luggage surface can also be obtained to identify the damaged state of the luggage and generate a corresponding buffer support plan, improving the transportation safety of damaged luggage, thereby effectively preventing further damage to the damaged luggage during transportation and ensuring the safety of luggage transportation.

[0037] First, the mesh point cloud data of the luggage surface is obtained through a high-precision three-dimensional scanning device, and at the same time, a pressure sensor array is used to collect the pressure distribution data of the contact surface between the luggage and the conveyor belt. The three-dimensional scanning device uses structured light scanning technology. By projecting a grating pattern with a specific code and collecting the deformed pattern, it is converted into high-density three-dimensional point cloud data. For irregularly shaped luggage, the scanning density and range are adjusted to ensure complete surface data is obtained. The pressure sensor array continuously collects the pressure data of the contact surface between the luggage and the conveyor belt. These sensors use advanced induction technologies such as capacitive or piezoelectric, and can capture minute pressure changes. The collected pressure data forms a complete pressure distribution map after spatial interpolation processing.

[0038] Subsequently, a surface reconstruction algorithm, such as Poisson reconstruction or B-spline surface fitting, is used to fuse and reconstruct the point cloud data and pressure distribution data to generate a complete surface model of the luggage.

[0039] By analyzing the continuity and integrity of the surface, possible breaks or missing parts can be identified. Specifically, by calculating the normal vector and principal curvature of each point on the surface, the degree of mutation of the normal vector in adjacent regions and the abnormal change of curvature are evaluated to locate potential damaged positions. At the same time, the force values in each region of the pressure distribution map are analyzed. When the force value in a certain region exceeds the preset threshold range, combined with the surface analysis results, it can be accurately determined whether there is structural damage in this region.

[0040] For the regions confirmed to be damaged, the unit area pressure value of the force concentration region is further calculated, which is obtained by calculating the ratio of the force value to the force area in the force concentration region, to evaluate the force condition of each damaged region. Based on the unit area pressure value, a preset buffer material pressure grade corresponding table is queried. This corresponding table contains various buffer materials and the corresponding unit area pressure value ranges, so as to select the most suitable type of buffer material.

[0041] Finally, based on the stress area of the damaged area and the selected type of cushioning material, the specific dimensions and quantity of the required cushioning material are determined through material mechanics calculations. This calculation process takes into account the deformation characteristics and stress distribution laws of the material to ensure that the cushioning material can provide sufficient support strength.

[0042] After selecting the appropriate cushioning material, an optimal layout plan for the cushioning material is designed according to the geometric characteristics and stress conditions of the damaged location. Specifically, for the area with concentrated stress, the best placement angle and position of the cushioning material are calculated to ensure effective dispersion and absorption of the impact force. At the same time, considering possible vibrations, collisions, etc. during transportation, additional protective layers are added at key stress points. Generate cushioning operation instructions, including information such as the usage method of the material and the position of the fixing points, to guide the operator to accurately execute the protection measures, thereby providing comprehensive transportation protection for the damaged luggage.

[0043] Through the above technical steps, a complete technical chain from damage detection to protection plan generation is constructed, which can formulate personalized protection plans for different types and degrees of damage situations. While improving the transportation safety of damaged luggage, it also optimizes the material usage efficiency, realizing the intelligence and precision of damaged luggage protection.

[0044] S102: Based on the three-dimensional size parameters and the external contour data of the luggage, compare with the preset sports equipment feature database, calculate the equipment feature matching probability values between the unlabeled luggage and various sports equipment, and filter out the equipment types whose equipment feature matching probability values exceed the preset first feature matching probability threshold to obtain the candidate equipment set.

[0045] Specifically, first, in the form of multi-dimensional feature vectors, feature extraction and quantization processing are performed on the obtained three-dimensional size parameters and external contour data of the luggage. For the three-dimensional size parameters, not only the specific numerical values of the three-dimensional size parameters are considered, but also derivative feature parameters such as the length-width ratio and width-height ratio are calculated; for the external contour data, geometric features of the contour are extracted through methods such as Fourier descriptors, including curvature distribution, symmetry, concave-convex features, etc. These features together constitute a high-dimensional vector space for describing the luggage features.

[0046] The preset sports equipment feature database stores standard feature templates of various sports equipment. These templates contain the range of feature parameters and variation laws of the equipment in different states (such as complete state, separated state, packaged state, etc.). The database adopts a hierarchical organizational structure, and different categories of equipment are grouped according to feature similarity for quick retrieval and matching.

[0047] When performing feature matching, a multi-level matching strategy is adopted. First is the coarse-grained matching, which quickly filters out the possible major categories of equipment through the proportional relationship of three-dimensional dimensions; then the fine-grained matching is carried out, using detailed shape features for precise comparison. The improved cosine similarity algorithm is used to calculate the similarity between the luggage feature vector and each equipment template in the database, and combined with the feature weight system to generate a comprehensive matching probability value.

[0048] S103. Obtain the list of checked equipment for all flights within a preset time period before the current time.

[0049] Specifically, first, a multi-level data acquisition architecture is established to obtain and update flight check-in information in real time. In the time dimension, taking the current moment as the reference point, all flight data within a preset time period (such as 6 hours) is traced back. For each flight, detailed information on the checked equipment of the corresponding flight is collected, including classification data of equipment types, the specific quantity of each type of equipment, and information such as the quantity of equipment that has been confirmed by the intelligent recognition system and the quantity of equipment to be recognized.

