Information identification method based on express sheet

Abstract

The invention discloses an express waybill information identification method, which comprises the following steps of: recording an express box on a conveying belt by adopting high-definition photographing equipment to obtain picture information of the express box; when express items are sorted or put in storage, bar codes on the express items are generally detected, names and address information of addressees are printed on the bar codes, so that elements forming basic information of the express items are Chinese characters, English and numbers, and the bar codes on pictures may be in an inclined state during photographing; therefore, the method also introduces the correction of the bar code, enables the fonts in the bar code to be clearer, employs an image processing technology and a neural network model to carry out the accurate recognition of the express sheet information, carries out the retrieval of the recognized express sheet information, and judges whether the express sheet information accords with the specification or not. The requirements of workers are met, the workers do not need to check and check one by one, convenience and rapidness are achieved, and the express delivery speed is greatly increased.

Description

Technical Field

[0001] The present invention relates to the technical field of express delivery bill information recognition, and in particular to an express delivery bill information recognition method. Background Art

[0002] The rapid development of e-commerce has led to an explosive growth in the volume of express delivery. As a direct information carrier of express delivery business, the express delivery bill is not only a commercial voucher to prove the express delivery business, but also a source of information for express delivery business personnel to contact buyers and sellers. The rational and efficient use of express delivery bill information is of great significance and application value to optimize the workflow of express delivery companies and improve their work efficiency.

[0003] The current sorting speed has been improved from the traditional method of manually scanning express orders with a scanner to a method that can automatically scan by setting up a camera and performing image analysis to obtain the barcode and destination. The image analysis and processing is performed using traditional digital image and pattern recognition methods. Although the image recognition function is realized, the image recognition efficiency is low and the error rate is high, which seriously affects the speed and efficiency of sorting. Summary of the invention

[0004] In order to solve the above technical problems, a method for identifying express delivery bill information is provided. This technical solution solves the current sorting speed mentioned in the above background technology. It has improved the traditional method of manually scanning express delivery bills with a scanner to a method that can automatically scan by setting up a camera and performing image analysis to obtain the barcode and destination. The traditional digital image and pattern recognition methods are used to analyze and process the image. Although the image recognition function is realized, the image recognition efficiency is low and the error rate is high, which seriously affects the speed and efficiency of sorting.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is: A method for identifying express delivery bill information, comprising: Use high-definition camera equipment to record the express boxes on the conveyor belt to obtain the picture information of the express boxes; Perform denoising on the express box image to obtain a pre-processed express box image; Identify the barcode area in the express box image, perform rectangular positioning and tilt angle correction on it, and obtain the pre-processed barcode; Acquire Chinese character feature information, English letter feature information and number feature information from the database, and establish an express delivery bill information recognition model based on the Chinese character feature information, English letter feature information and number feature information; Perform feature extraction on the preprocessed barcode information to obtain barcode feature information; Input the obtained barcode feature information into the express delivery bill information recognition model, and output the express delivery bill information; Search the output express delivery bill information to determine whether the express delivery bill information complies with the specifications.

[0006] Preferably, the express boxes on the conveyor belt are recorded using a high-definition camera to obtain the express box image information, which specifically includes the following steps: Use sensors to detect whether a courier box passes under the high-definition camera; If there is, high-definition camera equipment is activated to record the passing express box; If not, the high-definition camera device will not start.

[0007] Furthermore, in the present solution, sensors are used to control the startup and shutdown of the high-definition camera device, so that the high-definition camera device is not always in working state, which has the effect of saving power and avoiding the situation where the device is prone to malfunction due to long-term operation.

[0008] Preferably, the denoising process of the express box image to obtain the pre-processed express box image specifically comprises the following steps: Decompose the high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.

