CN114742088A

CN114742088A - Bar code image processing method, device and equipment

Info

Publication number: CN114742088A
Application number: CN202210385695.4A
Authority: CN
Inventors: 孙彦飞; 周俊; 王俊宇
Original assignee: Fudan University; Zhuhai Fudan Innovation Research Institute
Current assignee: Fudan University; Zhuhai Fudan Innovation Research Institute
Priority date: 2022-04-13
Filing date: 2022-04-13
Publication date: 2022-07-12

Abstract

The application relates to a barcode image processing method, a barcode image processing device and barcode image processing equipment. The method comprises the following steps: acquiring a bar code image to be processed; inputting the bar code image to be processed into a pre-constructed preset neural network so as to identify a target bar code object in the bar code image to be processed in the preset neural network, and generating a newly-constructed bar code image according to the target bar code object; and outputting the newly-built bar code image so that a decoding device can decode the newly-built bar code image. The scheme provided by the application can improve the success rate and the accuracy of the decoding equipment for recognizing and decoding the bar code image.

Description

Bar code image processing method, device and equipment

Technical Field

The present application relates to the field of barcode technologies, and in particular, to a barcode image processing method, apparatus, and device.

Background

Bar codes (also known as bar codes) are world-wide identification cards unique to goods and provide machine-readable, printable language as compared to machine-readable language loaded on magnetic disks, magnetic tapes and optical disks. Barcode technology, one of the automatic identification technologies, has been developed for nearly thirty years, and barcode labels basically cover all products. Barcodes can be printed on packaging and articles, scanned by a digital camera or camera-equipped cell phone, digitized, and then information (e.g., product lot, date of manufacture, etc.) can be extracted. The bar code has the characteristic of easy realization of informatization management, and is widely applied to the product traceability field.

In the related art, a camera or a mobile phone photographs an article to which a barcode is attached to obtain a barcode image, and a decoding device can recognize and decode the barcode image to obtain information of the barcode in the barcode image.

However, at present, the success rate and accuracy of decoding and recognizing of the decoding device are not high for the barcode image obtained by photographing the article attached with the barcode with a camera or a mobile phone.

Disclosure of Invention

In order to solve or partially solve the problems in the related art, the application provides a method, a device and equipment for processing a barcode image, which can improve the success rate and the accuracy of the decoding equipment for recognizing and decoding the barcode image.

The first aspect of the present application provides a barcode image processing method, including:

acquiring a bar code image to be processed;

inputting the bar code image to be processed into a pre-constructed preset neural network so as to identify a target bar code object in the bar code image to be processed in the preset neural network, and generating a newly-constructed bar code image according to the target bar code object;

and outputting the newly-built bar code image so that a decoding device can decode the newly-built bar code image.

In one embodiment, the generating a new barcode image according to the target barcode object includes:

when the target bar code object is judged to be distorted, correcting the target bar code object, and generating a newly-built bar code image with a preset pixel size according to the corrected target bar code object; or the like, or, alternatively,

and when the target bar code object is judged not to be distorted, generating a new bar code image with a preset pixel size according to the target bar code object.

In one embodiment, the generating a new barcode image with a preset pixel size includes:

and generating a new bar code image with a preset pixel size in a zooming mode.

In one embodiment, the predetermined pixel size is determined according to the pixel requirements of the decoding device.

In one embodiment, the preset neural network is pre-constructed according to a preset semantic segmentation network; or the preset neural network is obtained by pre-constructing according to a preset semantic segmentation network and a preset space transformation network.

In one embodiment, the preset semantic segmentation network comprises a BiSeNet network.

The second aspect of the present application provides a barcode image processing apparatus, including:

the acquisition module is used for acquiring a bar code image to be processed;

the processing module is used for inputting the bar code image to be processed into a pre-constructed preset neural network so as to identify a target bar code object in the bar code image to be processed in the preset neural network and generate a newly-built bar code image according to the target bar code object;

and the output module is used for outputting the newly-built bar code image so as to enable decoding equipment to decode the newly-built bar code image.

