A code reading method, device, equipment, system and storage medium

Abstract

Embodiments of the present application provide a code reading method, device, equipment, system and storage medium. The method comprises: acquiring an image to be identified collected by a first image collection device; detecting whether a feature pattern exists in the current image to be identified, wherein the feature pattern is used to represent the existence of a code in the image; in the case that the feature pattern exists in the current image to be identified, judging whether the feature pattern meets a preset identification condition; and in the case that the feature pattern meets the preset identification condition, identifying the code in the image to be identified collected by the first image collection device to obtain an identification result. Since the code in the image is identified only after the feature pattern representing the existence of the code is detected and it is confirmed that the feature pattern meets the preset identification condition, the accuracy of code identification is improved, and the power loss caused by invalid identification is avoided.

Description

Technical Field

[0001] This application relates to the field of machine vision technology, and in particular to a code reading method, apparatus, device, system and storage medium. Background Technology

[0002] Currently, in many applications, barcodes or QR codes containing product data are printed or affixed to products for recording and traceability. Subsequently, the product data is obtained by scanning the barcode or QR code for recording and traceability.

[0003] For example, medicine packaging typically includes barcodes or QR codes. When dispensing medicine, each item is scanned to identify its specific data, allowing for recording and traceability. Similarly, in logistics, barcodes or QR codes are affixed to the outer packaging of goods to be mailed. When delivering goods, each item is scanned to identify its specific data, enabling recording and traceability.

[0004] In related technologies, the recognition of barcodes or QR codes mainly employs the following two methods:

[0005] The first method involves measuring distance. When a change in distance is detected between the camera and the platform being measured, the barcode or QR code is identified.

[0006] This method, because it only measures distance, cannot determine whether the distance change is caused by the target object. Therefore, if a non-target object (such as a hand, a container, etc.) accidentally enters the measured platform and causes a distance change, it will still trigger the recognition of the barcode or QR code, resulting in recognition errors and insufficient accuracy.

[0007] The second method involves determining whether barcode or QR code recognition is triggered by observing changes in the brightness of the captured image.

[0008] This method is highly dependent on light; in addition, when the product is small, the change in brightness is not obvious, which may lead to recognition errors and insufficient recognition accuracy. Summary of the Invention

[0009] The purpose of this application is to provide a code reading method, apparatus, device, system, and storage medium to improve the accuracy of code recognition. The specific technical solution is as follows:

[0010] In a first aspect, embodiments of this application provide a code reading method, the method comprising:

[0011] Acquire the image to be identified captured by the first image acquisition device;

[0012] For the acquired current image to be identified, detect whether there is a feature pattern in the current image to be identified, wherein the feature pattern is used to characterize the presence of a code in the image;

[0013] If the feature pattern exists in the current image to be identified, determine whether the feature pattern meets the preset recognition conditions;

[0014] When the feature pattern meets the preset recognition conditions, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain the recognition result.

[0015] Optionally, determining whether the feature image meets the preset recognition conditions includes:

[0016] Based on the confidence level and / or motion state of the feature patterns in the current image to be identified, it is determined whether the feature patterns meet the preset conditions.

[0017] Optionally, determining whether the feature pattern meets preset conditions based on the confidence level and / or motion state of the feature pattern in the current image to be identified includes:

[0018] Determine whether the confidence level of the feature pattern in the current image to be identified is greater than a preset confidence threshold;

[0019] If the confidence level of the feature graphic is greater than the preset confidence threshold, the positions of the feature graphics in the subsequent multiple images to be identified are continuously obtained;

[0020] Based on the position of the feature graphic in the image to be identified, determine the positional changes of the feature graphic in multiple consecutive images to be identified;

[0021] Based on the positional change, if the positional change in multiple consecutive images to be identified is less than a preset change threshold, the feature graphic is determined to meet the preset recognition conditions.

[0022] Optionally, if multiple feature patterns exist in the current image to be recognized,

[0023] For each feature graphic, the following steps are performed: determining whether the confidence level of the feature graphic in the current image to be identified is greater than a preset confidence threshold; if the confidence level of the feature graphic is greater than the preset confidence threshold, continuously acquiring the position of the feature graphic in multiple subsequent images to be identified; based on the position of the feature graphic in the images to be identified, determining the position change of the feature graphic in multiple consecutive images to be identified; and, based on the position change, determining that the feature graphic meets the preset recognition conditions if the position change amount in multiple consecutive images to be identified is less than a preset change amount threshold.

[0024] When the feature graphic satisfies the preset recognition conditions, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain a recognition result, including:

[0025] For each feature pattern, if the feature pattern meets the preset recognition conditions, the code corresponding to the feature pattern in the image to be recognized acquired by the first image acquisition device is recognized to obtain the recognition result of the code.

[0026] Optionally, the step of continuously acquiring the position of the feature graphic in the subsequent multiple images to be identified; determining the position change of the feature graphic in the consecutive multiple images to be identified based on the position of the feature graphic in the images to be identified; and determining that the feature graphic meets the preset recognition condition according to the position change, if the position change in the consecutive multiple images to be identified is less than a preset change threshold, includes:

[0027] The system continuously acquires the positions of the bounding boxes of feature graphics in multiple subsequent images to be identified; based on the positions of the bounding boxes of feature graphics in the images to be identified, it determines the position changes of the bounding boxes of feature graphics in multiple consecutive images to be identified; and based on the position changes, if the position change of the bounding boxes in multiple consecutive images to be identified is less than a preset change threshold, it determines that the feature graphics meet preset recognition conditions.

[0028] Optionally, if the positional change in the multiple consecutive images to be identified is less than the preset change threshold, the method further includes:

[0029] Turn on the light source to provide supplemental lighting for the field of view of the first image acquisition device;

[0030] The step of recognizing the code in the image to be recognized acquired by the first image acquisition device to obtain the recognition result includes:

[0031] The code in the image to be identified acquired by the first image acquisition device under supplementary lighting conditions is identified to obtain the identification result, and the light source is controlled to be turned off after obtaining the identification result.

[0032] Optionally, the step of recognizing the code in the image to be recognized acquired by the first image acquisition device to obtain the recognition result includes:

[0033] Based on the position of the feature pattern in the consecutive multiple images to be identified, the target position of the code in the image to be identified acquired by the first image acquisition device is determined;

[0034] The target location in the image to be identified acquired by the first image acquisition device is identified to obtain the identification result.

[0035] Optionally, when the feature pattern is the code itself, determining the target position of the code in the image to be recognized acquired by the first image acquisition device based on the position of the feature pattern in the consecutive multiple images to be recognized includes:

[0036] The position of the feature graphic in the consecutive multiple images to be identified is determined as the target position of the code in the image to be identified acquired by the first image acquisition device;

[0037] When the feature graphic is an identification pattern located on the same target object as the code and having a fixed relative positional relationship with the code; determining the target position of the code in the image to be identified acquired by the first image acquisition device based on the position of the feature graphic in the consecutive multiple images to be identified includes:

[0038] Based on the position of the feature graphic in the consecutive multiple images to be identified, and the fixed relative positional relationship between the code and the identification pattern, the target position of the code in the image to be identified acquired by the first image acquisition device is determined.

[0039] Optionally, detecting whether a feature pattern exists in the current image to be identified includes:

[0040] Detect whether there are feature patterns in the preset region of interest in the current image to be identified.

[0041] Optionally, the method further includes:

[0042] The system receives configuration information of a feature graphic sent by a host computer to determine whether the feature graphic is a code or an identification pattern. The configuration information of the feature graphic is configured by the user in the interface of the host computer and includes a code or identification pattern.

[0043] Optionally, before acquiring the image to be identified acquired by the first image acquisition device, the method further includes:

[0044] The motion path of the feature image is obtained, and based on the motion path, at least one first image acquisition device that can subsequently acquire the feature image is predicted;

[0045] The control ensures that images are continuously acquired solely by the at least one first image acquisition device.

[0046] The acquisition of the image to be identified by the first image acquisition device includes:

[0047] The image to be identified is acquired by the at least one first image acquisition device.

[0048] Optionally, obtaining the motion path of the feature image includes:

[0049] Acquire the image to be detected captured by the second image acquisition device;

[0050] If the feature pattern exists in the acquired image to be detected, the motion path of the feature pattern is determined based on the acquired multiple images to be detected.

[0051] Optionally, the control to continue acquiring images solely by the at least one first image acquisition device includes:

[0052] Determine the time when the feature graphic arrives at the at least one first image acquisition device;

[0053] Based on the determined time, control is exercised to allow only the at least one first image acquisition device to continue acquiring images.

[0054] Optionally, when the feature graphic satisfies the preset recognition conditions, the step of recognizing the code in the image to be recognized acquired by the first image acquisition device to obtain the recognition result includes:

[0055] When the feature graphic satisfies the preset recognition conditions, the code in the current image to be recognized is recognized to obtain the recognition result;

[0056] or,

[0057] If the feature image satisfies the preset recognition conditions, it is detected whether there is a code in the current image to be recognized. If there is a code in the current image to be recognized, the code in the image to be recognized obtained after the current image to be recognized is recognized to obtain the recognition result.

