Method and apparatus for reducing the risk of mis-sorting baggage

Abstract

The embodiment of the specification discloses a method and equipment for reducing luggage mis-sorting hidden dangers, which are applied to a whole-process luggage tracking system including a luggage tracking system, a security check information system and a luggage sorting system. The method comprises the following steps: the luggage tracking system acquires current luggage information in an RFID luggage tag read by an RFID reading station and a luggage case image corresponding to the current luggage information; the current luggage information is compared with flight data corresponding to the RFID reading station to determine mis-sorting luggage as target luggage; according to the current luggage case image in the RFID luggage tag, the luggage features of the target luggage are acquired, and the luggage case contour of the current luggage case image in the RFID luggage tag is extracted; according to the luggage features and the luggage case contour, a plurality of suspected target luggage collected by a preset camera monitoring point is determined; and the suspected target luggage is labeled as the target luggage according to the similarity between the suspected target luggage and the target luggage, so as to realize luggage tracking.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of civil aviation, and in particular to a method and device for reducing the risk of luggage mis-sorting. BACKGROUND

[0002] With the rapid development of the civil aviation transportation industry, more and more passengers who pursue efficiency choose to travel by air. The continuous increase in airport passenger and cargo volume has led to a series of problems such as luggage mis-sorting, mis-loading, and loss. When luggage is found in the wrong bin, it is often necessary to search all the checked-in luggage to find one piece of luggage. During the peak period of luggage, the work of the sorting personnel is very saturated, and the efficiency of searching piece by piece is low, time-consuming and labor-intensive. Not only can it affect the current flight guarantee task and cause the flight to be delayed, but it can also have an adverse impact on the subsequent flight guarantee task. Therefore, during the luggage transportation process, the luggage needs to be tracked and identified to determine the real-time position of the luggage during the transportation process.

[0003] The traditional technology of manually scanning luggage with barcodes is not only slow and error-prone, but also cannot effectively track the luggage. Therefore, in the prior art, radio frequency identification technology is introduced to automatically obtain the full-process information of the luggage without increasing the number of staff and workload. However, since the RFID reading station corresponding to the radio frequency identification technology is generally installed at the luggage transportation nodes of each carousel station of the sorting carousel, the detection of each luggage transportation node is realized, which makes it difficult to track the luggage transported between the luggage transportation nodes, that is, between the carousel stations. Further, the lost luggage caused by falling or missing in the sorting conveyor, or the mis-sorting luggage caused by flight delays, etc., can only be traced back after receiving the user's report, resulting in low reliability and low query efficiency of the luggage transportation. SUMMARY

[0004] To solve the above technical problems, one or more embodiments of the present specification provide a method, device, and medium for reducing the risk of luggage mis-sorting.

[0005] One or more embodiments of the present specification adopt the following technical solutions:

[0006] One or more embodiments of the present specification provide a method for reducing the risk of luggage mis-sorting, the method comprising:

[0007] The luggage tracking system obtains the current luggage information in the RFID luggage tag read by the RFID reading station, and the luggage box image corresponding to the current luggage information; wherein the RFID luggage tag includes RFID information; the RFID information includes passenger data and flight data;

[0008] The luggage tracking system compares the current luggage information with current flight information corresponding to the RFID reading station to preliminarily identify the misclassified luggage as the target luggage.

[0009] The luggage tracking system acquires the luggage feature of the target luggage according to the current luggage case image corresponding to the RFID luggage tag, and extracts the luggage case contour of the current luggage case image based on a preset convex curve contour mode.

[0010] The luggage tracking system determines a plurality of suspected target luggage collected by a preset camera monitoring point according to the luggage feature and the luggage case contour; the preset camera monitoring point is located on a luggage transmission path corresponding to the RFID reading station.

[0011] The luggage tracking system acquires suspected image features of the suspected target luggage and current image features of the current luggage image, and serializes the suspected image features and the current image features to determine the similarity between the suspected target luggage and the target luggage according to the serialized features.

[0012] If it is determined that the similarity is greater than a preset threshold, the luggage tracking system labels the suspected target luggage as the target luggage to achieve luggage tracking.

[0013] Optionally, in one or more embodiments of the present specification, the luggage sorting system comprises an RFID mobile terminal placed in a signal rectifier box made of metal; the signal rectifier box is used to constrain the RFID signal of the RFID mobile terminal to a controllable range and to concentrate and strengthen the normal RFID signal.

[0014] Optionally, in one or more embodiments of the present specification, before the luggage tracking system acquires current luggage information in the RFID luggage tag read by the RFID reading station and the current luggage case image corresponding to the current luggage information, the method further comprises:

[0015] The security information system acquires a check-in luggage image collected by a preset check-in camera, and receives an RFID luggage tag identified by a reading station, to load the check-in luggage image into a user terminal corresponding to the RFID luggage tag, to synchronize the current luggage image to the luggage tracking system based on the user terminal; the preset check-in camera is arranged in front of a departure X machine or on a check-in counter, and the RFID luggage tag is placed on each luggage.

[0016] The security information system acquires a security identification uploaded by a luggage security node; the security identification comprises a security pass identification or a secondary security identification.

[0017] If the security information system determines that the security identification is a secondary security identification, a security area for secondary security is determined according to the identification type of the secondary security identification, and the security area is uploaded to a user terminal corresponding to the RFID luggage tag, so that the RFID luggage tag is identified by an RFID mobile terminal, and the luggage corresponding to the RFID luggage tag is sorted into the corresponding security area.

[0018] If it is determined that the security identification is a security pass identification, the security information system identifies the flight data in the RFID luggage tag through an RFID mobile terminal, so as to determine the loading vehicle corresponding to the luggage according to the flight data in the RFID luggage tag, and perform luggage sorting and transportation.

