A method, system, device and program product for automatically reading steel coil labels

Abstract

The present invention belongs to the field of label collection technology, and specifically discloses a method, system, device, and program product for automatically reading steel coil labels. The method determines the two-dimensional coordinates of each steel coil in the loading and unloading area by collecting monitoring images from the top perspective of the loading and unloading area. Then, each steel coil is matched with a corresponding side monitoring camera in the loading and unloading area to collect the side image of the steel coil, so as to detect and determine the side area of ​​the steel coil and the position of the label image area within the side area of ​​the steel coil. Then, the side monitoring camera is adjusted according to the position of the label image area to specifically collect the label image for target detection. Finally, the label image is extracted for code scanning and recognition to obtain the label reading information of the corresponding steel coil, which can realize automated steel coil label reading. The present invention can use monitoring cameras to realize unmanned label reading of steel coils entering and leaving the warehouse, thereby improving the reading efficiency of the labels of steel coils entering and leaving the warehouse.

Description

Technical Field

[0001] The present invention belongs to the technical field of label collection, and in particular relates to a method, system, equipment and program product for automatically reading steel coil labels. Background Art

[0002] In an intelligent steel coil warehouse, based on management needs, it is necessary to read the label information (i.e. barcode or QR code data) of all steel coils on the transport vehicles when the steel coils enter and leave the warehouse, and pass the label information of the steel coils to the upper information system. The traditional way to read steel coil labels is to manually use a scanning device (such as a handheld device with decoding function) to scan and read the code, or to use a fixed scanning device to read at a fixed position; the manual scanning method cannot meet the requirements of unmanned operation of the warehouse, and the labor cost is high and the efficiency is low; and the fixed scanning method, due to the parking accuracy of the steel coil transport vehicles in the loading and unloading area, the position of the steel coil and the position of the label on the side of the steel coil are not fixed. The fixed scanning equipment often cannot effectively scan and read the steel coil label information due to misalignment, and it is difficult to meet the actual needs of automatic reading of steel coil labels. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, system, device and program product for automatically reading steel coil labels to solve the above-mentioned problems existing in the prior art.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] In a first aspect, a method for automatically reading a steel coil label is provided, comprising:

[0006] Obtaining a monitoring image of the loading and unloading area captured by a monitoring camera at the top of the loading and unloading area, wherein the monitoring image of the loading and unloading area includes a plurality of steel coils;

[0007] Performing a first target detection on the monitoring image of the loading and unloading area to determine each steel coil in the monitoring image of the loading and unloading area and the two-dimensional coordinates of the top surface corresponding to each steel coil;

[0008] Matching the side monitoring cameras corresponding to the steel coils according to the two-dimensional coordinates of the top surfaces of the steel coils, and determining the first shooting orientation adjustment information of the side monitoring cameras matched to the steel coils according to the two-dimensional coordinates of the top surfaces of the steel coils and the two-dimensional coordinates of the side surfaces of the side monitoring cameras matched to the steel coils;

[0009] Sending the first shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, shoots the side of the corresponding steel coil to obtain a side shooting image, and obtains the side shooting image fed back by the corresponding side monitoring camera;

[0010] Performing a second target detection on the side-view image to determine a side area of ​​the steel coil in the side-view image, and determining whether an area ratio of the side area of ​​the steel coil in the side-view image exceeds a first ratio threshold;

[0011] When it is determined that the area ratio of the side area of ​​the steel coil in the side image exceeds the first ratio threshold, a third target detection is performed on the side area of ​​the steel coil to determine the first label image area in the side area of ​​the steel coil and the image coordinate information of the first label image area in the side image;

[0012] Generate second shooting orientation adjustment information based on the image coordinate information, and send the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, shoots the label on the side of the corresponding steel coil, obtains the label shooting image, and obtains the label shooting image fed back by the corresponding side monitoring camera;

[0013] Performing a third target detection on the tag shot image, determining a second tag image region in the tag shot image, and determining whether an area ratio of the second tag image region in the tag shot image exceeds a second ratio threshold;

[0014] When it is determined that the area ratio of the second label image area in the label shot image exceeds the second ratio threshold, the second label image area is extracted, and the second label image area is scanned and identified to obtain the label reading information of the corresponding steel coil.

[0015] In one possible design, the method further includes: aggregating the label reading information corresponding to each steel coil and uploading it to a management system.

