High-precision automatic scribing constant volume identification method

Abstract

The invention relates to the technical field of electronic information, and provides a high-precision automatic scribing constant volume identification method which comprises the following steps: initializing a device and carrying out quantitative test calibration on a liquid adding assembly; activating a camera assembly to carry out scribed line positioning, and monitoring that a liquid adding needle in a liquid adding assembly reaches a preset liquid level positioning threshold value range so as to determine that the liquid adding needle reaches a liquid pre-adding position; the liquid adding needle is controlled to conduct coarse liquid adding operation till the liquid level reaches the coarse liquid adding control threshold value range, and then medium-speed liquid adding operation is switched till the liquid level approaches the target scribed line and reaches the medium-speed liquid adding control threshold value range; and a liquid adding needle is executed to carry out fine liquid adding operation until the liquid level is aligned with the target scribed line and reaches the fine liquid adding adjustment control threshold value range, and liquid level positioning and constant volume recognition are completed. Scale marks, a liquid adding needle and the liquid level height can be accurately recognized, meanwhile, the liquid adding speed and the single-time liquid adding amount are accurately adjusted in combination with cooperative control of a straight-through control valve and a peristaltic pump, the requirements for the high liquid adding amount and the high precision are considered in different liquid adding stages, and excessive liquid adding is prevented.

Description

Technical Field

[0001] The invention relates to the field of electronic information technology, and in particular to a high-precision automatic marking line constant volume recognition method. Background Art

[0002] In many industries and scenarios, the extraction and volume determination of liquid reagents are common laboratory and industrial processes. This type of work requires high skills from the operator, and it is necessary to ensure the safety, pollution-free and high precision of the liquid drop process during operation. Therefore, the entire line determination process is highly repetitive and requires high stability.

[0003] In the prior art, the traditional method of setting volume by engraved lines mainly relies on manual operation. When the operator uses a volumetric flask to prepare a solution of accurate concentration, when adding a liquid reagent, there is still 1 to 2 cm from the scale line, and the liquid reagent is sucked into the volumetric flask with a rubber-tipped dropper. The line of sight is level with the lowest point of the concave liquid surface, and the liquid finally reaches the scale line. However, this traditional volume setting process requires the naked eye to judge the engraved lines, resulting in inaccurate precision of the volume setting. Secondly, when using highly toxic, explosive, volatile or corrosive liquids or gases, the operator is easily exposed to the liquid, which causes serious harm to the body.

[0004] In order to avoid the danger and low precision of manual volume determination, the market has begun to try to use automated equipment for line determination. However, due to the high environmental requirements of high-precision cameras, existing automatic volume determination technology may have problems such as recognition errors and excessive liquid addition. Therefore, the existing technology has certain limitations and technical difficulties. Summary of the invention

[0005] In view of the above problems, the present invention proposes a high-precision automatic line volume recognition method to solve the problems of insufficient accuracy, complex operation, and easy overflow or insufficient filling in traditional liquid filling technology. The above-mentioned invention object of the present invention is achieved through the following technical solutions: The present invention provides a high-precision automatic marking line constant volume recognition method, which performs automatic liquid addition constant volume recognition through a liquid adding device based on machine vision technology. The liquid adding device includes a drive control component, a camera component and a liquid adding component. The camera component and the liquid adding component are driven by the drive control component to adjust the position and perform liquid addition operations respectively, including: Initialize the device and perform quantitative test and calibration on the liquid adding components; Activate the camera component to perform marking line positioning, monitor the liquid adding needle in the liquid adding component to reach a preset liquid level positioning threshold range, and determine that the liquid adding needle reaches the pre-liquid adding position; Control the dosing needle to perform a rough dosing operation until the liquid level reaches the rough dosing control threshold range, and then switch to a medium-speed dosing operation until the liquid level approaches the target scale line and reaches the medium-speed dosing control threshold range; Execute the liquid adding needle to perform fine liquid adding operation until the liquid level is aligned with the target scale line and reaches the fine liquid adding adjustment control threshold range, completing the liquid level positioning and fixed volume identification.