[0019] Optionally, in one or more embodiments of the present specification, the luggage tracking system obtains the luggage features of the target luggage according to the current luggage box image corresponding to the RFID luggage tag, and extracts the luggage box contour of the current luggage box image based on a preset convex curve contour mode, specifically comprising:

[0020] The luggage tracking system obtains a gray image of the current luggage image, so as to down-sample each pixel point in the gray image to obtain a down-sampled gray image;

[0021] The luggage tracking system determines the gray scale distribution of the gray image based on the image histogram of the gray image, and determines the gray scale threshold of the gray image according to the gray scale analysis, so as to convert the gray image into a binary image based on the gray scale threshold;

[0022] The luggage tracking system determines the convex curve contour search initial pixel point of the binary image, so as to traverse the adjacent pixel points of the convex curve contour search initial pixel point, and obtain the pixel points connected with the convex curve contour search initial pixel point as the next convex curve contour search initial pixel point;

[0023] The luggage tracking system iteratively calculates the next convex curve contour search initial pixel point, and if it is determined that the next convex curve contour search initial pixel point is the curve contour search initial pixel point, the curve contour search initial pixel point is sequentially connected with each next convex curve contour search initial pixel point to obtain the luggage box contour of the current luggage image;

[0024] The luggage tracking system coarsely segments the current luggage box image based on the luggage box contour to obtain a coarsely segmented image, and converts the coarsely segmented image to HIS color space and Lab color space based on RGB color space, respectively, so as to obtain the luggage features of the target tracking luggage based on the color channel information of each pixel point in the coarsely segmented image in different color spaces.

[0025] Optionally, in one or more embodiments of the present specification, the suspected image features of the suspected target luggage and the current image features of the current luggage image are acquired, and the suspected image features and the current image features are serialized to determine the similarity between the suspected target luggage and the target luggage according to the serialized features, specifically comprising:

[0026] The luggage tracking system divides the monitoring image of the suspected target luggage and the current luggage image based on a preset size window to obtain a monitoring sub-image of the monitoring image and a current luggage sub-image of the current luggage image;

[0027] The luggage tracking system determines the suspected image features of the monitoring image and the current image features of the current monitoring image based on the monitoring sub-image and the current luggage sub-image, respectively; wherein the suspected image features include suspected image local features and suspected image global features; the current image features include current image local features and current image global features;

[0028] The luggage tracking system sorts the suspected image features of the monitoring sub-image based on the position of each monitoring sub-image in the monitoring image to obtain serialized suspected image features;

[0029] The luggage tracking system sorts the current image features of the current monitoring sub-image based on the position of each current monitoring sub-image in the current monitoring image to obtain serialized current image features;

[0030] The luggage tracking system determines the corresponding mapping relationship of each feature according to the position of each feature after serialization in the serialized suspected image features and the serialized current image features, and sequentially compares the suspected image local features and the current image local features based on the corresponding mapping relationship to obtain the local similarity between the suspected target luggage and the target luggage;

[0031] The luggage tracking system sequentially compares each suspected image global feature and each current image global feature according to the corresponding mapping relationship to obtain the global similarity between the suspected target luggage and the target luggage;

[0032] The luggage tracking system takes the product of the local similarity and the global similarity as the similarity between the suspected target luggage and the target luggage.

[0033] Optionally, in one or more embodiments of the present specification, the suspected image features of the monitoring image and the current image features of the current monitoring image are determined based on the monitoring sub-image and the current luggage sub-image, specifically comprising:

[0034] The luggage tracking system segments the monitoring image of the suspected target luggage based on a preset rectangular segmentation size to obtain a monitoring sub-image;

[0035] The luggage tracking system obtains the variance of each pixel point and the mean value of each pixel point in the current luggage sub-image of the RFID luggage tag, and determines the binary image of the monitoring sub-image based on the variance and the mean value, and the binary image of the current luggage sub-image.

[0036] The luggage tracking system determines the suspected image local feature of the monitoring image according to the binary image of the monitoring sub-image, and determines the current image local feature of the current luggage image according to the binary image of the current luggage sub-image.

[0037] The suspected image global feature of the monitoring image and the current image global feature of the current luggage image are extracted by a preset two-dimensional discrete wavelet transform.

[0038] Based on the suspected image local feature and the suspected image global feature, the suspected image feature of the monitoring image is determined, and based on the current image local feature and the current image global feature, the current image feature of the current luggage image is determined.

[0039] Optionally, in one or more embodiments of the present specification, the luggage tracking system compares the current luggage information with the current flight information corresponding to the RFID reading station to preliminarily identify misclassified luggage as target luggage, specifically comprising:

[0040] The luggage sorting system determines the position area corresponding to the corresponding luggage box according to the flight data, to determine whether the position of the RFID reading station is within the coverage range of the position area; wherein the position area is the area that should be passed through in the shipping process of the corresponding luggage box, including: sorting carousel station, loading vehicle;

[0041] If not, the luggage sorting system determines that the corresponding luggage box is misclassified luggage, and transmits the misclassified luggage to the luggage tracking system as target luggage.

[0042] If yes, the luggage tracking system determines whether the current luggage box has been delayed based on the flight data and the current time, and if it has been delayed, the corresponding luggage box is taken as target luggage.

[0043] Optionally, in one or more embodiments of the present specification, after the luggage tracking system obtains the position of the RFID reading station and the RFID luggage tag information read by the RFID reading station, the method further comprises:

[0044] The baggage sorting system determines current baggage data according to the RFID information, to perform secondary confirmation on the current baggage data;

[0045] If it is confirmed that the current baggage data is inconsistent with the baggage data read by the RFID reading station, the baggage sorting system returns the current baggage data to the baggage tracking system.