[0016] In one possible design, the side monitoring camera corresponding to each steel coil is matched according to the two-dimensional coordinates of the top surface of each steel coil, and the shooting orientation adjustment information of the side monitoring camera matched to each steel coil is determined according to the two-dimensional coordinates of the top surface of each steel coil and the two-dimensional coordinates of the side monitoring camera matched to each steel coil, including:

[0017] Determine the first coordinate position of each steel coil in the reference direction of the loading and unloading area according to the two-dimensional coordinates of the top surface of each steel coil, and determine the second coordinate position of each side monitoring camera in the reference direction of the loading and unloading area according to the two-dimensional coordinates of the side surface of each side monitoring camera;

[0018] Determine the second coordinate position closest to the first coordinate position of the corresponding steel coil in the reference direction of the loading and unloading area, and use the side monitoring camera corresponding to the second coordinate position closest to the first coordinate position of the corresponding steel coil as the side monitoring camera matched with the corresponding steel coil;

[0019] The azimuth difference is determined based on the first coordinate position of the steel coil in the reference direction of the loading and unloading area and the second coordinate position of the side monitoring camera matched to it in the reference direction of the loading and unloading area, and the shooting azimuth adjustment information of the side monitoring camera matched to the corresponding steel coil is generated based on the azimuth difference.

[0020] In one possible design, before performing second target detection on the side-shot image, the method further includes:

[0021] A. Determine a first center point of the side image and a first center coordinate of the first center point in the side image, and determine a second center point of the side area of ​​the steel coil and a second center coordinate of the second center point in the side image;

[0022] B. when it is determined based on the first center coordinates and the second center coordinates that the second center point is not within a first error range around the first center point, generating first orientation fine-tuning information based on the first center coordinates and the second center coordinates;

[0023] C. Sending the first orientation fine-tuning information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first orientation fine-tuning information, re-shoots the side of the corresponding steel coil, and obtains a new side shot image;

[0024] The above steps AC are repeatedly performed using the new side-view image until the second center point is within a first error range around the first center point, thereby obtaining an adjusted side-view image.

[0025] In one possible design, when it is determined that the area ratio of the side region of the steel coil in the side image does not exceed a first ratio threshold, the method further includes:

[0026] generating first zoom information according to the area ratio of the side area of ​​the steel coil in the side-shot image;

[0027] Sending the first zoom information to the corresponding side monitoring camera, so that the corresponding side monitoring camera re-zooms and shoots the side of the corresponding steel coil according to the first zoom information to obtain a zoomed side image;

[0028] The zoomed side shot image is used as the updated side shot image, and the second target detection is re-performed on the updated side shot image to determine the side area of ​​the steel coil and the area ratio of the side area of ​​the steel coil.

[0029] In one possible design, before performing third target detection on the tag-captured image, the method further includes:

[0030] a. Determine the third center point of the label image, and the third center coordinates of the third center point in the label image, and determine the fourth center point of the second label image area, and the fourth center coordinates of the fourth center point in the label image;

[0031] b. when it is determined based on the third center coordinates and the fourth center coordinates that the fourth center point is not within the second error range around the third center point, generating second orientation fine-tuning information based on the third center coordinates and the fourth center coordinates;

[0032] c. Sending the second position fine-tuning information to the corresponding side surveillance camera so that the corresponding side surveillance camera adjusts the shooting position according to the second position fine-tuning information, re-shoots the label on the side of the corresponding steel coil, and obtains a new label captured image;

[0033] The above steps ac are repeatedly performed using the new label captured image until the fourth center point is within the second error range around the third center point, thereby obtaining an adjusted label captured image.

[0034] In one possible design, when it is determined that the area ratio of the second label image region in the label captured image does not exceed the second ratio threshold, the method further includes:

[0035] generating second zoom information according to the area ratio of the second tag image region in the tag shot image;

[0036] Sending the second zoom information to the corresponding side monitoring camera, so that the corresponding side monitoring camera zooms and shoots the label on the side of the corresponding steel coil according to the second zoom information to obtain a zoom label shooting image;

[0037] The zoomed tag-shot image is used as the updated tag-shot image, and the third target detection is performed again on the updated tag-shot image to determine the second tag image area and the area ratio of the second tag image area.

[0038] In a second aspect, a steel coil label automatic reading system is provided, comprising a first acquisition unit, a first detection unit, a shooting and matching unit, a second acquisition unit, a second detection unit, a third detection unit, a third acquisition unit, a fourth detection unit, and a label reading unit, wherein:

[0039] A first acquisition unit is configured to acquire a monitoring image of the loading and unloading area acquired by a monitoring camera at the top of the loading and unloading area, wherein the monitoring image of the loading and unloading area includes a plurality of steel coils;

[0040] a first detection unit, configured to perform first target detection on the monitoring image of the loading and unloading area, and determine each steel coil in the monitoring image of the loading and unloading area, and the two-dimensional coordinates of the top surface corresponding to each steel coil;