[0009] Further, in this scheme, the high-frequency information and low-frequency information of the image are processed separately through wavelet transform, and the low-frequency component is restored using the LPG-PCA method. In the process of restoring the high-frequency component, the natural image is sparsely represented in the group domain, and the inherent local sparsity and non-local self-similarity of the natural image are clearly and effectively characterized in a unified manner, and the high-frequency part is extracted from the image without being affected by noise. While effectively removing noise, the image details are retained, high sparsity and high restoration quality are achieved, and the visual performance of the image as well as the peak signal-to-noise ratio and structural similarity are improved.

[0010] Preferably, the identifying of the barcode area in the express box image, performing rectangular positioning and tilt angle correction on it, and obtaining the pre-processed barcode specifically comprises the following steps: Perform edge detection on the image in the horizontal and vertical directions, and sum the detection results in the horizontal and vertical directions; The barcode part of the image is processed based on the first morphological closing operation to obtain the connected area, and then the single line interference on the image is removed by the second morphological closing operation; Filling holes in the image processed by the first morphological closing operation; Calculate the minimum bounding rectangle according to each white area in the image after filling the holes, and obtain information of each bounding rectangle, wherein the rectangle information includes coordinates of the four vertices of the rectangle, an inclination angle, and the number of white pixels in the rectangular area; According to the calculated rectangular tilt angles, their distribution within the right angle range is counted, and the angles with more concentrated distribution are used to perform preliminary tilt angle correction.

[0011] Preferably, the calculation formula of the minimum enclosing rectangle is: Smin = min (S1, S2, . . . Sn) In the formula, Smin is the minimum enclosing rectangle; S1, S2, . . . Sn are the areas of several groups of enclosing rectangles.

[0012] Preferably, the step of obtaining the preprocessed barcode further comprises the following steps: In the image after the initial tilt angle correction, the barcode is located in the 0 degree direction or the 90 degree direction, and the error is within the range of plus or minus 10 degrees; The candidate rectangles of the barcode are screened at 0 degree and 90 degree directions respectively to determine the final position and final angle of the barcode; Perform a final rotation to obtain the final rectified barcode image.

[0013] Preferably, the step of extracting features from the preprocessed barcode information to obtain barcode feature information specifically comprises the following steps: Perform feature extraction on the final corrected barcode image to obtain Chinese character feature information, English letter feature information and digital feature information; Based on the acquired Chinese character feature information, English letter feature information and number feature information, the barcode feature information is obtained.

[0014] Preferably, the express delivery bill information recognition model based on Chinese character feature information, English letter feature information and digital feature information specifically includes the following steps: Based on a feedforward neural network, a courier bill information recognition model is constructed, wherein the input data of the courier bill information recognition model includes the acquired barcode feature information; Based on machine learning, the plurality of sample Chinese character feature information, the plurality of sample English letter feature information, and the plurality of sample number feature information are labeled and divided to obtain a training set, a validation set, and a test set; The training set, validation set and test set are used to perform supervised training, validation and testing on the license plate recognition model to obtain the express delivery bill information recognition model whose accuracy meets the preset accuracy requirements.

[0015] Further in this scheme, the preferred ratio of this application is 8:1:1, wherein the training set is used to train the model, that is, to determine the model's weights and biases and other learning parameters; the validation set is used to verify each model using the validation set data after multiple models are trained in the training set, and to record the model accuracy, and then select the best model and its corresponding parameters; the test set is used only once, that is, when evaluating the final model after the training is completed, and it does not participate in the process of learning parameters or the process of hyperparameter selection, but is only used for model evaluation. Based on machine learning, the model is supervised and trained through the training set, the obtained model is verified through the validation set, and the trained model is evaluated through the test set, and the hyperparameters are continuously adjusted until the accuracy meets the preset requirements, and the express delivery bill information recognition model is output.

[0016] Preferably, the step of inputting the obtained barcode feature information into the express delivery bill information recognition model and outputting the express delivery bill information specifically comprises the following steps: Input the obtained Chinese character feature information into the express delivery bill information recognition model that meets the preset accuracy requirement to obtain Chinese character information; Input the obtained English letter feature information into the express delivery bill information recognition model that meets the preset accuracy requirement to obtain the English letter information; Input the obtained digital feature information into the express delivery bill information recognition model that meets the preset accuracy requirements to obtain digital information; Output express delivery label information based on Chinese character information, English letter information and digital information.

