Railway wagon loading quality intelligent detection system

Abstract

The invention discloses an intelligent detection system for the loading quality of a railway wagon, and the system comprises a data collection module which is used for collecting detection data for the same quality detection purpose of the wagon through a plurality of collection units, and obtaining a corresponding quality state based on the detection data; the data storage and management module is used for storing the detection data and presetting a weight value of each collected detection data for each quality detection purpose; the data fusion and processing module is used for comparing the detection data to obtain a quality state result; and the intelligent identification and judgment module carries out fault judgment according to the quality state result. According to the method, the limitation of a single detection means can be avoided, the reliability of a result can be ensured during data collision, the false alarm rate of a single sensor is effectively reduced, the accuracy and reliability of the detection result are improved, manual intervention is not needed, the labor intensity and the safety risk are reduced, and powerful support is provided for efficient management of truck loading quality.

Description

Technical Field

[0001] This manual relates to the field of railway transportation safety detection technology, and in particular to an intelligent detection system for the loading quality of railway freight cars. Background Technology

[0002] Currently, the inspection of railway freight cars mainly relies on manual observation and traditional inspection equipment. Manual observation has many drawbacks, such as low accuracy, high labor intensity, and safety hazards. Workers need to climb ladders to the top of the freight car for visual inspection or rely on a freight inspection system to judge via video feed on a computer terminal. This method is not only inefficient but also prone to missed or false inspections. For example, manual inspection may fail to accurately detect the amount of bulk frozen cargo inside the freight car, leading to overloading or uneven loading during subsequent loading, affecting train operation safety; it may also fail to promptly remove frozen cargo, causing tonnage discrepancies during subsequent transportation, affecting vehicle turnover and inventory statistics, resulting in direct economic losses.

[0003] Traditional inspection equipment has relatively limited functionality, often only detecting a single indicator of a truck. For example, off-center load measuring devices can only detect whether a truck is overloaded or off-center, but cannot comprehensively inspect the truck's loading status or vehicle information. Furthermore, data from different inspection devices cannot be effectively integrated and shared, making it impossible to verify the reliability of the inspection results. This affects the accuracy and reliability of the results; for instance, using an off-center load measuring device alone may not accurately determine the empty or loaded status of a truck. In some cases, multiple people and multiple devices are involved in receiving and inspecting trucks, inevitably leading to missed detections, measurement errors, and even necessitating the presence of fixed personnel on-site for inspection and observation.

[0004] With the continuous increase in railway freight volume and the continuous improvement of transportation requirements, relevant policies and regulations are placing increasingly higher demands on the inspection of railway freight cars. Traditional inspection methods can no longer meet the needs of modern railway transportation. Therefore, developing a comprehensive intelligent inspection system capable of fully, intelligently, and efficiently inspecting the condition of railway freight cars is of significant practical importance. Summary of the Invention

[0005] This specification provides one or more embodiments of an intelligent detection system for the loading quality of railway freight cars, which is used to solve at least one of the above-mentioned technical problems.

[0006] One or more embodiments of this specification employ the following technical solution: One or more embodiments of this specification provide an intelligent detection system for the loading quality of railway freight cars, comprising: The data acquisition module includes multiple acquisition units, which respectively collect test data for the same quality inspection purpose of the truck, and obtain the corresponding quality status based on each of the test data; The data storage and management module is used to store the detection data and preset the weight values ​​of each collected detection data for each quality detection purpose. The data fusion and processing module is used to compare the detection data collected by the multiple acquisition units. When the multiple quality states are consistent for the same quality detection purpose, a quality state result is obtained. When the multiple quality states are inconsistent for the same quality detection purpose, a core acquisition unit is determined based on the preset weight values ​​of the multiple acquisition units for the quality detection purpose, and the corresponding quality state of the detection data of the core acquisition unit is used as the quality state result. The intelligent identification and judgment module determines the fault based on the quality status results.

[0007] In one possible implementation of this application, the quality inspection objective includes residue detection; the plurality of acquisition units respectively acquire detection data for the residue detection, the detection data including two-dimensional image data and three-dimensional image data, and the quality status includes the residue retention status obtained based on the two-dimensional image data and the three-dimensional image data respectively.

[0008] In one possible implementation of this application, the data fusion and processing module takes the corresponding quality status of the three-dimensional image data as the quality status result, and the intelligent identification and judgment module verifies the quality status result. When the retention status of the corresponding residue in the two-dimensional image data is "present" while the retention status of the corresponding residue in the three-dimensional image data is "absent", an alarm is issued based on the quality status result; when the retention status of the corresponding residue in the two-dimensional image data is "absent" while the retention status of the corresponding residue in the three-dimensional image data is "present", a fault is determined based on the quality status result.