[0046] Optionally, in one or more embodiments of the present specification, after the suspected target baggage is labeled as the target baggage if it is determined that the similarity is greater than the preset threshold, the method further comprises:

[0047] The baggage tracking system uploads the preset detection point corresponding to the target baggage on the RFID baggage tag, and binds the user terminal of the RFID baggage tag;

[0048] If it is determined that the user terminal uploads a query instruction, the baggage tracking system synchronizes the monitoring image of the target baggage and the corresponding preset camera to the user terminal, so as to facilitate the user terminal to determine the tracking of the target baggage.

[0049] One or more embodiments of the present specification provide a device for reducing the risk of baggage misclassification, the device comprising: at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any of the above-mentioned methods.

[0050] The above-mentioned at least one technical solution adopted by the embodiments of the present specification can achieve the following beneficial effects:

[0051] Based on the RFID reading station reading the current luggage information in the RFID luggage tag, the problem of low identification and sorting efficiency caused by manual scanning of barcodes is avoided. By binding the current luggage information corresponding to the luggage image and the RFID luggage tag, the search efficiency of lost luggage is greatly improved, and by determining the target luggage between two RFID reading stations, the luggage that may have a mis-sorting problem is pre-tracked to avoid the problem of low reliability and query efficiency caused by backtracking. By determining the suspected target luggage box based on the current luggage image, and then based on the similarity between the suspected target luggage and the target luggage, the target luggage tracking is realized, which saves the analysis and calculation cost and improves the accuracy of identification and tracking. Based on the RFID mobile terminal arranged in the signal rectifier box, the normal RFID signal is concentrated and strengthened, so that the RFID signal is read within a controllable range through the opening of the signal rectifier box, solving the problem of misidentification caused by the wide range and sensitive reaction of the mobile device RFID signal. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present specification or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the present specification, and those skilled in the art can also obtain other drawings according to these drawings without creative labor. In the drawings:

[0053] Figure 1 A method flow diagram for reducing luggage mis-sorting risks provided by the present specification embodiment;

[0054] Figure 2 A display diagram of the carousel station of the carousel provided by the present specification embodiment;

[0055] Figure 3 A display diagram of the RFID mobile terminal provided by the present specification embodiment;

[0056] Figure 4 A display diagram of the signal rectifier box provided by the present specification embodiment;

[0057] Figure 5 A display diagram of the RFID mobile device of the loading node provided by the present specification embodiment;

[0058] Figure 6 An operation interface of a user terminal adding a label provided by the present specification embodiment;

[0059] Figure 7 An operation interface of a user terminal signing and checking provided by the present specification embodiment;

[0060] Figure 8 An internal structure diagram of a device for reducing luggage mis-sorting risks is provided in the embodiments of the present specification. DETAILED DESCRIPTION

[0061] The embodiments of the present specification provide a method and device for reducing luggage mis-sorting risks.

[0062] In order for those skilled in the art to better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be described clearly and completely in conjunction with the drawings in the embodiments of the present specification. Obviously, the described embodiments are only part of the embodiments of the present specification, not all the embodiments. Based on the embodiments of the present specification, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present specification.

[0063] As shown in the embodiments of the present specification, a flowchart of a method for reducing luggage mis-sorting risks is provided, which comprises the following steps: Figure 1 It can be known that a method for reducing luggage mis-sorting risks is applied to a whole-process luggage tracking system comprising a luggage tracking system, a security check information system and a luggage sorting system as shown in the embodiments of the present specification, and the method comprises the following steps: Figure 1 It can be known that a method for reducing luggage mis-sorting risks is applied to a whole-process luggage tracking system comprising a luggage tracking system, a security check information system and a luggage sorting system as shown in the embodiments of the present specification, and the method comprises the following steps: Figure 2 It can be known that a method for reducing luggage mis-sorting risks is applied to a whole-process luggage tracking system comprising a luggage tracking system, a security check information system and a luggage sorting system as shown in the embodiments of the present specification, and the method comprises the following steps:

[0064] S101: The luggage tracking system acquires current luggage information in an RFID luggage tag read by an RFID reading station, and a luggage case image corresponding to the current luggage information; wherein the RFID luggage tag comprises RFID information; the RFID information comprises luggage data and flight data.

[0065] The traditional luggage processing mode through manual inspection, infrared code scanning and loading has low efficiency and cannot guarantee the correctness of luggage sorting during the luggage peak period, which can easily lead to luggage misloading and missing. It is impossible to track each piece of luggage throughout the whole process, and after the passenger completes the luggage consignment, no one knows where the luggage goes. This not only seriously affects the service quality of ground support, but also causes difficulties in finding luggage. For example, in order to find a piece of luggage, all the checked-in luggage needs to be searched, which is low in efficiency and time-consuming and laborious during the luggage peak period. This not only may affect the current flight support task and cause flight delay, but also may have an adverse effect on the subsequent flight support task. In addition, the lack of luggage node information based on manual inspection makes the luggage transportation process opaque, and the passenger can only know whether the luggage consignment is normal after arriving at the luggage pickup hall. If the passenger does not receive the luggage, it is impossible to confirm the problem in the luggage transportation process due to the lack of luggage node information. Since there are many reasons for the missing luggage, the search for the luggage is not only time-consuming but also extremely complicated. This not only makes the passenger's air travel experience extremely poor, but also causes economic losses to the airline and the airport due to the compensation for the missing luggage of the passenger. Moreover, it is impossible to real-time master the loading progress of the luggage of the flight, and only after the check-in is completed, the sorting personnel can receive the total number of luggage of the flight. If there is a difference between the total number of luggage and the number of loaded luggage, it is necessary to confirm the luggage on each sorting carousel to check whether there is unloaded luggage. If no unloaded luggage is found, it is necessary to check whether there is dropped or missing luggage on the sorting transmission line, and so on. After the check-in is completed, the time left for the sorting personnel is very limited. Such processing mode is not targeted and is extremely time-consuming, which can easily cause flight delay and other problems.