[0041] a shooting matching unit, configured to match the side monitoring camera corresponding to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil, and determine first shooting orientation adjustment information of the side monitoring camera matched to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil and the two-dimensional coordinates of the side surface of the side monitoring camera matched to each steel coil;

[0042] a second acquisition unit, configured to send the first shooting orientation adjustment information to a corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, shoots the side of the corresponding steel coil to obtain a side shot image, and obtains the side shot image fed back by the corresponding side monitoring camera;

[0043] a second detection unit, configured to perform second target detection on the side-view image, determine a side area of ​​the steel coil in the side-view image, and determine whether an area ratio of the side area of ​​the steel coil in the side-view image exceeds a first ratio threshold;

[0044] a third detection unit, configured to perform a third target detection on the side area of ​​the steel coil when it is determined that the area ratio of the side area of ​​the steel coil in the side image exceeds a first ratio threshold, and determine a first label image area in the side area of ​​the steel coil and image coordinate information of the first label image area in the side image;

[0045] a third acquisition unit, configured to generate second shooting orientation adjustment information based on the image coordinate information, and send the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, shoots the label on the side of the corresponding steel coil, obtains the label shooting image, and obtains the label shooting image fed back by the corresponding side monitoring camera;

[0046] a fourth detection unit, configured to perform a third target detection on the tag-shot image, determine a second tag image region in the tag-shot image, and determine whether an area ratio of the second tag image region in the tag-shot image exceeds a second ratio threshold;

[0047] The label reading unit is used to extract the second label image area when it is determined that the area ratio of the second label image area in the label shooting image exceeds a second ratio threshold, and scan and identify the second label image area to obtain the label reading information of the corresponding steel coil.

[0048] In a third aspect, a device for automatically reading steel coil labels is provided, comprising:

[0049] a memory for storing instructions;

[0050] A processor is used to read the instructions stored in the memory and execute any one of the methods described in the first aspect according to the instructions.

[0051] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform any one of the methods described in the first aspect. Furthermore, a computer program product is provided that, when executed on a computer, performs any one of the methods described in the first aspect.

[0052] Beneficial effects: The present invention determines the two-dimensional coordinates of each steel coil in the loading and unloading area by collecting monitoring images from the top perspective of the loading and unloading area, and then matches each steel coil with a corresponding side monitoring camera in the loading and unloading area to collect the side image of the steel coil, so as to detect and determine the side area of ​​the steel coil and the position of the label image area in the side area of ​​the steel coil, and then adjusts the side monitoring camera according to the position of the label image area to collect the label image for target detection, and finally extracts the label image for code scanning and recognition to obtain the label reading information of the corresponding steel coil, which can realize automated steel coil label reading. The present invention can use monitoring cameras to realize unmanned label reading of steel coils entering and leaving the warehouse, improve the reading efficiency of steel coil labels entering and leaving the warehouse, meet the requirements of intelligent and unmanned management of steel coils entering and leaving the warehouse, and the present invention can save a lot of manpower and financial resources, has a fast steel coil label reading speed, and is stable in operation, which is conducive to promotion and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 This is a schematic diagram of the layout of the top surveillance cameras in the loading and unloading area;

[0055] Figure 2 A schematic diagram of the surveillance cameras on each side of the loading and unloading area;

[0056] Figure 3 Schematic diagram of the steps of the method in Example 1 of the present invention;

[0057] Figure 4 This is a schematic diagram of the structure of the device in Example 3 of the present invention. DETAILED DESCRIPTION

[0058] It should be noted that the description of these embodiments is intended to help understand the present invention, but does not constitute a limitation of the present invention. The specific structural and functional details disclosed herein are merely intended to describe exemplary embodiments of the present invention. However, the present invention may be embodied in a variety of alternative forms, and should not be construed as being limited to the embodiments set forth herein.

[0059] It should be understood that, unless otherwise expressly specified or limited, the corresponding terms should be understood in a broad sense. For example, "connection" can refer to a fixed connection, a detachable connection, or an integral connection; it can refer to a direct connection, an indirect connection through an intermediate medium, or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the embodiments based on specific circumstances.

[0060] In the following description, certain details are provided to facilitate a thorough understanding of the example embodiments. However, one skilled in the art will appreciate that the example embodiments may be practiced without these specific details. For example, devices may be shown in block diagrams to avoid obscuring the examples with unnecessary detail. In other embodiments, well-known processes, structures, and techniques may be shown without unnecessary detail to avoid obscuring the embodiments.

