A digital dangerous goods transportation supervision system and method

Abstract

The application discloses a kind of digital dangerous goods transport supervision system and method, belong to vehicle supervision technical field, to solve the authenticity of digital dangerous goods transport in loading and unloading link is insufficient, transport process abnormal fluctuation is difficult real-time identification and abnormal disposal response lag problem.After obtaining dangerous goods transport electronic transport note, by demarcating compliance area, shooting initial scene image, dynamic construction image flow sequence is completed loading and unloading scene verification, generates carrying password for goods encryption storage and unloading unlocking;To the section of route, establish vehicle operation baseline based on historical data, by monitoring the compliance of transport data and baseline and transmission interval, distinguish data delay and loss and identify abnormal fluctuation, trigger early warning;In early warning, based on positioning selection target check position, guide vehicle to check area to complete abnormal check.Eliminate the cheating behavior of loading and unloading link, identify transport risk and dispose abnormality quickly, effectively improve dangerous goods transport safety and supervision efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle supervision, more particularly to a digital dangerous goods transportation supervision system and method. BACKGROUND

[0002] As a high-risk link in the logistics field, dangerous goods transportation still faces many deep-seated common problems when using technical means for supervision: in the management of loading and unloading, the existing technology has obvious shortcomings in verifying the authenticity of operation, the conventional verification method is simple, only basic information interaction is confirmed, it is difficult to fully investigate human irregularities, leading to the inconsistency between actual loading and unloading and declared content, loading and unloading in unrecorded places, and other situations frequently occur, which poses a hidden danger to subsequent transportation safety; the communication guarantee problem in the transportation process restricts the supervision effect, in complex geographical environment such as mountainous areas, deserts or areas with lagging infrastructure, data transmission is often unstable or even interrupted, so that the supervision end cannot real-time master the vehicle running state and specific position, forming a supervision blind area, hindering dynamic management of the transportation process; the existing technical system lacks effective verification mechanism, the data uploaded to the supervision system often deviates from the actual situation, and some may be tampered with by humans, so that the supervision department cannot accurately judge the transportation real state, affecting the scientificity and effectiveness of supervision decision; these common problems seriously restrict the development of intelligent and precise dangerous goods transportation supervision. Therefore, in order to overcome these limitations, the present application provides a digital dangerous goods transportation supervision system and method. SUMMARY

[0003] In view of the deficiencies of the prior art, the purpose of the present application is to provide a digital dangerous goods transportation supervision system and method to solve the problems of loading and unloading cheating, communication interruption, data distortion, incomplete track tracing and inefficient emergency response in dangerous goods transportation, and to build an intelligent supervision system throughout the whole process to ensure transportation safety.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0005] A digital dangerous goods transportation supervision system, comprising a scene verification module, a transportation verification module and a track tracing module:

[0006] The scene verification module acquires the electronic waybill of dangerous goods transportation for scene verification, including: performing basic positioning verification by delimiting a compliant operation area, performing initial scene verification by shooting an initial scene image, and dynamically controlling the collection interval to build an image flow sequence containing time sequence for dynamic change verification;

[0007] According to the scene verification result, a carrying password is generated and sent, which is used for encrypted storage of goods and unlocking of unloading;

[0008] The transportation verification module divides the transportation route of the electronic waybill into road segments, establishes a vehicle operation baseline for each road segment based on historical transportation records, judges whether the transportation data received by the monitoring platform conforms to the vehicle operation baseline, and combines the collection time of the transportation data to distinguish between data delay and data loss, respectively performs deviation analysis and loss anomaly diagnosis, identifies transportation abnormal fluctuations, and judges whether to perform transportation abnormal early warning;

[0009] When the transportation abnormal early warning is triggered, the trajectory tracing module selects a target verification position based on the positioning information, divides a verification area, and guides the transportation vehicle to the verification position for abnormal verification.

[0010] Specifically, the specific steps of identifying transportation abnormal fluctuations and judging whether to perform transportation abnormal early warning include:

[0011] A monitoring time interval is configured, and vehicle transportation data is collected according to the monitoring time interval and transmitted to the monitoring platform; the vehicle transportation data includes collection time, position, speed, and start-stop state;

[0012] After the monitoring platform receives the transportation data, integrity verification is performed, and if the verification fails, the data is marked as invalid and a retransmission instruction is triggered for the on-board terminal;

[0013] If the verification is successful, the reference speed interval, the passing time interval, the conventional stop point coordinates, and the reference stop duration of the current road segment are obtained according to the vehicle operation baseline, and the received transportation data is verified to determine whether it conforms to the vehicle operation baseline; if it does not conform to the vehicle operation baseline, transportation abnormal early warning is performed;

[0014] If it conforms to the vehicle operation baseline, the transmission time interval since the last transportation data reception is monitored, and if it is greater than the interval time threshold, it is determined that there is potential data delay, and delay early warning is performed;

[0015] The abnormal road segment of the potential data delay is obtained, the driving trajectory and speed change trend before the potential data delay of the abnormal road segment are interpolated to complete the transportation data at the collection time point of the potential data delay, and the transportation data at the collection time point of the potential data delay is marked as delay completion data.

[0016] Specifically, the specific steps of identifying transportation abnormal fluctuations and judging whether to perform transportation abnormal early warning further include:

[0017] The collection time of the monitoring data subsequently received by the monitoring platform and the collection time point of the delay completion data are obtained, and it is judged whether the collection times are the same;

[0018] If they are the same, the delay completion data and the monitoring data received with a delay are subjected to deviation analysis in terms of position deviation, speed deviation, and start-stop state difference, it is judged whether there is an abnormal situation during the delay, and if there is, transportation abnormal early warning is performed;

[0019] If different, the delayed complete data is re-labeled as missing complete data; the missing section information and missing duration of the missing complete data are counted, if the missing duration is greater than a preset missing threshold, missing abnormal diagnosis is performed, potential transport paths of the missing section are screened, and it is judged whether to perform transport abnormal early warning;

[0020] The real-time trajectory sequence of each section is constructed in time sequence, including collection time, position, speed and start-stop state, and the electronic waybill information is archived.

[0021] Specifically, the specific steps of missing abnormal diagnosis include:

[0022] Based on the starting point coordinates and the end point coordinates of the missing section, the map data from the starting point to the end point is called, and the potential path between the starting point coordinates and the end point coordinates of the missing section is identified;

[0023] In combination with the rated transport speed range of the transport vehicle, the potential transport path conforming to the rated transport speed range is screened from the potential path;

[0024] The potential transport path is compared with the vehicle operation baseline, if there is a potential transport path that does not conform to the vehicle operation baseline, it is determined that the missing abnormal diagnosis is abnormal, and transport abnormal early warning is performed.