[0058] Secondly, embodiments of this application provide a code reading device, the device comprising:

[0059] The image acquisition module is used to acquire the image to be recognized acquired by the first image acquisition device;

[0060] The feature pattern detection module is used to detect whether there are feature patterns in the acquired current image to be identified, wherein the feature patterns are used to characterize the presence of codes in the image;

[0061] The judgment module is used to determine whether the feature pattern meets the preset recognition conditions when the feature pattern exists in the current image to be recognized.

[0062] The code recognition module is used to recognize the code in the image to be recognized acquired by the first image acquisition device when the feature graphic meets the preset recognition conditions, and to obtain the recognition result.

[0063] Optionally, the determination module is specifically used for:

[0064] Based on the confidence level and / or motion state of the feature patterns in the current image to be identified, determine whether the feature patterns meet the preset conditions.

[0065] Optionally, the determination module includes:

[0066] The confidence judgment unit is used to determine whether the confidence of the feature pattern in the current image to be recognized is greater than the preset confidence threshold.

[0067] The position change determination unit is used to continuously acquire the position of the feature graphic in the image to be identified in multiple subsequent images when the confidence of the feature graphic is greater than a preset confidence threshold, and determine the position change of the feature graphic in multiple consecutive images to be identified based on the position of the feature graphic in the image to be identified.

[0068] The position change judgment unit is used to determine that the feature graphic meets the preset recognition conditions when the position change amount in multiple consecutive images to be recognized is less than a preset change amount threshold.

[0069] Optionally, when multiple feature patterns exist in the current image to be identified, the judgment module is specifically used for:

[0070] For each feature graphic, the steps described above are performed: determining whether the confidence level of the feature graphic in the current image to be identified is greater than a preset confidence threshold; if the confidence level of the feature graphic is greater than the preset confidence threshold, continuously acquiring the position of the feature graphic in multiple subsequent images to be identified; based on the position of the feature graphic in the images to be identified, determining the position change of the feature graphic in multiple consecutive images to be identified; and based on the position change, if the position change in multiple consecutive images to be identified is less than a preset change threshold, determining that the feature graphic meets the preset recognition conditions.

[0071] The code recognition module is specifically used for:

[0072] For each feature graphic, if the feature graphic meets the preset recognition conditions, the code corresponding to the feature graphic in the image to be recognized acquired by the first image acquisition device is recognized to obtain the recognition result of the code.

[0073] Optionally, the determination module is specifically used for:

[0074] The system continuously acquires the positions of the bounding boxes of the feature graphics in multiple subsequent images to be recognized. Based on the positions of the bounding boxes of the feature graphics in the images to be recognized, it determines the position changes of the bounding boxes of the feature graphics in multiple consecutive images to be recognized. According to the position changes, if the position change of the bounding boxes in multiple consecutive images to be recognized is less than a preset change threshold, it determines that the feature graphics meet the preset recognition conditions.

[0075] Optionally, the device further includes:

[0076] The light source control module is used to control the light source to turn on when the positional change in the multiple consecutive images to be identified is less than the preset change threshold, so as to provide supplemental lighting for the acquisition field of the first image acquisition device.

[0077] The code recognition module is specifically used for:

[0078] The code in the image to be identified acquired by the first image acquisition device under supplementary lighting conditions is identified to obtain the identification result;

[0079] The light source control module is also used to control the light source to turn off after the code recognition module obtains the recognition result.

[0080] Optionally, the code recognition module includes:

[0081] The target location determination unit determines the target location of the code in the image to be identified acquired by the first image acquisition device based on the position of the feature graphic in the consecutive multiple images to be identified.

[0082] The recognition unit is used to recognize the target location in the image to be recognized acquired by the first image acquisition device and obtain the recognition result.

[0083] Optionally, the target location determination unit is specifically used for:

[0084] When the feature graphic is the code itself, the position of the feature graphic in the consecutive multiple images to be identified is determined as the target position of the code in the image to be identified acquired by the first image acquisition device.

[0085] When the feature graphic is an identification pattern located on the same target object as the code and having a fixed relative positional relationship with the code; based on the position of the feature graphic in the consecutive multiple images to be identified, and the fixed relative positional relationship between the code and the identification pattern, the target position of the code in the image to be identified acquired by the first image acquisition device is determined.

[0086] Optionally, the feature image detection module is specifically used for:

[0087] Detect whether there are feature patterns in the preset region of interest in the current image to be identified.

[0088] Optionally, the device further includes:

[0089] The configuration information receiving module is used to receive the configuration information of the feature graphic sent by the host computer, so as to determine whether the feature graphic is a code or an identification pattern. The configuration information of the feature graphic is configured by the user in the interface of the host computer, and the configuration information of the feature graphic includes: code or identification pattern.

[0090] Optionally, the device further includes:

[0091] The device prediction module is used to acquire the motion path of the feature image and, based on the motion path, predict at least one first image acquisition device that can subsequently acquire the feature image.

[0092] The control module is used to control that images are continuously acquired only by at least one first image acquisition device;

[0093] The image acquisition module is specifically used for:

[0094] Acquire at least one image to be identified from a first image acquisition device.

[0095] Optionally, the device prediction module is specifically used for:

[0096] The system acquires the image to be detected from the second image acquisition device. If a feature graphic exists in the acquired image to be detected, the system determines the motion path of the feature graphic based on the acquired multiple images to be detected.

[0097] Optionally, the control module is specifically used for:

[0098] The time when the feature graphic arrives at at least one first image acquisition device is determined, and based on the determined time, the system controls the acquisition of images to continue only by at least one first image acquisition device.

[0099] Optionally, the code recognition module is specifically used for

[0100] When the feature graphic satisfies the preset recognition conditions, the code in the current image to be recognized is recognized to obtain the recognition result;

[0101] or,

[0102] If the feature image satisfies the preset recognition conditions, it is detected whether there is a code in the current image to be recognized. If there is a code in the current image to be recognized, the code in the image to be recognized obtained after the current image to be recognized is recognized to obtain the recognition result.

[0103] Thirdly, embodiments of this application provide a code reading device, including: a base, a controller, an image acquisition device, and a support frame;

[0104] The base has a scanning platform for placing the object to be scanned, and the support frame is disposed on one side of the base;

[0105] The image acquisition device is mounted on top of the support frame, such that the image acquisition device faces the scanning table surface;

[0106] The controller is located inside the base and communicates with the image acquisition device to execute any of the code reading methods described in the first aspect above.

[0107] Fourthly, embodiments of this application provide a code reading system, including: a conveyor belt for placing the object to be scanned, a controller, and multiple image acquisition devices;

[0108] The plurality of image acquisition devices are located above the conveyor belt and are positioned facing the conveyor belt;

[0109] The controller communicates with the plurality of image acquisition devices to execute the code reading method described in any of the first aspects above.

[0110] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the code reading methods described in the first aspect above.

[0111] Beneficial effects of the embodiments in this application:

[0112] In the solution provided in this application embodiment, an image to be recognized is acquired by a first image acquisition device; for the acquired current image to be recognized, it is detected whether there is a feature pattern in the current image to be recognized, wherein the feature pattern is used to represent the presence of a code in the image; if there is a feature pattern in the current image to be recognized, it is determined whether the feature pattern meets a preset recognition condition; if the feature pattern meets the preset recognition condition, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain a recognition result. Since the code in the image is only recognized after the presence of a feature pattern representing the presence of a code in the image is detected and the preset recognition condition is confirmed, the accuracy of code recognition is improved, and the power consumption loss caused by invalid recognition is avoided.

[0113] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0114] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0115] Figure 1 A flowchart illustrating a code reading method provided in an embodiment of this application;

[0116] Figure 2 For based on Figure 1 A schematic diagram of the first specific process of the code reading method in the illustrated embodiment;

[0117] Figure 3 For based on Figure 2 A schematic diagram of a specific process for the code reading method in the embodiment shown;

[0118] Figure 4 For based on Figure 1 A schematic diagram of a second specific process for the code reading method in the illustrated embodiment;

[0119] Figure 5 For based on Figure 3 A schematic diagram of a specific process for the code reading method in the embodiment shown;

[0120] Figure 6 For based on Figure 5 A schematic diagram showing the position of the feature graphic's positioning box in each image according to the embodiment shown;

[0121] Figure 7 For based on Figure 1 A schematic diagram of the third specific process of the code reading method in the illustrated embodiment;

[0122] Figure 8 For based on Figure 3 Another specific flowchart of the code reading method in the illustrated embodiment is shown.