[0061] Example 1:

[0062] This embodiment provides a method for automatically reading steel coil labels, which can be applied to corresponding processing hosts, such as Figures 1 to 2 As shown, the processing host establishes data connections with the top monitoring camera and each side monitoring camera of the loading and unloading area, such as Figure 3 As shown, the method includes the following steps:

[0063] S1. Obtain a monitoring image of the loading and unloading area captured by a monitoring camera at the top of the loading and unloading area, wherein the monitoring image of the loading and unloading area includes a plurality of steel coils.

[0064] In specific implementation, steel coil products are transported by special vehicles. When the steel coils are in and out of the warehouse, the vehicles loaded with steel coils are parked in fixed loading and unloading areas. The steel coil labels are set on the vertical planes on both sides of the steel coils, but the positions are not fixed. The diameter of the steel coil is generally close to 1.5 meters, but the code on the steel coil label is very small. If a multi-camera stitching solution is used to capture images, a large number of cameras are required, and the stitching quality cannot be guaranteed. In addition, the number of hardware is large, the failure rate is high, and the overall reliability of the system is poor. If a line scan camera is used, it requires the coordination of a complex motion mechanism, which is expensive and has a high failure rate. There is also the problem of image stitching. Therefore, if Figures 1 to 2As shown, this method installs a top surveillance camera at the top of the loading and unloading area and a small number of side surveillance cameras on both sides of the loading and unloading area. The top surveillance camera in the loading and unloading area directly faces the vehicles and steel coils in the loading and unloading area below, first capturing a loading and unloading area surveillance image. The loading and unloading area surveillance image includes the vehicles in the loading and unloading area and the steel coils on the vehicles. The top surveillance camera transmits the captured loading and unloading area surveillance image to the processing host.

[0065] S2. Perform a first target detection on the monitoring image of the loading and unloading area to determine each steel coil in the monitoring image of the loading and unloading area and the two-dimensional coordinates of the top surface corresponding to each steel coil.

[0066] During specific implementation, the processing host performs a first target detection on the monitoring image of the loading and unloading area. A trained and tested target detection model, such as the ‌R-CNN series model or the ‌YOLO series model, can be used to perform a first target detection on the monitoring image of the loading and unloading area. Based on the target detection results, each steel coil in the monitoring image of the loading and unloading area is determined, and the corresponding two-dimensional coordinates of the top surface of each steel coil in the monitoring image of the loading and unloading area are determined.

[0067] S3. Match the side monitoring camera corresponding to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil, and determine the first shooting orientation adjustment information of the side monitoring camera matched to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil and the side two-dimensional coordinates of the side monitoring camera matched to each steel coil.

[0068] In a specific implementation, the processing host can match the side monitoring camera corresponding to each steel coil based on the top surface two-dimensional coordinates corresponding to each steel coil, and then use the top surface two-dimensional coordinates corresponding to each steel coil and the side two-dimensional coordinates of the side monitoring camera it matches to analyze and determine the first shooting orientation adjustment information of the side monitoring camera matched to each steel coil. For example, the processing host can first determine the first coordinate position of each steel coil in the reference direction of the loading and unloading area based on the top surface two-dimensional coordinates corresponding to each steel coil, and at the same time, determine the second coordinate position of each side monitoring camera in the reference direction of the loading and unloading area based on the side two-dimensional coordinates of each side monitoring camera; then determine the second coordinate position closest to the first coordinate position of the corresponding steel coil in the reference direction of the loading and unloading area, and use the side monitoring camera corresponding to the second coordinate position closest to the first coordinate position of the corresponding steel coil as the side monitoring camera matched to the corresponding steel coil; finally, determine the orientation difference based on the first coordinate position of the steel coil in the reference direction of the loading and unloading area and the second coordinate position of the side monitoring camera it matches in the reference direction of the loading and unloading area, and generate the shooting orientation adjustment information of the side monitoring camera matched to the corresponding steel coil based on the orientation difference.

[0069] S4. Send the first shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, shoots the side of the corresponding steel coil, obtains the side shooting image, and obtains the side shooting image fed back by the corresponding side monitoring camera.

[0070] In specific implementation, the processing host may send the first shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, that is, adjusts the rotation angle and pitch angle of the monitoring camera pan / tilt to shoot the side of the corresponding steel coil and obtain a side shot image. After initially obtaining the side shot image, the processing host also needs to determine whether the steel coil area is in the center position of the side shot image. If it is not in the center position, it needs to fine-tune the shooting angle so that the steel coil area is in the center position of the side shot image. This process includes:

[0071] A. Determine a first center point of the side image and a first center coordinate of the first center point in the side image, and determine a second center point of the side area of ​​the steel coil and a second center coordinate of the second center point in the side image;

[0072] B. when it is determined based on the first center coordinates and the second center coordinates that the second center point is not within a first error range around the first center point, generating first orientation fine-tuning information based on the first center coordinates and the second center coordinates;

[0073] C. Sending the first orientation fine-tuning information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first orientation fine-tuning information, re-shoots the side of the corresponding steel coil, and obtains a new side shot image;

[0074] The above steps AC are repeatedly performed using the new side-view image until the second center point is within a first error range around the first center point, thereby obtaining an adjusted side-view image.