A third aspect of the present application provides a decoding device comprising:

the processing unit is used for acquiring a bar code image to be processed; inputting the bar code image to be processed into a pre-constructed preset neural network so as to identify a target bar code object in the bar code image to be processed in the preset neural network, and generating a newly-constructed bar code image according to the target bar code object; outputting the newly-built bar code image;

and the decoding unit is used for decoding according to the newly-built bar code image output by the processing unit.

A fourth aspect of the present application provides an electronic device, comprising:

a processor; and

a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method as described above.

A fifth aspect of the present application provides a computer-readable storage medium having stored thereon executable code, which, when executed by a processor of an electronic device, causes the processor to perform a method as described above.

The technical scheme provided by the application can comprise the following beneficial effects:

according to the method, the bar code image to be processed is acquired, the bar code image to be processed is input into the pre-established preset neural network, so that the target bar code object in the bar code image to be processed is identified in the preset neural network, a new bar code image is generated according to the target bar code object, and the new bar code image is output, so that the decoding equipment decodes the new bar code image. Therefore, compared with the barcode image to be processed, the newly-built barcode image has less interference information amount, is beneficial to the decoding equipment to carry out recognition and decoding, and effectively improves the success rate and the accuracy of the decoding equipment for recognition and decoding.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The foregoing and other objects, features and advantages of the application will be apparent from the following more particular descriptions of exemplary embodiments of the application as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the application.

Fig. 1 is a schematic flowchart of a barcode image processing method according to an embodiment of the present application;

FIG. 2 is another schematic flow chart diagram illustrating a barcode image processing method according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a barcode image processing apparatus according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a decoding device shown in an embodiment of the present application;

fig. 5 is a schematic structural diagram of an electronic device shown in an embodiment of the present application.

Detailed Description

Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While embodiments of the present application are illustrated in the accompanying drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms "first," "second," "third," etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.

In the related art, for a barcode image obtained by photographing an article to which a barcode is attached with a camera or a mobile phone, a success rate and a correct rate of recognition and decoding of a decoding device are not high.

In view of the above problems, embodiments of the present application provide a barcode image processing method, which can improve the success rate and accuracy of decoding a barcode image by a decoding device.

The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1 is a schematic flowchart of a barcode image processing method according to an embodiment of the present application.

Referring to fig. 1, the method includes:

and S101, acquiring a bar code image to be processed.

The barcode image to be processed may refer to an image including a barcode, that is, the barcode exists in the barcode image to be processed. The number of barcodes present in the barcode image to be processed may be one or more. The barcode may be a one-dimensional barcode or a two-dimensional barcode, which is not limited in this application.

For example, the barcode image to be processed may be an image obtained by photographing an object (e.g., a package box of an article) to which a barcode is attached by a photographing device (e.g., a camera, a smart phone, etc.). For another example, the barcode image to be processed may be an image obtained by photographing a barcode displayed on the electronic screen by a photographing device (e.g., a camera, a smart phone, etc.).

Step S102, inputting the bar code image to be processed into a preset neural network which is constructed in advance, so that a target bar code object in the bar code image to be processed is identified in the preset neural network, and a new bar code image is generated according to the target bar code object.

In one embodiment, the preset neural network is pre-constructed according to a preset semantic segmentation network. The preset semantic Segmentation Network can be a BiSeNet Network, namely a Bilatel Segmentation Network, and a double-leaf Segmentation Network, wherein the BiSeNet Network can realize real-time semantic Segmentation and simultaneously improve the speed and the precision of the real-time semantic Segmentation. The preset neural network may be constructed according to a BiSeNet network, for example, the preset neural network is obtained by adjusting and modifying a network operation layer in the BiSeNet network. In the embodiment, a target bar code object in the bar code image to be processed is identified through a preset neural network, and a new bar code image is generated according to the target bar code object.

In another embodiment, the preset neural network is obtained by pre-constructing according to a preset semantic segmentation network and a preset spatial transformation network. The preset semantic segmentation network may be a BiSeNet network. In the embodiment, after the preset neural network identifies a target barcode object in a barcode image to be processed, when the target barcode object is judged to be distorted, the target barcode object is corrected, and a new barcode image with a preset pixel size is generated according to the corrected target barcode object; or when the target bar code object is judged not to be distorted, generating a new bar code image with a preset pixel size according to the target bar code object.