[0066] Therefore, in order to improve the sorting efficiency and reduce the probability of luggage misloading, in the embodiment of the present specification, the luggage tracking system obtains the current luggage information in the RFID luggage tag read by the reading station and the luggage box image corresponding to the current luggage information during the departure process. It should be noted that the RFID luggage tag includes RFID information, and the RFID information includes passenger data and flight data. Specifically, during the check-in process, the luggage tag printer prints the RFID luggage tag, so that the passenger writes the luggage information and passenger information into the RFID electronic tag of the luggage tag when checking in, so as to send the related data such as passenger data and flight data to the luggage tracking system according to the data exchange platform. By introducing the RFID radio frequency technology, the whole process information of the luggage can be obtained by the luggage tracking system without increasing the workload of the staff.

[0067] Further, in order to quickly find the lost luggage and improve the sales of luggage inquiry, the luggage can be photographed by the X-ray machine or the camera at the check-in counter before the luggage enters the X-ray machine in the embodiment of the present specification, and the appearance photo of the luggage is sent to the luggage tracking system through the data exchange protocol, so that the luggage can be found through the photo when the luggage is found later, which greatly improves the efficiency of luggage finding. Specifically, in one or more embodiments of the present specification, before the luggage tracking system obtains the current luggage information in the RFID luggage tag read by the RFID reading station and the luggage box image corresponding to the current luggage information, the method further includes the following process:

[0068] First, the security information system obtains the check-in luggage image collected by the preset check-in camera, and receives the RFID luggage tag identified by the reading station, so as to load the check-in luggage image into the user terminal corresponding to the RFID luggage tag, so that the user terminal synchronizes the current luggage image to the luggage tracking system. It should be noted that the preset check-in camera is arranged in front of the departure X-ray machine or on the check-in counter, and the RFID luggage tag is placed on each luggage. In order to ensure safety during transportation and avoid luggage backlog caused by security when passenger flow is large, the security information system obtains the security identification uploaded by the security node corresponding to the security node of the luggage after the luggage enters the X-ray machine. It can be understood that the security identification includes: security pass identification or secondary security identification marked for subsequent package inspection when there is suspicious object in security. If the security information system determines that the security identification is the secondary security identification, the security area of the secondary security is determined according to the identification type of the secondary security identification, and the security area is uploaded to the user terminal corresponding to the RFID luggage tag, so as to identify the RFID luggage tag through the RFID mobile terminal, and sort the luggage box corresponding to the RFID luggage tag to the corresponding security area for package secondary security. On the contrary, if it is determined that the security identification is the security pass identification, the security information system identifies the flight data in the RFID luggage tag through the RFID mobile terminal, so as to determine the loading vehicle corresponding to the luggage box according to the flight data in the RFID luggage tag, and sort the luggage.

[0069] S102: The luggage tracking system compares the current luggage information with the current flight information corresponding to the RFID reading station to preliminarily identify the misclassified luggage as the target luggage.

[0070] In addition, in order to obtain the flight corresponding to the loading vehicle and reduce the risk of luggage misdelivery, in some embodiments of the present specification, the staff will select the loading vehicle type and input the loading vehicle number based on the user terminal, and then bind the vehicle button, so as to bind the loading vehicle and the flight. At this time, when the luggage is transported to a certain carousel station through the sorting carousel, the current luggage information in the RFID luggage tag read by the RFID reading station is compared with the current flight information corresponding to the RFID reading station, that is, the current flight information corresponding to the current sorting carousel station, so as to preliminarily identify the luggage not belonging to the carousel station. For example, Figure 2 As shown in a carousel station display diagram, based on the display result of the carousel station display screen, the sorting personnel can perform loading operation on the luggage based on the display result, so that the luggage enters the RFID mobile terminal departure task list. Due to the large passenger flow in the airport, when multiple flights are concentrated for sorting, multiple carousel stations will exist in the sorting carousel with long transmission link, so that multiple sorting personnel can perform collaborative loading operation during the movement of the sorting carousel. Since personnel are involved, there is a risk of mis-sorting caused by luggage falling and other objective reasons during the transportation of the luggage between one carousel station and the next carousel station. Therefore, the luggage between two carousel stations which has not been sorted is identified as mis-sorting luggage, so that the mis-sorting luggage is tracked as target luggage for subsequent tracking, so as to reduce the possible luggage mis-sorting problem and improve the sorting accuracy of the sorting whole process node.

[0071] Further, it needs to be explained that, in order to enable the RFID signal to be read within a controllable range during sorting and loading, and to achieve the effect of signal enhancement and reduction of mis-sorting probability, in the luggage sorting system of the present specification, as shown in Figure 3 An RFID mobile terminal is added, which is placed in a signal rectifier box made of metal material as shown in Figure 4 This solves the problem of easy misloading caused by the cross operation of scanning and carrying when the staff holds the RFID mobile terminal for scanning. Also based on the signal rectifier box, the RFID signal is constrained within a controllable range, and the normal RFID signal is bundled and strengthened, so that the RFID signal is read within a controllable range through the opening of the signal rectifier box, solving the problem of misidentification caused by the wide range and sensitive reaction of the RFID signal of the mobile device. It needs to be explained that the signal rectifier box installed on the mobile terminal has a lifting function, that is, the staff can adjust the height of the signal rectifier box according to the height of the luggage in the loading vehicle to facilitate the identification of the information of each luggage in the loading vehicle.

