Logistics order tracking method and device based on video transmission, equipment and storage medium

Abstract

The invention relates to the technical field of logistics, in particular to a logistics order tracking method, device and equipment based on video transmission and a storage medium. The logistics order tracking method based on video transmission comprises the following steps: acquiring real-time logistics image data, and preprocessing the real-time logistics image data to obtain preliminary analysis data; acquiring historical logistics image data, and training the basic model by using the historical logistics image data to obtain an image recognition and analysis model; analyzing the preliminary analysis data by adopting an image recognition and analysis model to obtain an image analysis result, and constructing logistics track information according to the image analysis result; and obtaining query request information, extracting an image analysis result and logistics track information according to the query request information, and generating a visual map according to the logistics track information. The real-time logistics image data and the image recognition and analysis model are utilized to obtain the image analysis result, and more accurate and detailed logistics track information can be formed.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics, and particularly to a logistics order tracking method, device, equipment and storage medium based on video transmission. Background Art

[0002] In the modern logistics industry, with the booming development of e-commerce, consumers' attention to logistics information has been increasing day by day. Traditional logistics order tracking mainly relies on text information, such as text records of nodes like order pickup, transportation, transfer, delivery, etc. Consumers can only query through the logistics order number on the e-commerce platform or the websites and APPs of logistics enterprises. These text descriptions are simple in content and lack an intuitive feeling for the actual state of the package during transportation. For example, information such as whether the package encounters bumps during transportation, whether it is properly placed, and the driving environment of the transport vehicle cannot be known. This limitation of information affects consumers' logistics experience to a certain extent and is also not conducive to the logistics enterprise's more refined management and supervision of the transportation process.

[0003] It can be seen that the existing technology still needs to be improved. Summary of the Invention

[0004] In view of the deficiencies of the above-mentioned existing technology, the purpose of the present invention is to provide a logistics order tracking method, device, equipment and storage medium based on video transmission, aiming to improve the fineness of logistics transportation tracking and the user experience during logistics query.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] The first aspect of the present invention provides a logistics order tracking method based on video transmission, including the following steps: obtaining real-time logistics image data, preprocessing the real-time logistics image data to obtain preliminary analysis data; obtaining historical logistics image data, using the deep learning algorithm as the basic model, and training the basic model with the historical logistics image data to obtain an image recognition and analysis model; using the image recognition and analysis model to analyze the preliminary analysis data to obtain an image analysis result, and constructing logistics trajectory information according to the image analysis result; obtaining a query request information, extracting the image analysis result and logistics trajectory information according to the query request information, and generating a visualization map according to the logistics trajectory information.

[0007] Optionally, in the first implementation manner of the first aspect of the present invention, the obtaining of real-time logistics image data and the preprocessing of the real-time logistics image data to obtain preliminary analysis data specifically include: obtaining real-time logistics image data, compressing the real-time logistics image data to obtain compressed data; denoising the compressed data to obtain denoised data, enhancing the denoised data to obtain image enhancement data; using an image feature extraction algorithm to extract features from the image enhancement data to obtain preliminary analysis data.

[0008] Optionally, in the second implementation manner of the first aspect of the present invention, the denoising of the compressed data to obtain denoised data and the enhancement of the denoised data to obtain image enhancement data specifically include: obtaining the compressed data, using spatial domain filtering to denoise the compressed data to obtain denoised data; using gamma correction to adjust the brightness and contrast of the denoised data to obtain preliminary enhanced data; using a color conversion matrix to perform color correction on the preliminary enhanced data to obtain image enhancement data.

[0009] Optionally, in the third implementation manner of the first aspect of the present invention, the obtaining of historical logistics image data, using a deep learning algorithm as a basic model, and training the basic model with the historical logistics image data to obtain an image recognition and analysis model specifically include: obtaining historical logistics image data, preprocessing the historical logistics image data to obtain preprocessed historical data; dividing the preprocessed historical data into a training set and a validation set, constructing a basic model with a convolutional neural network, training the basic model with the training set to obtain a preliminary model; evaluating the preliminary model with the validation set, and adjusting the preliminary model according to the evaluation results to obtain an image recognition and analysis model.

[0010] Optionally, in the fourth implementation manner of the first aspect of the present invention, the using of the image recognition and analysis model to analyze the preliminary analysis data to obtain an image analysis result and constructing logistics trajectory information according to the image analysis result specifically include: using the image recognition and analysis model to analyze the preliminary analysis data to obtain the real-time position information, attitude information, and logistics link information of the package in the image; summarizing and correlating the real-time position information, attitude information, and logistics link information of the package to obtain an image analysis result; extracting transmission position information from the real-time logistics image data, and constructing logistics trajectory information according to the transmission position information and the image analysis result.

