Multi-bar code recognition system

Abstract

The invention belongs to the technical field of barcode recognition, and particularly relates to a multi-barcode recognition system, which comprises an image acquisition module used for acquiring a target image comprising at least one candidate region which can be segmented into independent barcodes; the processing module is provided with a memory and an operation core, and the processing module is used for positioning and segmenting each bar code candidate area in the target image and distributing a unique identifier for the bar code candidate area; each bar code candidate area with the unique identifier is preliminarily decoded, the processing state of the bar code candidate area is recorded according to a decoding result, and the processing state comprises successful decoding or to-be-optimized; and establishing a to-be-optimized queue for the bar code candidate areas in the to-be-optimized state, selecting a target bar code candidate area from the to-be-optimized queue according to a predetermined strategy, and processing the target bar code candidate area until the to-be-optimized queue is empty or meets a preset global termination condition.

Description

Technical Field

[0001] The present invention relates to the technical field of barcode recognition, and in particular to a multi-barcode recognition system. Background Art

[0002] As a key information carrier, barcode recognition requires fast and accurate recognition. However, practical application scenarios often feature complex lighting conditions and diverse target material materials, presenting numerous challenges for barcode recognition. For example, when exposed to high-intensity or angled light, the surface of a target object (such as metal, coated paper, or transparent plastic) can easily produce specular reflections or localized highlights, resulting in overexposure of portions of the barcode image and loss of critical black and white module information. Furthermore, when multiple barcodes are present in the same field of view, they may exhibit varying exposure states and clarity due to their position and depth of field. Global exposure adjustment or simple HDR (high dynamic range) techniques struggle to address these local issues and may compromise overall image quality or real-time processing. Furthermore, existing barcode processing workflows are often linear or simple parallel processes, lacking consideration for the interplay between multiple barcodes and effective utilization of processed information, potentially resulting in low barcode recognition accuracy. Therefore, there is an urgent need for a recognition method and system that can intelligently sense and proactively optimize lighting conditions, precisely adjust focus, selectively repair image defects, and efficiently manage the processing status of multiple barcodes to improve overall recognition performance. Summary of the Invention

[0003] The present invention provides a multi-barcode recognition method and system, which are used to solve the problem in the prior art of poor multi-barcode recognition performance caused by different positions and depths of field of multiple barcodes in the same field of view.

[0004] In one aspect, the present invention provides a multi-barcode recognition method, comprising:

[0005] Under a first lighting mode, acquiring a target image including at least one region that can be segmented into independent barcode candidate regions;

[0006] Locate and segment each barcode candidate region in the target image and assign it a unique identifier;

[0007] Performing preliminary decoding on each barcode candidate region with a unique identifier, and recording its processing status according to the decoding result, wherein the processing status includes successfully decoded or to be optimized;

[0008] Establishing a queue for barcode candidate areas to be optimized, selecting target barcode candidate areas from the queue according to a predetermined strategy and processing them until the queue is empty or a preset global termination condition is met;

[0009] The decoding information of the barcode candidate areas corresponding to all unique identifiers and whose final processing status is successfully decoded is summarized and output.

[0010] According to a multi-barcode recognition method provided by the present invention, the processing status further includes decoding confidence. When the decoding is successful but the confidence is lower than a preset threshold, the barcode candidate area is also marked as to be optimized.

[0011] According to a multi-barcode recognition method provided by the present invention, the target barcode candidate area is selected from the queue to be optimized according to a predetermined strategy and processed, including:

[0012] Select a barcode candidate area from the queue to be optimized as the target barcode candidate area;

[0013] Perform imaging quality analysis on the selected target barcode candidate area, and process the target barcode candidate area using one or more combination strategies of intelligent fill light optimization, local image restoration, and automatic focusing based on the analysis results;

[0014] The processed target barcode candidate area is decoded and its processing status is updated according to the decoding result.

[0015] According to a multi-barcode recognition method provided by the present invention, when an intelligent fill light optimization strategy is adopted, images are recaptured under a second lighting mode, and the potential imaging impact on non-target barcode candidate areas in the field of view under this lighting mode is evaluated. The processing status of the affected non-target barcode candidate areas is selectively updated based on the evaluation results.

[0016] According to a multi-barcode recognition method provided by the present invention, when local image restoration is adopted, or when image-level refinement processing is still required after intelligent fill light optimization, at least one restoration operation selected from image enhancement, multi-exposure fusion, or texture reconstruction based on a generative adversarial network is performed on the target barcode candidate area.