And step S103, outputting the newly-built bar code image so that the decoding equipment can decode the newly-built bar code image.

It can be understood that the newly-created bar code image is only generated according to the target bar code object, so that the interference factor in the bar code image to be processed can be removed, and the bar code can be recognized and decoded by decoding equipment.

It can be seen from this embodiment that, in the method provided in this embodiment of the present application, the barcode image to be processed is input to the pre-established preset neural network by obtaining the barcode image to be processed, so that the target barcode object in the barcode image to be processed is identified in the preset neural network, a newly-created barcode image is generated according to the target barcode object, and the newly-created barcode image is output, so that the decoding device decodes the newly-created barcode image. Therefore, compared with the barcode image to be processed, the newly-built barcode image has less interference information amount, is beneficial to the decoding equipment to carry out recognition and decoding, and effectively improves the success rate and the accuracy of the decoding equipment for recognition and decoding.

Fig. 2 is another schematic flow chart of a barcode image processing method according to an embodiment of the present application. Fig. 2 depicts the solution of the present application in more detail with respect to fig. 1.

Referring to fig. 2, the method includes:

step S201, acquiring a bar code image to be processed.

This step can be referred to the description of step S101, and is not described herein again.

Step S202, inputting the bar code image to be processed into a pre-established neural network so as to identify a target bar code object in the bar code image to be processed in the pre-established neural network, correcting the target bar code object when the target bar code object is judged to be distorted, and generating a new bar code image with a preset pixel size according to the corrected target bar code object; or when the target bar code object is judged not to be distorted, generating a new bar code image with a preset pixel size according to the target bar code object.

The preset neural network is obtained by constructing in advance according to a preset semantic segmentation network and a preset spatial transformation network. The preset semantic segmentation network can be a BiSeNet network, and the preset neural network can identify a target bar code object from the bar code image to be processed by utilizing the network structure of the BiSeNet network, namely, the area occupied by the target bar code object in the bar code image to be processed is determined. The method can be understood that the BiSeNet network can effectively retain the space information of an original image and generate the advantages of high-resolution features, the preset neural network is constructed according to the BiSeNet network, after the preset neural network is trained, the preset neural network can identify the region where a target bar code object in a bar code image to be processed is located, not only is real-time semantic segmentation realized, but also the performance of the semantic segmentation is effectively enhanced through the feature fusion module and the attention optimization module in the network structure based on the BiSeNet network, and the speed and the precision of executing a semantic segmentation task are greatly improved.

The preset space transformation network is used for correcting and repairing the detected and identified target bar code object. The preset Spatial transform Network may be an STN (Spatial Transformer Network). The preset neural network can be constructed according to a BiSeNet network and an STN network, the network structure of the preset neural network can comprise the network structure of the BiSeNet network and the network structure of the STN network, the output of the BiSeNet network in the preset neural network is the input of the STN network, namely, in the preset neural network, the BiSeNet network outputs a target bar code object which is detected and identified, and the STN network corrects and repairs the target bar code object. It should be noted that the preset spatial transformation network may extract features through a positioning network by convolution operation, and then use a hidden network layer to infer a parameter θ of spatial transformation, then obtain a correspondence between pixel coordinates before and after transformation according to the parameter θ by a lattice generator, and finally generate a feature map after spatial transformation by using a sampler in a bilinear interpolation manner. The preset space transformation network has better identification and detection capability on the distorted and deformed target in the image, and can correct and repair the distorted and deformed target in a space transformation mode.

In the step, after the target barcode object in the barcode image to be processed is identified in the preset neural network, the target barcode object is processed according to the following two conditions (1) and (2).

(1) And when the target bar code object is judged to be distorted, correcting the target bar code object, and generating a new bar code image with a preset pixel size according to the corrected target bar code object.

