A weight detection and identification system

Abstract

The present invention relates to the technical field of weight detection, and discloses a weight detection and identification system. The key points of the technical solution are as follows: it includes a first device, a second device and a third device; both the first device and the second device establish a network connection with the third device, and the network connection is used for data transmission; the first device includes a number of barcode positions, and at least one barcode can be placed in each barcode position. The first device acquires images of the barcodes in the number of barcode positions and generates barcode image information. The present invention provides a first device, a second device and a third device, and the three devices can work together. Through the first device, all the barcodes in all barcode positions can be scanned at one time, and through the second device, an image of the weight set on the detection position can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of weight detection, and more specifically, to a weight detection and identification system. Background Art

[0002] At present, when measuring and testing instruments in China are sent to the measurement institution for inspection, a unique barcode is usually assigned to an instrument to ensure that the instrument uses a unique identity information during the process of measurement and testing circulation. The scanning and identification of the barcode is an important necessary link, and currently, the identification of the barcode basically still uses a barcode scanner to obtain the barcode information. With the advancement of automated production line detection, it is inefficient to scan one barcode at a time with a barcode scanner.

[0003] Especially in the process of automatically detecting weights by an automatic detection system, the traditional solution is to install a barcode scanner on the robotic arm. During the automatic detection process, the corresponding barcode is scanned before each weight detection to achieve the one-to-one correspondence between the barcode and the weight. This solution requires first sticking the barcodes to the barcode positions one by one before detection, and during the detection process, the robotic arm has to move to the barcode position to scan and read it. This solution is time-consuming for both personnel and the robotic arm for detection. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a weight detection and identification system to overcome the above-mentioned defects in the existing technology.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A weight detection and identification system includes a first device, a second device, and a third device;

[0006] Both the first device and the second device establish a network connection with the third device, and the network connection is used for data transmission;

[0007] The first device includes a plurality of barcode positions, and at least one barcode can be placed on each barcode position. The first device acquires the images of the barcodes in the plurality of barcode positions and generates barcode image information;

[0008] The second device includes a detection position and a plurality of positions to be inspected. Weighing sets can be placed on both the detection position and the plurality of positions to be inspected. The weighing set includes at least one weight to be inspected. The second device can send the weight to be inspected to the detection position, acquire the image of the weight on the detection position and generate weight image information, and the second device is also used to weigh the weight data of the detection position and generate weight data information;

[0009] The third device includes a barcode recognition model, a text recognition model, and a database. It acquires the barcode image information and performs image processing to obtain several groups containing at least one barcode partial image and at least one text partial image, and respectively substitutes them into the barcode recognition model for recognition and the text recognition model for recognition. The barcode recognition results and text recognition results of several groups are respectively stored in the database. It acquires the weight image information, matches the weight group in the detection position with one of the barcode positions, and matches each weight image in the weight group with the barcode recognition result and text recognition result in the matched barcode position one by one to obtain the weight data information, and records the weight data information into the barcode recognition result that has been matched one by one and stores it in the database.

[0010] The present invention provides a first device, a second device, and a third device, and the three devices can work collaboratively. Through the first device, all barcodes in all barcode positions can be scanned at one time. Through the second device, the weight group image on the detection position can be acquired, and the third device can match the images collected by the first device and the second device through the network. Compared with the barcode matching method in the prior art, the matching efficiency of the present invention is greatly improved. At the same time, the second device can detect the data of the weights to be inspected and record the detected data into the barcode recognition result through the third device. Compared with the existing barcode scanning method, the present invention also recognizes the text around the barcode. When the staff identifies the barcode in the database, they can quickly determine whether it is the required barcode directly through the text, thus improving the efficiency of barcode confirmation.

[0011] As a further improvement of the present invention, the first device is configured with an initial image, a preset time threshold, a preset image threshold, and a barcode shooting strategy. The barcode shooting strategy includes an image recognition step, an image threshold comparison step, and a time threshold comparison step.

[0012] In the image recognition step, when the first device starts, it acquires the images of several barcode positions in real time and enters the image threshold comparison step.

[0013] In the image threshold comparison step, the image of the barcode position is compared with the initial image in real time for difference. If the comparison result is greater than the preset image threshold, it enters the time threshold comparison step. If the comparison result changes and is greater than the preset image threshold, it re-enters the time threshold comparison step.