[0050] In terms of data storage, a distributed database architecture is adopted, and the check-in data is indexed and stored according to the timestamp and flight number. The database design uses the characteristics of a time series database, supporting time range queries and aggregation statistics. At the same time, similar or related equipment types are classified into the same major category for multi-dimensional statistical analysis.

[0051] When performing data statistics, a sliding time window mechanism is adopted. This mechanism can dynamically update the statistical results and timely reflect the latest check-in trends. For each time window, key indicators such as the quantity distribution and recognition rate of various types of equipment are calculated, and through time series analysis methods, the periodic patterns and short-term change trends of equipment check-in are identified.

[0052] S104. Extract the intersection of the candidate equipment set and the equipment types in the checked equipment list of each flight to obtain the candidate checked equipment set and the corresponding candidate flights of the candidate checked equipment set.

[0053] Specifically, a mapping relationship between the candidate equipment set and flight check-in records is established. For each equipment type in the candidate equipment set, the checked equipment lists of all flights are retrieved, and the set intersection operation is used to find the equipment types that appear in both sets. This process uses data structures such as bitmap indexes or hash tables to complete the intersection calculation of large-scale data.

[0054] When performing set intersection operations, a multi-level matching strategy is adopted. First is the precise matching of equipment types to find exactly the same equipment types; second is fuzzy matching, considering the category hierarchy relationship of the equipment. For example, "professional snowboard" can be matched with the "snow sports equipment" category. Synonyms and variant forms of the equipment are also considered, and by maintaining a correspondence table of equipment names, it is ensured that valid matches are not missed due to differences in appellations.

[0055] For each successfully matched equipment type, record the flight information corresponding to the equipment, including key data such as flight number, departure and arrival times, and the quantity of checked luggage. This information is organized into a candidate set of checked equipment and forms an associated mapping with the corresponding candidate flights. Through this mapping relationship, relevant flights can be quickly located based on the equipment type, and the distribution of relevant equipment can be queried based on flight information.

[0056] S105. For each equipment type in the candidate set of checked equipment, calculate the matching probability of the quantity of equipment corresponding to all candidate flights respectively.

[0057] Specifically, for each candidate flight, obtain two key data of this type of equipment in the checked list corresponding to the candidate flight: the quantity of equipment that has passed system identification (identified quantity) and the total quantity of checked-in equipment (total quantity). By calculating the ratio of the identified quantity to the total quantity, the quantity matching probability of this type of equipment for this flight is obtained. This probability value reflects the completion degree of equipment identification for the flight and also indirectly indicates the potential association degree between the flight and the luggage to be identified.

[0058] A dynamic weight mechanism is adopted when calculating the matching probability. When the identified quantity of a flight is close to the total quantity, it indicates that the equipment identification work for this flight is nearly completed. At this time, if there is unlabeled luggage of the same type, the possibility that it belongs to this flight is relatively low. On the contrary, if the identified quantity is much less than the total quantity, it indicates that there are still many pieces of equipment to be identified for this flight, and at this time the unlabeled luggage is more likely to belong to this flight.

[0059] In some embodiments, to improve the accuracy of probability calculation, the influence of time factors is also considered. For flights with adjacent time periods, the timeliness of their data is relatively high, and the calculated probability values will be given a larger weight; while for flights in earlier time periods, the credibility of their data is relatively low, and the weight of the probability values will be correspondingly reduced.

[0060] S106. Based on all the equipment quantity matching probabilities and equipment feature matching probability values, combine them in pairs respectively and sum them up with the preset weight coefficients to obtain the final set of matching probability values.

[0061] Specifically, first, construct a probability combination matrix to pair the equipment feature matching probability values and the equipment quantity matching probabilities pairwise. For each pair of probability values, apply a preset weight coefficient for weighted calculation. The weight coefficient reflects the importance of different matching dimensions in the final decision-making and can generally be expressed as: the final matching probability value = w1 × equipment quantity matching probability + w2 × equipment feature matching probability, where w1 and w2 are weight coefficients and w1 + w2 = 1. This weighted calculation method ensures that the final probability value takes into account both the similarity of the external features of the equipment and the rationality of the quantity distribution.

[0062] In some embodiments, for equipment with special shapes or non-standard sizes (such as special-shaped snowboards, customized golf club sets, specially made surfboards, etc.), such equipment usually has unique external features and size parameters, and the feature matching results have strong recognition value. By evaluating the standardization degree index of the equipment (obtained by calculating the deviation degree of the equipment size and shape from the standard specifications), when the standardization degree index exceeds the preset standardization degree index threshold, the weight of the feature matching probability will be increased accordingly. Conversely, for equipment with a higher degree of standardization (such as standard basketballs, volleyballs, etc.), due to the lower distinguishability of the external features, more reliance will be placed on the quantity matching probability for judgment, and at this time, the weight ratio will be adjusted to favor the quantity matching probability. This adaptive weight adjustment mechanism based on the standardization degree of the equipment can make full use of the significant features of different types of equipment and improve the accuracy of recognition.

[0063] S107. Sort the set of final matching probability values to obtain the maximum matching probability value.