[0006] Furthermore, the drive control component includes a PLC controller, a camera displacement motor controller, and a dosing needle displacement motor controller; wherein the camera displacement motor controller is used to control the displacement and rotation of the camera component; and the dosing needle displacement motor controller is used to control the displacement and rotation of the dosing component; The liquid adding component includes a liquid adding needle, a straight-through valve and a peristaltic pump. The straight-through valve is used to control the flow rate and dripping amount of the liquid; the peristaltic pump is used to adjust the delivery rate of the liquid.

[0007] Further, the device is initialized and the liquid adding component is quantitatively tested and calibrated, including, After the camera displacement motor controller and the dosing needle displacement motor controller are zero-calibrated, they are adjusted to the waiting area to enter the ready state and the peristaltic pump is activated to clean the pipeline; at the same time, the dosing component is quantitatively tested and calibrated based on the initialized device.

[0008] Further, a quantitative test and calibration of the liquid adding component is performed based on the initialized device, including: The delivery rate of the peristaltic pump is set based on the number of revolutions of the peristaltic pump to achieve a fixed droplet volume, so as to control the droplet volume delivered by the peristaltic pump per revolution; The opening and closing size of the through valve is adjusted based on the set amount of droplets delivered by the peristaltic pump per revolution to control the single droplet volume of the dosing needle, and the operation is continued to ensure that the amount of droplets delivered by the peristaltic pump per revolution and the single droplet volume of the dosing needle are within the preset error range to complete the quantitative test calibration.

[0009] Further, activating the camera assembly to perform scribed line positioning includes, Position the camera in the camera assembly to the preset starting position of the container mouth, and establish a rectangular coordinate system with the lower left corner of the image obtained by the camera as the origin; Activate the camera, acquire images in real time and perform median filtering to eliminate the corresponding image noise; perform grayscale conversion on the median filtered image and apply the Canny edge detection algorithm to highlight the edges in the image; use Hough line detection to process the edge acquisition image to obtain the long straight line as the target line, and adjust the vertical position of the camera based on the result of Hough line detection to align the target line with the center of the image acquired by the camera until the error between the target line and the center of the image is maintained within a preset range.

[0010] Further, monitoring the liquid adding needle in the liquid adding component to reach a preset liquid level positioning threshold range to determine that the liquid adding needle reaches the pre-liquid adding position, including, During the image acquisition process, a mask is applied to cover the target scribed line and the area below, and the image is binarized. At the same time, the element boundaries of the image are clarified through corrosion and expansion operations. The dosing needle displacement motor controller controls the dosing needle to move downward according to a fixed step length until the position of the dosing needle reaches a preset liquid level positioning threshold range; the position of the dosing needle is determined as the white proportion value in the image.

[0011] Further, controlling the liquid adding needle to perform a rough liquid adding operation until the liquid level reaches a rough liquid adding control threshold range includes: Grayscale processing is performed on the image captured by the camera in real time, and the delivery rate of the peristaltic pump is adjusted to perform a rough liquid addition operation and the proportion of black pixels in the grayscale processed image is monitored in real time; When the black pixel ratio reaches the preset rough liquid addition control threshold range, the rough liquid addition operation is stopped and switched to the medium speed liquid addition operation.

[0012] Further, switching to a medium-speed liquid addition operation until the liquid level approaches the target mark and reaches a medium-speed liquid addition control threshold range includes, The dosing needle displacement motor controller is used to control the dosing needle to lift a fixed distance, and the delivery rate of the peristaltic pump is reduced to a preset speed for medium-speed dosing operation; Grayscale processing is performed based on the image obtained in real time by the camera, and mask processing is performed on the target scribed line and the part above it. Real-time monitoring is performed until the liquid level reaches the medium-speed liquid addition control threshold range, and the medium-speed liquid addition is stopped.

[0013] Further, the liquid adding needle is executed to perform fine liquid adding operation until the liquid level is aligned with the target scale line and reaches the fine liquid adding adjustment control threshold, completing the liquid level positioning and fixed volume identification, including, Grayscale processing, edge detection and opening operation are performed on the image captured by the camera to obtain the edge of the liquid level height and subtract the preset pixel value to determine the corresponding concave liquid surface position; at the same time, the spray mode is adjusted to a single-drop spray mode, and the concave liquid surface position is monitored in real time until the concave liquid surface position reaches the fine liquid addition adjustment control threshold, and the fine liquid addition adjustment control threshold is a preset range exceeding the target scale line.