[0014] In the time threshold comparison step, if the time when entering the time threshold comparison step is greater than the preset time threshold, the barcode image information is generated.

[0015] The present invention sets a barcode license plate strategy in the first device. By comparing the real-time captured image with the initial image, the generation of the barcode image information can be automatically started. In actual measurement work, a large number of weights to be inspected often need to be detected. Manually taking pictures for confirmation is very time-consuming. Through this strategy, pictures can be automatically taken according to the change of the image, and by using the time threshold comparison step, there is a certain time for placing the barcode, preventing the generation of the barcode image information before the barcode is not completely placed.

[0016] As a further improvement of the present invention, the third device includes an image processing strategy, and the image processing strategy includes an image preprocessing step, an image correction step, and an image segmentation step;

[0017] In the image preprocessing step, the barcode image information is obtained, and the outline of the image is highlighted through an algorithm, and then it enters the image correction step;

[0018] In the image correction step, the highlighted image is used for border detection and condition selection to locate and correct the barcode part image area and the text part image area, and then it enters the image segmentation step;

[0019] In the image segmentation step, the barcode part image area and the text part image area in the corrected image are segmented by the projection method, and the background noise is removed and the center position is determined to obtain the barcode part image and the text part image.

[0020] The present invention sets an image processing strategy in the third device, and the obtained barcode image information can be processed in sequence, and finally clear barcode part images and text part images are obtained.

[0021] As a further improvement of the present invention, the barcode recognition model includes a first learning strategy, and the first learning strategy includes training the barcode recognition model by real-time obtaining the barcode recognition result data in the database.

[0022] Through the setting of the first learning strategy, along with the operation of the whole system, the recognition success rate and accuracy of the barcode recognition model can be gradually enhanced, improving the practicability and applicability of the present system.

[0023] As a further improvement of the present invention, the text recognition model includes a second learning strategy, and the second learning strategy includes training the text recognition model by real-time obtaining the text recognition result data in the database.

[0024] Through the setting of the second learning strategy, along with the operation of the whole system, the recognition success rate and accuracy of the text recognition model can be gradually enhanced, improving the practicability and applicability of the present system.

[0025] As a further improvement of the present invention, the second learning strategy includes training the character recognition model by crawling data through a web crawler in real time.

[0026] By means of crawling data through the web crawler, during the process of not recognizing barcodes, the training of character recognition can be enhanced in real time, further improving the recognition success rate and accuracy of the character recognition model.

[0027] As a further improvement of the present invention, the first device includes a base, a first adjustment component, a second adjustment component, a third adjustment component and a first camera. A plurality of the barcode positions are all arranged on the base, the first camera is aligned with the plurality of barcode positions, the first adjustment component is used to adjust the movement of the first camera in the vertical direction, the second adjustment component is used to adjust the movement of the first camera in the first horizontal direction, and the third adjustment component is used to adjust the movement of the first camera in the second horizontal direction.

[0028] The present invention provides the first camera and the base equipped with a plurality of the barcode positions, so that simultaneous photographing of the plurality of barcode positions can be achieved. Moreover, three adjustment components are provided, which can move and rotate in multiple directions, making the position adjustment of the first camera more diversified and improving the accuracy and applicability of the photographing by the first camera.

[0029] As a further improvement of the present invention, the first device further includes at least one fill light, and the fill light is located on one side of the first camera.

[0030] Through the provision of the fill light, the first device can be normally used under the condition of insufficient light.

[0031] As a further improvement of the present invention, the second device further includes a robotic arm, a standard weight position, a mass comparator, a weight carrier vehicle, an automatic transport vehicle and a second camera. The weight sets are all arranged on the weight carrier vehicle, the weight carrier vehicle can be located within the position to be inspected and the detection position, a standard weight is arranged on the standard weight position, the second camera is aligned with the detection position, the detection position, the mass comparator and the standard weight position are all within the movement range of the robotic arm, the robotic arm is used to grab the standard weight and the weight to be inspected, and the automatic transport vehicle can drive the weight carrier vehicle to move, so that the weight carrier vehicle shuttles between the detection position and the position to be inspected.