[0123] Figure 9 For based on Figure 1 A schematic diagram of the fourth specific process of the code reading method in the illustrated embodiment;

[0124] Figure 10 This is a schematic diagram of the structure of a code reading device provided in an embodiment of this application;

[0125] Figure 11 This is a schematic diagram of the structure of a code reading device provided in an embodiment of this application;

[0126] Figure 12 This is a schematic diagram of the structure of a code reading system provided in an embodiment of this application. Detailed Implementation

[0127] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0128] To improve the accuracy of code recognition, embodiments of this application provide a code reading method, apparatus, device, system, computer-readable storage medium, and computer program product. The code reading method provided in this application embodiment will be described first below.

[0129] The code reading method provided in this application can be applied to a controller that communicates with an image acquisition device. The controller can be a processor, a computer, a server, etc., and is not specifically limited here. For clarity, it will be referred to as a controller below. It should be noted that the code to be identified in this application can be a one-dimensional barcode, a two-dimensional barcode, or other identifiable codes including content, and is not specifically limited here.

[0130] like Figure 1 As shown, the above code reading method may include:

[0131] S101, acquire the image to be recognized acquired by the first image acquisition device.

[0132] S102, For the acquired current image to be identified, detect whether there is a feature pattern in the current image to be identified. If the detection result is yes, that is, there is a feature pattern in the current image to be identified, then proceed to step S103.

[0133] Among them, feature graphics are used to characterize the presence of codes in an image.

[0134] S103, determine whether the feature image meets the preset recognition conditions. If the determination result is yes, that is, the feature image meets the preset recognition conditions, then proceed to step S104.

[0135] S104, the code in the image to be identified acquired by the first image acquisition device is identified to obtain the identification result.

[0136] In the solution provided in this application embodiment, an image to be recognized is acquired by a first image acquisition device; for the acquired current image to be recognized, it is detected whether there is a feature pattern in the current image to be recognized, wherein the feature pattern is used to represent the presence of a code in the image; if there is a feature pattern in the current image to be recognized, it is determined whether the feature pattern meets a preset recognition condition; if the feature pattern meets the preset recognition condition, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain a recognition result. Since the code in the image is only recognized after the presence of a feature pattern representing the presence of a code in the image is detected and the preset recognition condition is confirmed, the accuracy of code recognition is improved, and the power consumption loss caused by invalid recognition is avoided.

[0137] The first image acquisition device can acquire images of the scene to be recognized (such as a scanning table, conveyor belt, etc.). During this process, the controller can obtain the images acquired by the first image acquisition device through communication with it, and use them as the images to be recognized. The first image acquisition device can acquire images continuously or at certain time intervals, without specific limitations.

[0138] After obtaining the image to be recognized, the controller can detect whether there are feature patterns in the acquired image to determine whether there is a code in the image. Specifically:

[0139] On the one hand, the presence of a code in an image directly indicates its existence. On the other hand, because codes on objects are often accompanied by specific identifying patterns, such as logos or sharp edges, the presence of these patterns in an image can also indicate the possible presence of a code. Furthermore, since the code is located on an object, the presence of that object in an image can also indicate the possible presence of a code. Based on this, the code itself, the accompanying identifying patterns, or the object on which the code is located can be considered as feature graphics. By detecting the presence of these feature graphics in an image, the presence of a code can be determined.

[0140] Therefore, the controller can extract features from the acquired image to be identified, and then match the extracted image features with the features of the feature graphic to determine whether there is a part in the image to be identified that is similar to the feature graphic, thereby detecting whether there is a feature graphic in the image to be identified. The process of "detecting whether there is a feature graphic in the image to be identified" can be implemented by a pre-trained deep learning model, a pre-trained neural network, or an existing image recognition algorithm, without being specifically limited here.

[0141] If there are no feature patterns in the current image to be recognized, then it is highly likely that there is no code to be recognized in the current image. In this case, the controller can skip the subsequent code reading steps for the current image to be recognized, thereby improving processing efficiency.

[0142] If a feature pattern exists in the current image to be recognized, then it is highly likely that the image contains the code that needs to be recognized. In this case, to further improve the accuracy of code recognition, the controller can further filter out those images that truly meet the conditions for code recognition. Specifically:

[0143] Simply detecting the presence of a feature pattern does not guarantee accurate code recognition. This is because there may be interfering factors in the image that prevent it from meeting the requirements for accurate code recognition. For example, although a feature pattern may be detected, its confidence level may be low. Alternatively, the feature pattern may be in continuous motion, causing motion blur in the image. Or, the feature pattern may be located at the edge of the image, potentially causing distortion.

[0144] Therefore, after detecting feature patterns from the current image to be recognized, the controller can determine whether the feature patterns in the current image to be recognized meet the preset recognition conditions, thereby determining whether the code recognition conditions are met.

[0145] When determining whether a feature graphic meets preset conditions, the controller can base its judgment on the confidence level and / or motion state of the feature graphic. The specific judgment method will be described in detail in subsequent embodiments and will not be repeated here. Alternatively, based on the specific application scenario and the characteristics of the feature graphic, the controller can choose to determine whether the feature graphic meets the preset conditions based on other types of information. This is not specifically limited here. For example, the controller can determine whether the feature graphic meets the preset conditions based on its clarity and / or completeness, and determine whether the preset conditions are met if the feature graphic is relatively clear and / or relatively complete.

[0146] When the feature image meets the preset recognition conditions through the above steps, it indicates that the conditions for accurate code recognition are met. In this case, the controller can recognize the code in the image to be recognized acquired by the first image acquisition device and obtain the recognition result. Specifically:

[0147] In the first embodiment, when the controller determines that the feature pattern in the current image to be identified meets the preset recognition conditions, it can identify the code in the current image to be identified and obtain the recognition result.

[0148] In the second embodiment, if the controller determines that the feature patterns in the current image to be recognized meet the preset recognition conditions, it can detect whether a code exists in the current image to be recognized. If a code exists in the current image to be recognized, the controller can recognize the code in images to be recognized acquired after the current image to be recognized, and obtain a recognition result. The method for detecting whether a code exists in the current image to be recognized can be to match the features of a preset code with the image features of the current image to determine whether a code exists in the current image to be recognized, or it can be to decode the current image to be recognized and determine whether a code exists in the current image to be recognized based on whether the decoding is successful. No specific limitation is made here regarding the method for detecting whether a code exists in the current image to be recognized.

[0149] To further improve the accuracy of code recognition, the controller can turn on the light source to illuminate the scene after detecting the presence of a code in the current image to be recognized. Furthermore, for images acquired after the current image, the supplementary lighting enhances contrast, making the difference between the code and the surrounding background more pronounced and revealing the code's outline and lines more clearly. This facilitates the recognition algorithm's accurate extraction of the code's feature information, thus further improving recognition accuracy. To save power and reduce light pollution, the controller can also turn off the light source after receiving the recognition result.

[0150] Furthermore, if the detection result obtained by the controller in performing step S102 is negative or the judgment result obtained by performing step S103 is negative, that is, if there is no feature graphic in the current image to be identified or if the feature graphic does not meet the preset recognition conditions, the controller can update the current image to be identified to the next image to be identified and return to step S102.

[0151] For example, the first image acquisition device sequentially acquires image 1 and image 2 to be identified. When the controller acquires image 1 to be identified, it can determine image 1 to be identified as the current image to be identified and execute steps S102-S104. During this process, if the detection result obtained by executing step S102 is negative or the judgment result obtained by executing step S103 is negative, the controller can update the current image to be identified to image 2 and then execute steps S102-S104 based on image 2 to be identified.

[0152] In the solution provided in this application embodiment, the controller only identifies the code in the image after detecting the presence of a feature pattern representing the presence of a code and confirming that the feature pattern meets preset recognition conditions. This improves the accuracy of code recognition. Furthermore, when supplementary lighting is required for code reading, the controller in this application embodiment only controls the light source to turn on for code reading when the feature pattern meets the preset recognition conditions, or when the feature pattern meets the preset recognition conditions and a code exists in the image. Compared to existing methods that control the light source to turn on for code reading based on changes in detection distance or brightness, this reduces power consumption and light pollution.

[0153] As one implementation method of this application, such as Figure 2 As shown, when the controller executes step S103 above, it can be specifically implemented through step S201:

[0154] S201, based on the confidence level and / or motion state of the feature graphic in the current image to be identified, determine whether the feature graphic meets the preset conditions. If the determination result is yes, that is, the feature graphic meets the preset conditions, then execute step S104.

[0155] On the one hand, when the controller detects a feature pattern in the current image to be identified, it can further output the confidence level of the feature pattern. The confidence level of the feature pattern represents the degree of credibility of the currently detected feature pattern as a pre-set standard feature pattern. It can be in the form of rating, rating, score, percentage, etc. The quantification method of confidence level is not limited here.

[0156] On the other hand, in practical application scenarios, objects often enter the field of view of the first image acquisition device by means of manual handling or conveyor belt transport. Therefore, when the controller detects the presence of feature graphics in the current image to be identified, it can further determine the motion state of the feature graphics based on each image to be identified acquired by the first image acquisition device. The motion state of the feature graphics may include the position of the feature graphics in each image to be identified.