[0014] Compared with the prior art, the present invention has at least one of the following beneficial effects: The present invention can accurately identify scale lines, dosing needles and liquid level heights, and combines the coordinated control of the through control valve and the peristaltic pump to achieve precise adjustment of the dosing speed and the single dosing amount, taking into account both high dosing amount and high precision requirements at different dosing stages to prevent excessive dosing. At the same time, it effectively solves the error problem caused by traditional reliance on visual judgment, and comprehensively improves the reliability and efficiency of automation equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1It is a flowchart of the steps of the high-precision automatic marking line constant volume recognition method of the present invention; Figure 2 This is a flow chart of device initialization in the high-precision automatic marking line constant volume recognition method of the present invention; Figure 3 It is a flowchart of marking line positioning in the high-precision automatic marking line constant volume recognition method of the present invention; Figure 4 It is a flow chart of the identification of the liquid adding needle in the high-precision automatic marking line fixed volume identification method of the present invention; Figure 5 This is a flow chart of the rough liquid addition operation in the high-precision automatic marking line constant volume recognition method of the present invention; Figure 6 This is a flow chart of the medium-speed liquid adding operation of the high-precision automatic marking line constant volume recognition method of the present invention; Figure 7 The present invention is a flowchart of the precise liquid addition operation in the high-precision automatic marking line constant volume recognition method. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0018] First embodiment In order to overcome the problems of inconsistent liquid volume determination accuracy and high operation risk caused by manual operation in the prior art, the present invention proposes a high-precision automatic line determination method based on image recognition and precise fluid control. This method can solve some defects of traditional volume determination and existing automation equipment, and combine high-resolution cameras, machine learning algorithms and precision fluid control components to achieve full automation in the liquid processing process, accurately identify the liquid level and scale lines, and automatically adjust the liquid injection amount to ensure that each addition reaches the predetermined accuracy, effectively reduce human errors and improve operational safety. The specific implementation methods are as follows: like Figure 1As shown, the present invention provides a high-precision automatic scale line fixed volume recognition method, which performs automatic liquid addition and fixed volume recognition through a liquid adding device based on machine vision technology. The liquid adding device includes a drive control component, a camera component and a liquid adding component. The camera component and the liquid adding component are driven by the drive control component to adjust the position and perform liquid adding operations, respectively, to improve the accuracy of the liquid addition amount, quantitative control of a single droplet, effectively avoid excessive liquid addition, and improve the accuracy of scale line recognition and scale line fixed volume liquid addition under a fixed light source based on a high-precision physical zoom camera and a machine vision algorithm. The specific steps include: Initialize the device and perform quantitative test and calibration on the liquid adding components; Activate the camera component to perform marking line positioning, monitor the liquid adding needle in the liquid adding component to reach a preset liquid level positioning threshold range, and determine that the liquid adding needle reaches the pre-liquid adding position; Control the dosing needle to perform a rough dosing operation until the liquid level reaches the rough dosing control threshold range, and then switch to a medium-speed dosing operation until the liquid level approaches the target scale line and reaches the medium-speed dosing control threshold range; Execute the liquid adding needle to perform fine liquid adding operation until the liquid level is aligned with the target scale line and reaches the fine liquid adding adjustment control threshold range, completing the liquid level positioning and fixed volume identification.

[0019] Furthermore, the drive control component includes a PLC controller, a camera displacement motor controller, and a dosing needle displacement motor controller; wherein the camera displacement motor controller is used to control the displacement and rotation of the camera component; and the dosing needle displacement motor controller is used to control the displacement and rotation of the dosing component; The liquid adding component includes a liquid adding needle, a straight-through valve and a peristaltic pump. The straight-through valve is used to control the flow rate and dripping amount of the liquid; the peristaltic pump is used to adjust the delivery rate of the liquid.

[0020] Further, the device is initialized and the liquid adding component is quantitatively tested and calibrated, including, After the camera displacement motor controller and the dosing needle displacement motor controller are zero-calibrated, they are adjusted to the waiting area to enter the ready state and the peristaltic pump is activated to clean the pipeline; at the same time, the dosing component is quantitatively tested and calibrated based on the initialized device.