[0025] Specifically, the specific steps of scene verification include:

[0026] When the loading and unloading operation is triggered, the electronic waybill of dangerous goods transport is obtained, and the information of the electronic waybill includes the basic information of the goods, the transport path planning, and the transport subject information;

[0027] According to the transport path planning of the electronic waybill, the transport starting point and the transport end point are obtained, and by configuring the compliance range, the compliance operation area of the loading and unloading operation is delimited;

[0028] The real-time geographic position of the loading and unloading operation is obtained, if the real-time geographic position is in the compliance operation area, it is determined that the place is normal, the environmental feature collection is activated, the initial scene image is shot for initial scene verification, and based on the basic information of the goods and the transport subject information in the electronic waybill, the corresponding identification information set of the loading and unloading operation is configured;

[0029] The initial scene verification is performed information recognition, the initial scene recognition information is obtained, and it is matched with the identification information set, if the initial scene recognition information matches the identification information set, it is determined that the initial scene authentication is passed;

[0030] Specifically, the specific steps of scene verification also include:

[0031] If the initial scene authentication passes, the basic collection interval and the floating range are set according to the dangerous level classification of the dangerous goods, a random number in the floating range is generated by an encrypted random algorithm as a dynamic control value to regulate the basic collection interval, a dynamic collection interval is generated, panoramic images of the work site are taken, and an image stream sequence containing a time sequence is constructed;

[0032] The similarity of adjacent images in the image stream sequence is calculated, a similarity threshold is configured, if the similarity of adjacent images in the image stream sequence is greater than the similarity threshold, it is determined that the shooting is abnormal, otherwise it is determined that the real-time shooting is normal;

[0033] If it is determined that the shooting is abnormal, remote review is performed, if the remote review passes, the panoramic images of the work site are continuously taken at the dynamic collection interval, the image stream sequence containing the time sequence is constructed, and the loading and unloading operation is completed.

[0034] Specifically, the specific steps of generating and sending the carrying password include:

[0035] After the loading link is completed, random sampling is performed on the image stream sequence constructed during the loading link to obtain sample images;

[0036] The sample images are subjected to information recognition and comparison verification with the corresponding identification information set during the loading link, if the sample images are all verified, the carrying password generation process is triggered, the electronic waybill and the loading completion timestamp are combined, the carrying password is generated by an asymmetric encryption algorithm, the carrying password is composed of the loading location feature code, the encrypted segment of the cargo information and the dynamic check factor, and is used to encrypt and lock the cargo storage;

[0037] The effective period of the carrying password is dynamically set based on the dangerous level of the transported dangerous goods and the transportation mileage;

[0038] After the carrying password is generated, it is stored in an encrypted database, when the transport vehicle arrives at the unloading location recorded in the electronic waybill, it is determined that it is in a compliant operation area, and after the initial scene authentication passes, the carrying password is sent to the vehicle terminal and the destination authorized equipment of the transport vehicle through an encrypted communication channel;

[0039] When the unloading operation is performed, the carrying password is input, if the carrying password is correct and within the effective period, the encrypted lock of the cargo storage is unlocked;

[0040] If the carrying password is input incorrectly or has expired, the cargo storage remains in a locked state, a verification threshold is configured, if the number of incorrect password inputs is greater than the verification threshold, a password verification abnormality early warning is sent to the supervision platform, and personnel review is performed.

[0041] Specifically, the specific steps of establishing the vehicle running baseline of each section include:

[0042] The planned transportation route in the electronic waybill is called, the road section type and mileage data contained in the transportation route are extracted, and the transportation route is divided into road sections according to the road section type; and the key nodes of the road sections are recorded as road section identifiers;

[0043] The historical transportation records of each road section are obtained, the invalid transportation records with abnormal records are eliminated, the baseline data set is constructed, and the average driving speed, start-stop position, parking duration and passing time of each road section are included;

[0044] The baseline data set is analyzed according to the road section, the average driving speed of each road section is calculated, the speed fluctuation interval is calculated according to the time period, and the speed fluctuation interval is taken as the reference speed interval of each road section;

[0045] According to the reference speed interval, the passing time interval of each road section is calculated, and the clustering analysis of the start-stop position is performed, the regular parking point is screened, and the average parking duration of the regular parking point is calculated as the reference parking duration of the regular parking point;

[0046] The reference speed interval, passing time interval, regular parking point coordinates and reference parking duration of each road section are integrated into a structured vehicle operation baseline.

[0047] Specifically, the specific steps of guiding the transportation vehicle to the verification position for abnormal verification include:

[0048] When the transportation abnormality early warning is triggered, the current warning road section is determined based on the positioning information, the vehicle operation baseline and the key node marker of the warning road section are combined to screen the potential verification position of the warning road section, the straight line distance and the expected driving time of each potential verification position and the current vehicle are calculated, the distance and time priority are sorted, the target verification position is selected, and the verification area of the target verification position is divided;

[0049] The verification area information is sent to the driver through the vehicle terminal, and the driving route is monitored to determine whether the vehicle drives to the target verification position, if not, the supervision early warning is performed;

[0050] When the target verification position is reached, it is verified whether the vehicle is in the verification area of the target verification position, if yes, the abnormal verification process is triggered; if not, the vehicle position is adjusted to the verification area;

[0051] If the abnormal verification process is triggered, the transportation subject information of the electronic waybill is called, the information verification request is sent to the vehicle terminal through the encrypted communication channel, and the verification information is uploaded according to the verification request;

[0052] The uploaded verification information is extracted and matched with the transport main body information of the electronic waybill, if the verification request information is matched successfully, it is determined that the transport main body information verification is passed, the transport abnormality early warning is released, otherwise the unmatched verification request information is marked, the surrounding environment verification is started, the environment image containing the name of the target verification position is shot, the similarity of the environment image and the pre-stored target verification position standard environment image is calculated, if the similarity is greater than the preset verification similarity threshold, it is determined that the surrounding environment verification is passed, the transport abnormality early warning is released, otherwise, the transport abnormality early warning is maintained and synchronized to the supervision platform, and manual verification is carried out.

[0053] A digital dangerous goods transportation supervision method, comprising:

[0054] Step S1: In the loading link, the electronic waybill of dangerous goods transportation is acquired for scene verification, including: performing basic positioning verification by delimiting a compliance operation area, performing initial scene verification by shooting an initial scene image, dynamically adjusting and controlling collection intervals to construct an image flow sequence containing time sequences for dynamic change verification; generating and sending a carrying password according to the loading link scene verification result, for encrypted storage of goods and unlocking of unloading;

[0055] Step S2: In the transportation link, the transportation route of the electronic waybill is divided into road sections, and the vehicle running baseline of each road section is established based on historical transportation records, for judging whether the transportation data received by the monitoring platform conforms to the vehicle running baseline, and combining the collection time of the transportation data to compare and distinguish data delay and data loss, respectively through deviation analysis and loss anomaly diagnosis, identifying transportation abnormal fluctuations, and judging whether to perform transportation abnormality early warning;

[0056] Step S3: When the transportation abnormality early warning is triggered, a target verification position is selected based on positioning information, and a verification area is divided, guiding the transportation vehicle to the verification position for abnormality verification;

[0057] Step S4: In the unloading link, the electronic waybill of dangerous goods transportation is acquired for basic positioning verification and initial scene verification, after the verification is passed, the monitoring platform sends the carrying password to the vehicle-mounted terminal and the authorized equipment through an encryption channel, for unlocking the encrypted storage of goods, and continuing to perform dynamic change verification, and updating the carrying password state.