[0009] In one possible implementation of this application, the quality inspection objective includes residue detection; the multiple acquisition units respectively acquire detection data for the residue condition detection, and the quality status includes the weight status of the residue; the detection data includes three-dimensional image data obtained by the acquisition units scanning the reflection lines on the surface of objects inside the wagon of the moving truck; the data fusion and processing module obtains the volume of the free space inside the wagon based on the three-dimensional image data; the data storage and management module presets the factory wagon volume and residue density, obtains the residue volume based on the factory wagon volume and the volume of the free space inside the wagon, and calculates the residue weight based on the residue density to obtain the first residue weight status.

[0010] In one possible implementation of this application, the data storage and management module has preset factory-exit wagon weight data for freight cars, and the detection data also includes total wagon weight data. The data fusion and processing module calculates the residual weight based on the total wagon weight data and the factory-exit wagon weight data to obtain a second residual weight state. The data fusion and processing module compares the first residual weight state and the second residual weight state to obtain a residual weight state result.

[0011] In one possible implementation of this application, the detection data further includes the factory-exit wagon weight data and the total wagon weight data. The data fusion and processing module obtains the residual weight based on the total wagon weight data and the factory-exit wagon weight data, as a second residual weight state. The data fusion and processing module compares the first residual weight state and the second residual weight state to obtain the residual weight state result.

[0012] In one possible implementation of this application, the data storage and management module has a preset car number corresponding to the weight of the car body leaving the factory. The detection data also includes car number data. The data fusion and processing module obtains the corresponding weight of the car body leaving the factory based on the car number data, and calculates the weight of the residue based on the total weight data of the car body and the weight data of the car body leaving the factory, as a second residue weight state.

[0013] In one possible implementation of this application, the quality inspection objective includes vehicle body damage detection; the plurality of acquisition units respectively acquire detection data for the vehicle body damage detection, the detection data includes two-dimensional image data and three-dimensional image data, and the quality status includes the vehicle body damage status obtained based on the two-dimensional image data and the three-dimensional image data respectively.

[0014] In one possible implementation of this application, the quality inspection objective includes detecting overload and off-center loading of a truck; the plurality of acquisition units respectively acquire detection data for the overload and off-center loading detection of the truck, and the detection data includes two-dimensional image data, three-dimensional image data, and weight data of each wheel; the quality status includes the overload status and off-center loading status of the truck obtained based on the two-dimensional image data, three-dimensional image data, and weight data of each wheel, respectively.

[0015] In one possible implementation of this application, the system further includes: a data transmission module connected between the data acquisition module and the data storage and management module, the data transmission module being used to receive data acquired by the data acquisition module and transmit it to the data storage and management module; and a network interface module being used to enable data interaction between the intelligent identification and judgment module and an external network.

[0016] In one possible implementation of this application, the detection data includes environmental data, two-dimensional image data, and three-dimensional image data, and the data fusion and processing module performs environmental error correction on the two-dimensional image data and the three-dimensional image data based on the environmental data.

[0017] The intelligent detection system for the loading quality of railway freight cars provided in this application has the following beneficial effects: 1. The intelligent inspection system for railway freight car loading quality provided in this application collects inspection data from multiple acquisition units for the same quality inspection purpose. A data storage and management module presets the weight values ​​of each acquisition unit, and a data fusion and processing module compares and merges the multi-source inspection data. When the quality status is consistent, the result is directly determined; when inconsistent, the result of the core acquisition unit is selected based on the weight values. Finally, an intelligent identification and judgment module determines the fault. This management system, through mutual verification and supplementation of multi-source data and a weight mechanism to clarify the priority of core data, avoids the limitations of single inspection methods and ensures the reliability of results when data conflicts occur. It effectively reduces the false alarm rate of single sensors and improves the accuracy and reliability of inspection results. Compared with traditional manual observation and single-sensor detection methods, it greatly improves inspection accuracy and efficiency, eliminates the need for manual intervention, reduces labor intensity and safety risks, and provides strong support for the efficient management of freight car loading quality.