[0011] Optionally, in the fifth implementation manner of the first aspect of the present invention, the extracting the transmission location information from the real-time logistics image data and constructing the logistics trajectory information according to the transmission location information and the image analysis result specifically include: extracting the transmission location information from the real-time logistics image data, and associating the transmission location information and the image analysis result with the same coordinates; using the transmission location information as the basic points of the logistics trajectory, sorting multiple pieces of transmission location information in chronological order to form a preliminary trajectory point sequence, and correcting the preliminary trajectory point sequence according to the image analysis result to generate a corrected trajectory; connecting the corrected trajectories to obtain a connected trajectory, and performing smoothing and optimization processing on the connected trajectory to obtain the logistics trajectory information.

[0012] Optionally, in the sixth implementation manner of the first aspect of the present invention, the obtaining the query request information, extracting the image analysis result and the logistics trajectory information according to the query request information, and generating a visualization map according to the logistics trajectory information specifically include: obtaining the query request information, and extracting the image analysis result and the logistics trajectory information according to the query request information; obtaining map information, projecting the trajectory points in the logistics trajectory information onto the map information to generate a preliminary map; generating detailed description information according to the image analysis result, and associating the detailed description information with the preliminary map to generate a visualization map.

[0013] The second aspect of the present invention provides a logistics order tracking device based on video transmission, including: a preprocessing module, configured to obtain real-time logistics image data and perform preprocessing on the real-time logistics image data to obtain preliminary analysis data; a construction module, configured to obtain historical logistics image data, use a deep learning algorithm as a basic model, and train the basic model with the historical logistics image data to obtain an image recognition and analysis model; an analysis module, configured to analyze the preliminary analysis data by using the image recognition and analysis model to obtain an image analysis result, and construct logistics trajectory information according to the image analysis result; a generation module, configured to obtain query request information, extract the image analysis result and the logistics trajectory information according to the query request information, and generate a visualization map according to the logistics trajectory information.

[0014] Optionally, in the first implementation manner of the second aspect of the present invention, the preprocessing module includes: a compression unit, configured to obtain real-time logistics image data and compress the real-time logistics image data to obtain compressed data; an enhancement unit, configured to perform noise reduction on the compressed data to obtain noise-reduced data, and perform image enhancement on the noise-reduced data to obtain image-enhanced data; a feature extraction unit, configured to extract features from the image-enhanced data by using an image feature extraction algorithm to obtain preliminary analysis data.

[0015] Optionally, in the second implementation manner of the second aspect of the present invention, the enhancement unit includes: a noise reduction subunit, configured to obtain compressed data and perform noise reduction on the compressed data by using spatial domain filtering to obtain noise-reduced data; an adjustment subunit, configured to adjust the brightness and contrast of the noise-reduced data by using gamma correction to obtain preliminarily enhanced data; and a correction subunit, configured to perform color correction on the preliminarily enhanced data by using a color conversion matrix to obtain image enhanced data.

[0016] Optionally, in the third implementation manner of the second aspect of the present invention, the construction module includes: a preprocessing unit, configured to obtain historical logistics image data and perform preprocessing on the historical logistics image data to obtain preprocessed historical data; a training unit, configured to divide the preprocessed historical data into a training set and a validation set, construct a basic model by using a convolutional neural network, and train the basic model by using the training set to obtain a preliminary model; and an optimization unit, configured to evaluate the preliminary model by using the validation set and adjust the preliminary model according to the evaluation result to obtain an image recognition and analysis model.

[0017] Optionally, in the fourth implementation manner of the second aspect of the present invention, the analysis module includes: an analysis unit, configured to analyze the preliminary analysis data by using the image recognition and analysis model to obtain real-time position information, attitude information, and logistics link information of the packages in the image; an association unit, configured to summarize and associate the real-time position information, attitude information, and logistics link information of the packages to obtain an image analysis result; and a trajectory construction unit, configured to extract transmission position information from the real-time logistics image data and construct logistics trajectory information according to the transmission position information and the image analysis result.

[0018] Optionally, in the fifth implementation manner of the second aspect of the present invention, the trajectory construction unit includes: an association subunit, configured to extract transmission position information from the real-time logistics image data and associate the transmission position information and the image analysis result with the same coordinates; a correction subunit, configured to use the transmission position information as a basic point of the logistics trajectory, sort multiple pieces of transmission position information in chronological order to form a preliminary trajectory point sequence, and correct the preliminary trajectory point sequence according to the image analysis result to generate a corrected trajectory; and an optimization subunit, configured to connect the corrected trajectory to obtain a connected trajectory, and perform smoothing and optimization processing on the connected trajectory to obtain logistics trajectory information.

[0019] Optionally, in the sixth implementation manner of the second aspect of the present invention, the generation module includes: an information extraction unit, configured to obtain query request information and extract an image analysis result and logistics track information according to the query request information; a projection unit, configured to obtain map information and project track points in the logistics track information onto the map information to generate a preliminary map; a generation unit, configured to generate detailed description information according to the image analysis result, associate the detailed description information with the preliminary map, and generate a visualization map.

[0020] The third aspect of the present invention provides a logistics order tracking device based on video transmission, including a memory and at least one processor, where computer-readable instructions are stored in the memory; the at least one processor calls the computer-readable instructions in the memory to execute each step of the above-mentioned logistics order tracking method based on video transmission.