[0017] According to a multi-barcode recognition method provided by the present invention, when automatic focusing is adopted, an automatic focusing optimization operation is performed on the target barcode candidate area.

[0018] According to a multi-barcode recognition method provided by the present invention, if the evaluation result shows that the adjustment of the illumination mode has a significant negative imaging impact on the non-target barcode candidate area that has been successfully decoded before, the processing status of the non-target barcode candidate area is updated to be optimized and re-added to the queue for optimization.

[0019] According to a multi-barcode recognition method provided by the present invention, a target image is acquired through an image acquisition module, and the first lighting mode is that all the lamp beads of the annular adjustable light source array module arranged around the image acquisition module are lit.

[0020] According to a multi-barcode recognition method provided by the present invention, the second lighting mode is that the annular adjustable light source array module only lights up the lamp beads on the side close to the target barcode candidate area, or reduces or increases the current of the lamp beads based on the first lighting mode.

[0021] On the other hand, the present invention also provides a multi-barcode recognition method and system, comprising:

[0022] An image acquisition module, configured to acquire a target image containing at least one region that can be segmented into independent barcode candidate regions;

[0023] The annular adjustable light source array module is used to provide fill light when the image acquisition module acquires the target image. Its lighting mode is dynamically adjusted by the control command;

[0024] An automatic focusing module, linked with the image acquisition module, for adjusting the imaging focal length;

[0025] A processing module is connected to the image acquisition module, the annular adjustable light source array module and the automatic focusing module. The processing module is configured with a memory and a computing core for executing the method described above.

[0026] The multi-barcode recognition method and system provided by the present invention achieve efficient and accurate recognition of multiple codes by establishing a unique identifier for each barcode candidate area and dynamically updating the processing status, combined with conditionally triggered intelligent fill light, automatic focus adjustment and local image repair, while minimizing duplication of work and negative processing interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0028] Figure 1 1 is a flow chart of the multi-barcode recognition method provided by the present invention;

[0029] Figure 2 It is a schematic diagram of the processing flow of the barcode candidate area in the state to be optimized;

[0030] Figure 3 It is a module block diagram of the multi-barcode recognition system provided by the present invention;

[0031] Figure 4 Schematic diagram of the present invention applied to a high-speed camera;

[0032] Figure 5It is a schematic diagram of the front structure of the image acquisition module. DETAILED DESCRIPTION

[0033] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0034] The following combination Figure 1-Figure 5 The present invention describes a multi-barcode recognition method and system. The method and system are applied to a barcode scanning device with a barcode reading function (such as a high-definition scanner, an industrial camera, etc.). When scanning, fill light components can be used to provide fill light to increase the clarity of the barcode image.

[0035] In one aspect, the present invention provides a multi-barcode recognition method, comprising:

[0036] State initialization;

[0037] In the first lighting mode, a target image containing at least one area that can be divided into independent barcode candidate areas is obtained; wherein, the target image can be obtained through a camera (i.e., an image acquisition module), and the first lighting mode is that all the multiple lamp beads arranged around the camera are lit (i.e., uniform lighting of the entire ring), and the multiple lamp beads constitute a ring-shaped adjustable light source array module. The material and ambient lighting parameters of the target object can be obtained through the environmental perception module in the state initialization stage. The processing module calculates the light intensity, current size, etc. of the lamp beads based on the above parameters, which is the origin of the size of each initial parameter in the first lighting mode. In the first lighting mode, the camera acquires the target image for the first time.

[0038] Locate and segment each barcode candidate region in the target image and assign it a unique identifier;

[0039] Each barcode candidate area with a unique identifier is preliminarily decoded, and its processing status is recorded according to the decoding result. The processing status includes successful decoding or to be optimized (which can be understood as decoding failure). Furthermore, the processing status also includes decoding confidence. When the decoding is successful but the confidence is lower than the preset threshold, the barcode candidate area is also marked as to be optimized. In this embodiment, the preset threshold of the confidence is set to 0.98.

[0040] A queue for optimization is established for the candidate barcode regions to be optimized. Target barcode candidate regions are selected from the queue according to a predetermined strategy and processed until the queue is empty or a preset global termination condition is met. In other words, the queue is iteratively optimized. The global termination condition can be set to the maximum number of attempts.

[0041] Summarizes the decoding information of all barcode candidate regions corresponding to unique identifiers that have been successfully decoded and outputs them. For barcode candidate regions that still fail to decode after multiple optimizations, you can choose not to output them or mark them as recognition failures.