[0018] 2. One possible implementation of the intelligent detection system for railway freight car loading quality provided in this application involves using a two-dimensional image acquisition unit and a three-dimensional image acquisition unit to acquire two-dimensional and three-dimensional image data respectively during residue detection and car body damage detection, thereby obtaining the corresponding residue retention status. Two-dimensional image data can intuitively present the appearance, distribution, and other information of the residue from a planar perspective, while three-dimensional image data can accurately reflect the spatial morphology and even volume characteristics of the residue from a three-dimensional perspective. The combination of the two can achieve multi-dimensional and all-round detection of the residue. Furthermore, through the data fusion and processing module, subsequent intelligent analysis can fully integrate the advantages of two-dimensional and three-dimensional data under different environments. For example, insufficient lighting may cause two-dimensional images to be blurry, while three-dimensional images can compensate for the above-mentioned defects of two-dimensional images by scanning in a multi-dimensional, three-dimensional, and dynamic manner. Also, high humidity may affect the scanning accuracy of three-dimensional images, and three-dimensional images may have poor clarity when the vehicle speed changes too quickly, which can be compensated for by using two-dimensional images. Therefore, by flexibly determining the weight ratio of detection units according to different detection purposes and environments, the residual status can be determined more accurately and comprehensively, providing a strong guarantee for the accuracy and reliability of residual detection and helping to make the quality inspection and management of railway freight car loading more precise and efficient.

[0019] Similarly, in detecting overload and off-center loading of trucks, the overload and off-center loading states of the trucks are obtained based on two-dimensional image data, three-dimensional image data, and weight data of each wheel, respectively. The weight ratio of the detection units can be flexibly determined according to different detection purposes and environments, so as to more accurately and comprehensively determine the overload and off-center loading state of the trucks.

[0020] 3. One possible implementation of this application provides an intelligent detection system for the loading quality of railway freight cars. The intelligent identification and judgment module verifies the quality status result. During residue detection, the data fusion and processing module uses the quality status corresponding to the three-dimensional image data as the result, while the intelligent identification and judgment module verifies this result in conjunction with two-dimensional image data. When there is a conflict between the presence and absence of residue presented by the two-dimensional and three-dimensional image data, alarm prompts or fault determinations are issued respectively. In this way, the core result is determined by leveraging the detection advantages of three-dimensional image data in terms of spatial morphology, while two-dimensional image data is used for auxiliary verification from a planar perspective. Through mutual verification, potential anomalies during the detection process can be detected in a timely manner, thereby effectively improving the reliability of residue detection results and providing stronger support for the precise management of railway freight car loading quality.

[0021] 4. The intelligent detection system for railway freight car loading quality provided in one possible implementation of this application acquires three-dimensional image data by scanning the reflective lines on the surface of objects inside the moving freight car using a data acquisition unit. The data fusion and processing module accurately obtains the volume of the empty space inside the freight car based on this three-dimensional image data. Simultaneously, the data storage and management module pre-stores the freight car's factory-delivered volume and residual material density. Based on the factory-delivered volume and the volume of the empty space, the residual material volume can be calculated. Combined with the residual material density, the first residual material weight state is further derived. This fully utilizes the precise spatial morphology capture capability of three-dimensional image data, combined with preset freight car basic parameters and material density parameters, to achieve indirect but reliable detection of the residual material weight state inside the moving freight car. This effectively avoids the misjudgment of residual material weight that may occur in traditional detection methods due to limitations of direct weighing or blind spots in the viewing angle. It provides a technical path for residual material detection that is more in line with actual transportation scenarios, thereby improving the accuracy and applicability of residual material detection in railway freight car loading quality inspection and management.

[0022] 5. One possible implementation of the intelligent detection system for railway freight car loading quality provided in this application pre-stores the weight data of the freight cars leaving the factory through a data storage and management module, and simultaneously acquires the total weight data of the freight cars through an acquisition unit. A data fusion and processing module calculates a second residual weight state based on the difference between the two values, and then compares this with a first residual weight state derived from three-dimensional image data to determine the final residual weight state. This process indirectly calculates the residual weight from a spatial volume dimension using three-dimensional image data, and also obtains the residual weight from a direct weighing dimension based on the difference between the total weight of the freight cars and the factory weight. These two different detection logics complement and verify each other, effectively avoiding potential errors in single detection methods (such as spatial measurement deviations in three-dimensional image data calculations, and interference from impurities attached to the freight cars in direct weighing). This results in a more accurate and reliable determination of the residual weight state, providing dual assurance for the accurate determination of residual weight in intelligent detection of railway freight car loading quality.

[0023] 6. One possible implementation of the intelligent inspection system for railway freight car loading quality provided in this application includes a data transmission module connected between the data acquisition module and the data storage and management module. This module specifically receives various types of inspection data collected by the data acquisition module and transmits them to the data storage and management module. This avoids data loss or delays caused by unclear transmission paths or lack of dedicated transmission carriers after acquisition, ensuring that the inspection data can be stored completely and promptly to support subsequent processing. Simultaneously, a network interface module enables data interaction between the intelligent identification and judgment module and the external network. By linking with the railway freight inspection system through the network interface, photos of problematic freight cars are automatically uploaded to the three-level networked platform of the freight inspection system. The system correctly matches and accurately labels the train information, achieving information sharing, comprehensive statistics, and integration of machine inspection information. It also facilitates external configuration and control of the system's inspection logic and parameters via the network.