[0021] The fourth aspect of the present invention provides a computer-readable storage medium, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by a processor, each step of the above-mentioned logistics order tracking method based on video transmission is implemented.

[0022] Beneficial effects: The present invention provides a logistics order tracking method based on video transmission. The logistics order tracking method based on video transmission first obtains real-time logistics image data, preprocesses the real-time logistics image data to obtain preliminary analysis data, so as to provide clear image data as supplementary materials for logistics tracking; then obtains historical logistics image data, uses a deep learning algorithm as a basic model to obtain an image recognition and analysis model, and uses the image recognition and analysis model to analyze the preliminary analysis data, identify the appearance characteristics, location environment characteristics, and logistics operation behavior characteristics of the package, etc., to obtain an image analysis result, so as to obtain real-time information of the package from the image data, and then constructs logistics track information according to the image analysis result, so that users can query the transportation progress; finally, obtains query request information, extracts the image analysis result and logistics track information according to the query request information, generates a visualization map according to the logistics track information, improves the accuracy and experience of the visualization map, enables users to view the transportation track and relevant images of each node at the same time, and increases user satisfaction. Description of the Drawings

[0023] Figure 1 It is the first flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0024] Figure 2 It is the second flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0025] Figure 3 The third flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0026] Figure 4 The fourth flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0027] Figure 5 The fifth flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0028] Figure 6 The sixth flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0029] Figure 7 The seventh flowchart of the logistics order tracking method based on video transmission provided by the embodiment of the present invention;

[0030] Figure 8 A structural schematic diagram of the logistics order tracking device based on video transmission provided by the embodiment of the present invention;

[0031] Figure 9 Another structural schematic diagram of the logistics order tracking device based on video transmission provided by the embodiment of the present invention;

[0032] Figure 10 A structural schematic diagram of the logistics order tracking device based on video transmission provided by the embodiment of the present invention. Detailed implementation manners

[0033] The present invention provides a logistics order tracking method, device, equipment and storage medium based on video transmission. The present invention first obtains real-time logistics image data, preprocesses the real-time logistics image data to obtain preliminary analysis data, so as to facilitate subsequent analysis using a model and improve the accuracy of model analysis; then trains a basic model based on a deep learning algorithm using historical logistics image data to obtain an image recognition and analysis model, which can automatically and quickly analyze complex logistics image data; then inputs the previously obtained preliminary analysis data into the image recognition and analysis model to obtain a reliable and detailed image analysis result, and constructs logistics trajectory information according to the image analysis result; finally, obtains the query request information of the user, extracts the image analysis result and logistics trajectory information according to the query request information, generates a visualization map according to the logistics trajectory information, so that the query requester can intuitively see the transportation situation of the package, and thus can manage the logistics transportation more precisely.

[0034] It should be noted that the following data collection of the user equipment has obtained the prior permission of the user.

[0035] For ease of understanding, the specific process of the embodiments of the present invention will be described below. Please refer to Figure 1 , the first embodiment of the logistics order tracking method based on video transmission in the embodiments of the present invention includes:

[0036] S101. Obtain real-time logistics image data, and preprocess the real-time logistics image data to obtain preliminary analysis data;

[0037] For the acquisition of real-time logistics image data, high-definition cameras and intelligent image acquisition systems can be deployed in logistics transport vehicles, warehouses, and key logistics nodes (such as sorting centers, transfer stations, etc.). The cameras have panoramic shooting, intelligent focusing, and image stabilization functions, and can capture image information of packages in all logistics links in all directions and with high clarity. By preprocessing the real-time logistics image data, the data transmission volume can be reduced and valuable image feature information can be extracted.

[0038] Specifically, when the package corresponding to the waybill number is scanned or registered at a logistics node, its registration time and location can be matched with the images captured by the corresponding camera, so as to obtain the real-time logistics image data associated with the waybill number.

[0039] S102. Obtain historical logistics image data, use the deep learning algorithm as the basic model, and train the basic model with the historical logistics image data to obtain an image recognition and analysis model;

[0040] By training and constructing an image recognition and analysis model with historical logistics image data, the model can be made to have a more accurate analysis ability.

[0041] S103. Analyze the preliminary analysis data with the image recognition and analysis model to obtain an image analysis result, and construct logistics track information according to the image analysis result;

[0042] Analyzing with the image recognition and analysis model can effectively handle a large amount of data. By constructing logistics track information according to the image analysis result, not only can the logistics track be more accurate, but also users can see the relevant package image information when querying, realizing the visualization of logistics transport tracking.

[0043] S104. Obtain query request information, extract the image analysis result and logistics track information according to the query request information, and generate a visualization map according to the logistics track information.

[0044] Generate and output a visual map according to requirements, which allows users to intuitively see the location of logistics parcels and the transportation progress, and can also view the images of parcels during transportation in the visual map to understand the transportation quality of parcels, which helps to improve user satisfaction and also helps to resolve subsequent transportation disputes.