[0021] like Figure 2 As shown, in an embodiment of the present invention, the initialization stage of the device includes the following key steps: zeroing the upper and lower motors and the rotary motor and selecting the rotary motor to the waiting area, while the upper and lower motors are moved downward to the waiting area. In addition, the peristaltic pump will discharge the waste liquid and suck it back into the clean pipeline to ensure the cleanliness of the liquid adding device and prevent cross contamination. Through this step, all key parameters are reset before accurate liquid volume determination and identification to ensure consistency and repeatability of the operation. Therefore, the purpose of initializing the device in this application is to first ensure the accuracy of liquid addition to prevent excessive liquid addition.

[0022] Further, a quantitative test and calibration of the liquid adding component is performed based on the initialized device, including: The delivery rate of the peristaltic pump is set based on the number of revolutions of the peristaltic pump to achieve a fixed droplet volume, so as to control the droplet volume delivered by the peristaltic pump per revolution; The opening and closing size of the through valve is adjusted based on the set amount of droplets delivered by the peristaltic pump per revolution to control the single droplet volume of the dosing needle, and the operation is continued to ensure that the amount of droplets delivered by the peristaltic pump per revolution and the single droplet volume of the dosing needle are within the preset error range to complete the quantitative test calibration.

[0023] Since the viscosity and tension of different liquids are different, and the refined addition of liquids needs to ensure the consistency of the amount of liquid dropped each time. Therefore, the pressure nozzle, through valve and peristaltic pump need to be calibrated before each use. Take a certain amount of liquid and calibrate the quantitative liquid weighing to obtain the number of rotations required for quantitative liquid addition. Secondly, accurately control the opening and closing of the nozzle to test and collect the value of the single drop amount. After adjusting the hardware parameters, when the numerical results are basically consistent and within the acceptable error range, the hardware calibration step is completed. In terms of image recognition, in order to avoid the influence of external light sources, the device equipment is equipped with a light-shielding shell. The visual recognition equipment is carried out in an environment without external light. A fixed light source is set inside to ensure the accuracy of machine vision recognition of the scale and liquid level, set the image reading range of the camera, adjust and fix the appropriate contrast and brightness to make the picture content stable, and the image will not darken or lose pixel information as the picture content changes.

[0024] Further, activating the camera assembly to perform scribed line positioning includes, Position the camera in the camera assembly to the preset starting position of the container mouth, and establish a rectangular coordinate system with the lower left corner of the image obtained by the camera as the origin; Activate the camera, acquire images in real time and perform median filtering to eliminate the corresponding image noise; perform grayscale conversion on the median filtered image and apply the Canny edge detection algorithm to highlight the edges in the image; use Hough line detection to process the edge acquisition image to obtain the long straight line as the target line, and adjust the vertical position of the camera based on the result of Hough line detection to align the target line with the center of the image acquired by the camera until the error between the target line and the center of the image is maintained within a preset range.

[0025] like Figure 3As shown, the accuracy of liquid addition is ensured by accurately positioning the scale line of the container, unlocking the camera and moving the camera motor upward to a fixed position above the scale line, adjusting the background light source information according to the ambient light source conditions, adjusting the contrast based on the information, setting the number of steps for the camera motor to move downward, initializing the total descent distance, moving the camera motor downward slowly once to record the descent distance, acquiring and storing the image in real time, and processing the image to see if the scale line appears in the picture range. If not, the camera motor moves downward slowly once to record the descent distance. If yes, reduce the descent rate, rise or descend a fixed distance based on the current position of the scale line, and descend downward based on the new descent rate while recording the descent distance. The camera acquires and stores the image in real time, processes the image to identify whether the scale line appears in the middle of the picture. If not, the camera motor moves upward to a fixed position, and then returns to the descent distance initialization to start secondary line finding. If yes, perform liquid addition needle identification.