[0058] The beneficial effects of the present application are:

[0059] The present application solves the problem of insufficient authenticity verification in loading and unloading links from the root, eliminates cheating behaviors such as virtual code scanning and remote operation, and ensures traceability of operation; by segmenting the route to construct the vehicle operation baseline, combined with the differential identification of data delay and loss, the problem of real-time identification of abnormal fluctuations in the transportation process is accurately solved, and timely warning of speed, path and other risks is realized; by means of selecting target verification position based on positioning and guiding verification, the problem of abnormal disposal response lag is effectively solved, and abnormal review and disposal are quickly completed, the three form a whole-process closed-loop supervision, and the technical difficulties of each link are solved, which significantly improves the safety and supervision accuracy of dangerous goods transportation. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 FIG. 1 is a structural schematic diagram of a digital dangerous goods transportation supervision system of the present application;

[0061] Figure 2 FIG. 3 is a flowchart of the steps of constructing a multi-layer verification mechanism for scene verification of the present application;

[0062] Figure 3 FIG. 4 is a flowchart of generating and sending a transport password according to the scene verification result of the present application;

[0063] Figure 4 FIG. 5 is a flowchart of the specific steps of identifying transportation abnormal fluctuations and determining whether to perform transportation abnormal early warning of the present application;

[0064] Figure 5 FIG. 6 is a flowchart of a digital dangerous goods transportation supervision method of the present application. DETAILED DESCRIPTION

[0065] Please refer to Figure 1 The present embodiment introduces a digital dangerous goods transportation supervision system, which comprises a scene verification module, a transportation verification module and a trajectory tracking module;

[0066] The scene verification module focuses on solving the identity and location authenticity problems of the loading and unloading links, and based on the electronic waybill of dangerous goods transportation, a multi-layer verification mechanism is constructed by integrating positioning perception, image recognition and cargo information interaction technology to conduct scene verification, including: through the basic positioning verification by delimiting the compliance operation area, the initial scene verification by shooting the initial scene image, and the dynamic interval control to collect the image flow sequence containing time sequence for dynamic change verification; and according to the scene verification result, the carrying password is generated and sent, which is only valid within the specified time window after the scene verification is passed, and automatically invalid beyond the time. When the loading and unloading operation starts, the environmental feature collection function is automatically triggered, the physical identification and surrounding environmental features of the operation site are captured by high-definition cameras and environmental sensors, and real-time multi-dimensional comparison is carried out with the pre-stored compliance operation area feature library. Only when the matching degree reaches the set threshold, the actual operation site is determined to be consistent with the recorded information. Through the cross verification of multi-dimensional information, the cheating behaviors such as virtual scanning code, off-site operation and inconsistent waybill are avoided from the root, and each operation in the loading and unloading link is traceable and verifiable, which builds a strong safety line for the starting and terminal links of dangerous goods transportation.

[0067] Please refer to Figure 2 , preferably, the step of constructing a multi-layer verification mechanism for scene verification comprises:

[0068] When the loading and unloading operation is triggered, the electronic waybill of this dangerous goods transportation is obtained, and the information of the electronic waybill includes cargo basic information, transportation path planning, and transportation subject information; the cargo basic information includes dangerous goods category, UN number, dangerous characteristics, quantity, and packaging specification, which is used for bidirectional verification with the information in the electronic identification of the cargo; the transportation path planning includes transportation starting point, transportation ending point, transportation route, and loading and unloading time, which is used to determine the geographical range and time constraint of the loading and unloading site; the transportation subject information includes transportation enterprise qualification number, vehicle identification code, driver and escort qualification certificate number, which is used to verify whether the enterprise, vehicle and personnel participating in the loading and unloading operation have the corresponding qualifications; the consistency is ensured from the information source, and the true benchmark is established for subsequent verification.

[0069] The basic positioning verification is started, the transportation starting point and the transportation ending point are obtained according to the transportation path planning of the electronic waybill, and the compliance range is configured, which is used to dynamically delimit different radius of virtual electronic fence according to the site area of the transportation starting point and the transportation ending point and the dangerous level of the cargo, so as to delimit the compliance operation area of the loading and unloading operation; the fixed area restriction is broken, and the compliance range is delimited as needed, which avoids the rigidity of one-size-fits-all and provides accurate geographical constraint for position verification.

[0070] The real-time geographic position of the loading and unloading operation is obtained through the positioning perception technology, and is preliminarily compared with the transportation path planning of the electronic waybill. If the real-time geographic position is in the compliance operation area, it is determined that the site is normal, otherwise, it is determined that the site is abnormal, and the subsequent verification is terminated. According to the distance between the real-time geographic position and the center of the compliance operation area, a hierarchical abnormality warning is performed, so that the supervision response is more hierarchical and efficient.

[0071] If it is determined that the site is normal, the environment feature collection is activated, the initial scene image is shot for initial scene verification, the identification information set corresponding to the loading and unloading operation is configured based on the cargo basic information and the transportation subject information in the electronic waybill, which is used to limit the cargo basic information identification elements and the transportation subject information identification elements that must be included in the initial scene image during the loading and unloading process, such as dangerous goods class label, UN number identification on the cargo packaging, and part of the license plate of the transportation vehicle, to ensure that the image collection can fully cover the key information; the image collection is forced to cover the key evidence to ensure that the subsequent verification has complete and effective original materials.

[0072] The initial scene image is subjected to information recognition through a pre-trained target detection model, initial scene recognition information is obtained, and it is compared with the identification information set; if the initial scene recognition information matches the identification information set, it is determined that the initial scene authentication is passed; otherwise, a shooting abnormality warning is given, and the initial scene image is prompted to be re-shot and subjected to initial scene verification. The fake behavior of posing is quickly screened out, and the cost of manual verification is reduced.

[0073] If the initial scene authentication is passed, the initial scene shooting setting parameters are maintained, including the shooting angle and the shooting position, the corresponding basic collection interval and the floating range are set according to the hierarchical dangerous level of the dangerous goods, a random number in the floating range is generated through an encryption random algorithm as a dynamic control value to control the basic collection interval, a dynamic collection interval is generated to avoid cheaters arranging false scenes in advance by predicting the shooting time, to ensure that the time point of each image collection is unpredictable, panoramic images of the operation site are shot according to the dynamic collection interval to construct an image flow sequence containing time sequence; the cheaters cannot disguise real-time scenes by posing at regular intervals, and the time randomness and authenticity of the image flow are ensured.

[0074] The dynamic change verification is based on an image stream sequence, similarity of adjacent images in the image stream sequence is calculated to determine whether the work scene is in a dynamic change state, a similarity threshold is configured, and the similarity threshold is set according to the work type and the characteristics of the goods. If the similarity of adjacent images in the image stream sequence is greater than the similarity threshold, it indicates that the work scene has no obvious dynamic change, and there may be a situation of using fixed picture loop or static photo to replace real-time shooting, and it is determined that the shooting is abnormal. Otherwise, it is determined that real-time shooting is performed, and no processing is performed; static cycle fraud is identified, and whether the scene is a real dynamic change is detected through comparison of the similarity of adjacent images from the time and space dimensions to prevent fraud.