[0045] The present invention effectively overcomes the limitations of traditional logistics tracking information. Without installing video equipment on the parcels, it uses advanced algorithms and models to greatly improve the visualization and accuracy of logistics information, which not only satisfies consumers' right to know about logistics details but also helps logistics enterprises optimize operation management and service quality improvement.

[0046] Please refer to Figure 2 , the second embodiment of the logistics order tracking method based on video transmission in the embodiments of the present invention includes:

[0047] S201. Obtain real-time logistics image data, and compress the real-time logistics image data to obtain compressed data;

[0048] Specifically, a lossless compression algorithm can be used to compress the real-time logistics image data. When extracting image features subsequently, higher image quality can improve data quality. The lossless compression algorithm reduces the data volume by re-encoding the image data and other means while ensuring that the image can be fully restored.

[0049] S202. Denoise the compressed data to obtain denoised data, and enhance the denoised data to obtain image-enhanced data;

[0050] In the denoising stage, spatial domain denoising or frequency domain denoising methods can be used to achieve this. Among them, for the frequency domain denoising method, specifically, a Fourier transform can be used to perform a two-dimensional Fourier transform on the image to convert the image from the spatial domain to the frequency domain. The fast Fourier transform (FFT) algorithm can be used to improve the calculation efficiency. In the frequency domain, noise usually appears as high-frequency parts. By setting a suitable threshold, the low-frequency parts below the threshold are retained, and the high-frequency parts (usually containing noise) above the threshold are attenuated or set to zero, and then an inverse Fourier transform is performed to convert the image from the frequency domain back to the spatial domain to obtain the denoised image.

[0051] S203. Use an image feature extraction algorithm to extract features from the image-enhanced data to obtain preliminary analysis data.

[0052] The features of the image can include shape features (used to identify whether the shape of the logistics parcel is regular, whether there is damage or deformation, etc.), color features (check whether the color of the logistics label is correct, whether the color of the parcel meets the expectations), position features of objects (such as parcels, transportation tools), etc.

[0053] For shape feature extraction, edge detection algorithms such as Canny edge detection can be used. It determines the edges of objects by finding places in the image where the pixel intensity changes drastically.

[0054] For the position features of objects, object detection algorithms can be used. For example, using deep learning-based object detection algorithms (such as Faster R-CNN), after these algorithms are trained on a large amount of labeled image data, they can identify the position of the object in the image (usually represented in the form of a bounding box), and the coordinate information of the bounding box (such as the coordinates of the upper left corner and the lower right corner) is the position feature of the object.

[0055] Please refer to Figure 3 , the third embodiment of the logistics order tracking method based on video transmission in the embodiments of the present invention includes:

[0056] S301. Obtain compressed data, and perform noise reduction on the compressed data using spatial domain filtering to obtain noise-reduced data;

[0057] Specifically, mean filtering can be used to perform noise reduction on the compressed data. During implementation, first determine the size of the filter kernel, such as 3x3, 5x5, etc. The size of the filter kernel determines the range of pixels participating in the operation around each pixel in the image. For each pixel in the image, calculate the average value of the pixels within the area covered by the filter kernel with it as the center, and replace the value of the central pixel with this average value. For example, for a 3x3 filter kernel, add the values of the 8 pixels around the central pixel and its own value and then divide by 9, and use the resulting value as the new central pixel value.

[0058] S302. Perform gamma correction on the brightness and contrast of the noise-reduced data to obtain preliminary enhanced data;

[0059] Specifically, first select an appropriate gamma value according to the characteristics of the image and the desired enhancement effect, then construct a gamma correction curve, and by traversing each pixel in the noise-reduced data, convert its brightness value according to the gamma correction curve to obtain the adjusted pixel value. For a color image, the same gamma correction can be applied to the red, green, and blue channels respectively, or different gamma values can be used for each channel according to needs for adjustment to achieve more flexible color and brightness control.

[0060] S303. Use a color conversion matrix to perform color correction on the preliminary enhanced data to obtain image enhanced data. Select an appropriate color conversion matrix according to the type of color deviation to be corrected. Commonly used ones include matrices for correcting white balance (such as making white objects appear truly white under different lighting conditions) and matrices for adjusting color saturation. For example, for white balance correction, in the case where it is known that the image has a color cast under a certain light source (such as the image being yellowish under tungsten light), a matrix that corrects the yellowish color towards white can be obtained through measurement or calculation. By applying the color conversion matrix to each pixel in the preliminary enhanced data, the color correction of the image is achieved, and finally, image enhanced data is obtained, which is optimized in terms of brightness, contrast, and color and is more suitable for subsequent analysis or processing tasks.

[0061] Please refer to Figure 4 , the fourth embodiment of the logistics order tracking method based on video transmission in the embodiments of the present invention includes:

[0062] S401. Obtain historical logistics image data, perform preprocessing on the historical logistics image data to obtain preprocessed historical data;

[0063] The preprocessing of the historical logistics image data may include steps such as image format unification and inspection, image cropping, correction, and noise reduction to improve data quality.