[0075] S5. Perform a second target detection on the side-shot image, determine the side area of ​​the steel coil in the side-shot image, and determine whether the area ratio of the side area of ​​the steel coil in the side-shot image exceeds a first ratio threshold.

[0076] In a specific implementation, after determining the final adjusted side view image, the processing host may perform a second object detection on the final adjusted side view image (using a trained and tested object detection model, such as an R-CNN model or a YOLO model), detect the side region of the steel coil in the side view image, then determine the area ratio of the steel coil side region in the side view image, and determine whether the area ratio of the steel coil side region in the side view image exceeds a first ratio threshold. If the processing host determines that the area ratio of the steel coil side region in the side view image does not exceed the first ratio threshold, it may generate first zoom information based on the area ratio of the steel coil side region in the side view image. The first zoom information may then be transmitted to the corresponding side monitoring camera, causing the corresponding side monitoring camera to re-zoom the side of the corresponding steel coil based on the first zoom information to obtain a zoomed side view image. The zoomed side view image is then used as the updated side view image, and the second object detection is re-performed on the updated side view image to determine the side region of the steel coil and the area ratio of the steel coil side region. This process is repeated until the area ratio of the steel coil side region in the final side view image exceeds the first ratio threshold.

[0077] S6. When it is determined that the area ratio of the side area of ​​the steel coil in the side-shot image exceeds the first ratio threshold, a third target detection is performed on the side area of ​​the steel coil to determine the first label image area in the side area of ​​the steel coil and the image coordinate information of the first label image area in the side-shot image.

[0078] In specific implementation, when the processing host determines that the area ratio of the side area of ​​the steel coil in the side-shot image exceeds the first ratio threshold, a third target detection can be performed on the side area of ​​the steel coil (a trained and tested target detection model, such as the ‌R-CNN series model or the ‌YOLO series model, etc.) to detect the first label image area in the side area of ​​the steel coil and determine the image coordinate information of the first label image area in the side-shot image.

[0079] S7. Generate second shooting orientation adjustment information based on the image coordinate information, and send the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, shoots the label on the side of the corresponding steel coil, obtains the label shooting image, and obtains the label shooting image fed back by the corresponding side monitoring camera.

[0080] During specific implementation, the processing host generates second shooting orientation adjustment information based on the image coordinate information of the first label image area in the side-shot image, and sends the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, that is, adjusts the rotation angle and pitch angle of the monitoring camera pan / tilt to shoot the label on the side of the corresponding steel coil to obtain the label shot image. After initially obtaining the label shot image, the processing host also needs to determine whether the label area is in the center position of the label shot image. If it is not in the center position, it needs to fine-tune the shooting angle so that the label area is in the center position of the label shot image. This process includes:

[0081] a. Determine the third center point of the label image, and the third center coordinates of the third center point in the label image, and determine the fourth center point of the second label image area, and the fourth center coordinates of the fourth center point in the label image;

[0082] b. when it is determined based on the third center coordinates and the fourth center coordinates that the fourth center point is not within the second error range around the third center point, generating second orientation fine-tuning information based on the third center coordinates and the fourth center coordinates;

[0083] c. Sending the second position fine-tuning information to the corresponding side surveillance camera so that the corresponding side surveillance camera adjusts the shooting position according to the second position fine-tuning information, re-shoots the label on the side of the corresponding steel coil, and obtains a new label captured image;

[0084] The above steps ac are repeatedly performed using the new label captured image until the fourth center point is within the second error range around the third center point, thereby obtaining an adjusted label captured image.

[0085] S8. Perform third target detection on the tag shot image, determine the second tag image area in the tag shot image, and determine whether the area ratio of the second tag image area in the tag shot image exceeds a second ratio threshold.

[0086] In a specific implementation, after obtaining the final adjusted label-captured image, the processing host performs a third object detection on the label-captured image (using a trained and tested object detection model, such as an R-CNN model or a YOLO model) to detect the second label image region in the label-captured image and determine whether the area ratio of the second label image region in the label-captured image exceeds a second ratio threshold. If the area ratio of the second label image region in the label-captured image does not exceed the second ratio threshold, the processing host may generate second zoom information based on the area ratio of the second label image region in the label-captured image. The second zoom information is then transmitted to the corresponding side monitoring camera, causing the corresponding side monitoring camera to zoom in on the label on the corresponding side of the steel coil based on the second zoom information to obtain a zoomed label-captured image. The zoomed label-captured image is then used as the updated label-captured image, and the third object detection is re-performed on the updated label-captured image to determine the second label image region and the area ratio of the second label image region. This cycle is repeated until the area ratio of the second label image region in the final obtained label-captured image exceeds the second ratio threshold.