[0024] The data fusion and processing module uses the collected environmental data to correct environmental errors in two-dimensional and three-dimensional image data. This process fully considers the interference that environmental factors (such as light, temperature, humidity, and weather) may cause to image acquisition. For example, it enhances image contrast when the light intensity is insufficient and performs humidity compensation on weight data when the humidity is high, thus offsetting the deviation caused by environmental factors. This makes the two-dimensional image data present the details of the detected object more clearly and the three-dimensional image data more accurately reflect the spatial morphological characteristics. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is an overall structural diagram of an intelligent detection system for the loading quality of railway freight cars, provided for one or more embodiments of this specification. Detailed Implementation

[0026] This specification provides an intelligent detection system for the loading quality of railway freight cars.

[0027] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0028] Figure 1 This diagram illustrates the overall structure of an intelligent detection system for the loading quality of railway freight cars, provided for one or more embodiments of this specification. Certain input parameters or intermediate results in the detection process can be manually adjusted to help improve accuracy.

[0029] like Figure 1 As shown in the figure, the structure of an intelligent detection system for the loading quality of railway freight cars according to an embodiment of this specification includes a data acquisition module and a data processing center. The data processing center includes a data transmission module, a network interface module, a data storage and management module, a data fusion and processing module, an intelligent identification and judgment module, and an alarm and display module. The data acquisition module includes multiple acquisition units, which may optionally be a two-dimensional image acquisition unit, a three-dimensional image acquisition unit, a dynamic electronic track scale, an overload and off-center load data acquisition unit, a car number data acquisition unit, and an environmental data acquisition unit.

[0030] In one specific embodiment, the aforementioned multiple acquisition units collect detection data for the same quality inspection purpose of the freight cars, and obtain the corresponding quality status based on each of the detection data. It is understood that the data acquisition module of the intelligent detection system for railway freight car loading quality provided in this application may optionally include any two or three of the aforementioned acquisition units.

[0031] More specifically, in this embodiment, the two-dimensional image acquisition unit is installed on a gantry along the railway line to acquire high-definition two-dimensional image data of the freight car's appearance in real time during operation. Combined with image processing algorithms such as edge detection and feature matching from the data processing center, as well as convolutional neural networks (CNNs), it automatically identifies defects such as door status, car body damage, foreign objects inside the car, personnel climbing on the car, and abnormal car body appearance, achieving a detection accuracy of ≥95%, ensuring that no abnormalities in the freight car's appearance are overlooked.

[0032] The 3D image acquisition unit is mounted on the gantry and is responsible for acquiring 3D point cloud data of the truck. It uses deep learning algorithms (such as PintNet) to analyze the loading pattern, achieving high-precision measurements with a flatness error ≤2cm and an over-limit detection accuracy ≤5cm. It can detect vehicle body damage and foreign objects inside the vehicle. The 3D image acquisition unit can also detect the volume and weight of residual coal inside the vehicle, providing reliable data for coal loading quality inspection.

[0033] The dynamic electronic track scale and overload / off-center load data acquisition unit are used to measure the weight data of the freight car in real time, including the total weight of the vehicle and the axle load, in order to determine whether the freight car is overloaded or off-center loaded.

[0034] The train number data acquisition unit uses RFID electronic tag identification technology and is installed at specific locations along the railway line to identify information such as the train number, speed, and order of carriage. Simultaneously, the system also has the function of 2D intelligent recognition of train numbers and weight. Image processing technology from the data processing center is used to perform image recognition of the train number and weight markings, improving the accuracy and reliability of the identification.

[0035] The environmental data acquisition unit includes temperature sensors, humidity sensors, wind speed and direction sensors, rain gauges, light sensors, and sound sensors. It is used to detect the environmental conditions at the site of the railway freight car integrated inspection device, such as temperature, humidity, wind speed and direction, and rainfall, providing environmental parameters for the normal operation of the system.

[0036] The structure of the intelligent inspection system for loading quality of railway freight cars is described in detail, and its management method is explained in detail. The data acquisition module includes multiple acquisition units, which collect inspection data for the same quality inspection purpose of the freight cars and obtain the corresponding quality status based on the inspection data. The data storage and management module stores the detection data and presets weight values ​​for each collected detection data for each quality detection purpose. The data fusion and processing module compares the detection data collected by the multiple acquisition units. When multiple quality states are consistent for the same quality detection purpose, a quality state result is obtained. When multiple quality states are inconsistent for the same quality detection purpose, a core acquisition unit is determined based on the preset weight values ​​of multiple acquisition units for the quality detection purpose, and the corresponding quality state of the detection data of the core acquisition unit is used as the quality state result. The intelligent identification and judgment module performs fault judgment based on the quality state result, such as determining whether there are residues, overloads, or off-center loads.