[0064] S402. Divide the preprocessed historical data into a training set and a validation set, construct a basic model with a convolutional neural network, and use the training set to train the basic model to obtain a preliminary model;

[0065] Specifically, 70% of the preprocessed historical data can be used for the training set and 30% for the validation set; or 80% for the training set and 20% for the validation set, etc. It can be adjusted according to the actual data volume.

[0066] The construction process of the Convolutional Neural Network (CNN) model is as follows:

[0067] Input layer:

[0068] Receive the preprocessed historical data, and assume the image size is 224×224×3, where 3 represents the three RGB color channels;

[0069] Convolutional layer 1:

[0070] Use 64 convolutional kernels with a size of 3×3, a stride of 1, and a padding of 1;

[0071] The output feature map size is 224×224×64;

[0072] Calculation formula:

[0073]

[0074] Among them, Output i，j，k is the output value of the output feature map at position (i, j) for the k-th convolutional kernel, W m，n，p，k is the weight of the k-th convolutional kernel at position (m, n, p), Input i+m，j+n，p is the pixel value of the input image at the corresponding position, b k is the bias of the k-th convolutional kernel;

[0075] Then followed by a ReLU activation function: ReLU(x) = max(0, x), which performs a non-linear transformation on the convolution result;

[0076] Then perform a max pooling operation with a pooling kernel size of 2×2 and a stride of 2, and the size of the output feature map becomes 112×112×64;

[0077] Convolutional layer 2

[0078] Use 128 convolutional kernels with a size of 3×3, a stride of 1, and a padding of 1;

[0079] The size of the output feature map is 112×112×128;

[0080] The calculation method is similar to that of convolutional layer 1. After passing through the ReLU activation function and max pooling (pooling kernel size of 2×2, stride of 2), the size of the output feature map becomes 56×56×128.

[0081] S403. Use the validation set to evaluate the preliminary model, and adjust the preliminary model according to the evaluation results to obtain an image recognition and analysis model.

[0082] According to the evaluation results of the validation set, such as indicators like accuracy, recall rate, F1 value, etc., the advantages and disadvantages of the model can be identified, and thus the hyperparameters of the model can be adjusted accordingly. For example, if the accuracy on the validation set is low, parameters such as the learning rate, the number of convolutional kernels, and the number of network layers may need to be adjusted, and then retrained and evaluated to find the optimal parameter combination to improve the model performance.

[0083] Please refer to Figure 5 , the fifth embodiment of the logistics order tracking method based on video transmission in the embodiments of the present invention includes:

[0084] S501. Use the image recognition and analysis model to analyze the preliminary analysis data to obtain the real-time position information, pose information, and logistics link information of the packages in the image;

[0085] Analyzing the preliminary analysis data through an image recognition and analysis model can improve the data acquisition efficiency, enabling managers to quickly obtain the real-time location information, attitude information, and logistics link information of the packages. These information can be fed back in real time, thereby enabling precise control of the logistics links.

[0086] S502. Summarize and correlate the real-time location information, attitude information, and logistics link information of the package to obtain an image analysis result;

[0087] Specifically, each piece of information can be correlated based on the logistics order number. Through the logistics order number, it can be ensured that the various pieces of information of each package are accurately corresponding, and there will be no situation of information confusion. For example, when querying the information of a certain package, the location, attitude, and logistics link of the package at a specific time can be quickly obtained through the logistics order number.

[0088] S503. Extract the transmission location information from the real-time logistics image data, and construct logistics trajectory information based on the transmission location information and the image analysis result.

[0089] By combining the transmission location information and the image analysis result to construct the logistics trajectory information, the positioning of the package can be made more accurate, and the image analysis result can play a role in position correction during the construction process. In addition, the respective data in the image analysis result can be used as supplementary content for the logistics trajectory information, such as the image of the package at the logistics node, the status of the package, etc.

[0090] Please refer to Figure 6 , the sixth embodiment of the logistics order tracking method based on video transmission in the embodiment of the present invention includes:

[0091] S601. Extract the transmission location information from the real-time logistics image data, and correlate the transmission location information and the image analysis result with the same coordinates;

[0092] S602. Use the transmission location information as the basic point of the logistics trajectory, sort multiple pieces of transmission location information in chronological order to form a preliminary trajectory point sequence, and correct the preliminary trajectory point sequence according to the image analysis result to generate a corrected trajectory;

[0093] The preliminary trajectory point sequence constructed based on the transmitted location information may deviate due to various factors (such as signal interference, equipment failures, etc.). However, through the results of image analysis, such as the information on the actual placement location of goods and vehicle numbers captured by the surveillance cameras in the warehouse, the preliminary trajectory point sequence can be corrected to make the logistics trajectory more in line with the actual situation, thereby improving the accuracy of the trajectory. Image analysis can detect abnormal situations during the transportation of goods, such as whether the goods are damaged or illegally opened. While correcting the trajectory, the location and time of these abnormalities can also be recorded. For example, for the transportation of some high-value electronic products, by installing cameras in the transportation vehicle, when an image of the torn packaging of the goods is found, combined with the corrected trajectory points, the location where the event occurred can be accurately located.