[0064] Specifically, perform a sorting process on the set of final matching probability values. The sorting process uses an improved quicksort algorithm, which has high efficiency when dealing with large-scale probability value sets. Arrange all probability values in descending order while maintaining the associated mapping of the probability values with the corresponding equipment types and flight information to ensure that data relevance is not lost during the sorting process.

[0065] To improve the reliability of the sorting results, establish a multiple verification mechanism. First is the probability value validity verification, which checks whether the probability value is within the valid range (between 0 and 1) and eliminates outliers; second is the probability value significance verification. When the difference between the highest probability value and the second-highest probability value is less than the preset threshold, this situation will be marked for further analysis; finally is the probability value consistency verification, which checks whether the probability distribution of the same equipment type on different flights is reasonable.

[0066] S108. When the maximum matching probability value exceeds the preset matching probability threshold, associate the unlabeled luggage with the flight and equipment type corresponding to the maximum matching probability value and add an electronic tag.

[0067] Specifically, first, a dynamic matching probability threshold mechanism is set. This matching probability threshold is an adaptive parameter that is dynamically adjusted according to different scenarios and conditions. By analyzing historical recognition data, the recognition accuracy rates of different types of equipment at different times are calculated to establish a benchmark threshold. At the same time, the characteristics of the current operating environment (such as passenger flow, time urgency, etc.) are also considered to adjust the benchmark threshold in real time.

[0068] When it is determined that the maximum matching probability value exceeds the preset threshold, a multi-step association confirmation process will be initiated. First is the information verification step, where the characteristic data of the luggage and the checked baggage information of the corresponding flight are double-checked to ensure data consistency. Second is the conflict check step, where the equipment list of the flight is checked to ensure that adding a new association relationship will not cause data contradictions.

[0069] After verification is completed, an electronic tag is generated and an association relationship is established. The electronic tag contains a multi-level information structure: the basic information layer contains key information such as flight number, destination, equipment type, etc.; the extended information layer contains auxiliary information such as recognition time, matching probability value, characteristic parameters, etc.; the status information layer is used to record the status changes of the luggage during transportation.

[0070] In some embodiments, for high-value or special requirement sports equipment, an enhanced version of the recognition and verification mechanism is adopted. First, the value level and special nature are judged according to the characteristic parameters of the equipment (such as the specific dimensions, material characteristics, brand logos of professional-level equipment, etc.). When high-value or special requirement equipment is recognized, the matching probability threshold is increased, and a multiple verification process is initiated: including the accuracy review of equipment characteristics, the cross-verification of historical checked baggage records, the associated verification of relevant event information, etc. When generating the electronic tag, multiple special attribute marks are added to the tag information, such as "high-value equipment", "special handling requirements", etc. At the same time, a real-time monitoring mechanism is established to continuously track the transportation status of this type of equipment through Internet of Things devices, and an alarm is immediately issued once an abnormal situation (such as severe vibration, abnormal temperature, etc.) occurs to ensure the safety and reliability of the entire transportation process. This differentiated processing strategy effectively guarantees the transportation safety of high-value sports equipment by raising the recognition standard, increasing the verification links, and strengthening the monitoring measures.

[0071] In the above embodiments, first, the candidate equipment type is determined through feature matching, and then combined with the flight consignment list information, through the weighted calculation of the equipment feature matching probability and the quantity matching probability, the classification of unlabeled luggage is achieved. During large-scale sports events, when the professional equipment of multiple teams needs to be transferred at the same transfer airport, this multi-dimensional matching analysis can be used to associate unlabeled equipment with the correct flight. This intelligent recognition method based on feature matching and quantity statistics improves the accuracy and processing efficiency of luggage recognition. At the same time, by establishing the association between equipment features and flight information, the processing capacity of a large number of professional equipment is enhanced, and the recognition efficiency of unlabeled luggage is improved during large-scale sports events.

[0072] In some other embodiments of the present application, during large-scale sports events, when athletes disassemble and consign large equipment for easy transportation, it may occur that the part labels fall off, resulting in the inability to identify the complete equipment. By using the intelligent management method for checked luggage data provided in the present application, the identification and association of the equipment in the separated state can be realized through steps such as surface feature analysis, separated part identification, and complementary matching reconstruction.

[0073] As Figure 2 shown, it is another process schematic diagram of the intelligent management method for checked luggage data provided by the embodiments of the present application, including the following steps: S201. When unlabeled luggage with no valid label detected appears, detect the reflectance of the surface of the unlabeled luggage.

[0074] Specifically, a multi-spectral reflectance detection device is used to perform an all-round reflectance scan on the surface of the luggage. The device includes multiple light sources with different wavelengths (such as visible light, near-infrared light, etc.) and a high-precision photoelectric sensor array. The light sources irradiate the surface of the luggage according to the preset time sequence and angle, and the sensors synchronously collect the reflected light signals. By calculating the ratio of the incident light intensity to the reflected light intensity, the reflectance data at different positions and different wavelengths are obtained.

[0075] A dynamic compensation mechanism is adopted when collecting the reflectance data. First is the ambient light compensation, by real-time monitoring the ambient light intensity, eliminating the interference of external light sources; second is the angle compensation, by establishing the mapping relationship between the reflection angle and the reflectance, correcting the measurement errors caused by different incident angles; finally is the surface state compensation, considering the influence of factors such as the roughness and dirt of the luggage surface on the reflectance.