[0075] If it is determined that the shooting is abnormal, the current loading and unloading operation is immediately suspended, the operation permission of the related equipment is locked, and warning information including the abnormal image segment, the timestamp, and the geographic location is sent to the supervision platform to notify the supervisor to perform remote review. If the remote review is passed, panoramic images of the work site are continuously shot according to the dynamic collection interval, an image stream sequence including a time sequence is constructed, and the loading and unloading operation is completed. Both risks are strictly controlled, and false positives are avoided through remote review, balancing compliance and efficiency.

[0076] Please refer to Figure 3 Preferably, the step of generating and sending the carrying password according to the scene verification result comprises:

[0077] After the loading link is completed, sample images are obtained by randomly sampling from the image stream sequence including a time sequence constructed during the loading link. The sample images are subjected to information recognition by a pre-trained target detection model, and the goods basic information recognition elements and the transportation subject information recognition elements are extracted and compared with the corresponding identification information set in the loading link. If all sample images are verified, it indicates that the image recording of the whole loading process is real and effective, and the carrying password generation process is triggered. If there is a sample verification failure, it is determined that the loading image stream is abnormal, and manual review is triggered. If the manual review is passed, the carrying password generation process is triggered. Cheaters are prevented from only forging key frames, such as only shooting compliant pictures and replacing goods in the middle, to ensure that the images in the whole loading period are real and effective.

[0078] If the carrying password generation process is triggered, a unique carrying password is generated by combining the electronic waybill of the transportation and the loading completion timestamp through an asymmetric encryption algorithm. The carrying password includes a loading location feature code, a cargo information encrypted segment, and a dynamic check factor. The loading location feature code is a feature string generated by an encryption algorithm based on the latitude and longitude of the loading location and the unique identification code of the site, which is used to identify the actual loading location of the cargo and ensure consistency with the transportation starting point information in the electronic waybill. The cargo information encrypted segment is a piece of information formed by encrypting the basic information of the cargo in the electronic waybill, which is used to bind the core attributes of the cargo at the password level to prevent the cargo from being exchanged or tampered with. The dynamic check factor is a dynamically changing value generated by an encryption algorithm based on the loading completion timestamp and a random number generated by a random number generator, which is used to enhance the uniqueness and timeliness of the password and avoid repeated use or forgery of the password. Through hash operation and deep binding with the electronic waybill and the loading scene verification result, an unalterable association is formed, and the generated password is used to encrypt and lock the on-board cargo storage. Without the unlocking of the carrying password, any cargo operation cannot be performed.

[0079] The effective period of the carrying password is dynamically set based on the dangerous level of the transported dangerous goods and the transportation mileage, which is used to ensure that the password does not expire too early to affect normal unloading operations, and does not expire too long to increase the risk of password leakage or misuse. For example, dangerous goods are divided into multiple intervals according to the dangerous level, and multiple intervals are divided according to the transportation mileage. A two-dimensional mapping table is established to match the corresponding effective period for each combination of dangerous level interval and transportation mileage interval. The actual dangerous level and transportation mileage are matched to obtain the corresponding value from the mapping table as the effective period. The effective period covers the entire transportation stage from loading completion to unloading completion. The effective period can be automatically extended according to real-time positioning and new estimated arrival time, but it needs to be reported to the supervision platform at the same time.

[0080] The carrying password is stored in an encrypted database and is not pushed to any terminal in advance. When the transportation vehicle arrives at the unloading location recorded in the electronic waybill and is determined to be in the unloading compliance operation area through basic positioning verification, the unloading scene verification is automatically started. After the initial scene authentication is passed, the carrying password is sent to the on-board terminal of the transportation vehicle and the authorized device of the person in charge of the destination operation through an encrypted communication channel. The sending process uses end-to-end encryption technology, and records the password sending time, receiving terminal identifier and vehicle location information at that time to form a password pushing log.

[0081] During the unloading operation, the operator needs to input the received carrying password on the on-board terminal or unloading device. If the password is correct and within the effective period, the terminal device unlocks the encrypted lock of the cargo storage, and records the password verification passing time as the time node of the formal start of the unloading operation.

[0082] If the password is input incorrectly or has expired, the cargo storage remains in a locked state, a verification threshold is configured, if the number of times the password is input incorrectly is greater than the verification threshold, an abnormal password verification early warning is sent to the supervision platform, and real-time images of the unloading scene are synchronously called to provide a review for the supervisor.

[0083] After the unloading operation is completed, the carrying password state is updated to be used, and an operation voucher containing the password use time, unlocking duration, operation personnel, and full-process verification record of loading and unloading is generated, which is associated with the electronic waybill and full-scene verification data and is archived as a basis for subsequent tracing and supervision.

[0084] The transportation verification module focuses on solving the problem of real-time monitoring of the transportation vehicle in the transportation process. Based on the transportation route planned by the electronic waybill, the historical transportation records of the route are integrated to construct a vehicle operation baseline. It is judged whether the transportation data received by the monitoring platform conforms to the vehicle operation baseline, and the data delay and data loss are distinguished by comparing the collection time of the transportation data. The transportation abnormal fluctuation is identified by deviation analysis and potential path matching respectively, and it is judged whether to perform transportation abnormal early warning;

[0085] In the process of dangerous goods transportation, through the configured monitoring time interval, the vehicle terminal collects the vehicle positioning information, instantaneous driving speed, start-stop state and collection time in real time, which are transmitted to the monitoring platform through an encrypted wireless communication protocol. Through integrity verification and transportation data verification based on the vehicle operation baseline, invalid data can be quickly identified and retransmission is triggered, and whether the speed, travel time, and stop state conform to the baseline standard is accurately checked. For potential delays that may occur during data transmission, interpolation completion is performed first and the delayed completion data is marked. After receiving the subsequent data, the data delay and data loss are distinguished by time comparison: when the real delay occurs, the abnormality is identified by the deviation between the completed trajectory and the real trajectory, and when the data is lost, the missing road segment information and time are counted, a potential transportation path conforming to the rated speed range of the vehicle is generated, and the abnormal risk is detected. Finally, real-time monitoring of the whole process of dangerous goods transportation is realized, transportation abnormal fluctuations are accurately identified, and abnormal early warning is timely performed, which provides full-cycle and multi-dimensional verification guarantee for transportation safety, effectively improving the standardization and safety of dangerous goods transportation.

[0086] Preferably, the specific steps of constructing the vehicle operation baseline include:

[0087] The transportation route planned in the electronic waybill is called, the road segment type and mileage data contained in the transportation route are extracted, the transportation route is divided into road segments according to the road segment type, including highway segments, ordinary highway segments, mountainous road segments, and urban road segments; and the key nodes of each road segment are recorded as road segment identifiers, including rest areas and gas stations; the spatial segmentation and monitoring anchor points of vehicle operation analysis are determined.