[0037] The intelligent inspection system for railway freight car loading quality provided in this application collects inspection data from multiple acquisition units targeting the same quality inspection objective. A data storage and management module pre-sets weight values ​​for each acquisition unit, and a data fusion and processing module compares and integrates the multi-source inspection data. When the quality status is consistent, the result is directly determined; when inconsistent, the result of the core acquisition unit is selected based on the weight values. Finally, an intelligent identification and judgment module determines the fault. This management system, through mutual verification and supplementation of multi-source data and a weighting mechanism to clarify the priority of core data, avoids the limitations of single inspection methods and ensures the reliability of results in the event of data conflicts. It effectively reduces the false alarm rate of single sensors and improves the accuracy and reliability of inspection results. Compared with traditional manual observation and single-sensor detection methods, it significantly improves inspection accuracy and efficiency, eliminates the need for manual intervention, reduces labor intensity and safety risks, and provides strong support for the efficient management of freight car loading quality.

[0038] Example 1 This embodiment describes a scenario for detecting residual debris inside trucks and residual coal after unloading from coal trucks.

[0039] In this embodiment, a two-dimensional image acquisition unit and a three-dimensional image acquisition unit are used to acquire two-dimensional image data and three-dimensional image data respectively for residue detection, to obtain the corresponding quality status, i.e., residue present or residue absent. According to the preset weight values ​​of each acquired detection data in the data storage and management module, for example, in one environment, considering the omnidirectional nature of the three-dimensional image compared to the two-dimensional image, the preset weight value of the three-dimensional image data is greater than the weight value of the two-dimensional image data. For example, the preset weight value of the three-dimensional image data is 1, and the preset weight value of the two-dimensional image data is 0.5. After comprehensive processing, if the two-dimensional image shows residue in the corresponding car body, and the three-dimensional image shows residue in the corresponding exit car body, then the corresponding quality status of the two is consistent. The comprehensive processing result is that residue is present. Based on the quality status result, further fault determination is performed.

[0040] It is understandable that both two-dimensional and three-dimensional images can detect residual objects inside the vehicle body. However, two-dimensional images have higher planar image clarity. Therefore, in detection scenarios such as high vehicle speed and the residual object being a tool, the weight of two-dimensional images can be optionally set to be greater than that of three-dimensional images.

[0041] Furthermore, the data fusion and processing module uses the corresponding quality status of the three-dimensional image data as the quality status result, and the intelligent recognition and judgment module verifies the quality status result. If the retention status of the corresponding residue in the two-dimensional image data is "present" while the retention status of the corresponding residue in the three-dimensional image data is "absent", an alarm is issued based on the quality status result; if the retention status of the corresponding residue in the two-dimensional image data is "absent" while the retention status of the corresponding residue in the three-dimensional image data is "present", a fault is determined based on the quality status result.

[0042] For example, if there are residues in the corresponding car body in the 2D image, but no residues remain in the corresponding car body in the 3D image, then an alarm will be triggered in the quality status result. This indicates that the 3D image acquisition module with a weight of 1 may be damaged, causing a missed detection, and prompts the user to check the condition of the 3D image acquisition module and repair it.

[0043] Furthermore, in this embodiment, based on three-dimensional image data and two-dimensional image data, it is determined whether there is residual cargo inside the truck, and the weight and volume of the residual cargo can also be calculated, thereby expanding the corresponding weight status information of the three-dimensional image data and expanding the detection range of the three-dimensional image acquisition unit. Specifically, the quality status also includes the weight status of the residue; the detection data includes three-dimensional image data obtained by the acquisition unit scanning the reflection lines on the surface of objects inside the truck's body; the data fusion and processing module obtains the volume of the free space inside the truck body based on the three-dimensional image data; the data storage and management module presets the truck's factory truck body volume and residue density, obtains the residue volume based on the factory truck body volume and the volume of the free space inside the truck body, and calculates the residue weight based on the residue density to obtain the first residue weight status, i.e., the residue weight. More preferably, the data storage and management module has a preset car number corresponding to the weight of the car body leaving the factory, and the detection data also includes car number data. The data fusion and processing module obtains the corresponding weight of the car body leaving the factory based on the car number, so that the data fusion and processing module calculates the weight of the residue based on the total weight data of the car body and the weight data of the car body leaving the factory, as the second residue weight state, that is, the specific tonnage of the residue.