[0072] Further, in order to avoid misloading in the luggage sorting and transportation process, after the luggage tracking system obtains the position of the RFID reading station and the RFID luggage tag information read by the RFID reading station, the method further comprises:

[0073] The luggage sorting system determines the current luggage data according to the RFID information, and performs secondary confirmation on the current luggage data. If it is confirmed that the current luggage data is inconsistent with the luggage data read by the RFID reading station, the luggage sorting system returns the current luggage data to the luggage tracking system. That is, as shown in the embodiment of the present specification, the rectifier box is placed at the end of the trailer, and after the RFID mobile terminal is placed in the rectifier box, the luggage can be loaded. The RFID mobile terminal automatically scans the information of each piece of luggage loaded, performs secondary confirmation on the loading result, reports the loading node, and feeds back the loading result to the sorting personnel through sound and light, so as to avoid misloading. Figure 5

[0074] Further, in one or more embodiments of the present specification, the luggage tracking system compares the current luggage information with the current flight information corresponding to the RFID reading station to preliminarily identify the misloaded luggage as the target luggage, which specifically includes the following process:

[0075] First, the luggage sorting system determines the position area corresponding to the corresponding luggage box according to the flight data, so as to determine whether the position of the RFID reading station, that is, the position of the current carousel station, is within the coverage range of the position area corresponding to the corresponding luggage box. It should be noted that the position area is the area that should be passed in the transportation process of the corresponding luggage box, including: sorting carousel station, loading vehicle. If the corresponding luggage box is not currently in the corresponding position area, the luggage sorting system determines that the corresponding luggage box is misloaded luggage, and transmits the misloaded luggage to the luggage tracking system as the target luggage. If it is in the position area, the luggage tracking system determines whether the current luggage box has been delayed according to the flight data and the current time, and if it has been delayed, the corresponding luggage box is determined as the target luggage. For example, the luggage that needs to be forced to load, such as abnormal flight, inconsistent luggage destination station, etc., is also the target luggage, wherein it should be noted that such luggage is scanned and transported by forced loading.

[0076] Further, in some application scenarios of the present specification, for the delayed luggage box, the method can further include the following process:

[0077] ​The current meteorological data is obtained according to the loading position coordinates of the loading vehicle, so that the meteorological data along the way of the luggage case is determined according to the flight data corresponding to the luggage case, to filter the current meteorological data and the meteorological data along the way based on the Pearson correlation coefficient method, and obtain data with higher relevance to flight delay as key meteorological data. Then, the initial long short-term memory network is parameterized according to the pre-set optimization algorithm, and the optimal parameter combination is obtained, so as to update the initial long short-term memory network based on the optimal parameter combination to obtain the preset long short-term memory network. It should be noted that the parameters include the number of neurons, the sliding time window width, and the initial learning rate value. Then, the flight data and the key meteorological data corresponding to the current meteorological data are input into the preset long short-term memory network for iterative prediction to obtain the flight delay prediction corresponding to the flight data. It should be noted that the above process of parameterizing the initial long short-term memory network according to the preset optimization algorithm to obtain the optimal parameter combination, and updating the initial long short-term memory network based on the optimal parameter combination to obtain the preset long short-term memory network, specifically includes the following process: first, determine the number of random combination population of each parameter in the initial long short-term network, and determine the maximum number of iterations of the preset optimization algorithm, and determine the search parameter node of the current random combination population. Then, according to the distance between the search parameter node and other nodes in the random combination population, the fitness of each node in the random combination population is determined, and the current optimal position in the random combination population is determined according to the fitness. It can be understood that the greater the fitness value, the better the position. Then, according to the preset coefficient vector and the maximum number of iterations of the preset optimization algorithm, the position of each combination parameter in the optimization population is updated, the optimal position in the updated random combination population is obtained, and the parameter combination of the node corresponding to the optimal position is taken as the optimal parameter combination.

[0078] S103: The luggage tracking system obtains the luggage features of the target luggage according to the current luggage case image corresponding to the RFID luggage tag, and extracts the luggage case contour of the current luggage case image based on a preset convex curve contour method.

[0079] Since the range of signal receiving of the RFID reading station is limited, in order to realize the whole-process transportation monitoring of the target luggage transported between two RFID reading stations, the problem of whether there is luggage falling or missing in the high-enclosed loop transportation line after sorting error in the prior art is solved, and so on. After check-in, the time left for the sorting personnel is very tight, and such processing mode is not targeted, which wastes time abnormally and is easy to cause flight delay and other problems. Therefore, in order to reduce the probability of sorting error, the luggage that may have sorting error can obtain corresponding tracking information in real time. In the embodiment of the present specification, the luggage tracking system obtains the luggage features of the target luggage according to the current luggage box image corresponding to the RFID luggage tag, that is, the luggage image stored during check-in, and then extracts the luggage box contour of the current luggage box image based on the preset convex curve contour mode, so as to facilitate subsequent real-time tracking of the target luggage box that may have sorting error according to the luggage features, the luggage box contour and the luggage box image taken during the check-in process of the luggage box.