[0088] Obtain the historical transportation records of each road section, eliminate invalid transportation records with abnormal records, and construct a baseline dataset containing the average driving speed, start-stop position, stop duration, and travel time of each road section, and deposit the basic feature data of the normal operation of vehicles on each road section;

[0089] Feature analysis is performed on the baseline dataset according to road sections. For the average driving speed of each road section, the speed fluctuation interval is calculated according to different time periods such as peak hours, flat peak hours, holidays, and night, as the reference speed interval of each road section. According to the reference speed interval, the travel time interval of each road section is calculated, and the start-stop position is analyzed by clustering. The regular stop points are screened, and the average stop duration of each regular stop point is calculated as the reference stop duration of the regular stop point. The speed, travel time, and stop rules are analyzed by time period, and the normal threshold system of the vehicle's time and space behavior on each road section is quantified.

[0090] The reference speed interval, travel time interval, regular stop point coordinates, and reference stop duration of each road section are integrated into a structured vehicle operation baseline. Through blockchain technology, the electronic waybill is associated, and the encrypted hash value containing the vehicle operation baseline parameters and dangerous goods information is generated and written into the blockchain distributed ledger. At the same time, a digital signature is added to the structured baseline data to realize real-time monitoring of data tampering.

[0091] Please refer to Figure 4 , preferably, the specific steps of identifying transportation abnormal fluctuations and determining whether to perform transportation abnormal warning include:

[0092] Configure a monitoring time interval to regulate the frequency of data collection by the vehicle terminal. According to the difference in dangerous goods level and road section risk type, the vehicle terminal collects vehicle transportation data according to the monitoring time interval, including collection time, position, speed, and start-stop state. Through encrypted wireless communication protocol, the data is transmitted to the monitoring platform.

[0093] Wireless transmission is prone to data packet loss and format errors. If used directly for analysis, it will cause abnormal judgment deviation. After receiving the transportation data, the monitoring platform performs integrity verification. The preset verification rule is used to check whether the key fields of the data packet are complete and the format is correct. If the verification fails, it is marked as invalid data and triggers the vehicle terminal to retransmit the instruction. At the same time, the timestamp, terminal identifier, and failure reason of the invalid data are recorded.

[0094] If the verification is successful, the reference speed interval, the passing time interval, the coordinates of the regular stop point and the reference stop duration of the current road section are obtained according to the vehicle operation baseline, the received transportation data is verified, and whether it conforms to the vehicle operation baseline is judged, including: comparing the speed in the transportation data with the reference speed interval to determine whether the speed is within a reasonable range; in combination with the collection time of the transportation data and the road section mileage, it is judged whether the current driving duration is within the passing time interval; if the transportation data shows that the vehicle is in a stop state, the stop position is matched with the coordinates of the regular stop point, and it is checked whether the stop duration meets the reference stop duration requirement; if it does not conform to the vehicle operation baseline, a transportation abnormality early warning is performed;

[0095] Data transmission lag, such as network congestion, can cause the monitoring platform to break more, and cover up real abnormalities. The monitoring distance from the last transportation data receiving transmission time interval is monitored. If it is greater than the interval time threshold, it is determined that there is potential data delay, a delay early warning is performed, indicating that the data transmission does not arrive according to the preset frequency, and there may be transmission lag;

[0096] During data delay, the vehicle running state is completely unknown, and it is impossible to determine whether there is a violation. The abnormal road section where the potential data delay is located is obtained, the driving track before the potential data delay on the abnormal road section and the speed change trend are interpolated to complete, the transportation data of the collection time point where the potential data delay is located is generated, and is marked as delay completed data; based on the track and speed before the delay, the expected running state during the delay period is simulated to provide a reference for subsequent comparison.

[0097] The delay and loss of subsequent received data have different abnormal properties. The delay may be a temporary network problem, and the loss may be a device failure or human interference. The collection time of the monitoring data received by the monitoring platform subsequently and the collection time point of the delay completed data are obtained, and it is judged whether the collection times are the same;

[0098] If they are the same, it indicates that it is a temporary data loss caused by data delay. The delay completed data and the monitoring data received with delay are analyzed in terms of position deviation, speed deviation and start-stop state difference. According to the preset position deviation threshold and speed deviation threshold, and the start-stop difference, it is judged whether there is an abnormal situation during the delay process. If there is, it indicates that there is a significant deviation between the actual running state of the vehicle during the delay period and the expected completed state, and there may be a violation of driving behavior, and a transportation abnormality early warning is performed;

[0099] If not, it indicates that the data is missing due to data transmission loss, and the delayed complete data is re-labeled as missing complete data. Short-time loss may be an occasional failure, and long-time loss is likely to hide a violation operation. The missing complete data is statistically analyzed to obtain the missing section information and the missing duration. The missing section information includes the start point coordinate, the end point coordinate, and the passing area range of the missing section. If the missing duration is greater than a preset missing threshold, loss anomaly diagnosis is performed to identify the potential transportation path of the missing section:

[0100] During data loss, the vehicle may deviate from the planned route, such as walking in a forbidden area, taking a detour to unload, but there is no real-time data to trace back. Based on the start point coordinate and the end point coordinate of the missing section, the map data from the start point to the end point is retrieved, all potential paths between the start point coordinate and the end point coordinate of the missing section are identified, all passable road connection methods are covered, including main roads, auxiliary roads, branch roads, etc., and the rated transportation speed range of the transportation vehicle is combined to filter out the potential transportation path in the rated transportation speed range from the potential paths, i.e. the vehicle can travel from the start point to the end point at the maximum speed and the whole journey is within the vehicle operating range.

[0101] Each potential transportation path is compared with the vehicle operation baseline. If there is a potential transportation path that does not meet the vehicle operation baseline, it indicates that the vehicle may deviate from the normal transportation route or operating state during data loss, and there is a transportation risk. Therefore, it is determined that the loss anomaly diagnosis is abnormal, and a transportation anomaly warning is given.

[0102] Based on the comparison results of the complete verification transportation data, the abnormal judgment results of the delay process, the abnormal diagnosis results of the data loss, and the potential transportation path, the real-time trajectory sequence of each section containing the collection time, position, speed, and start-stop state is constructed in time sequence, and the electronic waybill information is archived.

[0103] The abnormality verification module is started when the transportation anomaly warning is triggered. The current section is determined based on the real-time positioning information of the vehicle. The regular stop point is identified by associating the vehicle operation baseline, and the position of the rest area, gas station, etc. that allows temporary parking is identified as a verification position according to the key node marker, and a verification area is divided. The transportation vehicle is guided to the verification position for abnormality verification. The consistency of the transportation main body information and the electronic waybill record information is verified, and the environment features around the vehicle are collected by means of a high-definition camera and an environment sensor, and are compared with the pre-stored corresponding position environment feature library. If the transportation main body information verification is passed and the environment feature matching degree reaches a set threshold, it is determined that the abnormality verification is passed, the transportation anomaly warning is removed, otherwise the warning state is maintained and the verification failure details are pushed to the supervision platform, providing accurate basis for review and disposal of transportation anomalies.