[0032] The present invention uses the automatic transport vehicle to transport the weight-bearing vehicle, thus eliminating the step of manually carrying the weight-bearing vehicle. And through the setting of the robotic arm, the taking of the test weight and the standard weight can be achieved, so that it is not necessary to manually place the two weights on the mass comparator for comparative weighing, further improving the automation of the system.

[0033] As a further improvement of the present invention, the third device includes a processor and a memory. The memory stores several programs, the barcode recognition model, the character recognition model and the database. The processor can call several of the programs to control the first device, the second device and the third device.

[0034] Through the setting of the processor and the memory, the first device, the second device and the third device can be connected in series and controlled, ensuring the integrity and coordination of the present invention.

[0035] The beneficial effects of the present invention: The present invention provides a first device, a second device and a third device, and the three devices can work together. Through the first device, all barcodes in all barcode positions can be scanned at one time. Through the second device, an image of the weight set on the detection position can be obtained. And the third device can match the images collected by the first device and the second device through the network. Compared with the barcode matching method in the prior art, the matching efficiency of the present invention is greatly improved. At the same time, the second device can detect the data of the test weight and record the detected data into the barcode recognition result through the third device. And compared with the existing barcode scanning method, the present invention also recognizes the characters around the barcode. When the staff identifies the barcode in the database, they can directly determine whether it is the required barcode quickly through the characters, thus improving the efficiency of barcode confirmation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic diagram of the overall processing flow of the barcode image information of the present invention;

[0037] Figure 2 is the processing flow of the barcode recognition model of the present invention;

[0038] Figure 3 is the processing flow of the character recognition model of the present invention;

[0039] Figure 4 is a three-dimensional structure diagram of the first device of the present invention;

[0040] Figure 5 is a top view of the structure of the second device of the present invention;

[0041] Figure 6 is a schematic diagram of the barcode of the present invention.

[0042] Reference numerals: 1, first device; 11, base; 12, first adjustment component; 121, vertical rod; 122, first clamping arm; 13, second adjustment component; 131, cross bar; 14, third adjustment component; 141, second clamping arm; 142, short rod; 15, first camera; 16, bar code position; 161, character pattern; 162, bar code pattern; 17, fill light; 2, second device; 21, position to be inspected; 22, quality comparator; 23, standard weight position; 24, robotic arm; 25, positioning sensor; 26, inspection position; 27, auxiliary positioning cylinder; 28, weight carrier vehicle; 29, automatic transport vehicle. Detailed implementation manners

[0043] The present invention will be further described in detail below with reference to the drawings and embodiments. The same components are denoted by the same reference numerals. It should be noted that the terms "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to the directions in the drawings, and the terms "bottom surface" and "top surface", "inner" and "outer" refer to the directions towards or away from the geometric center of a specific component, respectively.

[0044] A weight detection and identification system according to this embodiment includes a first device 1, a second device 2 and a third device;

[0045] Both the first device 1 and the second device 2 are connected to the third device via a network, and the network connection is used for data transmission;

[0046] The first device 1 includes a plurality of bar code positions 16, and at least one bar code can be placed on each bar code position 16. The first device 1 acquires images of the bar codes in the plurality of bar code positions 16 and generates bar code image information;

[0047] The second device 2 includes an inspection position 26 and a plurality of positions to be inspected 21 (as shown in the attached Figure 4 figures, 5 positions to be inspected 21 are provided in the embodiment of the attached drawings). Weights can be placed on both the inspection position 26 and the plurality of positions to be inspected 21. The weight group includes at least one weight to be inspected. The second device 2 can send the weight to be inspected to the inspection position 26. The second device 2 acquires an image of the weight on the inspection position 26 and generates weight image information. The second device 2 is also used for weighing the weight data of the inspection position 26 and generating weight data information;

[0048] The third device includes a barcode recognition model, a text recognition model, and a database. It acquires barcode image information and performs image processing to obtain several groups each containing at least one barcode partial image and at least one text partial image, and respectively substitutes them into the barcode recognition model and the text recognition model for recognition. The barcode recognition results and text recognition results of several groups are respectively stored in the database. It acquires weight image information, matches the weight group in the detection position 26 with one of the barcode positions 16, and matches each weight image in the weight group with the barcode recognition result and text recognition result in the matched barcode position 16 one by one to obtain weight data information, and records the weight data information into the barcode recognition result that has been matched one by one and stores it in the database.