[0076] In order to improve the measurement accuracy, a multi-region scanning strategy is established. The surface of the luggage is divided into multiple detection regions, and independent reflectance measurements are performed on each region. Through statistical analysis, characteristic parameters such as the average value and variance of the reflectance of each region are calculated to generate a complete reflectance distribution map.

[0077] In some embodiments, an adaptive enhanced detection strategy is adopted for luggage with special surface features. For luggage with special coatings on the surface (such as waterproof coatings, scratch-resistant coatings, reflective coatings, etc.), first, the coating type is identified through pre-scanning, and then the scanning frequency and the density of sampling points are increased. At the same time, the most suitable light source wavelength is selected according to the optical properties of the coating (for example, ultraviolet light source is used for fluorescent coatings, and near-infrared light source is used for infrared reflective coatings) for precise scanning.

[0078] S202. When the detected reflectance exceeds the preset reflectance threshold, adjust the scanning light intensity of the scanning device to obtain the scanning data of the unlabeled luggage.

[0079] Specifically, first, an adaptive light intensity adjustment mechanism is established. When the reflectance of the luggage surface is detected to exceed the preset threshold, start the light intensity dynamic adjustment program. This program calculates the optimal incident light intensity by analyzing the intensity distribution of the reflected light signal in real time. The adjustment process uses an iterative optimization algorithm to find the best light intensity parameters through multiple fine-tuning to ensure obtaining a clear scanning image.

[0080] A sub-region control strategy is adopted when adjusting the light intensity. First, divide the scanning area into multiple sub-regions, and independently adjust the light intensity according to the reflection characteristics of each region. For regions with particularly high reflectance, significantly reduce the light source power; for regions with relatively low reflectance, maintain the standard light intensity, so as to ensure that appropriate exposure can be obtained for the entire surface. At the same time, the reference light intensity will also be adjusted in real time according to the ambient light conditions to maintain the stability of the scanning quality.

[0081] To improve the scanning effect, a multi-mode scanning strategy is adopted. It includes: standard mode for conventional scanning; low light intensity high frequency mode for processing high-reflective regions; pulse mode to reduce radiation interference through short-time high-frequency sampling. According to the preliminary scanning results, automatically select the most suitable combination of scanning modes.

[0082] S203. Perform reflectance compensation processing on the scanning data to obtain the three-dimensional size parameters of the luggage and the luggage outer contour data.

[0083] Specifically, first, a multi-level reflectance compensation mechanism is established. By analyzing the reflectance interference patterns in the scanning data, an adaptive filtering algorithm is used to eliminate the data noise caused by reflectance. This algorithm includes three main processing levels: the first level is local reflectance correction, which performs pixel-level compensation for small-scale high-brightness points; the second level is regional reflectance equalization, which processes the uneven reflectance in a larger range; the third level is global reflectance compensation to ensure the consistency of the overall data.

[0084] When performing specular reflection compensation, an intelligent reference point technology is adopted. By identifying reliable reference feature points (such as corners, textures, etc.) in the scanned data, a benchmark system for data compensation is established. Using these reference points to construct a three-dimensional grid model, the data missing areas caused by specular reflection are repaired through interpolation algorithms to ensure the acquisition of complete surface contour information.

[0085] To accurately extract the three-dimensional dimensions, a multi-view data fusion technology is used. First, the luggage is scanned from multiple angles to obtain the dimensional data from different perspectives. Then, through feature point matching and three-dimensional reconstruction algorithms, the data from multiple perspectives is integrated into a unified three-dimensional model. Based on this, the length, width, and height parameters of the luggage are extracted, and considering the irregular shape of the luggage, the most accurate outer dimensions are calculated.

[0086] In some embodiments, after obtaining the basic feature data of the luggage, the temperature distribution characteristics of the luggage can also be obtained through thermal imaging technology, and special luggage is identified based on the temperature anomaly areas, realizing the intelligent identification and classification of special luggage that needs to be processed preferentially.

[0087] First, a high-precision thermal imaging device is used to perform an all-round temperature scan on the luggage. This device is equipped with an infrared sensor array and can complete the temperature acquisition of the entire luggage surface. Through the multi-point temperature sampling technology, a dense sampling grid is established on the luggage surface to ensure that the spatial resolution of the temperature data reaches centimeter-level accuracy. At the same time, a real-time calibration technology is adopted to continuously adjust the measurement accuracy through a reference temperature source to eliminate the influence of ambient temperature fluctuations.

[0088] After obtaining the original temperature data, data partitioning processing is performed based on a preset temperature difference range. By setting a reasonable temperature interval (for example, every 5°C as an interval), the continuous temperature data is converted into discrete temperature regions. This partitioning method not only simplifies the data processing process but also highlights the temperature anomaly areas. An adaptive partitioning algorithm can dynamically adjust the interval division according to the actual temperature distribution characteristics to ensure the rationality of the partitioning result.

[0089] For the divided temperature regions, the area of each region is calculated through an image segmentation algorithm. This algorithm combines edge detection and region growing to accurately identify the boundaries of each temperature region. Through pixel statistics and area conversion, the actual area value of each temperature region is obtained, and its proportion in the entire luggage surface is calculated. This accurate area calculation method provides a reliable data basis for subsequent anomaly judgment.