[0157] Correspondingly, during image acquisition, factors such as lighting and shooting angle may affect the feature patterns detected by the controller, leading to issues like blurring or deviation, and consequently, lower confidence levels for the feature patterns. Furthermore, the rapid movement of feature patterns during image acquisition can cause them to appear as ghosting or blurring in the image to be recognized, and feature patterns may be located at the edge of the field of view of the first image acquisition device, resulting in distortion in the image to be recognized.

[0158] In this situation, directly performing subsequent code recognition based on these low-confidence feature images, fast-moving feature images, or feature images located at the edge of the field of view is likely to yield incorrect recognition results. Therefore, to further improve the accuracy of code recognition, the controller can determine whether the feature images meet preset conditions based on the confidence level and / or motion state of the feature images in the current image to be recognized. Specifically:

[0159] Regarding the confidence level of the feature image, the controller can determine whether the confidence level of the detected feature image is greater than the preset confidence threshold. If the confidence level of the detected feature image is greater than the preset confidence threshold, it means that the currently detected feature image is a pre-set standard feature image with a high degree of credibility, and it can reliably characterize the presence of a code in the image and meet the conditions for accurate code recognition.

[0160] The confidence threshold can be preset by the user based on the characteristics of the feature images and the number of feature images to be identified simultaneously. For example, when the features of the feature images are relatively clear and simple, the user can set a higher confidence threshold; when the number of feature images to be identified simultaneously is relatively small, the user can also set a higher confidence threshold.

[0161] Regarding the motion state of the feature image, the controller can determine whether the detected feature image is approaching stillness ("approaching stillness" here does not necessarily mean completely still, but can also mean that the speed of the feature image is less than a certain threshold), and / or whether the detected feature image is located in the central region of the image. If the detected feature image is approaching stillness, and / or the detected feature image is located in the central region of the image, it means that the recognition conditions of the currently detected feature image are good, and it can reliably represent the presence of a code in the image and has the conditions for accurate code recognition.

[0162] Therefore, when the controller determines that the feature graphic meets the preset conditions based on the confidence level and / or motion state of the feature graphic in the current image to be identified, it can execute step S104 to identify the code in the current image to be identified and obtain the identification result.

[0163] It should be noted that when the judgment result of step S201 is negative, the controller can determine that the feature pattern in the current image to be recognized does not meet the preset recognition conditions.

[0164] In the solution provided in this application embodiment, merely detecting the existence of a feature image is insufficient to ensure accurate code recognition. Further judgment is made based on the confidence level and / or motion state of the feature image. Subsequent code recognition operations are only performed when a feature image exists and its confidence level and / or motion state meet the requirements. This reduces misidentification, avoids invalid captures, and better ensures the accuracy of code recognition.

[0165] As one implementation method of this application, such as Figure 3 As shown, when the controller executes step S201 above, it can be specifically implemented through steps S301-S303:

[0166] S301, determine whether the confidence level of the feature image in the current image to be identified is greater than a preset confidence threshold. If the determination result is yes, that is, the confidence level of the feature image is greater than the preset confidence threshold, then proceed to step S302.

[0167] After the controller detects the presence of a feature pattern in the current image to be identified, it can determine whether the confidence level of the detected feature pattern is greater than a preset confidence threshold. If the confidence level of the detected feature pattern is greater than the preset confidence threshold, it means that the current detected feature pattern is a pre-set standard feature pattern with a high degree of credibility. It can reliably characterize the presence of a code in the image and has the conditions for accurate code recognition. Then the controller can execute step S202.

[0168] S302, continuously acquire the position of the feature graphic in the image to be identified in multiple subsequent images to be identified, and determine the position change of the feature graphic in multiple consecutive images to be identified based on the position of the feature graphic in the image to be identified.

[0169] In practical applications, objects often enter the field of view of the first image acquisition device by manual handling or conveyor belt transport. If the code on the object is recognized while it is moving, misidentification is likely. Therefore, if the controller determines that the confidence level of the feature image is greater than a preset confidence threshold, it can further determine whether the feature image is stable. Only when the feature image is stable can the code on it be recognized. Specifically:

[0170] Because the controller continuously acquires the images to be identified from the first image acquisition device, when the controller determines that there is a feature graphic in the current image to be identified and the confidence level of the feature graphic is greater than the preset confidence threshold, it can detect the feature graphic in the subsequent multiple images to be identified based on the feature information of the feature graphic in the current image to be identified, thereby determining the position of the feature graphic in the multiple images to be identified. Then, based on the difference between the position parameters of the feature graphic in the multiple images to be identified, it further determines the positional change of the feature graphic in the multiple images to be identified.

[0171] For example, the controller determines that feature graphic 'a' exists in image 1 (the current image to be identified), and the confidence level of feature graphic 'a' is greater than a preset confidence threshold. After image 1, the controller continuously acquires images 2, 3, ..., N. Then, based on the feature information of feature graphic 'a' in image 1, the controller can detect feature graphic 'a' in images 2, 3, ..., N respectively, and determine the position of feature graphic 'a' in these multiple images. Finally, based on the determined position, the controller can determine the positional changes of feature graphic 'a' in images 1, 2, ..., N.

[0172] It should be noted that multiple consecutive images to be identified are not necessarily multiple truly consecutive images; there can be at least one frame between these multiple images to be identified.

[0173] For example, if the controller acquires frames 2 through 10 after the current image to be recognized, the controller can extract frames from the acquired images and treat frames 2, 4, 6, 8, and 10 as a series of consecutive images to be recognized.

[0174] For example, after the current image to be identified, the controller acquires images from frame 2 to frame 10. The controller can first evaluate the image quality of these 10 images respectively. If the image quality of frames 2, 4, and 9 is higher than the preset image quality threshold, the controller can treat frames 2, 4, and 9 as multiple consecutive images to be identified.

[0175] S303, based on the position change, determine whether the position change of the feature graphic in multiple consecutive images to be identified is less than a preset change threshold. If the determination result is yes, that is, the position change of the feature graphic in multiple consecutive images to be identified is less than the preset change threshold, then proceed to step S104.

[0176] To determine whether a feature image is stable, a metric (i.e., a preset change threshold) can be set in advance. After obtaining the positional changes of the feature image in multiple consecutive images to be recognized, the controller can determine whether the change in position of the feature image in multiple consecutive images to be recognized is less than the preset change threshold.

[0177] If the positional change of the feature graphic in multiple consecutive images to be identified is less than the preset change threshold, it indicates that the positional change of the feature graphic in the consecutive images is not significant. At this time, it can be considered that the position of the feature graphic is relatively stable and meets the conditions for accurate identification code. Then the controller can execute step S104 to identify the code in the current image to be identified and obtain the identification result.

[0178] The number of consecutive images to be identified can be preset by the user based on the field of view of the image acquisition device and the moving speed of the feature image. For example, with a larger field of view of the image acquisition device, the user can set a larger number of consecutive images to be identified; similarly, with a faster moving speed of the feature image, the user can set a larger number of consecutive images to be identified.

[0179] It should be noted that when the judgment result of step S303 is negative, the controller can determine that the feature pattern in the current image to be recognized does not meet the preset recognition conditions.

[0180] Furthermore, in one embodiment, when the judgment result in step S303 is yes, that is, when the control device determines that the feature pattern is stable, it can control the light source to turn on to supplement the acquisition field of view of the first image acquisition device. Thus, the image to be recognized subsequently acquired by the control device is the image to be recognized acquired by the first image acquisition device under supplementary lighting conditions. The control device then identifies the code in the image to be recognized acquired after supplementary lighting, which can further improve the accuracy of code recognition. Moreover, to save power and reduce light pollution, the controller can further control the light source to turn off after obtaining the recognition result.

[0181] In the solution provided in this application, by analyzing the confidence level and positional changes of the feature graphics, it is ensured that the code is only recognized when the feature graphics have both high confidence and relatively stable position in continuous images. This effectively avoids code recognition errors caused by inaccurate or unstable feature graphics, further improving the accuracy and reliability of code recognition.

[0182] As one implementation of this application, the image to be identified may contain multiple feature patterns. In this case, such as Figure 4 As shown, when the controller executes steps S103-S104, it can be specifically implemented through steps S401-S402:

[0183] S401, for each feature graphic, determine whether the feature graphic meets the preset recognition conditions. If the determination result is yes, that is, the feature graphic meets the preset recognition conditions, then proceed to step S402.

[0184] In practical application scenarios, multiple objects may enter the field of view of the first image acquisition device at the same time, and each of these multiple objects has a feature pattern. In this case, the controller can determine whether each feature pattern meets the preset conditions individually, and when it is determined that the feature pattern meets the preset recognition conditions, step S402 is executed for the feature pattern.

[0185] When the controller determines whether each feature graphic meets the preset conditions individually, it may perform the above steps S301-S303 for each feature graphic to determine whether the confidence level of the feature graphic is greater than the preset confidence threshold and whether the feature graphic tends to be static, thereby determining whether the feature graphic meets the preset recognition conditions.