[0049] In one embodiment, as shown in the appendix Figure 6 A barcode includes a barcode pattern 162 and a text pattern 161.

[0050] Furthermore, after image processing is performed on each barcode position 16, at least one barcode partial image and at least one text partial image are obtained, and the barcode partial image and text partial image on each barcode position 16 form a group.

[0051] In one embodiment, the number of barcode positions 16 on the first device 1 is greater than or equal to the number of inspection positions 21, which can ensure that the number of weights to be inspected in the inspection positions 21 does not exceed the maximum set number of barcodes in several barcode positions 16.

[0052] Furthermore, the barcode positions 16 and the inspection positions 21 are set to be matched one by one, that is, one barcode position 16 corresponds to one inspection position 21. The number of barcodes set on each barcode position 16 is greater than or equal to the maximum number of weights to be inspected set on one inspection position 21, and the barcodes and the weights to be inspected are set to be matched one by one.

[0053] Specifically, numbers are set on the weight carriers 28 on the inspection positions 21, and the numbers on the weight carriers 28 in the detection position 26 can be quickly matched with the barcode positions 16 by shooting with the second camera.

[0054] Furthermore, when barcodes are printed, they are usually printed in batches. The barcode paper is adhered to the oil paper arranged in a roll shape. When it is necessary to place the barcode on the barcode position 16, tear off the same number of barcodes as the number of weights to be inspected in the inspection position 21 corresponding to the barcode position 16 and place them on the corresponding barcode position 16.

[0055] Furthermore, the barcodes in the barcode positions 16 are numbered along the length direction of the oil paper, such as numbers 1, 2, 3... The weights to be inspected in the inspection positions 21 and / or the detection position 26 are numbered in sequence along the left - right direction or along the length direction, such as numbers 1, 2, 3... And the barcode numbers and the weight numbers are set to be matched one by one.

[0056] In one embodiment, the first device 1 is configured with an initial image, a preset time threshold, a preset image threshold, and a barcode shooting strategy. The barcode shooting strategy includes an image recognition step, an image threshold comparison step, and a time threshold comparison step;

[0057] In the image recognition step, when the first device 1 is started, images of a plurality of barcode positions 16 are acquired in real time, and the image threshold comparison step is entered;

[0058] In the image threshold comparison step, the images of the barcode positions 16 are compared with the initial image in real time for differences. If the comparison result is greater than the preset image threshold, the time threshold comparison step is entered. If the comparison result changes and is greater than the preset image threshold, the time threshold comparison step is entered again;

[0059] In the time threshold comparison step, if the time when the time threshold comparison step is entered is greater than the preset time threshold, barcode image information is generated.

[0060] Specifically, the initial image can be set as an image without a barcode.

[0061] Specifically, the preset time threshold is usually 10S.

[0062] Furthermore, the preset time threshold can be modified according to actual needs.

[0063] Specifically, the preset image threshold is set to 10%. When the captured image has a difference of more than 10% compared with the initial image, the time threshold comparison step is entered. If this difference value changes in real time, the time threshold comparison step will be entered repeatedly, and the barcode shooting strategy will automatically delete the previous time threshold comparison step.

[0064] Furthermore, the image threshold can be modified according to actual needs.

[0065] Furthermore, a number of detection points are provided on the barcode positions 16, and the difference between the captured image and the initial image can be obtained by analyzing the images of the detection points.

[0066] In one embodiment, the third device includes an image processing strategy. The image processing strategy includes an image preprocessing step, an image correction step, and an image segmentation step;

[0067] In the image preprocessing step, barcode image information is acquired, and the outline of the image is highlighted through an algorithm, and the image correction step is entered;

[0068] [[ID=

[0069] The image segmentation step is to segment the barcode part image area and the text part image area in the corrected image by the projection method, remove background noise and determine the center position, so as to obtain the barcode part image and the text part image.

[0070] Specifically, a border is set for the barcode position 16, and the image content outside the barcode position 16 can be quickly excluded by detecting the border.

[0071] Specifically, the image preprocessing step will call the grayscale, filtering, enhancement, dilation, erosion, and binarization algorithm processing modules to process the image collected by the first camera 15.

[0072] Furthermore, an OpenCV algorithm library is set in the memory, and the image is preprocessed by this OpenCV.