[0017] Preferably, the step of retrieving the outputted express delivery bill information and determining whether the express delivery bill information complies with the specification specifically comprises the following steps: Get map information of each province and region; Retrieve the obtained express delivery bill information from the map information; If it can be retrieved, the express delivery bill information complies with the specifications; If the retrieval fails, the courier delivery note information is incorrect and needs to be rechecked.

[0018] Compared with the prior art, the present invention provides a method for identifying express delivery bill information, which has the following beneficial effects: When express delivery is being sorted or put into storage, the barcode on it is generally checked. The barcode is printed with the recipient's name and address information, so the elements that constitute the basic information of the express delivery are Chinese characters, English and numbers. When taking a photo, the barcode on the picture may be tilted. Therefore, the present invention also introduces a correction of the barcode to make the font in the barcode clearer. The present invention uses image processing technology and a neural network model to accurately identify the express delivery bill information, and retrieves the identified express delivery bill information to determine whether the express delivery bill information complies with the specifications, which meets the needs of the staff. There is no need for the staff to check and proofread one by one, which is convenient and fast, greatly improving the speed of express delivery, as well as the efficiency of processing warehousing or sorting, and further meeting the needs of the staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of a method for identifying express delivery bill information in the present invention; Figure 2 A schematic diagram of a method for obtaining express box image information in the present invention; Figure 3 A schematic diagram of a method for obtaining a pre-processed image of a courier box in the present invention; Figure 4 A schematic diagram of a method for obtaining a pre-processed barcode in the present invention; Figure 5 A schematic diagram of a method for obtaining a final corrected barcode image in the present invention; Figure 6 A schematic diagram of a method for obtaining barcode feature information in the present invention; Figure 7 It is a schematic diagram of a method for an express delivery bill information recognition model based on Chinese character feature information, English letter feature information and digital feature information in the present invention; Figure 8 A schematic diagram of a method for outputting express delivery bill information in the present invention; Fig. 9 It is a schematic diagram of the method for determining whether the express delivery bill information complies with the specifications in the present invention. DETAILED DESCRIPTION

[0020] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations. Example

[0021] Please refer to Figure 1-Figure 9As shown, a method for identifying express delivery bill information includes: Use high-definition camera equipment to record the express boxes on the conveyor belt to obtain the picture information of the express boxes; Perform denoising on the express box image to obtain a pre-processed express box image; Identify the barcode area in the express box image, perform rectangular positioning and tilt angle correction on it, and obtain the pre-processed barcode; Acquire Chinese character feature information, English letter feature information and number feature information from the database, and establish an express delivery bill information recognition model based on the Chinese character feature information, English letter feature information and number feature information; Perform feature extraction on the preprocessed barcode information to obtain barcode feature information; Input the obtained barcode feature information into the express delivery bill information recognition model, and output the express delivery bill information; Search the output express delivery bill information to determine whether the express delivery bill information complies with the specifications.

[0022] It is understandable to those skilled in the art that, when express delivery is being sorted or put into storage, the barcode on it is generally inspected. The barcode is printed with the name of the recipient and the address information of the recipient. Therefore, the elements that constitute the basic information of the express delivery are Chinese characters, English and numbers. When taking a photo, the barcode on the picture may be tilted. Therefore, the present invention also introduces correction of the barcode to make the font in the barcode clearer. The present invention uses image processing technology and a neural network model to accurately identify the express delivery bill information, and retrieves the identified express delivery bill information to determine whether the express delivery bill information complies with the specifications, thereby meeting the needs of the staff. There is no need for the staff to check and proofread one by one, which is convenient and fast, greatly improving the speed of express delivery, as well as the efficiency of processing warehousing or sorting, and further meeting the needs of the staff.