[0094] S603. Connect the corrected trajectory to obtain a connected trajectory, and perform smoothing and optimization processing on the connected trajectory to obtain logistics trajectory information. During the actual logistics transportation process, due to reasons such as signal fluctuations and equipment precision, the location information may have some small jitters or abnormal points. Through smoothing and optimization processing, these factors that may interfere with the judgment can be removed. The optimization processing not only includes removing abnormal points, but also can adjust the trajectory according to factors such as the characteristics of the transportation mode and traffic rules. For example, for road transportation, the trajectory should conform to the road direction and traffic flow.

[0095] Please refer to Figure 7 , the seventh embodiment of the logistics order tracking method based on video transmission in the embodiments of the present invention includes:

[0096] S701. Obtain query request information, and extract the image analysis result and logistics trajectory information according to the query request information;

[0097] According to the logistics order number in the query request information, the real-time image analysis result and logistics trajectory information can be obtained.

[0098] S702. Obtain map information, project the trajectory points in the logistics trajectory information onto the map information to generate a preliminary map;

[0099] The map data source can be determined as needed, such as a commercial map service or an open-source map project. A geographic information system (GIS) software or a related map programming interface (API) can be used to perform the projection operation. The coordinates of each logistics trajectory point are converted into the coordinate system of the map according to the selected projection method.

[0100] Visualize the projected trajectory points on the map. According to the chronological order of the trajectory points, they can be connected by lines to show the dynamic process of the logistics trajectory. At the same time, different visual attributes such as colors and thicknesses can be set for the trajectory points and the connecting lines to highlight different information. For example, use red to indicate that the goods are in transit, green to indicate that the goods have reached the transfer station, and blue to indicate that the goods have been finally delivered; the thickness of the connecting line can indicate the speed of transportation, etc.

[0101] S703. Generate detailed description information based on the image analysis result, associate the detailed description information with the preliminary map, and generate a visualized map.

[0102] The detailed description information generated from the image analysis result can include various aspects such as the status of the goods, the transportation environment, the vehicle situation, and image screenshots. After associating this detailed description information with the preliminary map, on the visualized map, users can not only see the transportation trajectory of the goods but also understand the specific situation of the goods and the transportation environment at each trajectory point, providing an all-round logistics view.

[0103] The above described the logistics order tracking method based on video transmission in the embodiments of the present invention. Next, the logistics order tracking device based on video transmission in the embodiments of the present invention will be described. Please refer to Figure 8 One embodiment of the logistics order tracking device based on video transmission in the embodiments of the present invention includes:

[0104] A preprocessing module 10, configured to obtain real-time logistics image data and preprocess the real-time logistics image data to obtain preliminary analysis data;

[0105] A construction module 20, configured to obtain historical logistics image data, use a deep learning algorithm as a basic model, and train the basic model with the historical logistics image data to obtain an image recognition and analysis model;

[0106] An analysis module 30, configured to analyze the preliminary analysis data by using the image recognition and analysis model to obtain an image analysis result, and construct logistics trajectory information according to the image analysis result;

[0107] A generation module 40, configured to obtain a query request information, extract the image analysis result and the logistics trajectory information according to the query request information, and generate a visualized map according to the logistics trajectory information.

[0108] Please refer to Figure 9 One embodiment of the logistics order tracking device based on video transmission in the embodiments of the present invention includes:

[0109] A preprocessing module 10, configured to obtain real-time logistics image data and preprocess the real-time logistics image data to obtain preliminary analysis data;

[0110] The building block 20 is used to obtain historical logistics image data, use a deep learning algorithm as the basic model, and train the basic model with the historical logistics image data to obtain an image recognition and analysis model;

[0111] The analysis module 30 is used to analyze the preliminary analysis data by using the image recognition and analysis model to obtain an image analysis result, and construct logistics track information according to the image analysis result;

[0112] The generation module 40 is used to obtain query request information, extract the image analysis result and logistics track information according to the query request information, and generate a visualization map according to the logistics track information;

[0113] In this embodiment, the preprocessing module 10 includes:

[0114] The compression unit 11 is used to obtain real-time logistics image data, compress the real-time logistics image data to obtain compressed data;

[0115] The enhancement unit 12 is used to denoise the compressed data to obtain denoised data, and enhance the denoised data to obtain image enhancement data;

[0116] The feature extraction unit 13 is used to extract features from the image enhancement data by using an image feature extraction algorithm to obtain preliminary analysis data;

[0117] In this embodiment, the enhancement unit 12 includes:

[0118] The denoising subunit 121 is used to obtain compressed data, and denoise the compressed data by using spatial domain filtering to obtain denoised data;