[0080] Specifically, in one or more embodiments of the present specification, the luggage tracking system obtains the luggage features of the target luggage according to the current luggage box image corresponding to the RFID luggage tag, and extracts the luggage box contour of the current luggage box image based on the preset convex curve contour mode, specifically including the following processes:

[0081] Firstly, in order to reduce the complexity and calculation amount of image processing, the luggage tracking system first obtains the gray image of the current luggage image, so as to realize the downsampling of each pixel point in the gray image to obtain the downsampled gray image. Then, the luggage tracking system determines the gray distribution of the gray image according to the image histogram of the gray image, and determines the gray threshold of the gray image according to the gray analysis, so as to convert the gray image into a binary image based on the gray threshold. Then, the luggage tracking system determines the convex curve contour search initial pixel point of the binary image, so as to traverse the adjacent pixel points of each convex curve contour search initial pixel point, and obtain the pixel point connected with the convex curve contour search initial pixel point as the next convex curve contour search initial pixel point. Then, the luggage tracking system iteratively calculates the next convex curve contour search initial pixel point, and if it can be determined that the next convex curve contour search initial pixel point is a curve contour search initial pixel point, the curve contour search initial pixel point is connected in sequence according to the preset curvature, and the next convex curve contour search initial pixel point is obtained, so as to obtain the luggage box contour of the current luggage image. Then, in order to obtain the luggage features in the luggage box contour, the luggage tracking system performs rough segmentation on the current luggage box image according to the luggage box contour, obtains a rough segmentation image, and converts the RGB color space to HIS color space and Lab color space respectively, so as to obtain the luggage features of the target tracking luggage according to the color channel information of each pixel point in the different color spaces of the rough segmentation image.

[0082] S104: The luggage tracking system determines a plurality of suspected target luggage collected by the preset camera monitoring point according to the luggage features and the luggage case contour.

[0083] In order to realize the tracking of luggage that may have a misclassification probability between the two RFID code reading stations, in the embodiment of the present specification, the luggage tracking system determines a plurality of suspected target luggage collected by the preset camera monitoring point on the luggage transmission path corresponding to the RFID code reading station according to the luggage features and the luggage case contour obtained in the above step S103. That is, after the current luggage features and the current luggage case contour of each luggage case in the image collected by each camera on the camera monitoring point on the transmission path between the two rotary table stations are extracted, and after the current luggage features and the luggage case contour of the corresponding luggage case in the current luggage case image obtained in the check-in process are compared, a plurality of luggage cases with high similarity are locked as suspected luggage cases for subsequent detailed tracking and identification. In order to save the calculation power in the tracking process based on the combination of coarse identification and fine identification, the calculation pressure is relieved while the accuracy of tracking is ensured.

[0084] S105: The luggage tracking system obtains suspected image features of the suspected target luggage and current image features of the current luggage image, and serializes the suspected image features and the current image features to determine the similarity between the suspected target luggage and the target luggage according to the serialized features.

[0085] After obtaining the suspected target luggage based on the above step S104, in order to accurately identify the target luggage in the suspected target luggage, the luggage tracking system in the embodiment of the present specification obtains suspected image features of the suspected target luggage and current image features of the current luggage image in the check-in process, so as to serialize the suspected image features and the current image features, and determine the similarity between the suspected target luggage and the target luggage according to the serialized features. In this process, through multi-dimensional analysis and comparison of the images of the suspected target luggage and the target luggage, the reliability of the luggage case tracking is ensured, and the accuracy of the tracking is improved.

[0086] Specifically, in one or more embodiments of the present specification, the suspected image features of the suspected target luggage and the current image features of the current luggage image are obtained, and the suspected image features and the current image features are serialized to determine the similarity between the suspected target luggage and the target luggage according to the serialized features, which specifically includes the following processes:

[0087] Firstly, the luggage tracking system divides the monitoring image of the suspected target luggage and the current luggage image according to the pre-set size window, so as to obtain the monitoring sub-image of the monitoring image and the current luggage sub-image of the current luggage image. Then the luggage tracking system determines the suspected image features of the monitoring image and the current image features of the current monitoring image according to the monitoring sub-image and the current luggage sub-image. It should be noted that the suspected image features include suspected image local features and suspected image global features, and the current image features include current image local features and current image global features. In order to realize the combination of local features and global features to determine the target luggage, the luggage tracking system sorts the suspected image features of the monitoring sub-image based on the position of each monitoring sub-image in the monitoring image, and obtains the serialized suspected image features. Then the luggage tracking system sorts the current image features of the current monitoring sub-image according to the position of each current monitoring sub-image in the current monitoring image, and obtains the serialized current image features. The luggage tracking system determines the comparison mapping relationship of each feature according to the position of each feature after serialization in the serialized suspected image features and the serialized current image features, so as to compare the suspected image local features and the current image local features in sequence based on the corresponding mapping relationship, and obtain the local similarity between the suspected target luggage and the target luggage. Then the luggage tracking system compares each suspected image global feature and each current image global feature in sequence based on the corresponding mapping relationship, and obtains the global similarity between the suspected target luggage and the target luggage. Then the luggage tracking system takes the product of the local similarity and the global similarity as the similarity between the suspected target luggage and the target luggage.

[0088] Further, in one or more embodiments of the present specification, based on the monitoring sub-image and the current luggage sub-image, the suspected image features of the monitoring image and the current image features of the current monitoring image are determined respectively, which specifically includes the following steps:

[0089] First, the luggage tracking system will monitor the image of the suspected target luggage, according to the pre-set rectangular segmentation size, after segmentation to obtain the corresponding monitoring sub-image. Then the luggage tracking system obtains each monitoring sub-image, and the variance of each pixel point and the mean value of each pixel point in the current luggage sub-image of the RFID luggage tag, and then determines the binary image of the monitoring sub-image according to the obtained variance and mean value, and the binary image of the current luggage sub-image. Then the luggage tracking system will determine the suspected image local feature of the monitoring image according to the binary image of the monitoring sub-image, and determine the current image local feature of the current luggage image according to the binary image of the current luggage sub-image. Through the pre-set two-dimensional discrete wavelet transform, the suspected image global feature and the current image global feature of the monitoring image and the current luggage image are extracted respectively. Then the suspected image local feature and the suspected image global feature are used as the suspected image feature of the monitoring image, and the current image local feature and the current image global feature are used as the current image feature of the current luggage image.