[0023] The express boxes on the conveyor belt are recorded with high-definition camera equipment, and the specific steps of obtaining the express box image information include the following: Use sensors to detect whether a courier box passes under the high-definition camera; If there is, high-definition camera equipment is activated to record the passing express box; If not, the high-definition camera device will not start.

[0024] It is understandable to those skilled in the art that by using sensors to control the startup and shutdown of high-definition camera equipment, the high-definition camera equipment is not always in working state, which saves power and avoids the situation where the equipment is prone to malfunction due to long-term operation.

[0025] De-noising the courier box image and obtaining the pre-processed courier box image specifically includes the following steps: Decompose the high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.

[0026] .It can be understood by those skilled in the art that, through wavelet transform, the high-frequency information and low-frequency information of the image are processed separately, and the low-frequency components are restored using the LPG-PCA method. In the process of restoring the high-frequency components, the natural image is sparsely represented in the group domain, and the inherent local sparsity and non-local self-similarity of the natural image are clearly and effectively characterized in a unified manner, and the high-frequency part is extracted from the image without being affected by noise. The image details are retained while effectively removing the noise, achieving high sparsity and high restoration quality, and improving the visual performance of the image as well as the peak signal-to-noise ratio and structural similarity.

[0027] The barcode area in the express box image is identified, and rectangular positioning and tilt angle correction are performed to obtain the pre-processed barcode, which specifically includes the following steps: Perform edge detection on the image in the horizontal and vertical directions, and sum the detection results in the horizontal and vertical directions; The barcode part of the image is processed based on the first morphological closing operation to obtain the connected area, and then the single line interference on the image is removed by the second morphological closing operation; Filling holes in the image processed by the first morphological closing operation; Calculate the minimum bounding rectangle for each white area in the image after filling the holes, and obtain the information of each bounding rectangle, which includes the coordinates of the four vertices of the rectangle, the tilt angle, and the number of white pixels in the rectangular area; According to the calculated rectangular tilt angles, their distribution within the right angle range is counted, and the angles with more concentrated distribution are used to perform preliminary tilt angle correction.

[0028] The calculation formula for the minimum enclosing rectangle is: Smin = min (S1, S2, . . . Sn) In the formula, Smin is the minimum enclosing rectangle; S1, S2, . . . Sn are the areas of several groups of enclosing rectangles.

[0029] The steps of obtaining the pre-processed barcode include: In the image after the initial tilt angle correction, the barcode is located in the 0 degree direction or the 90 degree direction, and the error is within the range of plus or minus 10 degrees; The candidate rectangles of the barcode are screened at 0 degree and 90 degree directions respectively to determine the final position and final angle of the barcode; Perform a final rotation to obtain the final rectified barcode image.

[0030] The pre-processed barcode information is subjected to feature extraction, and obtaining the barcode feature information specifically includes the following steps: Perform feature extraction on the final corrected barcode image to obtain Chinese character feature information, English letter feature information and digital feature information; Based on the acquired Chinese character feature information, English letter feature information and number feature information, the barcode feature information is obtained.

[0031] The express delivery bill information recognition model based on Chinese character feature information, English letter feature information and digital feature information specifically includes the following steps: Based on the feedforward neural network, a courier bill information recognition model is constructed. The input data of the courier bill information recognition model includes the acquired barcode feature information. Based on machine learning, multiple sample Chinese character feature information, multiple sample English letter feature information, and multiple sample digital feature information are labeled and divided to obtain a training set, a validation set, and a test set; The license plate recognition model is supervised trained, verified and tested using training sets, validation sets and test sets to obtain an express delivery bill information recognition model whose accuracy meets the preset accuracy requirements.