[0073] In one embodiment, the recognition steps of the barcode recognition model are as shown in the appendix Figure 2 as follows.

[0074] In one embodiment, the barcode recognition model includes a first learning strategy, and the first learning strategy includes training the barcode recognition model by obtaining the barcode recognition result data in the database in real time.

[0075] In one embodiment, the recognition steps of the text recognition model are as shown in the appendix Figure 3 as follows.

[0076] In one embodiment, the text recognition model includes a second learning strategy, and the second learning strategy includes training the text recognition model by obtaining the text recognition result data in the database in real time.

[0077] Specifically, the text recognition two-model is an OCR (character recognition) model, and the recognition result combines the database and the data crawled by the web crawler to comprehensively evaluate the model and then outputs the text result.

[0078] In one embodiment, the second learning strategy includes training the text recognition model by crawling data through a web crawler in real time.

[0079] In one embodiment, the first device 1 includes a base 11, a first adjustment component 12, a second adjustment component 13, a third adjustment component 14, and a first camera 15. A plurality of barcode positions 16 are all arranged on the base 11, the first camera 15 is aligned with the plurality of barcode positions 16, the first adjustment component 12 is used to adjust the movement of the first camera 15 in the vertical direction, the second adjustment component 13 is used to adjust the movement of the first camera 15 in the first horizontal direction, and the third adjustment component 14 is used to adjust the movement of the first camera 15 in the second horizontal direction.

[0080] Specifically, the first adjustment component 12 includes a vertical rod 121 and a first clamping arm 122. The vertical rod 121 is vertically installed on the base 11. The first clamping arm 122 is provided with a first opening. The first clamping arm 122 is sleeved on the vertical rod 121 through the first opening, and the first clamping arm 122 can move along the length direction of the vertical rod 121 or rotate axially along the vertical rod 121 through the first opening.

[0081] Further, a first knob is provided at the first opening. The first knob is threadedly connected to the first clamping arm 122. When the first knob is rotated, the clamping force acting on the vertical rod 121 at the first opening increases or decreases. Through the setting of the first knob, when it is necessary to move the first clamping arm 122 up and down or rotate it, the first knob can be rotated in the reverse direction. When it is necessary to fix the first clamping arm 122, the first knob can be rotated in the forward direction.

[0082] Specifically, the second adjustment component 13 includes a cross bar 131 and a second opening provided on the first clamping arm 122. The opening direction of the second opening is perpendicular to the opening direction of the first opening. The cross bar 131 is arranged in the second opening, and the cross bar 131 can move along the opening direction of the second opening or rotate axially along the cross bar.

[0083] Further, a second knob is provided at the second opening. The second knob is threadedly connected to the first clamping arm 122. When the second knob is rotated, the clamping force acting on the cross bar 131 at the second opening increases or decreases. Through the setting of the second knob, when it is necessary to move or rotate the cross bar 131, the second knob can be rotated in the reverse direction. When it is necessary to fix the cross bar 131, the second knob can be rotated in the forward direction.

[0084] Specifically, the third adjustment component 14 includes a short rod 142 and a second clamping arm 141. The short rod 142 is horizontally arranged on the cross bar 131, and the length direction of the short rod 142 is perpendicular to the length direction of the vertical rod 121. A third opening is provided on the second clamping arm 141. The second clamping arm 141 is sleeved on the short rod 142 through the third opening, and the second clamping arm 141 can move along the length direction of the short rod 142 or rotate axially along the short rod 142 through the third opening.

[0085] Further, a third knob is provided at the third opening. The third knob is threadedly connected to the second clamping arm 141. When the third knob is rotated, the clamping force acting on the short rod 142 at the third opening increases or decreases. Through the setting of the third knob, when it is necessary to move the second clamping arm 141 horizontally or rotate it, the third knob can be rotated in the reverse direction. When it is necessary to fix the second clamping arm 141, the third knob can be rotated in the forward direction.

[0086] Specifically, the first camera 15 is installed on the second clamping arm 141.

[0087] In one embodiment, the first device 1 further includes at least one fill light 17, and the fill light 17 is located on one side of the first camera 15.

[0088] Specifically, the fill light 17 is movably connected to the cross bar 131.

[0089] The fill light 17 arranged in this way can be adjusted at any angle.