[0026] Specifically, the camera first moves to the position below the bottle mouth according to the bottle capacity. Using machine recognition technology, a rectangular coordinate system is established for the image obtained by the camera. The point in the lower left corner is set as the origin, the horizontal axis is the x-axis, which gradually increases from left to right, and the vertical axis is the y-axis, which gradually increases from top to bottom. The up and down motion control motor drives the camera to move slowly downward at a fixed step length, and each frame is first subjected to median filtering to remove additional noise in the picture. The processed color image is converted to grayscale, and the Canny function is used for edge detection. The processed image is detected by Hough line and the threshold is changed to quickly find the longer straight line in the picture. Since some water droplets may hang on the wall during the process, it may affect the recognition. Therefore, the position of the screen where the engraved line is located is determined by finding the row where the widest white pixel point in the picture is located. When the pixel height of the widest white pixel row is within the preset range, the step number is stopped and the refined engraved line recognition step is started. At this time, if the pixel height of the widest white pixel row is less than the height of the center line of the picture, the camera automatically rises a certain height to ensure that it is raised above the center line height. Otherwise, the upper and lower motor speeds are lowered and the machine starts to slowly descend. The image recognition range is automatically narrowed, and the recognition focal length is adjusted to perform median filtering and grayscale conversion on the image. The Canny function is used for edge detection. By identifying each pixel row, the row with the most white pixels is found, which is the target engraved line position. When the corresponding height reaches the center of the image, the error stops when it is 2 pixels above and below. At this point, the camera is flush with the engraved line, which is used for accurate volume determination in the later stage.

[0027] Further, monitoring the liquid adding needle in the liquid adding component to reach a preset liquid level positioning threshold range to determine that the liquid adding needle reaches the pre-liquid adding position, including, During the image acquisition process, a mask is applied to cover the target scribed line and the area below, and the image is binarized. At the same time, the element boundaries of the image are clarified through corrosion and expansion operations. The dosing needle displacement motor controller controls the dosing needle to move downward according to a fixed step length until the position of the dosing needle reaches a preset liquid level positioning threshold range; the position of the dosing needle is determined as the white proportion value in the image.

[0028] like Figure 4 As shown, in the dosing needle recognition stage, the upper and lower motors are first used to find the zero and position the rotary motor to the container bottle mouth position. Then, the number of steps for the upper and lower motors to move downward is set to make the motor move downward according to the preset number of steps to gradually approach the target position. At the same time, the camera acquires and stores images in real time, and determines the position of the dosing needle through analysis to determine whether the dosing needle is found in the image. If the image analysis confirms that the dosing needle has been correctly positioned, the rough dosing operation is entered. If the dosing needle cannot be detected or the maximum step length has been reached and the positioning is not successful, the dosing needle position will be adjusted again. If the continuous adjustment fails to position successfully, the recognition operation will stop to avoid invalid or erroneous operations from continuing.

[0029] As for the process of identification and analysis of the dosing needle in image processing, it is as follows: the up and down motors of the dosing needle drive the dosing needle into the bottle and slowly descend. At the same time, the engraved line and the part below are masked by a mask, the influence of the engraved line and the droplets are ignored, the real-time image is binarized, and then eroded and expanded. The proportion of white area in the image is detected in real time. When the dosing needle appears on the image, the proportion of white will suddenly increase. When the proportion of white reaches the threshold, the dosing needle stops moving downward, and the dosing needle reaches the dosing position. Then it enters the dosing stage.

[0030] Further, controlling the liquid adding needle to perform a rough liquid adding operation until the liquid level reaches a rough liquid adding control threshold range includes: Grayscale processing is performed on the image captured by the camera in real time, and the delivery rate of the peristaltic pump is adjusted to perform a rough liquid addition operation and the proportion of black pixels in the grayscale processed image is monitored in real time; When the black pixel ratio reaches the preset rough liquid addition control threshold range, the rough liquid addition operation is stopped and switched to the medium speed liquid addition operation.