[0042] In this embodiment, specifically, selecting and processing the target barcode candidate region from the queue to be optimized according to a predetermined strategy includes:

[0043] Randomly select a barcode candidate area from the queue to be optimized as the target barcode candidate area;

[0044] Perform imaging quality analysis on the selected target barcode candidate area, especially evaluating the exposure status (such as overexposure, underexposure, and insufficient contrast) and clarity. Based on the analysis results, process the target barcode candidate area using a combination of one or more strategies including intelligent fill light optimization, local image restoration, and autofocus.

[0045] The processed target barcode candidate area is decoded and its processing status is updated according to the decoding result.

[0046] More specifically, based on the imaging quality analysis results of the target barcode candidate area, when adopting the intelligent fill light optimization strategy, it is necessary to adjust the illumination mode of the annular adjustable light source array module, that is, recapture the image under the second illumination mode, and evaluate the potential imaging impact of the illumination mode on the non-target barcode candidate areas in the field of view (especially the barcode candidate areas marked as "successfully decoded"), and selectively update the processing status of the affected non-target barcode candidate areas based on the evaluation results. If the evaluation results indicate that the adjustment of the illumination mode has a significant negative imaging impact on the non-target barcode candidate areas that have been successfully decoded (such as causing the original readable barcode to become unreadable), the processing status of the non-target barcode candidate area is updated to pending optimization and re-added to the queue for optimization; at the same time, the decoding results of the non-target barcode candidate area under various illumination conditions are recorded, or in subsequent processing, attempts are made to restore favorable illumination conditions for the non-target barcode candidate area. The second lighting mode is that the annular adjustable light source array module lights up only the lamp beads on the side close to the target barcode candidate area (for overexposure and insufficient contrast analysis results), that is, a low-angle lighting mode is adopted, or on the basis of the first lighting mode, the current of the lamp beads is reduced (for overexposure analysis results) or increased (for underexposure analysis results).

[0047] Based on the imaging quality analysis results of the target barcode candidate area, if local image restoration is used or if image-level refinement is still required after intelligent fill-light optimization, at least one restoration operation selected from image enhancement (such as CLAHE, Contrast Limited Adaptive Histogram Equalization), multi-exposure fusion, or texture reconstruction based on generative adversarial networks (GANs) is performed on the target barcode candidate area to improve image clarity. Specifically, image enhancement uses the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm, and texture reconstruction based on generative adversarial networks uses a U-Net architecture as the generator network and a PatchGAN architecture as the discriminator network.

[0048] Based on the image quality analysis results of the target barcode candidate area, when using autofocus, an autofocus optimization operation is performed on the target barcode candidate area to improve its image clarity, and the image is recaptured. The autofocus optimization utilizes an iterative focusing mechanism based on feedback from an image clarity evaluation function (e.g., Laplace operator variance). It is understood that when both intelligent fill light optimization and autofocus strategies are required, only the recapture step is required.

[0049] This application is applicable to the needs of simultaneous recognition of multiple barcodes on multiple objects of different sizes (such as medical supply packaging, industrial component labels) in scenarios such as pharmaceutical traceability, logistics management, and industrial automation. Figure 4-5 As shown, the method and system are applied to a high-definition camera. The camera has an image acquisition module 100, which includes a camera 200 for capturing images. Surrounding the camera 200 are several LEDs 300 for fill light. These LEDs 300 form a ring-shaped adjustable light source array module. During use, multiple objects are placed under the image acquisition module 100, with the barcodes on the objects facing the camera 200. This allows for simultaneous recognition of multiple barcodes.

[0050] The multi-barcode recognition system provided by the present invention is described below. The multi-barcode recognition system described below and the multi-barcode recognition method described above can be referenced to each other.

[0051] On the other hand, the present invention also provides a multi-barcode recognition method and system, comprising:

[0052] An image acquisition module, configured to acquire a target image containing at least one region that can be segmented into independent barcode candidate regions;

[0053] The annular adjustable light source array module is arranged around the outside of the image acquisition module and is used to provide fill light when the image acquisition module acquires the target image. Its lighting mode is dynamically adjusted by the control command;

[0054] The automatic focusing module is linked with the image acquisition module to adjust the imaging focal length;

[0055] Environmental perception module, used to obtain the material and ambient lighting parameters of the target object;

[0056] The processing module is connected to the image acquisition module, the annular adjustable light source array module, the automatic focusing module and the environmental perception module. The processing module is configured with a memory and a computing core for executing the multi-barcode recognition method described above, which will not be described in detail here.