[0090] Furthermore, a rotating shaft is installed on one side of the fill light 17, and the rotating shaft is rotationally connected to the cross bar 131 through a bearing.

[0091] Specifically, as shown in the appendix Figure 4 There are two fill lights 17, which are respectively located on both sides of the cross bar 131.

[0092] In one embodiment, the top surface of the base 11 is concave, so as to form a plurality of bar code positions 16. After being arranged like this, when placing the bar code, the bar code paper will not break away from the bar code position 16.

[0093] In one embodiment, the second device 2 further includes a robotic arm 24, a standard weight position 23, a mass comparator 22, a weight carrier 28, an automatic transport vehicle 29, and a second camera. The weight sets are all arranged on the weight carrier 28. The weight carrier 28 can be located in the position to be inspected 21 and the inspection position 26. Standard weights are arranged on the standard weight position 23. The second camera is aligned with the inspection position 26. The inspection position 26, the mass comparator 22, and the standard weight position 23 are all within the movement range of the robotic arm 24. The robotic arm 24 is used to grab the standard weight and the weight to be inspected. The automatic transport vehicle 29 can drive the weight carrier 28 to move, so that the weight carrier 28 shuttles between the inspection position 26 and the position to be inspected 21.

[0094] Specifically, a plurality of standard weights are arranged on the standard weight position 23, and the masses of the plurality of standard weights are all different.

[0095] Specifically, the automatic transport vehicle 29 is an AGV cart. It can move to the bottom of the weight carrier 28, abut against the bottom of the weight carrier 28 through the cylinder on its top, and then drive the weight carrier 28 to move.

[0096] In one embodiment, the second device 2 further includes a temperature and humidity sensor, a PLC programmable logic controller, which is connected to the third device through a serial port, a cable, and WiFi. During the detection, it is necessary to monitor the temperature and humidity of the environment in real time to ensure the accuracy of the detection results. At the same time, the PLC programmable logic controller is used to control the movement of the robotic arm.

[0097] In one embodiment, an auxiliary positioning cylinder 27 and a positioning sensor 25 are provided in the detection position 26. The positioning sensor 25 is used to detect the position of the weight bearing cart 28. The auxiliary positioning cylinders 27 are located on both sides of the detection position 26. The auxiliary positioning cylinders 27 push the weight bearing cart 28 to move so that the weight bearing cart 28 is aligned with the positioning sensor 25, facilitating the subsequent grasping operation of the robotic arm 24.

[0098] In one embodiment, the barcode on the corresponding barcode position 16 is read by the first camera 15, and graphic processing is performed by a third device. The final structure is placed in a database. Then, it is determined by the second camera which vehicle's weight is being detected. Next, the robotic arm detects in the order of the weight verification regulation ABA (A represents the standard weight and B represents the weight to be inspected). The robot sequentially picks up the weights and places them on the mass comparator 22. The mass comparator 22 real-time transmits the ABA data back to the third device. The third device fills the corresponding data into the barcode and then transmits it back to the database.

[0099] In one embodiment, the third device includes a processor and a memory. The memory stores several programs, a barcode recognition model, a character recognition model, and a database. The processor can call several programs to control the first device 1, the second device 2, and the third device.

[0100] Specifically, the third device includes but is not limited to a computer, a mobile phone, a tablet, etc.

[0101] The above are only the preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements should also be regarded as within the protection scope of the present invention.

Claims

1. A weight detection and identification system, characterized by: comprising a first device (1), a second device (2) and a third device; The first device (1) and the second device (2) both establish a network connection with the third device, and the network connection is used to transmit data; The first device (1) includes a plurality of barcode positions (16), each of the barcode positions (16) can be provided with at least one barcode, and the first device (1) obtains images of the barcodes in the plurality of barcode positions (16) and generates barcode image information; The second device (2) includes a detection position (26) and a plurality of positions to be detected (21), wherein the detection position (26) and the plurality of positions to be detected (21) can each be placed with a weight group, wherein the weight group includes at least one weight to be detected, and the second device (2) can deliver the weight to be detected to the detection position (26). The second device (2) obtains an image of the weight on the detection position (26) and generates weight image information. The second device (2) is also used to weigh the weight data of the detection position (26) and generate weight data information. The third device includes a barcode recognition model, a text recognition model and a database, obtains the barcode image information and performs image processing to obtain several groups of images containing at least one barcode part and at least one text part, and substitutes them into the barcode recognition model and the text recognition model for recognition respectively, stores the barcode recognition results and text recognition results of the several groups in the database respectively, obtains the weight image information, and matches the weight group in the detection position (26) with one of the barcode positions (16), matches each weight image in the weight group with the barcode recognition result and text recognition result in the matched barcode position (16) one by one, obtains weight data information, records the weight data information in the matched barcode recognition results and stores them in the database.