[0090] Finally, a special temperature region is identified through a dual-threshold judgment mechanism. First, check whether the area ratio of the temperature region exceeds a preset threshold, which is used to screen out temperature anomaly regions with significant areas. Then calculate the temperature mean difference between this region and adjacent regions. When the difference exceeds the preset temperature difference threshold, it is determined as a special temperature region. This dual judgment ensures the reliability of the identification result.

[0091] Once a special temperature region is detected, immediately mark the luggage as a priority handling object.

[0092] For the luggage with temperature anomalies marked as priority handling, its temperature change will also be continuously monitored. If it is found that the temperature anomaly further intensifies (such as the temperature difference continues to expand or the area of the anomaly region increases), the handling priority will be raised, and the corresponding emergency plan will be triggered.

[0093] Through the above technical steps, the intelligent identification and priority management of temperature-sensitive luggage are realized. Based on accurate temperature distribution analysis and multiple threshold judgments, it is possible to timely detect and prioritize the handling of luggage that may have temperature risks, effectively preventing damage to items caused by temperature problems and improving the safety of luggage handling.

[0094] S204. Calculate the matching degrees between the three-dimensional size parameters and the external contour data of the luggage and the complete form parameters and each separated form parameter of each sports equipment in the sports equipment feature database to obtain the complete form matching degree and multiple separated form matching degrees.

[0095] Specifically, first, establish a multi-level feature matching mechanism. When calculating the matching degree, adopt a hierarchical comparison strategy: initially screen the three-dimensional size parameters to establish a candidate equipment set; then conduct a fine match on the external contour data, including contour curve features, surface texture features, etc. Use an improved shape context algorithm for contour matching. This algorithm can effectively handle shape deformation and rotation problems by extracting key point features and calculating the similarity of point set distributions. For three-dimensional size matching, adopt a weighted Euclidean distance calculation method, and assign weights according to the importance of different size parameters to improve the matching accuracy.

[0096] When dealing with the separated form matching, adopt a modular identification strategy. First, decompose the complete form parameters of the sports equipment in the database into multiple standard component features to establish a component feature library. Divide the area of the luggage to be identified through a graphic segmentation algorithm to extract possible component features. Then use a hierarchical clustering algorithm to combine and match these features and calculate the similarity with the standard components. This component-based matching method can effectively identify the disassembled sports equipment.

[0097] A dynamic weight adjustment mechanism is also established. By analyzing historical matching data, the weights of different feature parameters will be automatically adjusted to adapt to the feature differences of different types of equipment. For example, for ball equipment, more attention is paid to the matching degree of the overall shape; while for rod-shaped equipment, more attention is paid to the matching degree of the length ratio.

[0098] S205. When the maximum value among the multiple separation form matching degrees of the candidate separated equipment is higher than the complete form matching degree, mark the unlabeled luggage as a separated part, and obtain the set of candidate separated equipment types.

[0099] Specifically, for each piece of luggage to be identified, calculate its matching degrees with the complete equipment and the separated parts in the database at the same time. By setting a unified scoring standard, the two matching degrees are made comparable. The normalization processing method is adopted to convert the feature parameters of different dimensions into the same numerical interval to ensure the effectiveness of the comparison results. When the maximum matching degree of the separation form exceeds the complete form matching degree, trigger the separated part recognition process.

[0100] During the recognition process, a multi-level judgment mechanism is adopted. First, quickly screen possible candidate equipment types through the comparison of main features (such as size ratio, shape contour, etc.). Then, perform fine feature matching, including surface texture, material characteristics, etc., to further verify the judgment results. A confidence evaluation system is also established, and a credibility level is assigned to the judgment results according to the significance of the matching degree difference.

[0101] To improve the recognition accuracy, a dynamic optimization mechanism for the set of candidate separated equipment types is established. By setting a matching degree threshold, multiple possible candidate types can be retained at the same time. These candidate types are sorted according to the matching degree, and their respective feature matching situations are recorded. Analyze the correlation between these candidate types. If it is found that multiple candidate types belong to different parts of the same equipment series, their credibility weights will be increased accordingly.

[0102] S206. According to each candidate separated equipment in the set of candidate separated equipment types, calculate the three-dimensional size parameters of the separated part of the luggage and the luggage outer contour data, and calculate the remaining separation shape parameters and the remaining three-dimensional size parameters of the remaining separated parts.

[0103] Specifically, obtain the actual parameters of the currently detected separated part, including three-dimensional dimensions (length, width, height) and outer contour features. Then, based on the complete form parameters of each piece of equipment in the set of candidate separated equipment types, calculate the theoretical parameters that the other parts of the equipment should have after separation except the currently detected separated part through mathematical modeling. This calculation is based on a pre-established equipment separation model, and by subtracting the current separated part parameters from the complete equipment parameters, the expected shape and size features of the remaining parts are obtained.

[0104] S207. Calculate the complementary matching degree for all the luggage that has not detected valid tags within the preset detection time period according to the remaining separation shape parameters and the remaining separation three-dimensional size parameters.

[0105] Specifically, in terms of shape complementary matching, the contour splicing algorithm is adopted to analyze the joint surface features of the two components. By extracting the edge curve feature points, calculate the complementary degree of the curves, and evaluate the matching degree of the two components in shape. The improved Hausdorff distance algorithm is used to calculate the similarity between the contours, and at the same time, the corresponding relationship of local features is considered to ensure the accuracy of shape complementary judgment.