[0090] S106: If it is determined that the similarity is greater than the preset threshold, the luggage tracking system will label the suspected target luggage as the target luggage to realize luggage tracking.

[0091] Based on the similarity obtained in step S105, if it is determined that the similarity is greater than the preset threshold, the luggage tracking system will label the suspected target luggage as the target luggage, thereby realizing the tracking process of the luggage that may have a misclassification risk. In one or more embodiments of the present specification, after determining that the similarity is greater than the preset threshold and labeling the suspected target luggage as the target luggage, the method further includes the following process:

[0092] First, the pre-set detection point corresponding to the target luggage is uploaded to the RFID luggage tag, and the user terminal of the RFID luggage tag is bound. If it is determined that the user terminal uploads a query instruction, the monitoring image of the target luggage and the corresponding pre-set camera are synchronized to the user terminal, so as to determine the tracking of the target luggage by the user terminal. Further, for the luggage box that has been loaded, such as Figure 6 The method also includes: long pressing the luggage selection to operate the loaded luggage. For the misloaded luggage, it can be deleted; for the transshipment luggage, it can be pulled down; for the damaged luggage, it can be selected to report the exception; for the fragile, valuable, overweight, etc. Luggage can be labeled; for two-cabin, small animals and other luggage that need to be highlighted, the service type of the luggage can be modified to two-cabin and small animals. As shown in Figure 7After the baggage is loaded onto the truck, sorting personnel will hand over the entire truckload to the loading personnel. This operation can be done via an RFID mobile terminal. The loading personnel will verify the quantity loaded and sign for receipt. Upon receiving the baggage, the loading personnel will load it onto the aircraft. For baggage loaded onto flatbed trucks, the loading process is the same as for truck loading: each piece of baggage to be loaded is automatically scanned using an RFID mobile terminal, and loading confirmation is performed. For baggage loaded onto containers, the entire container can be loaded with a single click on the mobile terminal. After loading is complete, the loading point is recorded. After passengers collect their baggage, staff at the baggage claim exit can use an RFID mobile terminal to compare the baggage tag with the passenger's boarding pass information to ensure that passengers have not taken the wrong baggage.

[0093] like Figure 8 As shown, this specification provides a schematic diagram of the internal structure of a device to reduce the risk of baggage missorting in one or more embodiments. Figure 8 It is understood that a device for reducing the risk of baggage missorting includes: at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores computer-executable instructions that can be executed by the at least one processor, and the computer-executable instructions are capable of performing the above-described method.

[0094] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0095] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0096] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method of reducing the risk of misrouting baggage, characterized by, The method comprises: The baggage tracking system acquires current baggage information in an RFID baggage tag read by an RFID reading station, and a baggage case image corresponding to the current baggage information; wherein the RFID baggage tag comprises RFID information; the RFID information comprises passenger data and flight data; The baggage tracking system compares the current baggage information with current flight information corresponding to the RFID reading station to preliminarily identify mis-sorting baggage as target baggage; The baggage tracking system acquires baggage features of the target baggage according to a current baggage case image corresponding to the RFID baggage tag, and extracts a baggage case contour of the current baggage case image based on a preset convex curve contour mode; The baggage tracking system determines multiple suspected target baggage collected by a preset camera monitoring point according to the baggage features and the baggage case contour; wherein the preset camera monitoring point is located on a baggage transmission path corresponding to the RFID reading station; The baggage tracking system acquires suspected image features of the suspected target baggage and current image features of the current baggage case image, and serializes the suspected image features and the current image features to determine similarity between the suspected target baggage and the target baggage according to serialized features; If it is determined that the similarity is greater than a preset threshold, the baggage tracking system labels the suspected target baggage as target baggage to achieve baggage tracking.

2. The method of claim 1, wherein, The baggage tracking system comprises an RFID mobile terminal placed in a signal rectifier box made of metal; the signal rectifier box is used to constrain RFID signals of the RFID mobile terminal to a controllable range, and to intensify normal RFID signals.

3. The method of claim 1, wherein, Before the baggage tracking system acquires current baggage information in an RFID baggage tag read by an RFID reading station, and a baggage case image corresponding to the current baggage information, the method further comprises: The security information system acquires check-in baggage images collected by a preset check-in camera, and receives an RFID baggage tag identified by a reading station to load the check-in baggage images into a user terminal corresponding to the RFID baggage tag, so as to synchronize the current baggage case image to the baggage tracking system based on the user terminal; wherein the preset check-in camera is arranged in front of a departure X machine or on a check-in counter, and the RFID baggage tag is placed on each baggage; The security information system acquires security inspection identifiers uploaded by a baggage security inspection node; wherein the security inspection identifiers comprise security inspection pass identifiers or secondary security inspection identifiers; If the security information system determines that the security inspection identifier is a secondary security inspection identifier, it determines a security inspection area for secondary security inspection according to an identifier type of the secondary security inspection identifier, and uploads the security inspection area to a user terminal corresponding to the RFID baggage tag, so as to identify the RFID baggage tag through an RFID mobile terminal and sort a baggage case corresponding to the RFID baggage tag to a corresponding security inspection area. If it is determined that the security check identifier is a security check pass identifier, the security check information system identifies flight data in the RFID baggage tag through the RFID mobile terminal, and determines a loading vehicle corresponding to the luggage case according to the flight data in the RFID baggage tag, so as to perform luggage sorting and transportation.