[0186] S402, the code corresponding to the feature pattern in the image to be identified acquired by the first image acquisition device is identified to obtain the identification result of the code.

[0187] In practical image recognition applications, it is common to encounter situations where the image to be recognized contains multiple feature patterns, and these feature patterns have a one-to-one correspondence with codes. When the controller determines that a certain feature pattern meets the preset conditions, it means that the code corresponding to that feature pattern is now in a state suitable for recognition, that is, the code corresponding to that feature pattern possesses all the elements that can be accurately recognized.

[0188] For example, if the feature image is a logo, and the preset conditions are "confidence threshold and stillness," then when the confidence level of the logo on the object is greater than the preset confidence threshold, it indicates that the logo on the object is in a good state in the current image to be recognized. Correspondingly, the code on the object will also be in a good state. When the logo on the object is in a still state, it indicates that the object is in a still state, and correspondingly, the code on the object is also in a still state. Therefore, when the logo on the object meets the preset conditions, it means that the code on the object possesses all the elements that can be accurately recognized.

[0189] Therefore, for a feature graphic, when the controller determines that the feature graphic meets the preset conditions, it can determine the code corresponding to the feature graphic in the current image to be recognized, and then recognize the code to obtain the recognition result.

[0190] In the solution provided in this application embodiment, when there are multiple feature graphics in an image, the controller can determine whether each feature graphic meets the preset conditions. When the preset conditions are met, the controller can identify the code corresponding to the feature graphic, thereby achieving refined identification processing. This avoids confusion or omissions caused by the coexistence of multiple feature graphics and effectively ensures that each code can be accurately extracted when it has all the factors required for accurate identification. This provides a reliable data foundation for subsequent code-based applications such as item traceability and data management.

[0191] As one implementation method of this application, such as Figure 5 As shown, when the controller executes the above steps S302-S303, it can be specifically implemented through steps S501-S502:

[0192] S501, continuously acquire the positions of the bounding boxes of the feature graphics in the subsequent multiple images to be recognized, and determine the position changes of the bounding boxes of the feature graphics in the consecutive multiple images to be recognized based on the positions of the bounding boxes of the feature graphics in the images to be recognized.

[0193] When the controller detects the presence of a feature graphic in the current image to be recognized, it can further determine the bounding box of the detected feature graphic. Specifically, when the controller detects the presence of a feature graphic in the current image to be recognized, it can further determine the position information of the feature graphic (such as the center coordinates, boundary coordinates, etc. of the feature graphic), and then determine the bounding box of the feature graphic based on the obtained position information of the feature graphic.

[0194] Accordingly, the controller can determine positional changes based on the bounding boxes of the feature graphics. Specifically, determining positional changes based on the bounding boxes of the feature graphics can be done by determining the position of the center point or vertices of the bounding boxes in each image to be recognized.

[0195] For example, if the controller determines that feature graphic 'a' exists in image 1 (the current image to be identified), and the confidence level of feature graphic 'a' is greater than a preset confidence threshold, and the controller subsequently acquires images 2, 3, ..., 6, the controller can then detect feature graphic 'a' in images 2, 3, ..., 6 based on the feature information of feature graphic 'a' in image 1, and determine the position of the bounding box of feature graphic 'a' in each of these multiple images. Finally, the controller can determine the positional changes of the bounding box of feature graphic 'a' in images 1, 2, ..., 6 based on the position of the center point or vertex of the determined bounding box in the images.

[0196] S502, based on the position change, determine whether the position change of the feature graphic's bounding box in multiple consecutive images to be recognized is less than a preset change threshold. If the determination result is yes, that is, the position change of the feature graphic's bounding box in multiple consecutive images to be recognized is less than the preset change threshold, then proceed to step S104.

[0197] Correspondingly, the controller can determine whether the feature graphic is stable based on the positional change of the positioning box (such as the positional change of the center point or vertex of the positioning box).

[0198] The preset change threshold can be set by the user in advance based on the accuracy of the localization box and the processing frame rate. For example, the user can set a higher preset change threshold when the localization box accuracy is lower, and the user can set a higher preset change threshold when the processing frame rate (i.e., the number of images processed per unit time) is lower.

[0199] Continuing with the example above, suppose the number of "multiple consecutive images to be identified" is 2, the preset change threshold is 2 pixels, and the feature image a (i.e. Figure 6The location of the QR code in the image is shown in the image 1, image 2, ..., image 6. Figure 6 As shown, the positional change of the bounding box of feature graphic a between image 1 and image 2 is 5 pixels, between image 2 and image 3 is 4 pixels, between image 3 and image 4 is 3 pixels, between image 4 and image 5 is 0 pixels, and between image 5 and image 6 is 0 pixels. Since the positional change of the bounding box of feature graphic a in all three consecutive images is less than 2 pixels, it can be determined that feature graphic a meets the preset condition.

[0200] In the solution provided in this application, the controller determines the positioning bounding box of the feature graphic and judges whether the feature graphic meets preset conditions based on the positional changes of the feature graphic's positioning bounding box in various images to be recognized. This allows for accurate determination of whether the feature graphic's position is stable in consecutive images. Code recognition is only performed when the feature graphic's position is relatively stable across multiple consecutive images. This effectively avoids code recognition errors caused by inaccurate or unstable feature graphics, further improving the accuracy and reliability of code recognition.

[0201] As one implementation method of this application, such as Figure 7 As shown, when the controller performs step S102 above, it can achieve the following through step S701:

[0202] S701, for the acquired current image to be recognized, detect whether there is a feature pattern in the preset region of interest in the current image to be recognized. If the detection result is yes, that is, there is a feature pattern in the preset region of interest in the current image to be recognized, then proceed to step S103.

[0203] On the one hand, in the optical system of an image acquisition device, the lens will produce a certain degree of distortion. Generally speaking, the lens distortion is more obvious at the edge of the image. When the code is close to the center of the image, the distortion effect is minimal, and it can maintain a more regular shape. On the other hand, in the scenario where the code is illuminated by a light source, the edge areas of the image may produce shadows or overly bright areas due to the angle of the light, while the central area can obtain more suitable lighting conditions.

[0204] Therefore, the region near the center of the image can be defined as the region of interest. This way, when the controller detects the presence of feature patterns in the current image to be recognized, it no longer needs to perform a full scan of the entire image, but instead focuses on the preset region of interest, i.e., detecting whether feature patterns exist within the preset region of interest in the current image to be recognized. If the detection result is yes, the controller can then proceed to step S103 to determine whether the detected feature patterns meet the preset recognition conditions.

[0205] In the solution provided in this application embodiment, since the controller only detects feature patterns in the region of interest (ROI) of the image, compared to detecting the entire image, unnecessary calculations and processing are reduced. This not only saves computing resources but also avoids problems such as image display interference that may result from excessive detection operations on the entire image. Furthermore, when the feature pattern is located in the ROI of the image, it is less affected by distortion and illumination. Detecting it at this time and triggering the subsequent code reading process can greatly improve the accuracy of the subsequent code reading.

[0206] As one embodiment of this application, when the controller determines through steps S301-S303 that the feature graphic meets the preset recognition conditions, such as Figure 8 As shown, when the controller executes step S104 above, it can be specifically implemented through steps S801-S802:

[0207] S801, based on the position of the feature graphic in multiple consecutive images to be identified, determines the target position of the code in the image to be identified acquired by the first image acquisition device.

[0208] When the controller obtains a judgment result of step S303 (that is, when the controller determines that the position change of the feature graphic in multiple consecutive images to be identified is less than the preset change threshold), it can determine the target position of the code in the image to be identified acquired by the first image acquisition device based on the position of the feature graphic in the multiple consecutive images to be identified.

[0209] The position of the feature graphic in the consecutive images to be identified can be the position of the feature graphic in the last image to be identified, the position of the feature graphic in any image to be identified, or the average position of the feature graphic in the consecutive images to be identified. No specific limitation is made here.

[0210] Regarding the specific method of determining the code position based on the feature graphic position, the controller can implement it through, but is not limited to, the following methods:

[0211] In one implementation, when the feature graphic is the code itself, the controller can directly determine the position of the feature graphic in multiple consecutive images to be recognized as the target position of the code in the image to be recognized acquired by the first image acquisition device.

[0212] For example, if there are multiple consecutive images to be identified, namely image 1, image 2 and image 3, and assuming that the center point of the feature graphic in the positioning box of image 3 is (4,5), then the controller can determine the pixel with coordinates (4,5) in the image to be identified acquired by the first image acquisition device as the center point of the positioning box of the code.

[0213] In another embodiment, when the aforementioned feature graphic is an identification pattern located on the same target object as the code and having a fixed relative positional relationship with the code, since there is a fixed relative positional relationship between the code and the identification pattern—for example, for objects of the same type, the code is always in a specific direction of the identification pattern and has a fixed distance from it—the controller can determine the target position of the code in the image to be recognized acquired by the first image acquisition device based on the position of the detected feature graphic (i.e., the identification pattern) in multiple consecutive images to be recognized and the pre-set fixed relative positional relationship between the identification pattern and the code.