[0032] It can be understood by those skilled in the art that a feedforward neural network is a type of artificial neural network that uses a unidirectional multi-layer structure, in which the 0th layer is called the input layer, the last layer is called the output layer, and the other intermediate layers are called hidden layers. The hidden layers can be one layer or multiple layers. Each hidden layer contains a number of neurons, each neuron can receive signals from the neurons in the previous layer and generate outputs to the next layer. There is no feedback in the entire network, and the signal propagates unidirectionally from the input layer to the output layer. The feature information of multiple sample Chinese characters, the feature information of multiple sample English letters and the feature information of multiple sample digits are marked as construction data, and the construction data is randomly divided into a training set, a validation set and a test set according to a certain ratio, and the divided training set, validation set and test set are annotated. For example, the preferred ratio of this application is 8:1:1, wherein the training set is used to train the model, that is, to determine the learning parameters such as the weight and bias of the model; the validation set is used to verify each model using the validation set data after training multiple models in the training set, and record the model accuracy, and then select the best model and its corresponding parameters; the test set is only used once, that is, when evaluating the final model after the training is completed, and it does not participate in the process of learning parameters or the process of selecting hyperparameters, but is only used for the evaluation of the model. Based on machine learning, the model is supervised and trained through the training set, the obtained model is verified through the validation set, and the trained model is evaluated through the test set, and the hyperparameters are continuously adjusted until the accuracy meets the preset requirements, and the express delivery bill information recognition model is output.

[0033] Inputting the obtained barcode feature information into the express delivery bill information recognition model, and outputting the express delivery bill information specifically includes the following steps: Input the obtained Chinese character feature information into the express delivery bill information recognition model that meets the preset accuracy requirements to obtain Chinese character information; Input the obtained English letter feature information into the express delivery bill information recognition model that meets the preset accuracy requirements to obtain the English letter information; Input the obtained digital feature information into the express delivery bill information recognition model that meets the preset accuracy requirements to obtain digital information; Output express delivery label information based on Chinese character information, English letter information and digital information.

[0034] The output express delivery bill information is searched to determine whether the express delivery bill information complies with the specification, which specifically includes the following steps: Get map information of each province and region; Retrieve the obtained express delivery bill information from the map information; If it can be retrieved, the express delivery bill information complies with the specifications; If the retrieval fails, the courier delivery note information is incorrect and needs to be rechecked.

[0035] In summary, when express delivery is being sorted or put into storage, the barcode on it is generally checked. The barcode is printed with the name of the recipient and the address information of the recipient. Therefore, the elements that constitute the basic information of the express delivery are Chinese characters, English and numbers. When taking a photo, the barcode on the picture may be tilted. Therefore, the present invention also introduces correction of the barcode to make the font in the barcode clearer. The present invention uses image processing technology and a neural network model to accurately identify the express delivery bill information, and retrieves the identified express delivery bill information to determine whether the express delivery bill information complies with the specifications, which meets the needs of the staff. There is no need for the staff to check and proofread one by one, which is convenient and fast, greatly improving the speed of express delivery, as well as the efficiency of processing warehousing or sorting, and further meeting the needs of the staff.

[0036] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for identifying express delivery bill information, characterized in that: include: Use high-definition camera equipment to record the express boxes on the conveyor belt to obtain the picture information of the express boxes; Perform denoising on the express box image to obtain a pre-processed express box image; Identify the barcode area in the express box image, perform rectangular positioning and tilt angle correction on it, and obtain the pre-processed barcode; Acquire Chinese character feature information, English letter feature information and number feature information from the database, and establish an express delivery bill information recognition model based on the Chinese character feature information, English letter feature information and number feature information; Perform feature extraction on the preprocessed barcode information to obtain barcode feature information; Input the obtained barcode feature information into the express delivery bill information recognition model, and output the express delivery bill information; Search the output express delivery bill information to determine whether the express delivery bill information complies with the specifications.

2. A method for identifying express delivery bill information according to claim 1, characterized in that: The express boxes on the conveyor belt are recorded using a high-definition camera to obtain the express box image information, which specifically includes the following steps: Use sensors to detect whether a courier box passes under the high-definition camera; If there is, high-definition camera equipment is activated to record the passing express box; If not, the high-definition camera device will not start.