2. A weight detection and identification system according to claim 1, characterized in that: The first device (1) is configured with an initial image, a preset time threshold, a preset image threshold and a barcode shooting strategy, wherein the barcode shooting strategy includes an image recognition step, an image threshold comparison step and a time threshold comparison step; In the image recognition step, when the first device (1) is started, images of a plurality of barcode positions (16) are acquired in real time, and the image threshold comparison step is entered; The image threshold comparison step is to compare the difference between the image of the barcode position (16) and the initial image in real time. If the comparison result is greater than the preset image threshold, the step of time threshold comparison is entered. If the comparison result changes and is greater than the preset image threshold, the step of time threshold comparison is entered again. The time threshold comparison step generates the barcode image information if the time of entering the time threshold comparison step is greater than the preset time threshold.

3. A weight detection and identification system according to claim 1, characterized in that: The third device includes an image processing strategy, which includes an image preprocessing step, an image correction step, and an image segmentation step; The image preprocessing step acquires the barcode image information and highlights the image outline through an algorithm, and then enters the image correction step; The image correction step uses border detection and conditional selection to locate and correct the barcode image area and the text image area of the highlighted image, and then enters the image segmentation step; The image segmentation step segments the barcode image area and the text image area in the corrected image by projection method, removes background noise and determines the center position to obtain the barcode image and the text image.

4. A weight detection and identification system according to claim 1, characterized in that: The barcode recognition model includes a first learning strategy, which includes acquiring barcode recognition result data in the database in real time to train the barcode recognition model.

5. The weight detection and identification system according to claim 1, characterized in that: The character recognition model includes a second learning strategy, which includes acquiring character recognition result data in the database in real time to train the character recognition model.

6. A weight detection and identification system according to claim 5, characterized in that: The second learning strategy includes training the text recognition model by crawling data through a web crawler in real time.

7. A weight detection and identification system according to any one of claims 1 to 6, characterized in that: The first device (1) comprises a base (11), a first adjusting component (12), a second adjusting component (13), a third adjusting component (14) and a first camera (15); a plurality of barcode positions (16) are arranged on the base (11); the first camera (15) is aligned with the plurality of barcode positions (16); the first adjusting component (12) is used to adjust the movement of the first camera (15) in a vertical direction; the second adjusting component (13) is used to adjust the movement of the first camera (15) in a first horizontal direction; and the third adjusting component (14) is used to adjust the movement of the first camera (15) in a second horizontal direction.

8. The weight detection and identification system according to claim 7, characterized in that: The first device (1) further includes at least one fill light (17), and the fill light (17) is located on one side of the first camera (15).

9. A weight detection and identification system according to any one of claims 1 to 6, characterized in that: The second device (2) further comprises a mechanical arm (24), a standard weight position (23), a mass comparator (22), a weight carrier (28), an automatic transport vehicle (29) and a second camera. The weight groups are all arranged on the weight carrier (28). The weight carrier (28) can be located in the position to be inspected (21) and the inspection position (26). A standard weight is arranged on the standard weight position (23). The second camera is aligned with the inspection position (26). The inspection position (26), the mass comparator (22) and the standard weight position (23) are all located within the movement range of the mechanical arm (24). The mechanical arm (24) is used to grab the standard weight and the weight to be inspected. The automatic transport vehicle (29) can drive the weight carrier (28) to move so that the weight carrier (28) can move back and forth between the inspection position (26) and the position to be inspected (21).

10. A weight detection and identification system according to any one of claims 1 to 6, characterized in that: The third device includes a processor and a memory, wherein the memory stores a plurality of programs, the barcode recognition model, the text recognition model and the database, and the processor can call a plurality of the programs to control the first device (1), the second device (2) and the third device.

Images