[0214] S802, identify the target location in the image to be identified acquired by the first image acquisition device, and obtain the identification result.

[0215] After the controller determines the target position of the code in the image to be recognized acquired by the first image acquisition device, when performing code recognition on the image to be recognized acquired by the first image acquisition device, it is no longer necessary to perform real-time detection on the image to be recognized to determine the position of the code. Instead, it can directly recognize the target position in the image to be recognized and obtain the recognition result.

[0216] In the solution provided in this application embodiment, when the controller determines the position of the feature graphic in multiple consecutive images to be recognized, it can further determine the position of the code based on the position of the feature graphic. In this way, when the controller recognizes the code in the image to be recognized, it does not need to determine the position of the code by fully detecting the image to be recognized. Instead, it can directly and accurately determine the position of the code in the image to be recognized based on the position of the feature graphic. Then, it can directly obtain the code from the determined position and perform recognition to obtain the result, which can greatly improve the efficiency of code recognition. At the same time, the position of the feature graphic can be used to locate the code more accurately, which can effectively improve the accuracy of code recognition.

[0217] As one embodiment of this application, the controller can communicate with a host computer, and in addition, the controller can perform the following steps:

[0218] Receive configuration information of the feature graphic sent by the host computer to determine whether the feature graphic is a code or an identification pattern.

[0219] In practical applications, users can configure the controller to use either a code or an identification pattern as the feature graphic when executing the code reading method provided in this application, based on their actual needs. For example, when the identification pattern structure on an object is relatively simple and the features are clear, detecting the identification pattern is more accurate and convenient than detecting a code. In this case, the user can configure the feature graphic to be an identification pattern. As another example, when the controller executes the code reading method for a variety of objects, and each object has a different identification pattern, detecting a code is more accurate and convenient than detecting an identification pattern. In this case, the user can configure the feature graphic to be a code.

[0220] Based on the above, users can determine the feature graphic as a code or identifier pattern through configuration operations in the host computer interface to generate configuration information. The configuration operation can involve manually selecting a feature graphic as a code or identifier pattern from the codes and identifier patterns displayed in the host computer interface, manually marking a pattern on an object image displayed in the host computer interface as an identifier pattern, or manually drawing a pattern in the host computer interface as an identifier pattern, etc. The specific form of the configuration operation is not limited here.

[0221] After the host computer generates the configuration information, it can send the configuration information to the controller. The controller can then determine whether the feature image to be detected is a code or a symbolic pattern based on this configuration information. Specifically, when the feature image is a code, the controller's process of "detecting whether a feature image exists in the current image to be recognized and determining whether the feature image meets the preset recognition conditions" can be specifically "detecting whether a code exists in the current image to be recognized and determining whether the code meets the preset recognition conditions." Similarly, when the feature image is a symbolic pattern, the controller's process of "detecting whether a feature image exists in the current image to be recognized and determining whether the feature image meets the preset recognition conditions" can be specifically "detecting whether a symbolic pattern exists in the current image to be recognized and determining whether the symbolic pattern meets the preset recognition conditions."

[0222] In addition, based on the controller's ability to communicate with the host computer, the controller can also perform the following steps:

[0223] The recognition results are sent to the host computer.

[0224] The host computer usually has a larger storage capacity and stronger computing power. Therefore, after the controller obtains the recognition result, it can further send the recognition result to the host computer. The host computer can store the recognition result to record and trace the object corresponding to the code, or to integrate, analyze and manage a large number of recognition results from different controllers.

[0225] In the solution provided in this application embodiment, the controller communicates with the host computer, and the user performs flexible and diverse configuration operations on the host computer interface to determine whether the feature graphic is a code or an identification pattern. This allows the controller to accurately determine the detection object based on the actual object situation and needs, and to select a more accurate and convenient detection method in different scenarios, thereby improving the accuracy and efficiency of code reading. Furthermore, the controller sends the recognition results to the host computer, which, with its larger storage capacity and stronger computing power, can record and trace the object corresponding to the code, and uniformly integrate, analyze, and manage numerous recognition results.

[0226] As one implementation method of this application, such as Figure 9 As shown, before executing step S101, the controller may also execute steps S901-S902:

[0227] S901, acquire the motion path of the feature image, and based on the motion path, predict at least one first image acquisition device that can subsequently acquire the feature image.

[0228] To achieve accurate code reading, a multi-area collaborative image acquisition strategy is adopted in some scenarios. This means that there are multiple first image acquisition devices, and each first image acquisition device is assigned to focus on the code recognition task in a specific area. This division of labor and cooperation among the various first image acquisition devices helps to improve the overall accuracy and efficiency of code reading.

[0229] Based on this, in order to minimize unnecessary image acquisition and reduce resource consumption, the controller can first acquire the motion path of the feature image, and then, based on the motion path of the feature image and the pre-recorded positions of each first image acquisition device, predict at least one first image acquisition device that can subsequently acquire the feature image.

[0230] For the motion path of the feature graphic, the controller can select the corresponding motion path determination method according to the specific application scenario. Several examples will be given below, but no specific limitation will be made on the motion path determination method.

[0231] In the first example, a second image acquisition device can be set before the first image acquisition device. The controller can acquire the image to be detected acquired by the second image acquisition device and detect whether there is a feature graphic in the acquired image to be detected. If there is, the controller can determine the position of the feature graphic in multiple subsequent images to be detected and determine the motion path of the feature graphic based on the position of the feature graphic in each image to be detected.

[0232] Of course, when the controller determines that there are feature patterns in the image to be detected, it can further determine whether the feature patterns meet preset conditions, such as whether the confidence level of the feature patterns is greater than a preset confidence threshold.

[0233] In the second example, since objects generally move in a near-linear manner when moved by means of a conveyor belt, a grating can be set in front of the first image acquisition device. The controller can acquire the real-time detection data of the grating. When the grating detects that an object has passed through a certain position, the controller can use the straight line that passes through that position and is perpendicular to the setting direction of the grating as the movement path of the feature image.

[0234] In the third example, the code reading scenario can have multiple independent detection lines, with a first image acquisition device set on each detection line. In this case, a detection device such as a grating or camera can be set in front of each first image acquisition device. When the detection device detects that an object is passing through a certain detection line, the controller can transmit the direction of the object on the detection line and determine it as the motion path of the feature graphic.

[0235] S902, control that only at least one first image acquisition device continues to acquire images.

[0236] After the controller determines that at least one first image acquisition device can subsequently acquire feature graphics, it can control only the predicted first image acquisition device to continue acquiring images, while the other first image acquisition devices can be controlled not to acquire images, thereby reducing the overall power consumption of each first image acquisition device and extending the service life of each first image acquisition device.

[0237] In one embodiment, after the controller identifies at least one first image acquisition device, it can directly control the at least one first image acquisition device to acquire images, thereby avoiding missing feature graphics.

[0238] In the second embodiment, as mentioned above, due to factors such as lens distortion and lighting, when the feature image is located at the center of the field of view of the first image acquisition device, the first image acquisition device can capture the details and features of the feature image more clearly. Therefore, after the controller identifies at least one first image acquisition device, it can first determine the time when the feature image arrives at at least one first image acquisition device, and then control at least one first image acquisition device to perform image acquisition within a certain time range around that time. This ensures that the details and features of the feature image can be captured clearly, while also minimizing invalid acquisition by the first image acquisition device and reducing resource consumption.

[0239] Specifically, regarding the "time when the feature image arrives at the first image acquisition device".

[0240] In some scenarios, the distance between the first and second image acquisition devices can be predetermined, as can the speed of the object's movement (e.g., a conveyor belt moving at a fixed speed). Based on this, the transmission time between the first and second image acquisition devices can be predetermined. Thus, after the feature image is acquired by the second image acquisition device, the arrival time of the feature image at the first image acquisition device can be directly determined based on the predetermined transmission time.

[0241] In other scenarios, the distance between the first image acquisition device and the second image acquisition device can be predetermined, but the moving speed of the object is not fixed (e.g., moving the object by hand). Based on this, the controller can determine the movement path of the feature graphic and the position information of each first image acquisition device, and calculate the time when the feature graphic arrives at the first image acquisition device in real time.

[0242] Accordingly, when the controller executes step S101, it can specifically achieve this through step S903:

[0243] S903, acquire at least one image to be identified acquired by a first image acquisition device.

[0244] Accordingly, based on the controller controlling only at least one first image acquisition device to continue acquiring images, when the controller acquires the image to be identified, it specifically acquires the image to be identified acquired by the aforementioned at least one first image acquisition device.

[0245] The solution provided in this application employs a multi-regional collaborative image acquisition strategy, utilizing multiple first image acquisition devices to improve overall code reading accuracy and efficiency through division of labor and cooperation. Furthermore, by predicting subsequent acquisition devices based on the acquired feature graphic movement path, unnecessary image acquisition is effectively reduced, resource consumption is lowered, the overall power consumption of each first image acquisition device is reduced, and their lifespan is extended. After determining the acquisition device, its acquisition can be directly controlled to avoid missed acquisitions, or the arrival time of the feature graphic can be determined first, and acquisition can be performed within a specific time range, ensuring clear capture of feature graphic details and features while reducing invalid acquisitions.

