A method and system for screening and classifying peony seed oil raw materials based on image processing

Abstract

The present invention relates to a peony seed oil raw material screening and classification method and system based on image processing, the method comprising: based on a first acquisition device, performing image acquisition on materials in a detection area on a conveyor belt, using a preset peony seed recognition model to perform preliminary recognition on the image, and preliminarily classifying the materials in the image into unqualified products and qualified products; reading the result of the preliminary recognition, and adjusting the transmission speed of the conveyor belt according to the proportion of unqualified products in the result of the preliminary recognition; based on a second acquisition device, performing image acquisition on materials falling from the conveyor belt, using a preset peony seed recognition model to perform secondary recognition on the image of the falling material, and using a sorting device based on the result of the secondary recognition to respectively send unqualified products and qualified products into different collection containers. The present invention realizes that when there are many unqualified products, the speed is reduced, which is conducive to improving the accuracy of the secondary recognition, and also avoids taking away too many qualified products when rejecting unqualified products, and can achieve accurate rejection and accurate classification.

Description

Technical Field

[0001] The invention relates to the technical field of raw material classification, and in particular to a method and system for screening and classifying peony seed oil raw materials based on image processing. Background Art

[0002] Peony seed oil is often used in high-end edible oils, health products, and cosmetics due to its high nutritional value and unique health functions (such as enhancing intelligence, improving vision, and lowering blood lipids). Peony seeds will automatically fall off from the peony seed pods after being picked and dried. During the drying process, the peony seeds will change from yellow to brown and finally to black. When they are just picked, many nutrients are stored in the outer skin of the pods. This process will "transfer" the nutrients in the epidermis to the peony seeds. After threshing, the peony seeds must be color-sorted and screened. The main purpose of peony seed screening is to remove unqualified seeds and ensure the purity and high quality of the raw materials to meet the needs of the high-end edible oil and health product markets. At present, more sophisticated screening technologies, such as image recognition, are usually used to accurately identify and remove unqualified seeds such as mold, damage, and immaturity. The existing peony seed oil raw material screening and classification method and system usually screen at a uniform speed during screening, which has certain limitations, so that after the screening is completed, there will still be a large number of qualified products in the collection container of unqualified products. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention proposes a method and system for screening and classifying peony seed oil raw materials based on image processing.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for screening and classifying peony seed oil raw materials based on image processing, the method comprising the following steps:

[0006] Based on the first acquisition device, images of materials in the detection area on the conveyor belt are acquired, and the images are preliminarily recognized using a preset peony seed recognition model, and the materials in the images are preliminarily classified into unqualified products and qualified products;

[0007] Read the results of the preliminary identification, and adjust the transmission speed of the conveyor belt according to the proportion of unqualified products in the preliminary identification results. The proportion of unqualified products is negatively correlated with the transmission speed;

[0008] The second acquisition device is used to acquire images of the materials falling from the conveyor belt, and a preset peony seed recognition model is used to perform secondary recognition on the images of the falling materials. Based on the results of the secondary recognition, a sorting device is used to send the unqualified products and qualified products into different collection containers respectively.

[0009] As a further technical solution of the present invention, the step of reading the result of the preliminary identification and adjusting the transmission speed of the conveyor belt according to the proportion of unqualified products in the result of the preliminary identification includes:

[0010] Counting the number of materials identified as defective and qualified products in the inspection area;

[0011] Determine the proportion of unqualified products in the inspection area according to the number of unqualified products and qualified products in the inspection area. The proportion of unqualified products = the number of unqualified products / (the number of unqualified products + the number of qualified products);

[0012] When the proportion of defective products in the detection area does not exceed the first preset threshold, adjusting the transmission speed of the conveyor belt to the first preset speed and maintaining it;

[0013] When the proportion of defective products in the detection area exceeds a first preset threshold, adjusting the transmission speed of the conveyor belt to a second preset speed and maintaining it;

[0014] Among them, 0≤first preset threshold<1, first preset speed>second preset speed.

[0015] As a further technical solution of the present invention, the method further includes: when the speed reduction is triggered, adjusting the speed reduction according to the position of the defective product.