[0106] In terms of size complementary matching, a multi-dimensional size correlation analysis model is established. By comparing the size ratio relationship of the two components in each dimension, calculate whether it meets the combination requirements of standard equipment. The special size requirements of different types of equipment are also considered, such as the standard length ratio of the racket head to the handle, and the arc matching of the front and rear sections of the ski board.

[0107] In order to improve the matching efficiency, a time window screening mechanism is adopted. By setting a reasonable detection time period (such as 30 minutes), only the tagless luggage within this time window is subjected to complementary matching calculation, which not only ensures the timeliness of pairing but also reduces the calculation amount.

[0108] In addition, a hierarchical matching strategy is also designed. First, perform a quick pre-screening to quickly exclude obviously mismatched combinations through basic size and shape features. Then, conduct a detailed complementary feature analysis on the potentially matching components, including the precise matching of the joint surface and the verification of material consistency.

[0109] In some embodiments, when performing complementary matching of separated components, weight parameters and material feature parameters are also introduced as matching bases. Through multi-dimensional feature analysis and distribution statistics, the accuracy and reliability of the pairing of separated components are further improved.

[0110] First, extract the standard parameters of the candidate equipment from the sports equipment feature database. For each candidate equipment, obtain the weight parameters and material parameters in the complete state, including detailed information such as the total weight, weight distribution characteristics, material type, and material composition ratio.

[0111] When calculating the parameters of the separated components, an intelligent parameter estimation method is adopted. By analyzing the shape parameters and size parameters of the separated components and combining the structural characteristics of the standard equipment, a weight distribution model is established. This model considers the weight distribution laws of different types of equipment, such as the standard weight ratio of the racket head to the handle and the weight distribution of different parts of the ski board. At the same time, a material feature mapping relationship is also established. By analyzing visual features such as surface texture and reflection characteristics, the material composition of the separated components is inferred.

[0112] Subsequently, a statistical analysis of all untagged baggage within the detection period was performed. By constructing a distribution histogram of weight and material characteristics, statistically significant distribution intervals were identified. This statistically based approach not only reveals general patterns but also identifies outliers, helping to improve matching accuracy. An adaptive interval partitioning strategy dynamically adjusts the interval range based on data distribution characteristics to ensure the rationality of the statistical results.

[0113] A multi-layered screening strategy is employed during the complementary matching calculation. First, potential matches are screened within the corresponding distribution intervals based on the estimated weight and material characteristics. This pre-screening significantly reduces the number of pairs requiring detailed calculation, improving processing efficiency. Then, a detailed complementary matching degree calculation is performed on the selected candidate pairs, comprehensively considering multiple factors such as shape complementarity and dimensional compatibility.

[0114] The matching results are verified by comparing the total weight of the matched parts with the standard weight of the complete equipment. Furthermore, the consistency of material characteristics is analyzed to ensure that the matched parts belong to the same equipment series. This multi-factor verification mechanism significantly improves the reliability of the matching results.

[0115] The above technical steps achieve precise matching based on multi-dimensional features. By introducing weight and material characteristics, combined with statistical analysis and a multi-layered screening strategy, the accuracy of separation component matching is improved. This approach not only overcomes the limitations of relying solely on shape and size matching, but also provides a more reliable matching basis through statistical analysis, providing more comprehensive technical support for the intelligent management of separation equipment.

[0116] S208: When there is untagged luggage whose complementary matching degree exceeds a preset matching threshold, mark it as a candidate complementary component.

[0117] Specifically, a multi-level judgment mechanism is employed when evaluating complementary matching. First, a basic complementary matching degree is calculated, including basic metrics such as shape complementarity and dimensional fit. A weighting system is then introduced, assigning different weights to different features based on their importance, to calculate the overall matching degree. Feature reliability is also considered, with weights appropriately reduced for features with high noise levels to improve the stability of the judgment.

[0118] When the complementary match exceeds the preset threshold, the intelligent tagging process is initiated. This process includes not only simple tagging operations but also detailed matching information recording, such as matching value and main matching features.

[0119] To improve identification reliability, a multi-verification mechanism has been established. Baggage marked as candidate complementary parts undergoes secondary feature verification, including material consistency checks and weight matching verification. This multi-verification mechanism effectively reduces the rate of false positives.

[0120] S209: Select the candidate complementary component with the highest complementary matching value among all candidate complementary components as the complementary component.

[0121] Specifically, the matching degrees of all candidate complementary components are preliminarily sorted to establish a priority queue, and then the candidate complementary component with the highest complementary matching degree value among all candidate complementary components is selected as the complementary component.

[0122] S210 , based on the candidate separation equipment type corresponding to the complementary component, reconstruct the three-dimensional size parameters and shape contour data of the separation component and the candidate complementary component to obtain reconstructed complete morphological parameters and complete three-dimensional size parameters.

[0123] Specifically, a reconstruction reference model is constructed based on the standard structural features of the candidate separation device. This reconstruction reference model includes the spatial relationships, connection characteristics, and combination rules between the device components, providing accurate guidance for the reconstruction process. Guided by the reference model, the relative positions and spatial relationships of each separation component within the complete device are accurately located.