[0087] S9. When it is determined that the area ratio of the second label image area in the label shot image exceeds the second ratio threshold, the second label image area is extracted, and the second label image area is scanned and identified to obtain the label reading information of the corresponding steel coil.

[0088] In specific implementations, when it is determined that the area ratio of the second label image region in the captured label image exceeds a second ratio threshold, a label image suitable for code reading is obtained. At this point, the processing host can extract the second label image region and scan and identify the second label image region to obtain the label reading information of the corresponding steel coil. After simultaneously scanning and reading the label reading information corresponding to each steel coil, the label reading information corresponding to each steel coil is summarized and uploaded to the management system.

[0089] This method can use surveillance cameras to realize unmanned label reading of steel coils entering and leaving the warehouse, improve the reading efficiency of steel coil labels entering and leaving the warehouse, and meet the needs of intelligent and unmanned management of steel coils entering and leaving the warehouse. In addition, the present invention can save a lot of manpower and financial resources, has a fast steel coil label reading speed, and stable operation, which is conducive to promotion and use.

[0090] Example 2:

[0091] This embodiment provides a steel coil label automatic reading system, including a first acquisition unit, a first detection unit, a shooting and matching unit, a second acquisition unit, a second detection unit, a third detection unit, a third acquisition unit, a fourth detection unit, and a label reading unit, wherein:

[0092] A first acquisition unit is configured to acquire a monitoring image of the loading and unloading area acquired by a monitoring camera at the top of the loading and unloading area, wherein the monitoring image of the loading and unloading area includes a plurality of steel coils;

[0093] a first detection unit, configured to perform first target detection on the monitoring image of the loading and unloading area, and determine each steel coil in the monitoring image of the loading and unloading area, and the two-dimensional coordinates of the top surface corresponding to each steel coil;

[0094] a shooting matching unit, configured to match the side monitoring camera corresponding to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil, and determine first shooting orientation adjustment information of the side monitoring camera matched to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil and the two-dimensional coordinates of the side surface of the side monitoring camera matched to each steel coil;

[0095] a second acquisition unit, configured to send the first shooting orientation adjustment information to a corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, shoots the side of the corresponding steel coil to obtain a side shot image, and obtains the side shot image fed back by the corresponding side monitoring camera;

[0096] a second detection unit, configured to perform second target detection on the side-view image, determine a side area of ​​the steel coil in the side-view image, and determine whether an area ratio of the side area of ​​the steel coil in the side-view image exceeds a first ratio threshold;

[0097] a third detection unit, configured to perform a third target detection on the side area of ​​the steel coil when it is determined that the area ratio of the side area of ​​the steel coil in the side image exceeds a first ratio threshold, and determine a first label image area in the side area of ​​the steel coil and image coordinate information of the first label image area in the side image;

[0098] a third acquisition unit, configured to generate second shooting orientation adjustment information based on the image coordinate information, and send the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, shoots the label on the side of the corresponding steel coil, obtains the label shooting image, and obtains the label shooting image fed back by the corresponding side monitoring camera;

[0099] a fourth detection unit, configured to perform a third target detection on the tag-shot image, determine a second tag image region in the tag-shot image, and determine whether an area ratio of the second tag image region in the tag-shot image exceeds a second ratio threshold;

[0100] The label reading unit is used to extract the second label image area when it is determined that the area ratio of the second label image area in the label shooting image exceeds a second ratio threshold, and scan and identify the second label image area to obtain the label reading information of the corresponding steel coil.

[0101] Example 3:

[0102] This embodiment provides a steel coil label automatic reading device, such as Figure 4 As shown, at the hardware level, it includes:

[0103] Data interface, used to establish data connection between the processor and the monitoring camera on the top of the loading and unloading area and the monitoring cameras on each side;

[0104] a memory for storing instructions;

[0105] The processor is used to read the instructions stored in the memory and execute the automatic reading method of the steel coil label in Example 1 according to the instructions.

[0106] Optionally, the device further includes an internal bus, through which the processor, memory, and data interface can be interconnected. The internal bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0107] The memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out memory (FIFO) and / or first-in-last-out memory (FILO), etc. The processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0108] Example 4:

[0109] This embodiment provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed on a computer, the computer is caused to execute the method for automatically reading steel coil labels in Example 1. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive, and / or a memory stick. The computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable device.

