Control method for identifying and removing water stains on surface of yeast drum
Automatically control the blade through camera recognition and image recognition technology, the problem of water traces on the surface of yeast drum affecting the adsorption efficiency of yeast milk is solved, and automated monitoring and precise control of the yeast production process is realized, and product quality and production efficiency are improved.
Patent Information
- Application Number
- CN202411108511.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-12
AI Technical Summary
During the yeast production process, the presence of water marks on the surface of the yeast drum will affect the efficiency of starch adsorption of yeast milk, and manual monitoring is difficult to accurately control the feeding time, resulting in waste of resources and unstable product quality.
Use a camera to identify the water traces on the surface of the yeast rotary drum, and issue a tool inlet command through the main control system. Use image recognition technology to automatically determine when to adjust the blade to realize automatic tool inlet operation, and combine depth compensation and advance calculation of tool outflow to ensure water traces. Completely scraped off.
It improves production efficiency and product quality stability, reduces quality problems caused by manual monitoring failure, avoids water-breaking accidents, optimizes production processes, and reduces resource waste.
Smart Images

Figure CN118931693A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of water mark removal of a yeast drum, and in particular to a water mark recognition and removal control method based on the surface of a yeast drum. Background Art
[0002] In the yeast production process, the drum cavity is full of starch, and the yeast milk is underneath. The yeast milk is adsorbed by starch. If there are water marks on the surface of the yeast drum, it will affect the efficiency of starch adsorption of yeast milk, so it is necessary to scrape it down with a scraper for drying. The selection of the feed time is crucial to the yeast production process. A reasonable feed time can ensure that the material is fully processed, while avoiding excessive processing that leads to waste of resources or reduced product quality.
[0003] By using intelligent image recognition technology to monitor the water marks on the drum, automatically determine when the blade needs to be adjusted, and realize automatic feed operation, it can reduce the interference of human factors, improve the stability and predictability of the production process, and at the same time reduce labor costs and avoid quality problems caused by improper operation. By accurately controlling the feed time and blade adjustment, the yeast production process can be optimized, and the overall production efficiency and product quality can be improved. It can also avoid drum water collapse accidents caused by failure of manual monitoring, as well as quality problems such as bed collapse and waste materials, ensuring the stability and consistency of yeast product quality. Summary of the invention
[0004] In order to solve the technical problems in the above background, the present invention proposes a method for identifying and removing water marks on the surface of a yeast drum, comprising the following steps: S1. Using a camera to identify water marks on the surface of the yeast drum; S2. The main control system sends a knife feed command, causing the knife head to contact the surface of the yeast drum and scrape out water marks.
[0005] In a preferred embodiment, the camera is installed on the top of the yeast drum.
[0006] In a preferred solution, the camera is a binocular camera.
[0007] In a preferred solution, step S1 further includes: collecting images and identifying water marks on the surface of the yeast drum by an image recognition method.
[0008] In a preferred solution, step S1 also includes: the camera position is fixed with the knife position, the knife position is used as a reference point, the water mark image detected by the camera is mapped onto the drum surface, and the actual physical position of the water mark relative to the knife position is determined.
[0009] In a preferred solution, the lower edge of the water mark is detected, and the moment when the lower edge position coincides with the knife position is used as the timing for knife feeding.
[0010] In the preferred scheme, first, the circumference information of the yeast drum is obtained, and the linear speed of the yeast drum is calculated. Then, the water marks in the image are mapped to the drum surface, and the length of the water marks in the image is converted into the actual physical size to determine the actual length of the water marks. Finally, the time when the lower edge of the water marks reaches the knife position is estimated in combination with the linear speed of the yeast drum to determine the timing of knife removal.
[0011] In a preferred solution, the upper edge of the water mark is detected, and the time when the lower edge of the water mark reaches the knife position is calculated according to the length of the water mark to determine the timing of retracting the knife.
[0012] In the preferred solution, the common feed depth is a fixed value; however, when the drum rotates to the second circle, the water marks have not been completely scraped off, so depth compensation is performed; Depth compensation refers to the feed position determined by the difference between the lower edge position when the water mark appears for the nth time and the lower edge position when it appears for the n+1th time, where n is greater than the set number of times threshold; The starting condition for depth compensation is: the number of water marks reaches the set threshold.