[0031] like Figure 4As shown, in the rough liquid addition operation stage, the required rough liquid addition amount is calculated according to the initial setting value provided by the PLC controller and the specific specifications of the container. And the rough liquid addition is started based on the calculated delivery rate of the peristaltic pump. Continuously monitor and judge whether the rough liquid addition process has reached the predetermined liquid volume. If the required liquid volume has not been reached, continue to push the peristaltic pump to work; once it is confirmed that the liquid addition amount has reached the standard, take a photo and record it immediately and automatically enter the medium-speed liquid addition stage. In order to improve the liquid addition efficiency, it is necessary to increase the liquid addition amount and liquid addition rate. The camera obtains real-time high-definition images at the previously adjusted position, monitors the real-time images in real time through grayscale processing, adjusts the peristaltic pump speed, and performs rapid liquid addition. The black pixel ratio value obtained from the screen is used as the real-time value until the detection reaches the set threshold to stop the rough liquid addition work, and the lowest position liquid level appears in the set threshold screen.

[0032] Further, switching to a medium-speed liquid addition operation until the liquid level approaches the target mark and reaches a medium-speed liquid addition control threshold range includes, The dosing needle displacement motor controller is used to control the dosing needle to lift a fixed distance, and the delivery rate of the peristaltic pump is reduced to a preset speed for medium-speed dosing operation; Grayscale processing is performed based on the image obtained in real time by the camera, and mask processing is performed on the target scribed line and the part above it. Real-time monitoring is performed until the liquid level reaches the medium-speed liquid addition control threshold range, and the medium-speed liquid addition is stopped.

[0033] like Figure 6As shown in the figure, in the medium-speed liquid addition operation stage, in order to reduce the interference of splashes that may be generated during the liquid addition process on image recognition, the upper and lower motors first lift the liquid addition needle by a certain number of steps. Subsequently, the delivery rate and movement steps of the peristaltic pump are set so that the peristaltic pump starts to deliver liquid at the adjusted medium rate, and the number of movement steps each time is accurately controlled to maintain the continuity and stability of liquid addition. During this process, the camera continuously acquires and stores images in real time, and monitors the position change of the liquid surface through image processing technology. When the liquid surface is detected to be clearly visible at the bottom of the image, it is evaluated whether it is close to the target liquid level. If the predetermined liquid level is not reached, the peristaltic pump will continue a single operation to fine-tune the liquid surface position; if the liquid surface is close enough to or reaches the target position, this stage ends, and then it is ready to enter a more refined liquid addition adjustment, that is, the fine liquid addition operation stage. As for the processing details, in the medium-speed liquid addition operation stage, the liquid addition needle is first lifted by the upper and lower motors for a distance to prevent splashes during liquid addition. After reducing the liquid addition rate, medium-speed liquid addition is performed. The picture of the liquid surface is recognized in real time and image processing is performed. First, the video frame is obtained for grayscale processing. Binarize the processed grayscale image and run it through the open loop to remove minor noise. Then, use a mask to block the engraved line and the part above it, because the engraved line will be considered as the longest line in the image, which will affect the recognition of the liquid level. After covering the engraved line, the liquid level can be judged more accurately. Since the liquid level of each frame will fluctuate during dripping, the liquid level needs to be stable after dripping to take the picture for judgment. When the row with the most white pixels appears, judge whether the corresponding height reaches the threshold range. If it is not reached, continue dripping; if it is reached, the liquid level reaches the target position and the medium-speed liquid addition stage ends.

[0034] Further, the liquid adding needle is executed to perform fine liquid adding operation until the liquid level is aligned with the target scale line and reaches the fine liquid adding adjustment control threshold, completing the liquid level positioning and fixed volume identification, including, Grayscale processing, edge detection and opening operation are performed on the image captured by the camera to obtain the edge of the liquid level height and subtract the preset pixel value to determine the corresponding concave liquid surface position; at the same time, the spray mode is adjusted to a single-drop spray mode, and the concave liquid surface position is monitored in real time until the concave liquid surface position reaches the fine liquid addition adjustment control threshold, and the fine liquid addition adjustment control threshold is a preset range exceeding the target scale line.