4. The method of claim 1, wherein, The luggage tracking system obtains the luggage feature of the target luggage according to the current luggage case image corresponding to the RFID baggage tag, and extracts the luggage case contour of the current luggage case image based on a preset convex curve contour mode, specifically including: The luggage tracking system obtains a gray image of the current luggage case image, so as to down-sample each pixel point in the gray image and obtain a down-sampled gray image; The luggage tracking system determines the gray scale distribution of the gray image based on the image histogram of the gray image, determines the gray scale threshold of the gray image according to the gray scale distribution, and converts the gray image into a binary image based on the gray scale threshold; The luggage tracking system determines a convex curve contour search initial pixel point of the binary image, traverses adjacent pixel points of the convex curve contour search initial pixel point, and obtains a pixel point connected with the convex curve contour search initial pixel point as a next convex curve contour search initial pixel point; The luggage tracking system performs iterative calculation on the next convex curve contour search initial pixel point, and if it is determined that the next convex curve contour search initial pixel point is the curve contour search initial pixel point, sequentially connects the curve contour search initial pixel point and each next convex curve contour search initial pixel point to obtain the luggage case contour of the current luggage case image; The luggage tracking system performs coarse segmentation on the current luggage case image based on the luggage case contour, obtains a coarse segmentation image, converts the coarse segmentation image to HIS color space and Lab color space based on RGB color space respectively, and obtains the luggage feature of the target tracking luggage based on color channel information of each pixel point in different color spaces of the coarse segmentation image.

5. The method of claim 1, wherein, The luggage tracking system obtains the suspected image feature of the suspected target luggage and the current image feature of the current luggage case image, serializes the suspected image feature and the current image feature, determines the similarity between the suspected target luggage and the target luggage according to the serialized features, and specifically includes: The luggage tracking system divides the monitoring image of the suspected target luggage and the current luggage case image based on a preset size window, obtains a monitoring sub-image of the monitoring image and a current luggage sub-image of the current luggage case image; The luggage tracking system determines the suspected image feature of the monitoring image and the current image feature of the current luggage sub-image based on the monitoring sub-image and the current luggage sub-image respectively; wherein, the suspected image feature includes suspected image local feature and suspected image global feature; the current image feature includes current image local feature and current image global feature; The luggage tracking system sorts the suspected image features of each of the monitoring sub-images based on the positions of the monitoring sub-images in the monitoring image, to obtain sequenced suspected image features; The luggage tracking system sorts the current image features of each of the current luggage sub-images based on the current monitoring image positions corresponding to the current luggage sub-images, to obtain sequenced current image features; The luggage tracking system determines the contrast mapping relationship of each feature according to the sequenced suspected image features and the positions of the sequenced features in the sequenced current image features, to sequentially contrast the local features of the suspected image and the current image based on the contrast mapping relationship, to obtain the local similarity of the suspected target luggage and the target luggage; The luggage tracking system sequentially contrasts each of the suspected image global features and each of the current image global features according to the contrast mapping relationship, to obtain the global similarity of the suspected target luggage and the target luggage; The luggage tracking system takes the product of the local similarity and the global similarity as the similarity of the suspected target luggage and the target luggage.

6. The method of claim 5, wherein, Based on the monitoring sub-image and the current luggage sub-image, the suspected image features of the monitoring image and the current image features of the current luggage sub-image are determined, specifically including: The luggage tracking system divides the monitoring image of the suspected target luggage based on a preset rectangular segmentation size to obtain monitoring sub-images; The luggage tracking system obtains the variance of each pixel point and the mean value of each pixel point in each of the monitoring sub-images and the current luggage sub-image of the RFID luggage tag, to determine the binary image of the monitoring sub-image and the binary image of the current luggage sub-image based on the variance and the mean value; The luggage tracking system determines the suspected image local features of the monitoring image according to the binary image of the monitoring sub-image, and determines the current image local features of the current luggage box image according to the binary image of the current luggage sub-image; The suspected image global features of the monitoring image and the current image global features of the current luggage box image are extracted through a preset two-dimensional discrete wavelet transform; Based on the suspected image local features and the suspected image global features, the suspected image features of the monitoring image are determined, and based on the current image local features and the current image global features, the current image features of the current luggage box image are determined.

7. The method of claim 1, wherein, The luggage tracking system contrasts the current luggage information with the current flight information corresponding to the RFID code reading station to preliminarily identify misclassified luggage as target luggage, specifically including: The luggage tracking system determines the position area corresponding to the corresponding luggage box according to the flight data, to determine whether the position of the RFID code reading station is within the coverage range of the position area; wherein the position area is the area that the corresponding luggage box should pass through in the shipping process of the corresponding luggage box, including: sorting carousel station, loading vehicle; If not, the baggage tracking system determines the corresponding luggage as mis-identified baggage to transmit the mis-identified baggage as target luggage to the baggage tracking system; If yes, the baggage tracking system determines whether the current luggage has been delayed based on the flight data and the current time, and if so, the corresponding luggage is taken as target luggage.

8. The method of claim 1, wherein, After the baggage tracking system acquires the position of the RFID reading station and the RFID luggage tag information read by the RFID reading station, the method further comprises: The baggage tracking system determines current luggage data according to the RFID information to perform secondary confirmation on the current luggage data; If it is confirmed that the current luggage data is inconsistent with the luggage data read by the RFID reading station, the baggage tracking system returns the current luggage data to the baggage tracking system.

9. The method of claim 1, wherein, After determining that the similarity is greater than the preset threshold, the method further comprises: The baggage tracking system uploads the preset detection point corresponding to the target luggage to the RFID luggage tag and binds the user terminal of the RFID luggage tag; If it is determined that the user terminal uploads a query instruction, the baggage tracking system synchronizes the monitoring image of the target luggage and the corresponding preset camera to the user terminal, so as to facilitate the user terminal to determine the tracking of the target luggage.

10. A device for reducing the risk of mis-sorting baggage, the device comprising: At least one processor; And a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-9.

Images