[0124] A multi-level reconstruction strategy is employed for data stitching. First, the contour stitching layer analyzes the joint surface features of separated components to determine the optimal stitching position and angle. Next, the dimension fusion layer intelligently combines the 3D dimension data of each component, taking into account the overlapped areas at the joints to calculate the accurate complete dimensions. Finally, the surface feature layer smoothes the surface features at the joints to ensure the continuity and naturalness of the reconstruction result.

[0125] A stitching quality verification mechanism has also been established. The accuracy of the reconstruction is verified by comparing the results with the parameters of standard equipment in a database. Key dimensions, proportions, and overall contour characteristics are checked to ensure that the reconstruction meets the requirements of the standard equipment. If any anomalies are found, the stitching parameters are adjusted until the preset quality standards are met.

[0126] S211. Based on the luggage's three-dimensional dimensions and the luggage's outline data, the system compares the data with a preset sports equipment feature database, calculates equipment feature matching probabilities between the untagged luggage and various types of sports equipment, and selects equipment types whose equipment feature matching probabilities exceed a preset first feature matching probability threshold to obtain a candidate equipment set.

[0127] S212: Obtain a list of checked equipment for all flights within a preset time period based on the current time.

[0128] S213. Extract the intersection of the types of equipment in the candidate equipment set and the equipment list for each flight's checked luggage to obtain the candidate checked luggage set and the corresponding candidate flights for the candidate checked luggage set.

[0129] S214. For each type of equipment in the candidate checked luggage set, calculate the equipment quantity matching probability for all corresponding candidate flights respectively.

[0130] S215. Based on all the equipment quantity matching probabilities and equipment feature matching probability values, combine them in pairs respectively and sum them up with the preset weight coefficients to obtain the final set of matching probability values.

[0131] S216. Sort the final set of matching probability values to obtain the maximum matching probability value.

[0132] S217. When the maximum matching probability value exceeds the preset matching probability threshold, associate the unlabeled luggage with the flight and equipment type corresponding to the maximum matching probability value and add an electronic label.

[0133] Steps S211 - S217 are similar to Figure 1 Steps S102 - S108 in the embodiment shown. For details, refer to the description in Steps S102 - S108 and will not be elaborated here.

[0134] In the above - mentioned embodiment, first, accurate basic data is obtained through reflectance detection and compensation processing. Then, the separated components are identified through separated form matching, and the complete equipment features are restored based on complementary matching and reconstruction technology. Finally, double - verification is carried out in combination with the flight checked - luggage list. This comprehensive identification method combining surface feature analysis, component reconstruction, and flight information improves the identification accuracy and processing efficiency of special equipment. At the same time, by establishing a complete data - processing chain, the adaptability of the system to complex situations is enhanced, providing a more accurate and efficient management solution for the checked - luggage management during the event.

[0135] Next, an exemplary checked - luggage data intelligent management system 300 provided by the embodiments of the present application is introduced. Figure 3 It is an exemplary hardware structure diagram of the checked - luggage data intelligent management system 300 provided by the embodiments of the present application.

[0136] In some embodiments, the intelligent management system 300 for checked luggage data is a computer device or the intelligent management system 300 for checked luggage data includes a computer device. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with other external terminals or servers through a network connection. In some embodiments, the network interface can be a wired network interface, and in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, it implements the method in the embodiments of the present application.

[0137] Those skilled in the art can understand that Figure 3 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0138] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0139] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", or "after...", or "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" can be interpreted as "if determining...", or "in response to determining...", or "when detecting (the stated condition or event)", or "in response to detecting (the stated condition or event)".

[0140] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive), etc.

[0141] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by a computer program instructing relevant hardware. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The foregoing storage medium includes: various media that can store program codes such as ROM or random access memory RAM, magnetic disks, or optical discs.

Claims

1. An intelligent management method for checked baggage data, characterized in that, Including: When there is unlabeled luggage for which no valid label can be detected, scan the three-dimensional size parameters and the external contour data of the unlabeled luggage; Based on the three-dimensional size parameters of the luggage and the external contour data of the luggage, compare with a preset sports equipment feature database, calculate the equipment feature matching probability values between the unlabeled luggage and various types of sports equipment, and screen out the equipment types for which the equipment feature matching probability values exceed a preset first feature matching probability threshold to obtain a candidate equipment set; Obtain the list of checked equipment for all flights within a preset time period before the current time; the list of checked equipment for the flight includes the types of checked equipment and the information on the identified equipment quantity and the total quantity of checked equipment for each type of checked equipment; Extract the intersection of the types of equipment in the candidate equipment set and the list of checked equipment for each flight to obtain a candidate checked equipment set and the corresponding candidate flights for the candidate checked equipment set; For each type of equipment in the candidate checked equipment set, calculate the equipment quantity matching probability for all the corresponding candidate flights respectively; the equipment quantity matching probability is the ratio of the identified equipment quantity of the corresponding equipment type in the current flight to the total quantity of checked equipment; Based on all the equipment quantity matching probabilities and the equipment feature matching probability values, combine them pairwise and sum them up with a preset weight coefficient to obtain a set of final matching probability values; Sort the set of final matching probability values to obtain the maximum matching probability value; When the maximum matching probability value exceeds a preset matching probability threshold, associate the unlabeled luggage with the flight and the equipment type corresponding to the maximum matching probability value and add an electronic label.