[0044] In summary, the two-dimensional image reflects the presence of remaining coal in the wagon, and the three-dimensional image also reflects the presence of remaining coal in the wagon. The weight of the three-dimensional image is 1, and the weight of the two-dimensional image is 0.5. In addition, the three-dimensional image also reflects the tonnage of the remaining coal. The comprehensive processing result shows that there are XX tons of remaining coal in the wagon. Record, alarm, and report.

[0045] Understandably, by detecting the car number data, the corresponding weight of the factory-exit car can be retrieved and matched using the car number data, thereby adapting to the inspection needs of railway freight cars of different models and different loads. Different intelligent recognition algorithms can be retrieved according to different car types and models, improving the flexibility and adaptability of the system.

[0046] Of course, as an alternative embodiment of this example, the data storage and management module has preset factory-issued wagon weight data for freight cars, and the detection data also includes total wagon weight data. The data fusion and processing module calculates the residual weight based on the total wagon weight data and the factory-issued wagon weight data to obtain a second residual weight state. The data fusion and processing module compares the first residual weight state and the second residual weight state to obtain the residual weight state result.

[0047] This application adopts the aforementioned implementation method. During residue detection, a two-dimensional image acquisition unit and a three-dimensional image acquisition unit are used to acquire two-dimensional image data and three-dimensional image data respectively, thereby obtaining the corresponding residue retention status. Two-dimensional image data can intuitively present the appearance, distribution, and other information of residues from a planar perspective, while three-dimensional image data can accurately reflect the spatial morphology and even volume characteristics of residues from a three-dimensional perspective. The combination of the two can achieve multi-dimensional and comprehensive detection of residues. Further intelligent analysis through a data fusion and processing module can fully integrate the advantages of two-dimensional and three-dimensional data under different environments. For example, insufficient lighting may cause two-dimensional images to be blurry, while three-dimensional images, through multi-dimensional, three-dimensional, and dynamic scanning, compensate for the aforementioned defects of two-dimensional images. Similarly, high humidity may affect the scanning accuracy of three-dimensional images, and three-dimensional images may exhibit poor clarity when the vehicle speed changes too rapidly; this can be compensated for by using two-dimensional images. Therefore, by flexibly determining the weight ratio of the detection units according to different detection purposes and environments, the residue retention status can be determined more accurately and comprehensively, providing strong support for the accuracy and reliability of residue detection and contributing to more precise and efficient railway freight car loading quality inspection and management.

[0048] The intelligent identification and judgment module verifies the quality status result. During residue detection, the data fusion and processing module uses the quality status corresponding to the 3D image data as the result, while the intelligent identification and judgment module verifies this result in conjunction with the 2D image data. When there is a conflict between the presence and absence of residue presented in the 2D and 3D image data, alarm prompts or fault determinations are issued respectively. In this way, the core result is determined by leveraging the detection advantages of 3D image data in terms of spatial morphology, while 2D image data is used for auxiliary verification from a planar perspective. Through mutual verification, potential anomalies during the detection process can be detected in a timely manner, thereby effectively improving the reliability of residue detection results and providing stronger support for the precise management of railway freight car loading quality.

[0049] The acquisition unit scans the reflective lines on the surface of objects inside the moving freight car to obtain three-dimensional image data. The data fusion and processing module uses this three-dimensional image data to accurately obtain the volume of the empty space inside the car. At the same time, the data storage and management module pre-stores the factory-delivered car volume and the density of the residue. Based on the factory-delivered car volume and the volume of the empty space, the volume of the residue can be calculated. Then, combined with the residue density, the first weight state of the residue is further obtained. By making full use of the accurate capture capability of three-dimensional image data of spatial shape, combined with the preset basic parameters of the car and the material density parameters, the indirect but reliable detection of the weight state of the residue inside the moving freight car is realized. This effectively avoids the misjudgment of the weight of the residue that may be caused by the limitations of direct weighing or blind spots in the traditional detection method. It provides a technical path for residue detection that is more in line with the actual transportation scenario, thereby improving the accuracy and applicability of residue detection in the loading quality inspection and management of railway freight cars.

[0050] The data storage and management module pre-stores the weight data of the freight cars leaving the factory, while the acquisition unit acquires the total weight data of the freight cars. The data fusion and processing module calculates the second residual weight state based on the difference between the two, and then compares it with the first residual weight state derived from the three-dimensional image data to determine the final residual weight state. This process uses three-dimensional image data to indirectly calculate the residual weight from the spatial volume dimension, and also relies on the difference between the total weight of the freight car and the factory weight to obtain the residual weight from the direct weighing dimension. The two different detection logics complement and verify each other, which can effectively avoid the errors that may exist in a single detection method (such as spatial measurement deviation in the calculation of three-dimensional image data, interference from impurities attached to the freight car in direct weighing, etc.), and thus more accurately and reliably determine the residual weight state, providing double protection for the accurate determination of residual weight in the intelligent detection of railway freight car loading quality.