[0246] Corresponding to the above-described code reading method, this application also provides a code reading device, such as... Figure 10 As shown, the device may include:

[0247] The image acquisition module 1001 is used to acquire the image to be recognized acquired by the first image acquisition device;

[0248] The feature pattern detection module 1002 is used to detect whether there are feature patterns in the acquired current image to be identified, wherein the feature patterns are used to characterize the presence of codes in the image;

[0249] The judgment module 1003 is used to determine whether the feature graphic meets the preset recognition conditions when there is a feature graphic in the current image to be recognized.

[0250] The code recognition module 1004 is used to recognize the code in the image to be recognized acquired by the first image acquisition device when the feature graphic meets the preset recognition conditions, and to obtain the recognition result.

[0251] In the solution provided in this application embodiment, an image to be recognized is acquired by a first image acquisition device; for the acquired current image to be recognized, it is detected whether there is a feature pattern in the current image to be recognized, wherein the feature pattern is used to represent the presence of a code in the image; if there is a feature pattern in the current image to be recognized, it is determined whether the feature pattern meets a preset recognition condition; if the feature pattern meets the preset recognition condition, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain a recognition result. Since the code in the image is only recognized after the presence of a feature pattern representing the presence of a code in the image is detected and the preset recognition condition is confirmed, the accuracy of code recognition is improved, and the power consumption loss caused by invalid recognition is avoided.

[0252] As one implementation of this application, the above-mentioned judgment module 1003 can be specifically used for:

[0253] Based on the confidence level and / or motion state of the feature patterns in the current image to be identified, determine whether the feature patterns meet the preset conditions.

[0254] As one embodiment of this application, the above-mentioned judgment module 1003 may include:

[0255] The confidence judgment unit is used to determine whether the confidence of the feature pattern in the current image to be recognized is greater than the preset confidence threshold.

[0256] The position change determination unit is used to continuously acquire the position of the feature graphic in the image to be identified in multiple subsequent images when the confidence of the feature graphic is greater than a preset confidence threshold, and determine the position change of the feature graphic in multiple consecutive images to be identified based on the position of the feature graphic in the image to be identified.

[0257] The position change judgment unit is used to determine that the feature graphic meets the preset recognition conditions when the position change amount in multiple consecutive images to be recognized is less than a preset change amount threshold.

[0258] As one embodiment of this application, when there are multiple feature patterns in the current image to be identified, the above-mentioned judgment module 1003 can be specifically used for:

[0259] For each feature graphic, the steps described above are performed: determining whether the confidence level of the feature graphic in the current image to be identified is greater than a preset confidence threshold; if the confidence level of the feature graphic is greater than the preset confidence threshold, continuously acquiring the position of the feature graphic in multiple subsequent images to be identified; based on the position of the feature graphic in the images to be identified, determining the position change of the feature graphic in multiple consecutive images to be identified; and based on the position change, if the position change in multiple consecutive images to be identified is less than a preset change threshold, determining that the feature graphic meets the preset recognition conditions.

[0260] The code recognition module 1004 described above can be specifically used for:

[0261] For each feature graphic, if the feature graphic meets the preset recognition conditions, the code corresponding to the feature graphic in the image to be recognized acquired by the first image acquisition device is recognized to obtain the recognition result of the code.

[0262] As one implementation of this application, the above-mentioned judgment module 1003 can be specifically used for:

[0263] The system continuously acquires the positions of the bounding boxes of the feature graphics in multiple subsequent images to be recognized. Based on the positions of the bounding boxes of the feature graphics in the images to be recognized, it determines the position changes of the bounding boxes of the feature graphics in multiple consecutive images to be recognized. According to the position changes, if the position change of the bounding boxes in multiple consecutive images to be recognized is less than a preset change threshold, it determines that the feature graphics meet the preset recognition conditions.

[0264] As one embodiment of this application, the above-described apparatus may further include:

[0265] The light source control module is used to control the light source to turn on when the positional change in the multiple consecutive images to be identified is less than the preset change threshold, so as to provide supplemental lighting for the acquisition field of the first image acquisition device.

[0266] The code recognition module 1004 described above can be specifically used for:

[0267] The code in the image to be identified acquired by the first image acquisition device under supplementary lighting conditions is identified to obtain the identification result;

[0268] The aforementioned light source control module is also used to control the light source to turn off after the code recognition module 1004 obtains the recognition result.

[0269] As one embodiment of this application, the code recognition module 1004 described above may include:

[0270] The target location determination unit determines the target location of the code in the image to be identified acquired by the first image acquisition device based on the position of the feature graphic in the consecutive multiple images to be identified.

[0271] The recognition unit is used to recognize the target location in the image to be recognized acquired by the first image acquisition device and obtain the recognition result.

[0272] As one embodiment of this application, the target location determination unit described above can be specifically used for:

[0273] When the feature graphic is the code itself, the position of the feature graphic in the consecutive multiple images to be identified is determined as the target position of the code in the image to be identified acquired by the first image acquisition device.

[0274] When the feature graphic is an identification pattern located on the same target object as the code and having a fixed relative positional relationship with the code; based on the position of the feature graphic in the consecutive multiple images to be identified, and the fixed relative positional relationship between the code and the identification pattern, the target position of the code in the image to be identified acquired by the first image acquisition device is determined.

[0275] As one embodiment of this application, the above-mentioned feature image detection module 1002 can be specifically used for:

[0276] Detect whether there are feature patterns in the preset region of interest in the current image to be identified.

[0277] As one embodiment of this application, the above-mentioned feature graphic is an identification pattern that is located on the same target object as the code and has a fixed relative positional relationship with the code;

[0278] The code recognition module 1004 described above can be specifically used for:

[0279] Based on the fixed relative positional relationship between the code and the identification pattern, the code is obtained from the image to be identified acquired by the first image acquisition device and then identified to obtain the identification result.

[0280] As one embodiment of this application, the above-described apparatus may further include:

[0281] The configuration information receiving module is used to receive the configuration information of the feature graphic sent by the host computer, so as to determine whether the feature graphic is a code or an identification pattern. The configuration information of the feature graphic is configured by the user in the interface of the host computer, and the configuration information of the feature graphic includes: code or identification pattern.

[0282] As one embodiment of this application, the above-mentioned apparatus further includes:

[0283] The device prediction module is used to acquire the motion path of the feature image and, based on the motion path, predict at least one first image acquisition device that can subsequently acquire the feature image.

[0284] The control module is used to control that images are continuously acquired only by at least one first image acquisition device;

[0285] The image acquisition module 1001 described above can be specifically used for:

[0286] Acquire at least one image to be identified from a first image acquisition device.

[0287] As one embodiment of this application, the above-described device prediction module can be specifically used for:

[0288] The system acquires the image to be detected from the second image acquisition device. If a feature graphic exists in the acquired image to be detected, the system determines the motion path of the feature graphic based on the acquired multiple images to be detected.

[0289] As one implementation of this application, the above-described control module can be specifically used for:

[0290] The time when the feature graphic arrives at at least one first image acquisition device is determined, and based on the determined time, the system controls the acquisition of images to continue only by at least one first image acquisition device.

[0291] As one embodiment of this application, the code recognition module 1004 described above can be specifically used for:

[0292] When the feature graphic satisfies the preset recognition conditions, the code in the current image to be recognized is recognized to obtain the recognition result;

[0293] or,

[0294] If the feature image satisfies the preset recognition conditions, it is detected whether there is a code in the current image to be recognized. If there is a code in the current image to be recognized, the code in the image to be recognized obtained after the current image to be recognized is recognized to obtain the recognition result.

[0295] This application also provides a code reading device, such as... Figure 11 As shown, Figure 11 This is a schematic diagram of the structure of a barcode reading device provided in an embodiment of this application. The barcode reading device may include: a base 1101, a controller (not shown in the figure), an image acquisition device 1102, and a support frame 1103.

[0296] The base 1101 has a scanning stage for placing the object to be scanned, and the support frame 1103 is disposed on one side of the base 1101;

[0297] Image acquisition device 1102 is mounted on top of support frame 1103, so that image acquisition device 1102 faces the scanning table surface;

[0298] The controller is located inside the base 1101 and communicates with the image acquisition device 1102 to execute the code reading method described in any of the above embodiments.

[0299] This application also provides a code reading system, such as... Figure 12 As shown, Figure 12 This is a schematic diagram of the structure of a code reading system provided in an embodiment of this application. The code reading system may include: a conveyor belt 1201 for placing the object to be scanned, a controller 1202, and multiple image acquisition devices 1203.