3. A method for identifying express delivery bill information according to claim 2, characterized in that: The denoising process of the express box image to obtain the pre-processed express box image specifically includes the following steps: Decompose the high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.

4. A method for identifying express delivery bill information according to claim 3, characterized in that: The steps of identifying the barcode area in the express box image, performing rectangular positioning and tilt angle correction on the barcode area, and obtaining the pre-processed barcode specifically include the following steps: Perform edge detection on the image in the horizontal and vertical directions, and sum the detection results in the horizontal and vertical directions; The barcode part of the image is processed based on the first morphological closing operation to obtain the connected area, and then the single line interference on the image is removed by the second morphological closing operation; Filling holes in the image processed by the first morphological closing operation; Calculate the minimum bounding rectangle according to each white area in the image after filling the holes, and obtain information of each bounding rectangle, wherein the rectangle information includes coordinates of the four vertices of the rectangle, an inclination angle, and the number of white pixels in the rectangular area; According to the calculated rectangular tilt angles, their distribution within the right angle range is counted, and the angles with more concentrated distribution are used to perform preliminary tilt angle correction.

5. A method for identifying express delivery bill information according to claim 4, characterized in that: The calculation formula of the minimum enclosing rectangle is: Smin = min (S1, S2, . . . Sn) In the formula, Smin is the minimum enclosing rectangle; S1, S2, . . . Sn are the areas of several groups of enclosing rectangles.

6. A method for identifying express delivery bill information according to claim 5, characterized in that: The step of obtaining the pre-processed barcode specifically includes the following steps: In the image after the initial tilt angle correction, the barcode is located in the 0 degree direction or the 90 degree direction, and the error is within the range of plus or minus 10 degrees; The candidate rectangles of the barcode are screened at 0 degree and 90 degree directions respectively to determine the final position and final angle of the barcode; Perform a final rotation to obtain the final rectified barcode image.

7. A method for identifying express delivery bill information according to claim 6, characterized in that: The process of extracting features from the preprocessed barcode information to obtain barcode feature information specifically includes the following steps: Perform feature extraction on the final corrected barcode image to obtain Chinese character feature information, English letter feature information and digital feature information; Based on the acquired Chinese character feature information, English letter feature information and number feature information, the barcode feature information is obtained.

8. A method for identifying express delivery bill information according to claim 7, characterized in that: The express delivery bill information recognition model based on Chinese character feature information, English letter feature information and digital feature information specifically includes the following steps: Based on a feedforward neural network, a courier bill information recognition model is constructed, wherein the input data of the courier bill information recognition model includes the acquired barcode feature information; Based on machine learning, the plurality of sample Chinese character feature information, the plurality of sample English letter feature information, and the plurality of sample number feature information are labeled and divided to obtain a training set, a validation set, and a test set; The training set, validation set and test set are used to perform supervised training, validation and testing on the license plate recognition model to obtain the express delivery bill information recognition model whose accuracy meets the preset accuracy requirements.

9. A method for identifying express delivery bill information according to claim 8, characterized in that: The step of inputting the obtained barcode feature information into the express delivery bill information recognition model and outputting the express delivery bill information specifically comprises the following steps: Input the obtained Chinese character feature information into the express delivery bill information recognition model that meets the preset accuracy requirement to obtain Chinese character information; Input the obtained English letter feature information into the express delivery bill information recognition model that meets the preset accuracy requirement to obtain English letter information; Input the obtained digital feature information into the express delivery bill information recognition model that meets the preset accuracy requirements to obtain digital information; Output express delivery label information based on Chinese character information, English letter information and digital information.

10. A method for identifying express delivery bill information according to claim 9, characterized in that: The step of searching the outputted express delivery bill information and determining whether the express delivery bill information complies with the specification specifically includes the following steps: Get map information of each province and region; Retrieve the obtained express delivery bill information from the map information; If it can be retrieved, the express delivery bill information complies with the specifications; If the retrieval fails, the courier delivery note information is incorrect and needs to be rechecked.

Images