[0013] The beneficial effects of the present invention are as follows: the method reduces the need for manual monitoring and improves production efficiency through automated image recognition and intelligent control; effectively avoids drum water collapse accidents that may be caused by manual monitoring failure, reduces quality problems such as bed collapse and waste materials, and ensures the stability and consistency of yeast product quality; automated monitoring and control can achieve more accurate water mark monitoring and knife feed control, improving the accuracy of the production process; optimizes the production process and reduces raw material waste; makes decisions based on real-time data, and improves the level of intelligent production. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 is a flow chart of a method for controlling removal of water marks on the surface of a yeast drum in one embodiment of the present invention; Figure 2 This is the interface diagram of water trace detection during yeast production. DETAILED DESCRIPTION
[0015] Example 1 like Figure 1~2 As shown, this embodiment provides a method for identifying and removing water marks on the surface of a yeast drum, including the following process: A1. During the rotation of the drum, a camera at a fixed position collects images of the drum surface in real time. The camera is installed on the top of the yeast drum (the top includes the top front side, the top middle side and the top back side).
[0016] The camera uses a binocular camera because: the binocular camera simulates human eyes, shoots the same object from different angles through two cameras, calculates the distance of the object using the principle of triangulation, and calculates the depth information of the water mark surface relative to the camera by comparing the difference (parallax) between the images captured by the two cameras. The binocular camera is used to monitor the three-dimensional morphological changes of the water mark in real time, and the cutting depth and advance amount of the cutting are dynamically adjusted according to the depth information. During the cutting process, the distance between the cutter head and the drum surface can also be detected in real time to avoid collision or damage caused by too deep cutting.
[0017] A2. Obtain the circumference information of the yeast drum and calculate its linear speed.
[0018] A3. Collect images of the yeast drum surface and pre-process the collected images, including denoising, contrast enhancement, grayscale, etc., to improve image quality and make the water mark features more obvious; use the YOLO algorithm to quickly identify water marks in the image. Use the edge detection algorithm (Canny edge detector) to identify the lower and upper edges of the water mark contour, then map the water marks in the image to the drum surface, convert the position and length of the water marks in the image into actual physical coordinates and physical dimensions, and determine the physical position and actual length of the water marks.
[0019] When the drum rotates for the first time, the camera detects the linear speed of the drum and the actual length of the water mark, and estimates the time it takes for the bottom point of the lower edge of the water mark to reach the knife position, thereby determining the timing of knife output.
[0020] The timing of blade movement refers to the moment when the blade starts to move towards the water marks on the surface of the drum.
[0021] The timing is determined based on the time required for the blade to start contacting the drum surface and the moment when the bottom edge of the drum reaches the knife position.
[0022] Calculate the time it takes for the lower edge of the water mark to reach the knife position to determine the time to retract the knife.
[0023] The timing of putting away the blade refers to the moment when scraping is finished and the blade is put away.
[0024] A5. Set the common feed depth as a fixed value. When the drum rotates to the second circle, if the water marks are not completely scraped off, adjust according to the depth compensation mechanism. The depth compensation mechanism determines the feed position based on the difference between the lower edge position when the water mark appears for the nth time and the lower edge position when it appears for the n+1th time, where n is greater than the set number of times threshold. When the number of water marks reaches the set number of times threshold, the depth compensation mechanism is activated.
[0025] A6. Based on the above steps and calculation results, the main control system issues a knife feed command to make the knife head contact the surface of the yeast drum and scrape out water marks.
[0026] A7. During the whole process, continuously monitor the water mark removal and drum status, and adjust the blade depth, blade exit timing and blade retraction timing in real time according to the actual scraping effect and system feedback.
[0027] Example 2 Depth compensation calculation methods include: 1. Initial depth setting: First set an initial fixed feed depth for the first scraping operation.
[0028] 2. Water mark monitoring: Use the camera to monitor the scraping effect of water marks. If the water marks are not completely scraped off, record the position of the bottom edge of the water marks.