[0104] Preferably, the specific steps of guiding the transportation vehicle to the verification position for abnormality verification include:

[0105] After the abnormality early warning, if the check location is not clear, the driver may randomly stop, such as in the emergency lane of the highway, the forbidden parking area, which has safety risks; or the selected check location is too far away, the site does not meet the dangerous goods parking standard, which leads to low check efficiency and safety. When the transportation abnormality early warning is triggered, based on the real-time positioning information of the vehicle, the current early warning section is determined, combined with the vehicle operation baseline and key node marking of the early warning section, the regular stopping point, rest area or gas station of the early warning section are selected as the potential check location, the straight-line distance and the expected driving time of each potential check location from the current vehicle are calculated, and the potential check location is prioritized according to the distance and the expected driving time, and the target check location is selected; and according to the site size, surrounding road layout and dangerous goods transportation safety distance requirement of the target check location, the verification range of the target check location is set to divide the check area.

[0106] The driver may ignore the abnormality early warning and refuse to go to the check location, such as continuing to drive to save time, which leads to the abnormality cannot be checked in time and the risk continues to expand. The vehicle terminal sends the check area information to the driver, including the name, coordinates and driving route of the target check location, and real-time tracks the driving route of the vehicle to monitor the driving route, and judges whether the vehicle is driving to the target check location, if not, the supervision early warning is carried out, the reminding information is continuously sent and synchronized to the supervision platform, and the remote intervention is carried out by the supervision personnel;

[0107] When the vehicle arrives at the target check location, the positioning sensing technology is used to verify whether the vehicle is in the check area of the target check location, if yes, the abnormality check process is triggered, if not, the vehicle position is adjusted to the check area until it arrives at the check area. The positioning verification forces the vehicle to enter the preset check area, ensures that the check is carried out in a safe and standardized environment, and ensures the effectiveness of the verification.

[0108] If the abnormality check process is triggered, the transportation subject information of the electronic waybill is called, including the transportation enterprise qualification number, vehicle identification code, driver and escort qualification certificate number, the information verification request is sent to the vehicle terminal through the encrypted communication channel, the information content is clear, and the verification information is required to be uploaded in real time according to the verification request; by comparing the uploaded verification request information, the abnormality caused by the inconsistency of the subject qualification is quickly investigated, and the risk reason is accurately located.

[0109] The uploaded verification information is extracted and matched with the transportation subject information of the electronic waybill, if the verification request information is matched successfully, it is determined that the transportation subject information verification is passed, the transportation abnormality early warning is released, and the transportation is allowed, otherwise the unmatched verification request information is marked and the reason is recorded, and the surrounding environment verification is started:

[0110] An environment image containing the name of the target verification location is photographed, the similarity of the environment image and the pre-stored target verification location standard environment image is calculated, if the similarity is greater than a preset verification similarity threshold, it is determined that the surrounding environment verification is passed, and the transportation abnormality early warning is released; if the similarity is less than or equal to the preset verification similarity threshold or the environment parameter exceeds the standard parameter range, the early warning state is maintained and the result is synchronized to the supervision platform for manual verification. Through the similarity comparison of the environment image, it is verified whether the vehicle is really at the target verification location, so as to avoid false locations to escape verification; manual verification is started for the mismatching condition, and the accuracy and fault tolerance are balanced.

[0111] Please refer to Figure 5 The embodiment introduces a digital dangerous goods transportation supervision method, which comprises the following steps:

[0112] Step S1: In the loading link, the dangerous goods transportation electronic waybill is obtained, the goods basic information, transportation path planning and transportation subject information are extracted, the basic positioning verification is first performed through the dynamically delimited compliance operation area, it is confirmed that the loading and unloading location is consistent with the record; then the initial scene image containing the key elements such as the goods label and the vehicle license plate is photographed, the initial scene verification is completed by comparing with the preset identification information set; then the dynamic acquisition interval is set according to the dangerous goods level, the image flow sequence containing the time sequence is constructed, the static picture cheating is prevented through the adjacent image similarity detection, and the dynamic change verification is completed; after the scene verification passes, the loading password containing the loading location feature code, the goods information encryption segment and the dynamic verification factor is generated in combination with the electronic waybill and the loading completion timestamp, the effective period is set based on the dangerous goods level and the transportation mileage, which is used for goods encryption storage, and the password is pushed only after the unloading verification passes.

[0113] Step S2: In the transportation link, the transportation route of the electronic waybill is divided according to the road section types such as expressway, ordinary highway, mountainous area and urban area, based on the historical transportation data excluding abnormal records, the vehicle running baseline of each road section containing the reference speed interval, the passing time interval, the conventional stopping point and the reference stopping time is constructed; the vehicle-mounted terminal collects the transportation data containing the collection time, position, speed and start-stop state according to the differentiated monitoring time interval, and encrypts and transmits the data to the monitoring platform; the monitoring platform first performs integrity verification, excludes invalid data and triggers retransmission, and then checks whether the valid data conforms to the vehicle baseline standard based on the vehicle running baseline; at the same time, the transmission time interval of the transportation data reception is judged to determine whether there is potential data delay, if yes, the delay completion data is generated by interpolation completion, and then the collection time of the received monitoring data is compared to distinguish data delay and data loss, and the transportation abnormality early warning is performed for the conditions such as non-conformance to the baseline standard, delay process abnormality and data loss risk.

[0114] Step S3: When the transport anomaly early warning is triggered, the current section is determined based on the real-time positioning of the vehicle, the nearest regular stop point, rest area or gas station is selected as the target verification location combined with the vehicle operation baseline and key node marking, the verification area is divided according to the site scale, safety distance, etc., the position information and driving route are pushed through the vehicle terminal and voice guidance is provided, the vehicle trajectory is tracked in real time, and the deviation is reminded to correct; after the vehicle arrives at the verification area, the transport main body information is verified first, the transport enterprise qualification, vehicle identification code and personnel certificate information in the electronic transport order are called and compared with the uploaded verification information; if the main body information verification fails, the surrounding environment verification is started, the environment image containing the verification location name is shot, the similarity with the pre-stored standard image is calculated, if the similarity is greater than the preset verification similarity threshold, it is determined that the surrounding environment verification is passed, and the transport anomaly early warning is removed; if the similarity is less than or equal to the preset verification similarity threshold or the environmental parameters are out of the standard parameter range, the early warning state is maintained and the result is synchronized to the supervision platform for manual verification.

[0115] Step S4: In the unloading link, after the vehicle arrives at the unloading location recorded in the electronic transport order, the basic positioning verification and initial scene verification in the loading link are repeated, after the verification, the monitoring platform sends the transport password to the vehicle terminal and the authorized equipment of the unloading person in charge through an encrypted channel, records the sending time, terminal identifier and vehicle location; the operator inputs the transport password, if the transport password is correct and within the valid period, the encrypted lock of the goods storage is unlocked, and the verification pass time is recorded as the unloading starting node; if the password is incorrect or invalid, the goods remain locked, and if the number of input errors exceeds the threshold, the supervision platform is sent a warning and real-time images are retrieved; after unlocking the encrypted lock of the goods storage, dynamic change verification is continued, after unloading is completed, the transport password is updated to the used state, an operation voucher containing the password use time, operator and full-process verification record is generated, and is archived in association with the electronic transport order and scene verification data.