[0110] This embodiment further provides a computer program product, which, when executed on a computer, executes the method for automatically reading steel coil labels in Example 1. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0111] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for automatically reading a steel coil label, characterized in that: include: Obtaining a monitoring image of the loading and unloading area captured by a monitoring camera at the top of the loading and unloading area, wherein the monitoring image of the loading and unloading area includes a plurality of steel coils; Performing a first target detection on the monitoring image of the loading and unloading area to determine each steel coil in the monitoring image of the loading and unloading area and the two-dimensional coordinates of the top surface corresponding to each steel coil; Matching the side monitoring cameras corresponding to the steel coils according to the two-dimensional coordinates of the top surfaces of the steel coils, and determining the first shooting orientation adjustment information of the side monitoring cameras matched to the steel coils according to the two-dimensional coordinates of the top surfaces of the steel coils and the two-dimensional coordinates of the side surfaces of the side monitoring cameras matched to the steel coils; Sending the first shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, shoots the side of the corresponding steel coil to obtain a side shooting image, and obtains the side shooting image fed back by the corresponding side monitoring camera; Performing a second target detection on the side-view image to determine a side area of ​​the steel coil in the side-view image, and determining whether an area ratio of the side area of ​​the steel coil in the side-view image exceeds a first ratio threshold; When it is determined that the area ratio of the side area of ​​the steel coil in the side image exceeds the first ratio threshold, a third target detection is performed on the side area of ​​the steel coil to determine the first label image area in the side area of ​​the steel coil and the image coordinate information of the first label image area in the side image; Generate second shooting orientation adjustment information based on the image coordinate information, and send the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, shoots the label on the side of the corresponding steel coil, obtains the label shooting image, and obtains the label shooting image fed back by the corresponding side monitoring camera; Performing a third target detection on the tag shot image, determining a second tag image region in the tag shot image, and determining whether an area ratio of the second tag image region in the tag shot image exceeds a second ratio threshold; When it is determined that the area ratio of the second label image area in the label shot image exceeds the second ratio threshold, the second label image area is extracted, and the second label image area is scanned and identified to obtain the label reading information of the corresponding steel coil.

2. The method for automatically reading a steel coil label according to claim 1, characterized in that: The method further includes: aggregating the label reading information corresponding to each steel coil and uploading the aggregating information to a management system.

3. The method for automatically reading a steel coil label according to claim 1, characterized in that: The method of matching the side monitoring camera corresponding to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil, and determining the shooting orientation adjustment information of the side monitoring camera matched to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil and the side two-dimensional coordinates of the side monitoring camera matched to each steel coil, includes: Determine the first coordinate position of each steel coil in the reference direction of the loading and unloading area according to the two-dimensional coordinates of the top surface of each steel coil, and determine the second coordinate position of each side monitoring camera in the reference direction of the loading and unloading area according to the two-dimensional coordinates of the side surface of each side monitoring camera; Determine the second coordinate position closest to the first coordinate position of the corresponding steel coil in the reference direction of the loading and unloading area, and use the side monitoring camera corresponding to the second coordinate position closest to the first coordinate position of the corresponding steel coil as the side monitoring camera matched with the corresponding steel coil; The azimuth difference is determined based on the first coordinate position of the steel coil in the reference direction of the loading and unloading area and the second coordinate position of the side monitoring camera matched to it in the reference direction of the loading and unloading area, and the shooting azimuth adjustment information of the side monitoring camera matched to the corresponding steel coil is generated based on the azimuth difference.

4. The method for automatically reading a steel coil label according to claim 1, characterized in that: Before performing second target detection on the side-shot image, the method further includes: A. Determine a first center point of the side image and a first center coordinate of the first center point in the side image, and determine a second center point of the side area of ​​the steel coil and a second center coordinate of the second center point in the side image; B. when it is determined based on the first center coordinates and the second center coordinates that the second center point is not within a first error range around the first center point, generating first orientation fine-tuning information based on the first center coordinates and the second center coordinates; C. Sending the first orientation fine-tuning information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first orientation fine-tuning information, re-shoots the side of the corresponding steel coil, and obtains a new side shot image; The above steps AC are repeatedly performed using the new side-view image until the second center point is within a first error range around the first center point, thereby obtaining an adjusted side-view image.

5. The method for automatically reading a steel coil label according to claim 1, characterized in that: When it is determined that the area ratio of the side area of ​​the steel coil in the side image does not exceed the first ratio threshold, the method further includes: generating first zoom information according to the area ratio of the side area of ​​the steel coil in the side-shot image; Sending the first zoom information to the corresponding side monitoring camera, so that the corresponding side monitoring camera re-zooms and shoots the side of the corresponding steel coil according to the first zoom information to obtain a zoomed side image; The zoomed side shot image is used as the updated side shot image, and the second target detection is re-performed on the updated side shot image to determine the side area of ​​the steel coil and the area ratio of the side area of ​​the steel coil.