[0300] Multiple image acquisition devices 1203 are located above the conveyor belt 1201 and are positioned facing the conveyor belt 1201;

[0301] The controller 1202 communicates with multiple image acquisition devices 1203 to execute the code reading method described in any of the above embodiments.

[0302] The controller may include: a memory for storing computer programs; and a processor for implementing the code reading method described in any of the above embodiments when executing the program stored in the memory.

[0303] Furthermore, the aforementioned controller may also include a communication bus and / or a communication interface, through which the processor, communication interface, and memory communicate with each other.

[0304] The communication bus mentioned in the controller above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. The communication interface is used for communication between the controller and other devices.

[0305] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0306] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0307] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of the code reading method described in any of the above embodiments.

[0308] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the code reading method described in any of the embodiments above.

[0309] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0310] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0311] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of code reading devices, code reading equipment, code reading systems, computer-readable storage media, and computer program products are basically similar to the code reading method embodiments, so the descriptions are relatively simple, and relevant parts can be referred to the descriptions of the method embodiments.

[0312] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A code reading method, characterized in that, The method includes: Acquire the image to be identified captured by the first image acquisition device; For the acquired current image to be identified, detect whether there is a feature pattern in the current image to be identified, wherein the feature pattern is used to characterize the presence of a code in the image; If the feature pattern exists in the current image to be identified, determine whether the feature pattern meets the preset recognition conditions; When the feature pattern meets the preset recognition conditions, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain the recognition result; The step of determining whether the feature image meets the preset recognition conditions includes: Determine whether the confidence level of the feature pattern in the current image to be identified is greater than a preset confidence threshold; If the confidence level of the feature graphic is greater than the preset confidence threshold, the positions of the feature graphics in the subsequent multiple images to be identified are continuously obtained; Based on the position of the feature graphic in the image to be identified, determine the positional changes of the feature graphic in multiple consecutive images to be identified; Based on the positional change, if the positional change in multiple consecutive images to be identified is less than a preset change threshold, the feature graphic is determined to meet the preset recognition conditions.

2. The method according to claim 1, characterized in that, When there are multiple feature patterns in the current image to be identified For each feature graphic, the process of determining whether the confidence level of the feature graphic in the current image to be identified is greater than a preset confidence threshold is performed; if the confidence level of the feature graphic is greater than the preset confidence threshold, the positions of the feature graphics in the subsequent multiple images to be identified are continuously obtained. Based on the position of the feature graphic in the image to be identified, determine the position change of the feature graphic in multiple consecutive images to be identified; and based on the position change, determine that the feature graphic meets the preset recognition conditions if the position change in multiple consecutive images to be identified is less than a preset change threshold. When the feature graphic satisfies the preset recognition conditions, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain a recognition result, including: For each feature pattern, if the feature pattern meets the preset recognition conditions, the code corresponding to the feature pattern in the image to be recognized acquired by the first image acquisition device is recognized to obtain the recognition result of the code.

3. The method according to claim 1, characterized in that, The process involves continuously acquiring the positions of feature graphics in subsequent images to be identified; determining the positional changes of the feature graphics in the consecutive images based on their positions; and determining that the feature graphics meet preset recognition conditions if the positional change in the consecutive images is less than a preset change threshold, based on the positional changes. The system continuously acquires the positions of the bounding boxes of feature graphics in multiple subsequent images to be identified; based on the positions of the bounding boxes of feature graphics in the images to be identified, it determines the position changes of the bounding boxes of feature graphics in multiple consecutive images to be identified; and based on the position changes, if the position change of the bounding boxes in multiple consecutive images to be identified is less than a preset change threshold, it determines that the feature graphics meet preset recognition conditions.

4. The method according to claim 1, characterized in that, If the positional change in the multiple consecutive images to be identified is less than the preset change threshold, the method further includes: Turn on the light source to provide supplemental lighting for the field of view of the first image acquisition device; The step of recognizing the code in the image to be recognized acquired by the first image acquisition device to obtain the recognition result includes: The code in the image to be identified acquired by the first image acquisition device under supplementary lighting conditions is identified to obtain the identification result, and the light source is controlled to be turned off after obtaining the identification result.

5. The method according to claim 1, characterized in that, The step of recognizing the code in the image to be recognized acquired by the first image acquisition device to obtain the recognition result includes: Based on the position of the feature pattern in the consecutive multiple images to be identified, the target position of the code in the image to be identified acquired by the first image acquisition device is determined; The target location in the image to be identified acquired by the first image acquisition device is identified to obtain the identification result.

6. The method according to claim 5, characterized in that, When the feature pattern is the code itself, determining the target position of the code in the image to be recognized acquired by the first image acquisition device based on the position of the feature pattern in the consecutive multiple images to be recognized includes: The position of the feature graphic in the consecutive multiple images to be identified is determined as the target position of the code in the image to be identified acquired by the first image acquisition device; When the feature graphic is an identification pattern located on the same target object as the code and having a fixed relative positional relationship with the code; determining the target position of the code in the image to be identified acquired by the first image acquisition device based on the position of the feature graphic in the consecutive multiple images to be identified includes: Based on the position of the feature graphic in the consecutive multiple images to be identified, and the fixed relative positional relationship between the code and the identification pattern, the target position of the code in the image to be identified acquired by the first image acquisition device is determined.

7. The method according to any one of claims 1-6, characterized in that, The detection of whether a feature pattern exists in the current image to be identified includes: Detect whether there are feature patterns in the preset region of interest in the current image to be identified.

8. The method according to any one of claims 1-6, characterized in that, The method further includes: The system receives configuration information of a feature graphic sent by a host computer to determine whether the feature graphic is a code or an identification pattern. The configuration information of the feature graphic is configured by the user in the interface of the host computer and includes a code or identification pattern.

9. The method according to any one of claims 1-6, characterized in that, Before acquiring the image to be identified by the first image acquisition device, the method further includes: The motion path of the feature image is obtained, and based on the motion path, at least one first image acquisition device that can subsequently acquire the feature image is predicted; The control ensures that images are continuously acquired solely by the at least one first image acquisition device. The acquisition of the image to be identified by the first image acquisition device includes: The image to be identified is acquired by the at least one first image acquisition device.

10. The method according to claim 9, characterized in that, The process of obtaining the motion path of the feature image includes: Acquire the image to be detected captured by the second image acquisition device; If the feature pattern exists in the acquired image to be detected, the motion path of the feature pattern is determined based on the acquired multiple images to be detected.

11. The method according to claim 10, characterized in that, The control is that only the at least one first image acquisition device continues to acquire images, including: Determine the time when the feature graphic arrives at the at least one first image acquisition device; Based on the determined time, control is exercised to allow only the at least one first image acquisition device to continue acquiring images.

12. The method according to any one of claims 1-6, characterized in that, When the feature graphic satisfies the preset recognition conditions, the code in the image to be recognized acquired by the first image acquisition device is recognized to obtain a recognition result, including: When the feature graphic satisfies the preset recognition conditions, the code in the current image to be recognized is recognized to obtain the recognition result; or, If the feature image satisfies the preset recognition conditions, it is detected whether there is a code in the current image to be recognized. If there is a code in the current image to be recognized, the code in the image to be recognized obtained after the current image to be recognized is recognized to obtain the recognition result.

13. A code reading device, characterized in that, The device includes: The image acquisition module is used to acquire the image to be recognized acquired by the first image acquisition device; The feature pattern detection module is used to detect whether there are feature patterns in the acquired current image to be identified, wherein the feature patterns are used to characterize the presence of codes in the image; The judgment module is used to determine whether the feature pattern meets the preset recognition conditions when the feature pattern exists in the current image to be recognized. The code recognition module is used to recognize the code in the image to be recognized acquired by the first image acquisition device when the feature graphic meets the preset recognition conditions, and to obtain the recognition result. Specifically, the judgment module is used for: Determine whether the confidence level of the feature graphic in the current image to be identified is greater than a preset confidence threshold; if the confidence level of the feature graphic is greater than the preset confidence threshold, continuously acquire the position of the feature graphic in the subsequent multiple images to be identified; based on the position of the feature graphic in the images to be identified, determine the position change of the feature graphic in the consecutive multiple images to be identified; according to the position change, if the position change in the consecutive multiple images to be identified is less than a preset change threshold, determine that the feature graphic meets the preset recognition conditions.

14. A code reading device, characterized in that, include: Base, controller, image acquisition equipment, and support frame; The base has a scanning platform for placing the object to be scanned, and the support frame is disposed on one side of the base; The image acquisition device is mounted on top of the support frame, such that the image acquisition device faces the scanning table surface; The controller is located inside the base and communicates with the image acquisition device to execute the code reading method according to any one of claims 1-12.

15. A code reading system, characterized in that, include: A conveyor belt, controller, and multiple image acquisition devices are used to place the objects being scanned. The plurality of image acquisition devices are located above the conveyor belt and are positioned facing the conveyor belt; The controller communicates with the plurality of image acquisition devices to execute the code reading method according to any one of claims 1-12.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the code reading method according to any one of claims 1-12.

Images