[0116] Working principle and effect:

[0117] The present application constructs a full-process closed-loop mechanism of loading and unloading verification, transport monitoring and abnormal disposal, takes multi-layer verification, real-time comparison and accurate response as the core, solves the safety supervision pain points of each link of dangerous goods transportation, and realizes the controllable and traceable of the whole transportation cycle.

[0118] In the loading and unloading link, a compliance area matching the record information is first demarcated, key information such as cargo label and vehicle identification is extracted through high-definition shooting to complete initial verification, and a time sequence image stream sequence is generated by dynamically regulating the collection interval, adjacent frame similarity detection is used to prevent static picture cheating, and a carrying password encrypted cargo storage is generated by binding the loading site features and cargo information. This process eliminates virtual code scanning, remote operation and other fraudulent behaviors from the root, solves the problem of insufficient authenticity verification in the loading and unloading link, ensures that every operation of cargo loading and unloading is real and traceable, and the password binding mechanism effectively prevents cargo from being exchanged.

[0119] In the transportation process, the route is divided according to road types such as high-speed and mountainous areas, and the historical data excluding abnormal records is used to build the running baseline of each road section, such as the reference speed interval and regular stop points; the vehicle terminal collects and encrypts the transmission of data such as position and speed at different intervals, the platform first verifies the data integrity, then compares it with the baseline, and simultaneously interpolates the potential delay of transmission interval exceeding the standard to distinguish between real delay and data loss through time comparison. This baseline-based real-time monitoring and abnormal identification mechanism accurately captures the transportation risks such as overspeed, illegal stopping and path deviation, solves the problem of real-time identification of abnormal fluctuations, and provides solid data support for risk early warning.

[0120] When the abnormal early warning is triggered, the nearest regular stop point or rest area is selected as the verification location based on real-time positioning, after the vehicle arrives, the consistency of the transportation subject certificate and the electronic waybill is verified, and the environment image is shot and compared with the pre-stored feature library, and the early warning is lifted after the double verification is passed. This standardized verification process avoids the blindness of abnormal disposal, solves the problem of response lag, ensures accurate abnormal judgment and efficient disposal.

[0121] In summary, the whole-process closed-loop supervision formed by the cooperation of each link not only standardizes the loading, driving and abnormal handling of dangerous goods transportation at the operation level, but also moves the safety risk control node forward, significantly improves the safety, operation standardization and supervision accuracy of dangerous goods transportation, and builds a full-cycle, multi-dimensional safety line for dangerous goods transportation.

[0122] The above only describes the preferred embodiments of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solutions falling within the scope of the present application should be considered within the protection scope of the present application. It should be noted that for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application should also be considered within the protection scope of the present application.

Claims

1. A digitalized dangerous goods transportation monitoring system, characterized in that, The scene verification module, the transportation verification module, and the trajectory tracking module are included: The scene verification module obtains an electronic waybill of dangerous goods transportation for scene verification, including: performing basic positioning verification by delimiting a compliance operation area, performing initial scene verification by shooting an initial scene image, and dynamically adjusting and controlling collection intervals to construct an image stream sequence containing a time sequence for dynamic change verification; According to the scene verification result, a transport password is generated and sent, which is used for encrypted storage of goods and unlocking of unloading; The transportation verification module divides the transportation route of the electronic waybill into road segments, establishes a vehicle operation baseline for each road segment based on historical transportation records, judges whether the transportation data received by the monitoring platform conforms to the vehicle operation baseline, and combines the collection time of the transportation data to compare and distinguish data delay and data loss, respectively performs deviation analysis and loss anomaly diagnosis, identifies transportation abnormal fluctuations, and judges whether to perform transportation abnormal warning; When the transportation abnormal warning is triggered, the trajectory tracking module selects a target verification location based on positioning information, divides a verification area, and guides the transportation vehicle to the verification location for abnormal verification; The specific steps of performing scene verification include: When the loading and unloading operation is triggered, an electronic waybill of dangerous goods transportation is obtained, and the information of the electronic waybill includes basic information of goods, transportation path planning, and transportation subject information; According to the transportation path planning of the electronic waybill, the transportation starting point and the transportation terminal are obtained, and the compliance operation area of the loading and unloading operation is delimited by configuring a compliance range; The real-time geographic position of the loading and unloading operation is obtained, if the real-time geographic position is in the compliance operation area, it is determined that the location is normal, the environmental feature collection is activated, the initial scene image is shot for initial scene verification, the identification information set corresponding to the loading and unloading operation is configured based on the basic information of goods and the transportation subject information in the electronic waybill; Information recognition is performed on the initial scene verification, initial scene recognition information is obtained, and it is matched with the identification information set, if the initial scene recognition information matches the identification information set, it is determined that the initial scene authentication is passed; If the initial scene authentication is passed, the basic collection interval and the floating range are set according to the dangerous level classification of dangerous goods, a random number in the floating range is generated by an encryption random algorithm as a dynamic control value to control the basic collection interval, a dynamic collection interval is generated, panoramic images of the operation site are shot, and an image stream sequence containing a time sequence is constructed; The similarity of adjacent images in the image stream sequence is calculated, a similarity threshold is configured, if the similarity of adjacent images in the image stream sequence is greater than the similarity threshold, it is determined that the shooting is abnormal, otherwise it is determined that the real-time shooting is normal; If it is determined that the shooting is abnormal, remote review is performed, if the remote review is passed, the panoramic images of the operation site are continuously shot at the dynamic collection interval, the image stream sequence containing the time sequence is constructed, and the loading and unloading operation is completed.

2. A digitalized dangerous goods transportation monitoring system as claimed in claim 1, wherein, The specific steps of identifying transportation abnormal fluctuations and judging whether to perform transportation abnormal warning include: A monitoring time interval is configured, vehicle transportation data is collected according to the monitoring time interval, and is transmitted to the monitoring platform; the vehicle transportation data includes collection time, position, speed, and start-stop state; After the monitoring platform receives the transportation data, integrity check is performed, if the check fails, it is marked as invalid data and triggers the vehicle terminal to retransmit the instruction; If the check is successful, the reference speed interval, the passing time interval, the coordinates of the regular stop point and the reference stop duration of the current road section are obtained according to the vehicle operation baseline, the received transportation data is checked to determine whether it conforms to the vehicle operation baseline, if it does not conform to the vehicle operation baseline, transportation abnormality early warning is performed; If it conforms to the vehicle operation baseline, the transmission time interval since the last transportation data reception is monitored, if it is greater than the interval time threshold, it is determined that there is potential data delay, delay early warning is performed; The abnormal road section where the potential data delay is located is obtained, the driving track before the potential data delay and the speed change trend of the abnormal road section are interpolated to complete, the transportation data of the collection time point where the potential data delay is located is generated, and it is marked as delay completed data.