[0119] The adjustment subunit 122 is used to adjust the brightness and contrast of the denoised data by using gamma correction to obtain preliminary enhanced data;

[0120] The correction subunit 123 is used to perform color correction on the preliminary enhanced data by using a color conversion matrix to obtain image enhancement data;

[0121] In this embodiment, the building block 20 includes:

[0122] The preprocessing unit 21 is used to obtain historical logistics image data, preprocess the historical logistics image data to obtain preprocessed historical data;

[0123] The training unit 22 is used to divide the preprocessed historical data into a training set and a validation set, construct a basic model with a convolutional neural network, and train the basic model with the training set to obtain a preliminary model;

[0124] An optimization unit 23, which is used to evaluate the preliminary model using a validation set and adjust the preliminary model according to the evaluation results to obtain an image recognition and analysis model;

[0125] In this embodiment, the analysis module 30 includes:

[0126] An analysis unit 31, which is used to analyze the preliminary analysis data using the image recognition and analysis model to obtain the real-time position information, attitude information, and logistics link information of the package in the image;

[0127] An association unit 32, which is used to summarize and associate the real-time position information, attitude information, and logistics link information of the package to obtain an image analysis result;

[0128] A trajectory construction unit 33, which is used to extract transmission position information from the real-time logistics image data and construct logistics trajectory information according to the transmission position information and the image analysis result;

[0129] In this embodiment, the trajectory construction unit 33 includes:

[0130] An association sub-unit 331, which is used to extract transmission position information from the real-time logistics image data and associate the transmission position information and the image analysis result with the same coordinates;

[0131] A correction sub-unit 332, which is used to use the transmission position information as the basic point of the logistics trajectory, sort multiple transmission position information in chronological order to form a preliminary trajectory point sequence, and correct the preliminary trajectory point sequence according to the image analysis result to generate a corrected trajectory;

[0132] An optimization sub-unit 333, which is used to connect the corrected trajectory to obtain a connected trajectory, and perform smoothing and optimization processing on the connected trajectory to obtain logistics trajectory information;

[0133] In this embodiment, the generation module 40 includes:

[0134] An information extraction unit 41, which is used to obtain query request information and extract the image analysis result and logistics trajectory information according to the query request information;

[0135] A projection unit 42, which is used to obtain map information and project the trajectory points in the logistics trajectory information onto the map information to generate a preliminary map;

[0136] A generation unit 43, which is used to generate detailed description information according to the image analysis result, associate the detailed description information with the preliminary map, and generate a visual map.

[0137] The logistics order tracking device based on video transmission of the present invention obtains real-time logistics image data through high-definition cameras and intelligent image acquisition systems pre-deployed in logistics transport vehicles, warehouses, and key logistics nodes (such as sorting centers, transfer stations, etc.), and uses historical logistics image data to construct an image recognition and analysis model based on deep learning in advance. When a user enters a logistics order number through the official website of a logistics enterprise, a mobile APP, or other authorized query platforms to initiate a query request. The present invention can extract the image analysis results and logistics track information of the corresponding order generated by the image recognition and analysis model according to the query request. It effectively overcomes the limitations of traditional logistics tracking information. Without the need to install video devices on packages, it greatly improves the visualization and accuracy of logistics information by using advanced algorithms and models, which not only satisfies consumers' right to know about logistics details but also helps logistics enterprises optimize operation management and service quality improvement.

[0138] The above is a detailed description of the logistics order tracking device based on video transmission in the embodiments of the present invention from the perspective of modular functional entities. The following is a detailed description of the logistics order tracking device based on video transmission in the embodiments of the present invention from the perspective of hardware processing.

[0139] Figure 10 FIG. is a schematic structural diagram of a logistics order tracking device based on video transmission provided by an embodiment of the present invention. The logistics order tracking device 900 based on video transmission may vary greatly due to configuration or performance differences, and may include one or more processors (central processing units, CPU) 910 (for example, one or more processors) and a memory 920, and one or more storage media 930 (for example, one or more mass storage devices) for storing application programs 933 or data 932. Among them, the memory 920 and the storage media 930 can be transient storage or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the logistics order tracking device 900 based on video transmission. Further, the processor 910 may be configured to communicate with the storage media 930 and execute a series of instruction operations in the storage media 930 on the logistics order tracking device 900 to implement the steps of the logistics order tracking method based on video transmission provided in the above method embodiments.

[0140] The logistics order tracking device 900 based on video transmission may further include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that Figure 10 The illustrated structure of the logistics order tracking device based on video transmission does not constitute a limitation on the logistics order tracking device based on video transmission, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0141] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are run on a computer, the computer is caused to execute the steps of the logistics order tracking method based on video transmission.

[0142] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices or apparatuses can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0143] If the integrated unit is implemented in the form of 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, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0144] It can be understood that for those of ordinary skill in the art, equivalent substitutions or changes can be made according to the technical solution of the present invention and its inventive concept, and all such changes or substitutions should fall within the protection scope of the appended claims of the present invention.