2. The method according to claim 1, characterized in that, Before the comparison with the preset sports equipment feature database based on the three-dimensional size parameters of the luggage and the external contour data of the luggage, it further includes: Calculate the matching degrees between the three-dimensional size parameters of the luggage and the external contour data of the luggage and the complete form parameters and each separated form parameter of each sports equipment in the sports equipment feature database respectively to obtain a complete form matching degree and multiple separated form matching degrees; When the maximum value among the multiple separated form matching degrees of the candidate separated equipment is higher than the complete form matching degree, mark the unlabeled luggage as a separated part and obtain a set of candidate separated equipment types; According to each candidate separated equipment in the set of candidate separated equipment types, calculate the three-dimensional size parameters of the luggage of the separated part and the external contour data of the luggage, and calculate the remaining separated shape parameters and the remaining three-dimensional size parameters of the remaining separated parts; According to the remaining separated shape parameters and the remaining three-dimensional size parameters, calculate the complementary matching degrees for all the luggage that has not detected a valid label within a preset detection time period; the complementary matching degrees include a shape complementary matching degree and a size complementary matching degree; When there is unlabeled luggage for which the complementary matching degree exceeds a preset matching threshold, mark it as a candidate complementary part; Take the candidate complementary part with the highest complementary matching degree value among all the candidate complementary parts as the complementary part; According to the candidate separation equipment types corresponding to the complementary components, splice and reconstruct the three-dimensional size parameters and external contour data of the separation component and the candidate complementary component to obtain the reconstructed complete form parameters and complete three-dimensional size parameters.

3. The method according to claim 2, wherein Calculating the complementary matching degree for all unlabeled luggage within a preset detection time period according to the remaining separation shape parameters and the remaining separation three-dimensional size parameters specifically includes: Obtaining the equipment weight parameter and equipment material parameter of each candidate equipment in the candidate equipment type set; Calculating the separation weight parameter and separation material characteristic parameter of the separation component according to the equipment weight parameter, the equipment material parameter, the separation shape parameter and the separation three-dimensional size parameter of the separation component; Conducting a distribution statistics on the weight parameters and material characteristic parameters of all unlabeled luggage within the detection time period to obtain a weight distribution interval and a material characteristic distribution interval; Calculating the complementary matching degree for the unlabeled luggage in the distribution interval corresponding to the expected weight parameter and the material characteristic parameter according to the remaining separation shape parameter and the remaining separation three-dimensional size parameter.

4. The method according to claim 1, characterized in that, After scanning the three-dimensional size parameter and external contour data of the unlabeled luggage, it further includes: Obtaining the grid point cloud data on the surface of the unlabeled luggage and the pressure distribution map of the contact surface with the conveyor belt, and reconstructing a surface model; When it is determined that there is a fracture or defect in the surface model or the force values in each area of the pressure distribution map are calculated, and the force value in any area exceeds the preset upper limit threshold or is lower than the preset lower limit threshold, the unlabeled luggage is determined as damaged luggage; Determining the force concentration area according to the pressure distribution map or the surface model, generating a buffer support plan for strengthening the force concentration area, and outputting a buffer operation instruction corresponding to the buffer support plan.

5. The method according to claim 4, wherein Determining the force concentration area according to the pressure distribution map or the surface model and generating a buffer support plan for strengthening the force concentration area specifically includes: Calculating the ratio of the force value to the force area in the force concentration area to obtain the unit area pressure value, and selecting a matching buffer material type from the preset buffer material pressure level correspondence table according to the unit area pressure value; Calculating the required buffer material size and quantity based on the force area and the buffer material type.

6. The method according to claim 5, wherein When an unlabeled luggage with no effective label detected appears, after scanning the three-dimensional size parameter and external contour data of the unlabeled luggage, it further includes: Obtaining the surface temperature distribution data of the unlabeled luggage; Dividing the surface temperature distribution data into multiple temperature regions according to a preset temperature difference range; Statistical the area proportion of each temperature region; When there is a special temperature region in the unlabeled luggage, mark the unlabeled luggage as a priority processing luggage; the special temperature region is the temperature region whose area proportion is greater than the temperature area proportion threshold and the difference between the average temperature of the temperature region and the average temperatures of all adjacent temperature regions is greater than the preset temperature difference threshold.

7. The method according to claim 1, characterized in that The three-dimensional size parameters and the outer contour data of the tagless luggage obtained by the scanning specifically include: Detecting the reflectance of the surface of the tagless luggage; When it is detected that the reflectance exceeds a preset reflectance threshold, adjusting the scanning light intensity of the scanning device to obtain the scanning data of the tagless luggage; Performing a reflectance compensation process on the scanning data to obtain the three-dimensional size parameters and the outer contour data of the luggage.

8. An intelligent management system for checked baggage data, characterized in that, The checked luggage data intelligent management system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the checked luggage data intelligent management system to execute the method according to any one of claims 1-7.

9. A computer program product comprising instructions, characterized in that, When the computer program product runs on the checked luggage data intelligent management system, it causes the checked luggage data intelligent management system to execute the method according to any one of claims 1-7.

10. A computer-readable storage medium, comprising instructions, characterized in that, When the instruction runs on the checked luggage data intelligent management system, it causes the checked luggage data intelligent management system to execute the method according to any one of claims 1-7.

Images