[0051] Preferably, in this embodiment, the system further includes: a data transmission module connected between the data acquisition module and the data storage and management module, which receives data collected by the data acquisition module and transmits it to the data storage and management module; and a network interface module for enabling data interaction between the intelligent identification and judgment module and an external network. More preferably, the system further includes an alarm and display module for issuing voice prompts or flashing lights. After the intelligent identification and judgment module issues a fault determination or alarm prompt, such as overloading, excessive residual goods, or issuing an alarm signal such as a voice prompt or flashing lights, the system alerts the staff. Simultaneously, the detection results and alarm information are displayed on a display terminal for easy viewing and processing by staff.

[0052] Preferably, the detection data includes environmental data, two-dimensional image data, and three-dimensional image data. The data fusion and processing module performs environmental error correction on the two-dimensional and three-dimensional image data based on the environmental data. For example, it enhances image contrast when the light intensity is insufficient and performs humidity compensation on the weight data when the humidity is high.

[0053] By setting up a data transmission module connecting the data acquisition module and the data storage and management module, various types of detection data collected by the data acquisition module can be specifically received and transmitted to the data storage and management module. This avoids data loss and delays caused by unclear transmission paths or lack of dedicated transmission carriers after acquisition, ensuring that detection data can be completely and promptly stored to support subsequent processing. Simultaneously, a network interface module enables data interaction between the intelligent identification and judgment module and the external network. Through the network interface, it links with the railway freight inspection system, automatically uploading photos of problematic vehicles to the three-level networked platform of the freight inspection system. The freight inspection system correctly matches and accurately labels the train information, achieving information sharing, comprehensive statistics, and integration of machine inspection information. It also facilitates external configuration and control of the system's detection logic and parameters via the network.

[0054] The data fusion and processing module uses the collected environmental data to correct environmental errors in two-dimensional and three-dimensional image data. This process fully considers the interference that environmental factors (such as light, temperature, humidity, and weather) may cause to image acquisition. For example, it enhances image contrast when the light intensity is insufficient and performs humidity compensation on weight data when the humidity is high, thus offsetting the deviation caused by environmental factors. This makes the two-dimensional image data present the details of the detected object more clearly and the three-dimensional image data more accurately reflect the spatial morphological characteristics.

[0055] Example 2 This embodiment describes a detection scenario for vehicle body damage detection.

[0056] In this embodiment, the quality inspection objective includes vehicle body damage detection. The multiple acquisition units collect inspection data for the vehicle body damage detection, and the inspection data includes two-dimensional image data and three-dimensional image data. The quality status includes the vehicle body damage status obtained based on the two-dimensional image data and the three-dimensional image data, including the outer side of the vehicle body being intact, slightly damaged, or severely damaged; or, the vehicle body damage status includes the top of the vehicle body being in good condition, containing foreign objects, or damaged; or, the vehicle body damage status includes the tanker opening being intact, not closed, or damaged.

[0057] Example 3 This embodiment describes a detection scenario for detecting overloading and off-center loading of trucks.

[0058] In this embodiment, the quality inspection objective includes detecting overloading and off-center loading of trucks. The multiple acquisition units collect inspection data for this purpose, including two-dimensional image data, three-dimensional image data, and wheel weight data. The quality status includes the truck's overload status and off-center loading status obtained based on the two-dimensional image data, three-dimensional image data, and wheel weight data, respectively. This application, during truck overloading and off-center loading detection, flexibly determines the weight ratio of the inspection units based on the overload and off-center loading status obtained from the two-dimensional image data, three-dimensional image data, and wheel weight data, according to different inspection objectives and environments, thereby more accurately and comprehensively determining the truck's overloading and off-center loading status.

[0059] 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.

[0060] 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 apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0061] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0062] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0063] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0064] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The aforementioned units can be implemented in hardware or software.

[0065] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0066] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A smart detection system for the loading quality of railway freight cars, characterized in that, include: The data acquisition module includes multiple acquisition units, which respectively collect test data for the same quality inspection purpose of the truck, and obtain the corresponding quality status based on each of the test data; The data storage and management module is used to store the detection data and preset the weight values ​​of each collected detection data for each quality detection purpose. The data fusion and processing module is used to compare the detection data collected by the multiple acquisition units. When the multiple quality states are consistent for the same quality detection purpose, a quality state result is obtained. When the multiple quality states are inconsistent for the same quality detection purpose, a core acquisition unit is determined based on the preset weight values ​​of the multiple acquisition units for the quality detection purpose, and the corresponding quality state of the detection data of the core acquisition unit is used as the quality state result. The intelligent identification and judgment module determines the fault based on the quality status results.