6. The method for automatically reading a steel coil label according to claim 1, characterized in that: Before performing third target detection on the tag-shot image, the method further includes: a. Determine the third center point of the label image, and the third center coordinates of the third center point in the label image, and determine the fourth center point of the second label image area, and the fourth center coordinates of the fourth center point in the label image; b. when it is determined based on the third center coordinates and the fourth center coordinates that the fourth center point is not within the second error range around the third center point, generating second orientation fine-tuning information based on the third center coordinates and the fourth center coordinates; c. Sending the second position fine-tuning information to the corresponding side surveillance camera so that the corresponding side surveillance camera adjusts the shooting position according to the second position fine-tuning information, re-shoots the label on the side of the corresponding steel coil, and obtains a new label captured image; The above steps ac are repeatedly performed using the new label captured image until the fourth center point is within the second error range around the third center point, thereby obtaining an adjusted label captured image.

7. The method for automatically reading a steel coil label according to claim 1, characterized in that: When it is determined that the area ratio of the second label image region in the label captured image does not exceed the second ratio threshold, the method further includes: generating second zoom information according to the area ratio of the second tag image region in the tag shot image; Sending the second zoom information to the corresponding side monitoring camera, so that the corresponding side monitoring camera zooms and shoots the label on the side of the corresponding steel coil according to the second zoom information to obtain a zoom label shooting image; The zoomed tag-shot image is used as the updated tag-shot image, and the third target detection is performed again on the updated tag-shot image to determine the second tag image area and the area ratio of the second tag image area.

8. A steel coil label automatic reading system, characterized in that: It includes a first acquisition unit, a first detection unit, a shooting and matching unit, a second acquisition unit, a second detection unit, a third detection unit, a third acquisition unit, a fourth detection unit and a tag reading unit, wherein: A first acquisition unit is configured to acquire a monitoring image of the loading and unloading area acquired by a monitoring camera at the top of the loading and unloading area, wherein the monitoring image of the loading and unloading area includes a plurality of steel coils; a first detection unit, configured to perform first target detection on the monitoring image of the loading and unloading area, and determine each steel coil in the monitoring image of the loading and unloading area, and the two-dimensional coordinates of the top surface corresponding to each steel coil; a shooting matching unit, configured to match the side monitoring camera corresponding to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil, and determine first shooting orientation adjustment information of the side monitoring camera matched to each steel coil according to the two-dimensional coordinates of the top surface corresponding to each steel coil and the two-dimensional coordinates of the side surface of the side monitoring camera matched to each steel coil; a second acquisition unit, configured to send the first shooting orientation adjustment information to a corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the first shooting orientation adjustment information, shoots the side of the corresponding steel coil to obtain a side shot image, and obtains the side shot image fed back by the corresponding side monitoring camera; a second detection unit, configured to perform second target detection on the side-view image, determine a side area of ​​the steel coil in the side-view image, and determine whether an area ratio of the side area of ​​the steel coil in the side-view image exceeds a first ratio threshold; a third detection unit, configured to perform a third target detection on the side area of ​​the steel coil when it is determined that the area ratio of the side area of ​​the steel coil in the side image exceeds a first ratio threshold, and determine a first label image area in the side area of ​​the steel coil and image coordinate information of the first label image area in the side image; a third acquisition unit, configured to generate second shooting orientation adjustment information based on the image coordinate information, and send the second shooting orientation adjustment information to the corresponding side monitoring camera, so that the corresponding side monitoring camera adjusts the shooting orientation according to the second shooting orientation adjustment information, shoots the label on the side of the corresponding steel coil, obtains the label shooting image, and obtains the label shooting image fed back by the corresponding side monitoring camera; a fourth detection unit, configured to perform a third target detection on the tag-shot image, determine a second tag image region in the tag-shot image, and determine whether an area ratio of the second tag image region in the tag-shot image exceeds a second ratio threshold; The label reading unit is used to extract the second label image area when it is determined that the area ratio of the second label image area in the label shooting image exceeds a second ratio threshold, and scan and identify the second label image area to obtain the label reading information of the corresponding steel coil.

9. An automatic reading device for steel coil labels, characterized in that: include: a memory for storing instructions; A processor is used to read the instructions stored in the memory and execute the automatic reading method of steel coil labels according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on a computer, the method for automatically reading steel coil labels according to any one of claims 1 to 7 is executed.

Images