In the embodiment of the application, the basis for judging whether the target bar code object is distorted can be the distortion degree of the target bar code object; and when the distortion degree of the target bar code object reaches a preset value, determining that the target bar code object is distorted. In the preset neural network, the target bar code object is corrected by using the preset space transformation network, that is, pixels corresponding to the distorted target bar code object are corrected and restored, so that a new bar code image with the size of a preset pixel is generated according to the corrected target bar code object. It can be understood that, a preset spatial transformation network (e.g., STN network) is introduced into the preset neural network, which provides a corresponding spatial transformation manner for the input in the network structure, and for the target barcode object that is recognized to be quadrilateral, the preset spatial transformation network can fit the quadrilateral to the convex hull of each possible information area in the network, and distort each quadrilateral back to the original pixels by calculating homography, thereby implementing the correction of the target barcode object.

(2) And when the target bar code object is judged not to be distorted, generating a new bar code image with a preset pixel size according to the target bar code object.

When the target bar code object is judged not to be distorted, the target bar code object does not need to be corrected by using a preset space transformation network, and a newly-built bar code image with a preset pixel size is generated directly according to the target bar code object.

It can be found that in both cases (1) and (2), a newly created barcode image of a preset pixel size is generated. In the embodiment of the application, the preset pixel size is determined according to the pixel requirement of the decoding device. In other words, the preset pixel size is determined by the pixel requirements of the decoding device receiving the newly created barcode image. The pixel requirements of the decoding device, i.e., the pixel requirements of the input image for the decoding device to be able to perform decoding.

In one embodiment, generating a new barcode image with a preset pixel size includes: and generating a new bar code image with a preset pixel size in a zooming mode. For example, the corrected target barcode object may be enlarged or reduced to generate a new barcode image of a preset pixel size.

It should be noted that the preset neural network is obtained after being trained in advance. In the embodiment of the application, the preset neural network is trained by using a pre-constructed training data set. The training data set comprises a plurality of different barcode images, and the image area proportion of the target barcode object in the different barcode images is different. For example, the target barcode object may be transformed (e.g., flipped, shifted) and placed in a high resolution image to obtain a barcode image used to construct the training data set. The preset neural network can adjust parameters of segmentation size in the semantic segmentation task by using a training suite in the related art during training.

And step S203, outputting the newly-built bar code image so that a decoding device can decode the newly-built bar code image.

It can be understood that the occupation ratio of the pixel area of the target barcode object in the barcode image to be processed may be small, and the target barcode object may be distorted to some extent in the barcode image to be processed due to the shooting angle, the performance of the camera, and the like. The decoding device in the related art cannot be used for processing complete detection in a large image, that is, for a barcode image to be processed with a small pixel ratio of a target barcode object in the image, the decoding device in the related art cannot perform decoding processing, and similarly, for a barcode image to be processed with a distorted target barcode object, the decoding success rate and the accuracy of the decoding device in the related art are low.

In the embodiment of the application, the newly-built bar code image is only generated according to the target bar code object, so that the interference factor in the bar code image to be processed can be removed, and the bar code can be recognized and decoded by decoding equipment. In one embodiment, the pixel occupancy (occupancy ratio) of the target barcode object in the new barcode image reaches a preset requirement, for example, the pixel occupancy of the target barcode object in the new barcode image reaches more than 70%.

According to the method provided by the embodiment of the application, compared with the barcode image to be processed, the newly-built barcode image has less interference information amount, so that the decoding equipment can recognize and decode the newly-built barcode image, and the success rate and the accuracy of the decoding equipment in recognizing and decoding are effectively improved.

Corresponding to the embodiment of the application function implementation method, the application also provides a bar code image processing device.

Fig. 3 is a schematic structural diagram of a barcode image processing apparatus according to an embodiment of the present application.

Referring to fig. 3, a barcode image processing apparatus 30 includes: an acquisition module 310, a processing module 320, and an output module 330.

The acquiring module 310 is configured to acquire a barcode image to be processed.

The processing module 320 is configured to input the barcode image to be processed into a pre-established preset neural network, so that a target barcode object in the barcode image to be processed is identified in the preset neural network, and a newly-established barcode image is generated according to the target barcode object.

And the output module 330 is configured to output the newly created barcode image, so that the decoding device decodes the newly created barcode image.