3. A digitalized dangerous goods transportation monitoring system as claimed in claim 2, wherein, The specific steps of identifying transportation abnormality fluctuation and determining whether to perform transportation abnormality early warning further include: The collection time of the monitoring data received subsequently by the monitoring platform and the collection time point of the delay completed data are obtained, and it is determined whether the collection times are the same; If they are the same, the delay completed data and the monitoring data received with delay are analyzed for deviation from position deviation, speed deviation and start-stop state difference, it is determined whether there is an abnormal situation during the delay, if there is, transportation abnormality early warning is performed; If they are different, the delay completed data is re-marked as missing completed data; the missing road section information and the missing duration of the missing completed data are counted, if the missing duration is greater than a preset missing threshold, missing abnormality diagnosis is performed, the potential transportation path of the missing road section is screened, and it is determined whether to perform transportation abnormality early warning; Real-time track sequences of each road section containing collection time, position, speed and start-stop state are constructed in time sequence, and electronic waybill information is archived.

4. A digitalized dangerous goods transportation supervision system as claimed in claim 3, characterized in that, The specific steps of missing abnormality diagnosis include: Based on the starting point coordinates and the end point coordinates of the missing road section, map data from the starting point to the end point is called, and the potential path between the starting point coordinates and the end point coordinates of the missing road section is identified; In combination with the rated transportation speed range of the transportation vehicle, the potential transportation path conforming to the rated transportation speed range is screened from the potential path; The potential transportation path is compared with the vehicle operation baseline, if there is a potential transportation path that does not conform to the vehicle operation baseline, it is determined that the missing abnormality diagnosis is abnormal, and transportation abnormality early warning is performed.

5. A digitalized dangerous goods transportation monitoring system as claimed in claim 1, wherein, The specific steps of generating and sending the carrying password include: After the loading link is completed, sample images are obtained by random sampling from the image stream sequence constructed during the loading link; Information recognition is performed on the sample images, and the sample images are verified by comparison with the corresponding identification information set during the loading link, if the sample images are all verified, the carrying password generation process is triggered, the carrying password is generated by asymmetric encryption algorithm in combination with the electronic waybill of the transportation and the loading completion timestamp, the carrying password is composed of the loading location feature code, the encrypted segment of the cargo information and the dynamic check factor, and is used for encrypting and locking the cargo storage; The effective period of the carrying password is dynamically set based on the danger level of the transportation dangerous goods and the transportation mileage; The carrying password is stored in an encrypted database after being generated. When the transport vehicle arrives at the unloading location recorded in the electronic waybill, it is determined that the vehicle is in a compliance operation area, and after the initial scene authentication is passed, the carrying password is sent to the on-board terminal of the transport vehicle and the destination authorized device through an encrypted communication channel; During the execution of the unloading operation, the carrying password is input. If the carrying password is correct and within the valid period, the encrypted lock of the cargo storage is unlocked. If the carrying password is input incorrectly or has expired, the cargo storage remains in a locked state. A verification threshold is configured. If the number of incorrect password inputs is greater than the verification threshold, a password verification anomaly early warning is sent to the supervision platform, and personnel review is performed.

6. A digitalized dangerous goods transportation supervision system as claimed in claim 1, characterized in that, The specific steps of establishing the vehicle operation baseline of each road section include: The planned transport route in the electronic waybill is called, the road section type and mileage data contained in the transport route are extracted, and the transport route is divided into road sections according to the road section type. The key nodes of the road sections are recorded as road section identifiers; The historical transport records of each road section are obtained, and invalid transport records with abnormal records are excluded to construct a baseline data set containing the average driving speed, start-stop position, stop duration, and transit time of each road section; The baseline data set is analyzed according to the road section. For the average driving speed of each road section, the speed fluctuation interval is calculated according to the time period, which is the baseline speed interval of each road section. According to the baseline speed interval, the transit time interval of each road section is calculated, and the start-stop position is analyzed by clustering. The average stop duration of the regular stop points is calculated as the baseline stop duration of the regular stop points. The baseline speed interval, transit time interval, regular stop point coordinates, and baseline stop duration of each road section are integrated into a structured vehicle operation baseline.

7. A digitalized dangerous goods transportation supervision system as claimed in claim 1, characterized in that, The specific steps of guiding the transport vehicle to the verification position for anomaly verification include: When the transport anomaly early warning is triggered, the current warning road section is determined based on the positioning information. The potential verification positions of the warning road section are selected by combining the vehicle operation baseline and key node markers of the warning road section. The straight-line distance and expected driving time of each potential verification position from the current vehicle are calculated, and the target verification position is selected according to the distance and time priority. The verification area of the target verification position is divided; The verification area information is sent to the driver through the on-board terminal, and the driving route is monitored to determine whether the vehicle is driving to the target verification position. If not, a supervision early warning is performed; When the target verification position is reached, it is verified whether the vehicle is in the verification area of the target verification position. If yes, the anomaly verification process is triggered. If not, the vehicle position is adjusted to the verification area; If the anomaly verification process is triggered, the transport subject information of the electronic waybill is called, and the information verification request is sent to the on-board terminal through an encrypted communication channel. The verification information is uploaded according to the verification request; The uploaded verification information is extracted and matched with the transport subject information of the electronic waybill. If the verification request information is matched successfully, it is determined that the transport subject information verification is passed, the transport abnormality early warning is released, otherwise the unmatched verification request information is marked, the surrounding environment verification is started, the environment image containing the name of the target verification position is shot, the similarity of the environment image and the pre-stored target verification position standard environment image is calculated, if the similarity is greater than the preset verification similarity threshold, it is determined that the surrounding environment verification is passed, the transport abnormality early warning is released, otherwise, the transport abnormality early warning is maintained and synchronized to the supervision platform, and manual verification is carried out.

8. A method for digitalized dangerous goods transportation supervision, implemented based on the digitalized dangerous goods transportation supervision system of any one of claims 1-7, characterized in that, Comprise: Step S1: In the loading link, the electronic waybill of dangerous goods transportation is acquired for scene verification, comprising: performing basic positioning verification by delimiting a compliant operation area, shooting an initial scene image for initial scene verification, dynamically adjusting and controlling the collection interval to build an image flow sequence containing time sequence for dynamic change verification; generating and sending a carrying password according to the loading link scene verification result, for encrypted storage of goods and unlocking of unloading; Step S2: The transport route of the electronic waybill is divided into road sections, and the vehicle running baseline of each road section is established based on historical transportation records, for judging whether the transportation data received by the monitoring platform conforms to the vehicle running baseline, and combining the collection time of the transportation data to compare and distinguish data delay and data loss, respectively through deviation analysis and loss anomaly diagnosis, identifying transportation abnormal fluctuations, and judging whether to perform transportation abnormality early warning; Step S3: When the transport abnormality early warning is triggered, a target verification position is selected based on the positioning information, and a verification area is divided, guiding the transport vehicle to the verification position for abnormality verification; Step S4: In the unloading link, the electronic waybill of dangerous goods transportation is acquired for basic positioning verification and initial scene verification, after the verification is passed, the monitoring platform sends the carrying password to the vehicle terminal and the authorized equipment through an encrypted channel, for unlocking the encrypted storage of goods, and continuing to perform dynamic change verification, and updating the carrying password state.

Images