[0057] This application effectively avoids repeated optimization and decoding operations on successfully decoded barcodes by assigning a unique identifier to each barcode and dynamically tracking its processing status (from unprocessed, pending optimization, to successfully decoded or decoded failed), significantly improving the overall processing efficiency of the system.

[0058] At the same time, this application only initiates targeted lighting adjustment, image restoration, or focus optimization for barcodes that fail to decode or have poor image quality, avoiding redundant "one-size-fits-all" processing for all barcodes and enabling a more reasonable allocation of computing and time resources.

[0059] Furthermore, when optimizing lighting for specific barcodes, we not only strive to improve the imaging quality of the target barcode, but also introduce an evaluation mechanism for the potential impact on other barcodes, striving to find the overall optimal or acceptable recognition effect in multi-code scenarios, reducing the problem of deterioration in other areas due to local optimization.

[0060] In summary, this application integrates multiple technical means such as active intelligent fill light, automatic fine focus, and diversified image restoration (from rapid enhancement to deep learning reconstruction), and enables them to work together under the dynamic state management framework, significantly enhancing the system's success rate and robustness in multiple barcode recognition under various complex lighting, material reflection and imaging conditions.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A multi-barcode recognition system, characterized in that: include: An image acquisition module, configured to acquire a target image containing at least one region that can be segmented into independent barcode candidate regions; A processing module is configured with a memory and a computing core. The processing module is used to locate and segment each barcode candidate area in the target image and assign a unique identifier to it; preliminarily decode each barcode candidate area with a unique identifier and record its processing status based on the decoding result, wherein the processing status includes successfully decoded or to be optimized; establish an optimization queue for the barcode candidate areas in the optimization state, select the target barcode candidate area from the optimization queue according to a predetermined strategy and process it until the optimization queue is empty or a preset global termination condition is met.

2. The multi-barcode recognition system according to claim 1, characterized in that: The processing status also includes decoding confidence. When the decoding is successful but the confidence is lower than a preset threshold, the barcode candidate area is also marked as to be optimized.

3. The multi-barcode recognition system according to claim 2, characterized in that: It also includes an annular adjustable light source array module, which is arranged around the outside of the image acquisition module and is used to provide fill light when the image acquisition module acquires the target image. Its lighting mode is dynamically adjusted by the control instructions issued by the processing module.

4. The multi-barcode recognition system according to claim 3, characterized in that: It also includes an automatic focusing module, which is linked with the image acquisition module to adjust the imaging focal length.

5. The multi-barcode recognition system according to claim 4, characterized in that: In the first illumination mode of the annular adjustable light source array module, the image acquisition module acquires the target image for the first time.

6. The multi-barcode recognition system according to claim 5, characterized in that: The step of selecting and processing target barcode candidate regions from the queue to be optimized according to a predetermined strategy includes: Select a barcode candidate area from the queue to be optimized as the target barcode candidate area; Perform imaging quality analysis on the selected target barcode candidate area, and process the target barcode candidate area using one or more combination strategies of intelligent fill light optimization, local image restoration, and automatic focusing based on the analysis results; The processed target barcode candidate area is decoded and its processing status is updated according to the decoding result.

7. The multi-barcode recognition system according to claim 6, characterized in that: When adopting the intelligent fill light optimization strategy, images are recaptured under the second illumination mode of the annular adjustable light source array module, and the potential imaging impact on the non-target barcode candidate area in the field of view under this illumination mode is evaluated, and the processing status of the affected non-target barcode candidate area is selectively updated based on the evaluation results; wherein, if the evaluation results show that the adjustment of the illumination mode has a significant negative imaging impact on the non-target barcode candidate area that has been successfully decoded before, the processing status of the non-target barcode candidate area is updated to be optimized and re-added to the queue for optimization.

8. The multi-barcode recognition system according to claim 6, characterized in that: When local image restoration is used, or when image-level refinement processing is still required after intelligent fill light optimization, at least one restoration operation selected from image enhancement, multi-exposure fusion, or texture reconstruction based on a generative adversarial network is performed on the target barcode candidate area.

9. The multi-barcode recognition system according to claim 6, wherein: When automatic focusing is adopted, an automatic focusing optimization operation is performed on the target barcode candidate area.

10. The multi-barcode recognition system according to claim 1, wherein: The processing module is further configured to aggregate and output decoding information of all barcode candidate regions corresponding to all unique identifiers and having a final processing status of successfully decoded.

Images