2. The intelligent detection system for railway freight car loading quality according to claim 1, characterized in that, The purpose of the quality inspection includes residue detection; The multiple acquisition units acquire detection data for the residue detection, and the detection data includes two-dimensional image data and three-dimensional image data. The quality status includes the retention status of the residue obtained based on the two-dimensional image data and the three-dimensional image data, respectively.

3. The intelligent detection system for railway freight car loading quality according to claim 2, characterized in that, The data fusion and processing module takes the corresponding quality status of the three-dimensional image data as the quality status result. The intelligent recognition and judgment module verifies the quality status result. When the retention status of the corresponding residue in the two-dimensional image data is "yes" and the retention status of the corresponding residue in the three-dimensional image data is "no", an alarm is issued based on the quality status result. When the retention status of the corresponding residue in the two-dimensional image data is zero, while the retention status of the corresponding residue in the three-dimensional image data is positive, a fault determination is made based on the quality status result.

4. The intelligent detection system for railway freight car loading quality according to claim 1, characterized in that, The purpose of the quality inspection includes residue detection; The multiple acquisition units respectively acquire detection data for the condition of the residue, and the quality status includes the weight status of the residue. The detection data includes three-dimensional image data obtained by the acquisition unit scanning the reflection lines on the surface of objects inside the wagon of the moving truck. The data fusion and processing module obtains the volume of the free space inside the wagon based on the three-dimensional image data. The data storage and management module has preset wagon volume and residual density of the truck at the factory. Based on the wagon volume at the factory and the volume of the free space inside the wagon, the residual volume is obtained, and the residual weight is calculated based on the residual density to obtain the first residual weight state.

5. The intelligent detection system for railway freight car loading quality according to claim 4, characterized in that, The data storage and management module has preset data on the weight of the freight cars leaving the factory. The detection data also includes the total weight data of the freight cars. The data fusion and processing module calculates the weight of the residue based on the total weight data of the freight cars and the weight data of the freight cars leaving the factory, and obtains the second residue weight status. The data fusion and processing module compares the weight status of the first residue with the weight status of the second residue to obtain the weight status result of the residue.

6. The intelligent detection system for railway freight car loading quality according to claim 4, characterized in that, The detection data also includes the factory-exit car weight data and the total car weight data of the truck. The data fusion and processing module obtains the residual weight based on the total car weight data and the factory-exit car weight data, as the second residual weight state. The data fusion and processing module compares the weight status of the first residue with the weight status of the second residue to obtain the weight status result of the residue.

7. The intelligent detection system for the loading quality of railway freight cars according to claim 5, characterized in that, The data storage and management module has a preset car number corresponding to the weight of the car body leaving the factory. The detection data also includes car number data. The data fusion and processing module matches the car number data to obtain the corresponding weight data of the car body leaving the factory. The residual weight is obtained based on the total weight data of the car body and the weight data of the car body leaving the factory, which serves as the second residual weight state.

8. The intelligent detection system for railway freight car loading quality according to claim 1, characterized in that, The quality inspection objective includes vehicle body damage detection. The multiple acquisition units collect detection data for the vehicle body damage detection, and the detection data includes two-dimensional image data and three-dimensional image data. The quality status includes the vehicle body damage status obtained based on the two-dimensional image data and the three-dimensional image data, respectively. Alternatively, the quality inspection objective includes truck overload and off-center load detection. The multiple acquisition units collect inspection data for the truck overload and off-center load detection, and the inspection data includes two-dimensional image data, three-dimensional image data, and weight data of each wheel. The quality status includes the truck overload status and off-center load status obtained based on the two-dimensional image data, three-dimensional image data, and weight data of each wheel, respectively.

9. The intelligent detection system for the loading quality of railway freight cars according to claim 1, characterized in that, The system also includes: A data transmission module is connected between the data acquisition module and the data storage and management module. The data transmission module is used to receive data acquired by the data acquisition module and transmit it to the data storage and management module. A network interface module is used to enable data interaction between the intelligent identification and judgment module and an external network.

10. The intelligent detection system for railway freight car loading quality according to claim 1, characterized in that, The detection data includes environmental data, two-dimensional image data, and three-dimensional image data. The data fusion and processing module performs environmental error correction on the two-dimensional image data and three-dimensional image data based on the environmental data.

Images