[0035] like Figure 7As shown in the figure, in the first stage of fine liquid addition, the upper and lower motors are moved upward by a certain number of steps to prevent splashing water from interfering with image recognition. Subsequently, the peristaltic pump delivery rate and movement steps are set, and the peristaltic pump works in a single operation to ensure the accuracy of each step of liquid addition. The camera acquires and stores image data in real time and continuously monitors the position of the liquid surface. Through image processing technology, the system evaluates whether the liquid surface is close to the scale line position and determines whether the first stage of fine liquid addition is completed, that is, whether the liquid surface reaches the vicinity of the target scale line. If it does not reach the predetermined position, the peristaltic pump will continue to operate in a single operation to fine-tune the liquid surface position. When entering the second stage of fine liquid addition, the key changes in the operation include adjusting the peristaltic pump delivery rate to cooperate with the micro valve to adjust the injection speed to achieve the ultimate accuracy of single drop liquid addition. At this time, the camera continues to monitor the liquid surface in real time and processes image data to confirm whether the liquid surface has accurately reached the position above the scale line. If it is detected that the liquid surface has not completely reached the preset position, the system will continue to perform a single injection operation. Once the liquid surface is accurately aligned with the scale line, this stage is completed, and then the system performs a zero reset operation, indicating that the high-precision scale line fixed volume process has been successfully completed. This series of precise control ensures the extremely high accuracy of the liquid addition process and greatly improves the efficiency and reliability of automated volume determination. It should be specifically noted that when entering the fine liquid addition stage, the speed of the peristaltic pump is adjusted to a low speed to reduce the rate of reagent addition. In the image processing part, the image is first grayscale processed, followed by edge detection and opening operations to make the liquid surface clearer. Subsequently, the number of white pixels in each row is counted. When the height of the row where the widest pixel is located is reduced by 3-4 pixels and is below the scale line (the concave liquid surface position is taken, and it has been tested and verified that it will be 3-4 pixels lower than the widest liquid surface position) and reaches the scale line, the second stage of fine liquid addition begins. The pressure nozzle will cooperate with the straight-through valve to change the reagent addition into a single drop of liquid. The amount of liquid droplets sprayed each time is 0.01ml, and the accuracy is much higher than the 0.05ml single drop of liquid output of the traditional rubber-tipped dropper. Until the pixel height at the concave liquid surface position is equal to the scale line, the fine liquid addition process is completed, and a photo is taken to record the volume determination completion photo, and the entire high-precision automatic scale determination and recognition method is completed.

[0036] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

[0037] It should be noted that the above embodiments can be freely combined as needed. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention.

Claims

1. A high-precision automatic marking line constant volume recognition method, which performs automatic liquid filling constant volume recognition through a liquid filling device based on machine vision technology, wherein the liquid filling device comprises a drive control component, a camera component and a liquid filling component, wherein the camera component and the liquid filling component are driven by the drive control component to adjust the position and perform liquid filling operations, respectively, and is characterized in that: include, The device is initialized and the liquid adding component is quantitatively tested and calibrated; Activating the camera assembly to perform marking line positioning, monitoring the liquid adding needle in the liquid adding assembly to reach a preset liquid level positioning threshold range, so as to determine that the liquid adding needle reaches a pre-liquid adding position; Controlling the liquid adding needle to perform a rough liquid adding operation until the liquid level reaches a rough liquid adding control threshold range, and then switching to a medium speed liquid adding operation until the liquid level approaches a target scale line and reaches a medium speed liquid adding control threshold range; The liquid adding needle is executed to perform a fine liquid adding operation until the liquid surface is aligned with the target scale line and reaches a fine liquid adding adjustment control threshold range, thereby completing the liquid surface positioning and fixed volume identification.

2. The high-precision automatic marking line volume recognition method according to claim 1 is characterized in that: The drive control component includes a PLC controller, a camera displacement motor controller, and a dosing needle displacement motor controller; wherein the camera displacement motor controller is used to control the displacement and rotation of the camera component; and the dosing needle displacement motor controller is used to control the displacement and rotation of the dosing component; The liquid adding component comprises a liquid adding needle, a through valve and a peristaltic pump. The through valve is used to control the flow rate and the dripping amount of the liquid; the peristaltic pump is used to adjust the delivery rate of the liquid.

3. The high-precision automatic marking line volume recognition method according to claim 2 is characterized in that: The device is initialized and the liquid adding component is quantitatively tested and calibrated, including: After the positions of the camera displacement motor controller and the liquid-filling needle displacement motor controller are zero-calibrated, they are adjusted to the waiting area to enter the ready state and the peristaltic pump is activated to clean the pipeline; at the same time, the liquid-filling component is quantitatively tested and calibrated based on the initialized device.