[0016] As a further technical solution of the present invention, the step of adjusting the speed reduction according to the position of the defective product when the speed reduction is triggered includes:

[0017] When deceleration is triggered, the initial speed when deceleration is triggered and the target speed after deceleration are obtained, wherein the target speed after deceleration is the second preset speed;

[0018] The detection area is divided into several unit grids of uniform size. During the conveyor belt transmission process, each unit grid is close to the end of the conveyor belt in turn. The size of each grid unit can be adjusted according to the actual situation to ensure the accuracy and operability of the measurement;

[0019] Count the number of materials identified as defective products in each unit grid, and determine whether each unit grid is a hot spot grid;

[0020] Acquire the position information of the hot spot grid, and determine the distance between the hot spot grid and the end of the conveyor belt according to the position information of the hot spot grid;

[0021] The target acceleration value is calculated according to the distance value between the hot spot grid and the end of the conveyor belt, the initial speed when the deceleration is triggered, and the target speed after the deceleration. The calculation formula is as follows:

[0022] v 2 -v0 2 =2ax;

[0023] Where v is the target speed after deceleration, v0 is the initial speed when deceleration is triggered, a is the target acceleration, and x is the distance between the hotspot grid and the end of the conveyor belt;

[0024] The uniform deceleration is performed according to the value of the target acceleration, so that when the hot spot grid reaches the end of the conveyor belt, the transportation speed of the conveyor belt is the target speed after deceleration.

[0025] As a further technical solution of the present invention, the method is used for a control module in a screening device, the screening device includes a main body, a feed port, a conveyor belt, a first collecting device, a second collecting device, a sorting device and a control module, the control module is electrically connected to the driving motor in the conveyor belt, the feed port is fixedly connected to the main body and is located above the conveyor belt, the conveyor belt is arranged in the main body, a first collecting device is arranged above the end of the conveyor belt, the first collecting device is fixedly connected to the main body, the main body is provided with a material trough on the bottom side of the end of the conveyor belt, second collecting devices are arranged on both sides of the material trough, a sorting device is arranged on one side of the material trough, two material distribution ports are arranged at the bottom of the material trough, and a collection container is placed on the bottom side of the material distribution port.

[0026] Another object of the present invention is to provide a peony seed oil raw material screening and classification system based on image processing, the system comprising:

[0027] A material preliminary identification module is used to collect images of materials in the detection area on the conveyor belt based on the first acquisition device, use a preset peony seed recognition model to perform preliminary identification on the images, and preliminarily classify the materials in the images into unqualified products and qualified products;

[0028] The transmission speed adjustment module is used to read the preliminary identification result and adjust the transmission speed of the conveyor belt according to the proportion of unqualified products in the preliminary identification result. The proportion of unqualified products is negatively correlated with the transmission speed.

[0029] The identification and sorting execution module is used to collect images of materials falling from the conveyor belt based on the second acquisition device, use a preset peony seed recognition model to perform secondary recognition on the images of the falling materials, and use a sorting device based on the results of the secondary recognition to send unqualified products and qualified products into different collection containers.

[0030] As a further technical solution of the present invention, the transmission speed adjustment module includes:

[0031] A quantity counting unit, used to count the number of materials identified as unqualified products and qualified products in the detection area;

[0032] A proportion calculation unit, used to determine the proportion of unqualified products in the inspection area according to the number of unqualified products and qualified products in the inspection area;

[0033] A first speed adjustment unit, used for adjusting the transmission speed of the conveyor belt to a first preset speed and maintaining it when the proportion of defective products in the detection area does not exceed a first preset threshold value;

[0034] A second speed adjustment unit, used for adjusting the transmission speed of the conveyor belt to a second preset speed and maintaining it when the proportion of defective products in the detection area exceeds a first preset threshold value;

[0035] Among them, 0≤first preset threshold<1, first preset speed>second preset speed.