[0029] 3. Depth compensation condition: Set a number threshold N. When the drum rotates to the Nth circle, if the water marks are still not completely scraped off, the depth compensation is started.
[0030] 4. Depth compensation calculation: Compare the positions of the bottom edges of the nth and n+1th water marks, and calculate the difference ΔP between the two.
[0031] The depth compensation amount Dc=ΔP*k, where k is a coefficient used to adjust the compensation amplitude.
[0032] 5. Update the feed depth: Add the calculated depth compensation Dc to the initial feed depth to obtain the updated feed depth.
[0033] Example 3 The calculation method of the advance amount of the tool includes: 1. Water mark length measurement: measure the length of the water mark along the rotating direction of the drum.
[0034] 2. Linear velocity calculation: Calculate the linear velocity V based on the rotation speed of the drum.
[0035] 3. Water mark arrival time estimation: According to the current position of the water mark and the linear velocity V, estimate the time T when the lower edge of the water mark reaches the tool position.
[0036] 4. Calculation of the advance amount of the cutter: A coefficient ε is set to ensure that the cutter head starts to contact before the water mark arrives.
[0037] Tool advance amount = ε* T.
[0038] Among them, the value of ε is a number greater than 1, ensuring that the cutter head has enough lead time to cope with any possible delays or errors. It is necessary to control the time required for the system to issue a command to the actual start of the cutter head action, and it is necessary to consider the sensor detection time, signal transmission time, control system calculation and decision time, etc., the time required for the cutter head to completely contact the drum surface from a stationary state, and the speed at which the water mark rotates along the drum, which is usually related to the linear speed of the drum.
[0039] 5. Adjust the timing of knife extension: According to the calculated advance amount of knife extension, adjust the timing of knife extension to ensure that the knife head just touches the bottom of the water mark.
[0040] Example 4 The water mark depth is estimated as follows: First, it is necessary to use a binocular camera to capture an image of the drum surface and extract the features of the water mark area, such as area, shape, color, etc. Then, stereo vision technology is used to analyze the extracted features and estimated depth to determine the thickness of the water mark. The analysis results are output in numerical form for use by the control system to adjust the feed depth.
[0041] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for identifying and removing water marks on the surface of a yeast drum, characterized in that: The following steps are involved: S1. Using a camera to identify water marks on the surface of the yeast drum; S2. The main control system sends a knife feed command, causing the knife head to contact the surface of the yeast drum and scrape out water marks.
2. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1, characterized in that: The camera is mounted on top of the yeast drum.
3. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1, characterized in that: The camera uses a binocular camera.
4. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1, characterized in that: Step S1 also includes: collecting images and identifying water marks on the surface of the yeast drum by an image recognition method.
5. The method for identifying and removing water marks on the surface of a yeast drum according to claim 4, characterized in that: Step S1 also includes: the camera position is fixed with the knife position, the knife position is used as a reference point, the water mark image detected by the camera is mapped onto the drum surface, and the actual physical position of the water mark relative to the knife position is determined.
6. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1, characterized in that: Detect the lower edge of the water mark, and the moment when the lower edge coincides with the tool position is used as the timing for tool feed.
7. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1 is characterized in that: first, the circumference information of the yeast drum is obtained, the linear speed of the yeast drum is calculated, and then the water marks in the image are mapped to the drum surface, the length of the water marks in the image is converted into an actual physical size, the actual length of the water marks is determined, and finally, the time when the lower edge of the water marks reaches the knife position is estimated in combination with the linear speed of the yeast drum to determine the timing of knife removal.
8. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1, characterized in that: Detect the upper edge of the water mark, calculate the time when the lower edge of the water mark reaches the knife position according to the length of the water mark, and determine the timing of retracting the knife.
9. The method for identifying and removing water marks on the surface of a yeast drum according to claim 1, characterized in that: The common feed depth is a fixed value; however, when the drum rotates to the second circle and the water marks have not been completely scraped off, depth compensation is performed; Depth compensation refers to the feed position determined by the difference between the lower edge position when the water mark appears for the nth time and the lower edge position when it appears for the n+1th time, where n is greater than the set number of times threshold; The starting condition for depth compensation is: the number of water marks reaches the set threshold.