Claims

1. A logistics order tracking method based on video transmission, characterized in that: The steps include: Acquire real-time logistics image data, and pre-process the real-time logistics image data to obtain preliminary analysis data; Obtain historical logistics image data, use the deep learning algorithm as the basic model, and use the historical logistics image data to train the basic model to obtain an image recognition and analysis model; Analyzing the preliminary analysis data using an image recognition and analysis model to obtain image analysis results, and constructing logistics trajectory information based on the image analysis results; Obtain query request information, extract image analysis results and logistics track information according to the query request information, and generate a visual map according to the logistics track information.

2. The method for tracking logistics orders based on video transmission according to claim 1, characterized in that: The real-time logistics image data is acquired and pre-processed to obtain preliminary analysis data, specifically including: Acquire real-time logistics image data, and compress the real-time logistics image data to obtain compressed data; Denoising the compressed data to obtain denoised data, and performing image enhancement on the denoised data to obtain image enhanced data; The image feature extraction algorithm is used to extract features from the image enhancement data to obtain preliminary analysis data.

3. The method for tracking logistics orders based on video transmission according to claim 2, characterized in that: The denoising of the compressed data to obtain denoised data, and image enhancement of the denoised data to obtain image enhanced data specifically include: Obtain compressed data, and use spatial domain filtering to reduce noise on the compressed data to obtain reduced noise data; Gamma correction is used to adjust the brightness and contrast of the noise-reduced data to obtain preliminary enhanced data; The color conversion matrix is ​​used to perform color correction on the preliminary enhanced data to obtain image enhanced data.

4. The method for tracking logistics orders based on video transmission according to claim 1, characterized in that: The acquisition of historical logistics image data, taking the deep learning algorithm as the basic model, uses the historical logistics image data to train the basic model to obtain the image recognition and analysis model, specifically includes: Acquire historical logistics image data, preprocess the historical logistics image data, and obtain preprocessed historical data; The preprocessed historical data is divided into a training set and a validation set, a basic model is constructed using a convolutional neural network, and the basic model is trained using the training set to obtain a preliminary model; The validation set is used to evaluate the preliminary model, and the preliminary model is adjusted according to the evaluation results to obtain the image recognition and analysis model.

5. The method for tracking logistics orders based on video transmission according to claim 1, characterized in that: The image recognition and analysis model is used to analyze the preliminary analysis data to obtain image analysis results, and logistics trajectory information is constructed according to the image analysis results, specifically including: Analyzing the preliminary analysis data using an image recognition and analysis model to obtain real-time location information, posture information, and logistics link information of the package in the image; Summarize and correlate the real-time location information, posture information, and logistics link information of the package to obtain image analysis results; The transmission location information is extracted from the real-time logistics image data, and the logistics trajectory information is constructed based on the transmission location information and image analysis results.

6. The method for tracking logistics orders based on video transmission according to claim 5, characterized in that: The extracting of transmission location information from real-time logistics image data and constructing logistics track information according to the transmission location information and image analysis results specifically include: Extract transmission location information from real-time logistics image data, and associate the transmission location information and image analysis results of the same coordinates; Taking the transmission position information as the basic point of the logistics trajectory, multiple transmission position information are sorted in chronological order to form a preliminary trajectory point sequence, and the preliminary trajectory point sequence is corrected according to the image analysis results to generate a corrected trajectory; The corrected trajectories are connected to obtain the connection trajectory, and the connection trajectory is smoothed and optimized to obtain the logistics trajectory information.

7. The method for tracking logistics orders based on video transmission according to claim 1, characterized in that: The obtaining of query request information, extracting image analysis results and logistics track information according to the query request information, and generating a visual map according to the logistics track information specifically includes: Obtain query request information, and extract image analysis results and logistics track information according to the query request information; Obtaining map information, and projecting the track points in the logistics track information onto the map information to generate a preliminary map; Generate detailed description information based on the image analysis results, associate the detailed description information with the preliminary map, and generate a visual map.

8. A logistics order tracking device based on video transmission, characterized in that: include: A preprocessing module is used to obtain real-time logistics image data and preprocess the real-time logistics image data to obtain preliminary analysis data; A construction module is used to obtain historical logistics image data, and use the deep learning algorithm as the basic model to train the basic model using historical logistics image data to obtain an image recognition and analysis model; An analysis module, used to analyze the preliminary analysis data using an image recognition and analysis model to obtain image analysis results, and construct logistics trajectory information according to the image analysis results; The generation module is used to obtain query request information, extract image analysis results and logistics trajectory information according to the query request information, and generate a visual map according to the logistics trajectory information.

9. A logistics order tracking device based on video transmission, characterized in that: comprising a memory and at least one processor, wherein the memory has computer-readable instructions stored therein; The at least one processor calls the computer-readable instructions in the memory to execute the various steps of the logistics order tracking method based on video transmission as described in any one of claims 1-7.

10. A computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the logistics order tracking method based on video transmission as described in any one of claims 1 to 7 are implemented.

Images