4. The high-precision automatic marking line volume recognition method according to claim 2 is characterized in that: The step of performing quantitative testing and calibration on the liquid adding component based on the initialized device includes: The delivery rate of the peristaltic pump is set based on the number of rotations of the peristaltic pump to achieve a fixed droplet volume, so as to control the droplet volume delivered by the peristaltic pump per rotation; The opening and closing size of the through valve is adjusted based on the set amount of droplets delivered by the peristaltic pump per revolution to control the single droplet amount of the dosing needle, and the operation is continued to ensure that the amount of droplets delivered by the peristaltic pump per revolution and the single droplet amount of the dosing needle are within a preset error range to complete the quantitative test calibration.

5. The high-precision automatic marking line volume recognition method according to claim 2 is characterized in that: Activating the camera assembly to perform reticle positioning includes, Positioning the camera in the camera assembly to a preset starting position of the container mouth, and establishing a rectangular coordinate system with the lower left corner of the image acquired by the camera as the origin; Activate the camera, acquire images in real time and perform median filtering to eliminate corresponding image noise; The image after the median filtering is subjected to grayscale conversion, and the Canny edge detection algorithm is applied to highlight the edges in the image; the edges are processed by using Hough line detection to obtain a long straight line in the image as the target scribed line, and the vertical position of the camera is adjusted based on the result of the Hough line detection, so that the target scribed line is aligned with the center of the image obtained by the camera until the error between the target scribed line and the center of the image is maintained within a preset range.

6. The high-precision automatic marking line volume recognition method according to claim 2 is characterized in that: Monitoring the liquid adding needle in the liquid adding component to reach a preset liquid level positioning threshold range to determine that the liquid adding needle reaches a pre-liquid adding position, including: During the image acquisition process, a mask is applied to cover the target scribed line and the area below, and the image is binarized, and at the same time, the element boundaries of the image are clarified through corrosion and expansion operations; The dosing needle displacement motor controller controls the dosing needle to move downward according to a fixed step length until the position of the dosing needle reaches the preset liquid level positioning threshold range; the position of the dosing needle is determined as the white proportion value in the image.

7. The high-precision automatic marking line volume recognition method according to claim 1 is characterized in that: Controlling the liquid adding needle to perform a rough liquid adding operation until the liquid level reaches a rough liquid adding control threshold range, including: Performing grayscale processing on the image captured by the camera in real time, adjusting the delivery rate of the peristaltic pump to perform the rough liquid addition operation, and monitoring the proportion of black pixels in the image after the grayscale processing in real time; When the black pixel ratio reaches the preset rough liquid adding control threshold range, the rough liquid adding operation is stopped and switched to the medium speed liquid adding operation.

8. The high-precision automatic marking line volume recognition method according to claim 1 is characterized in that: Switching to the medium-speed liquid addition operation until the liquid level approaches the target mark and reaches the medium-speed liquid addition control threshold range, including: The dosing needle displacement motor controller is used to control the dosing needle to lift a fixed distance, and the delivery rate of the peristaltic pump is reduced to a preset speed to perform the medium-speed dosing operation; Grayscale processing is performed based on the image acquired in real time, and mask processing is performed on the target scribed line and the portion above it. Real-time monitoring is performed until the liquid level reaches the medium-speed liquid addition control threshold range, and the medium-speed liquid addition is stopped.

9. The high-precision automatic marking line volume recognition method according to claim 1, characterized in that: Executing the liquid adding needle to perform fine liquid adding operation until the liquid surface is aligned with the target scale line and reaches the fine liquid adding adjustment control threshold, completing the liquid surface positioning and fixed volume identification, including: The liquid level height edge is obtained by grayscale processing, edge detection and opening operation on the captured image and the preset pixel value is subtracted to determine the corresponding concave liquid surface position; at the same time, the spray mode is adjusted to a single-drop spray mode, and the concave liquid surface position is monitored in real time until the concave liquid surface position reaches the fine liquid adjustment control threshold.

10. The high-precision automatic marking line volume recognition method according to claim 1, characterized in that: The fine liquid adjustment control threshold is a preset range exceeding the target scale line.

Images