It can be seen from this embodiment that, the barcode image processing apparatus 30 provided in the present application can obtain a newly created barcode image, and compared with a barcode image to be processed, the newly created barcode image has less interference information amount, and is beneficial to a decoding device to perform recognition and decoding, thereby effectively improving the success rate and accuracy of the decoding device to perform recognition and decoding.

With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Fig. 4 is a schematic structural diagram of a decoding apparatus according to an embodiment of the present application.

Referring to fig. 4, a decoding apparatus 40 includes: a processing unit 410 and a decoding unit 420.

The processing unit 410 is used for acquiring a barcode image to be processed; inputting the bar code image to be processed into a pre-constructed preset neural network so as to identify a target bar code object in the bar code image to be processed in the preset neural network, and generating a newly-built bar code image according to the target bar code object; outputting a newly-built bar code image;

and the decoding unit 420 is used for decoding according to the newly-built barcode image output by the processing unit.

The processing unit 410 may also be used to perform the method in the embodiments shown in fig. 1 or fig. 2, among others.

As can be seen from this embodiment, the decoding device 40 provided in the present application outputs the new barcode image through the processing unit 410, so that the decoding unit 420 decodes the new barcode image. Therefore, compared with the bar code image to be processed, the newly-built bar code image has less interference information amount, is beneficial to the implementation of recognition and decoding, and effectively improves the success rate and the accuracy of the recognition and decoding.

Referring to fig. 5, an electronic device 500 includes a memory 510 and a processor 520.

Processor 520 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 510 may include various types of storage units, such as system memory, Read Only Memory (ROM), and permanent storage. Wherein the ROM may store static data or instructions for the processor 520 or other modules of the computer. The persistent storage device may be a read-write storage device. The persistent storage may be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered down. In some embodiments, the persistent storage device employs a mass storage device (e.g., magnetic or optical disk, flash memory) as the persistent storage device. In other embodiments, the permanent storage may be a removable storage device (e.g., floppy disk, optical drive). The system memory may be a read-write memory device or a volatile read-write memory device, such as a dynamic random access memory. The system memory may store instructions and data that some or all of the processors require at runtime. Further, the memory 510 may comprise any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), magnetic and/or optical disks, as well. In some embodiments, memory 510 may include a removable storage device that is readable and/or writable, such as a Compact Disc (CD), digital versatile disc read only (e.g., DVD-ROM, dual layer DVD-ROM), Blu-ray disc read only, ultra-dense disc, flash memory card (e.g., SD card, min SD card, Micro-SD card, etc.), magnetic floppy disk, or the like. Computer-readable storage media do not contain carrier waves or transitory electronic signals transmitted by wireless or wired means.

The memory 510 has stored thereon executable code that, when processed by the processor 520, may cause the processor 520 to perform some or all of the methods described above.

Furthermore, the method according to the present application may also be implemented as a computer program or computer program product comprising computer program code instructions for performing some or all of the steps of the above-described method of the present application.

Alternatively, the present application may also be embodied as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) having executable code (or a computer program or computer instruction code) stored thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the various steps of the above-described method according to the present application.

Having described embodiments of the present application, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A barcode image processing method is characterized by comprising the following steps:

acquiring a bar code image to be processed;

2. The method of claim 1, wherein generating a new barcode image from the target barcode object comprises:

3. The method of claim 2, wherein the generating of the new barcode image with the preset pixel size comprises:

and generating a new bar code image with a preset pixel size in a zooming mode.

4. The method of claim 2, wherein:

and determining the preset pixel size according to the pixel requirement of the decoding device.

5. The method of claim 1, wherein:

the preset neural network is obtained by pre-constructing according to a preset semantic segmentation network; or the preset neural network is obtained by pre-constructing according to a preset semantic segmentation network and a preset space transformation network.

6. The method of claim 5, wherein:

the preset semantic segmentation network comprises a BiSeNet network.

7. A barcode image processing apparatus characterized by comprising:

the acquisition module is used for acquiring a bar code image to be processed;

8. A decoding device, characterized by comprising:

9. An electronic device, comprising:

a processor; and

a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method of any one of claims 1-6.

10. A computer-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to perform the method of any of claims 1-6.