[0036] As a further technical solution of the present invention, the system further includes a speed reduction execution module, and the speed reduction execution module includes:

[0037] A speed acquisition unit, used for acquiring an initial speed when the speed reduction is triggered and a target speed after the speed reduction when the speed reduction is triggered, wherein the target speed after the speed reduction is a second preset speed;

[0038] The area division unit is used to divide the detection area into a number of unit grids of uniform size. During the conveyor belt transmission process, each unit grid approaches the end of the conveyor belt in turn;

[0039] The hotspot grid determination unit is used to count the number of materials identified as defective products in each unit grid and determine whether each unit grid is a hotspot grid;

[0040] A position acquisition unit, used to acquire the position information of the hot spot grid, and determine the distance between the hot spot grid and the end of the conveyor belt according to the position information of the hot spot grid;

[0041] The acceleration calculation unit is used to calculate the target acceleration value according to the distance value between the hot spot grid and the end of the conveyor belt, the initial speed when the deceleration is triggered, and the target speed after the deceleration. The calculation formula is as follows:

[0042] v 2 -v0 2 =2ax;

[0043] Where v is the target speed after deceleration, v0 is the initial speed when deceleration is triggered, a is the target acceleration, and x is the distance between the hotspot grid and the end of the conveyor belt;

[0044] The deceleration execution unit is used to perform uniform deceleration according to the value of the target acceleration, so that when the hot spot grid reaches the end of the conveyor belt, the transportation speed of the conveyor belt is the target speed after deceleration.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a method and system for screening and classifying peony seed oil raw materials based on image processing, in which the materials in the detection area are preliminarily identified before the materials fall, and the transmission speed of the conveyor belt is adjusted based on the preliminary identification results, so that when there are many unqualified products, the speed is reduced, which is beneficial to improving the accuracy of secondary identification, and at the same time, it also avoids the situation where the transmission speed is too high when there are many unqualified products, resulting in too many qualified products being taken away when the unqualified products are removed, so that accurate removal and accurate classification can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flowchart of a peony seed oil raw material screening and classification method based on image processing.

[0047] Figure 2 This is a flowchart of the speed adjustment step in a method for screening and classifying peony seed oil raw materials based on image processing.

[0048] Figure 3 This is a structural block diagram of a peony seed oil raw material screening and classification system based on image processing.

[0049] Figure 4 It is a screening device in this embodiment.

[0050] Notes on the figure numbers: 1- conveyor belt, 2- first collecting device, 3- second collecting device, 4- sorting device, 5- feed port, 6- main body, 7- material trough, 8- material distribution port, 9- collection container. DETAILED DESCRIPTION

[0051] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] like Figure 1 As shown, an embodiment of the present invention provides a method for screening and classifying peony seed oil raw materials based on image processing, and the method comprises the following steps:

[0053] Step S100, based on the first acquisition device 2, the image of the material in the detection area on the conveyor belt 1 is acquired, and the image is preliminarily recognized using a preset peony seed recognition model, and the material in the image is preliminarily divided into unqualified products and qualified products; in the present invention, the first acquisition device 2 is arranged above the conveyor belt 1, and is aimed at the detection area of ​​the conveyor belt 1, and is used to acquire the material image in the detection area of ​​the conveyor belt 1. In the present invention, the acquisition device can be an industrial camera, a camera or a scanner, etc. The detection area is usually determined according to the actual situation of the production line and the characteristics of the material flow. The selection of the detection area should ensure that it can cover the key position at the end of the conveyor belt 1 or before the material processing area;

[0054] Among them, the preset peony seed recognition model is usually a deep learning model, which can automatically identify peony seeds from images and further distinguish qualified products from unqualified products; the deep learning model is usually a convolutional neural network (CNN), which performs well in the field of image recognition. The preset peony seed recognition model is obtained through the following steps: collect a large amount of peony seed image data, including qualified and unqualified images under different angles, lighting conditions, and background interference. Label the collected images to clarify which are qualified products and which are unqualified products. Use the labeled data set to train the model and adjust the model parameters to optimize the recognition performance. Use image data that did not participate in the training to verify the accuracy and generalization ability of the model. Until the preset requirements are met, the training is completed and the preset peony seed recognition model is obtained;

[0055] Step S200, read the result of the preliminary identification, and adjust the transmission speed of the conveyor belt 1 according to the proportion of unqualified products in the result of the preliminary identification. The proportion of unqualified products is negatively correlated with the transmission speed. When there are many unqualified products, the speed is reduced to improve the accuracy of the secondary identification. At the same time, it is also avoided that the transmission speed of many unqualified products is too high, resulting in too many qualified products being taken away when the unqualified products are removed, so that accurate removal and accurate classification can be achieved.

[0056] Step S300, based on the second acquisition device 3, the image of the material falling from the conveyor belt 1 is acquired, and the preset peony seed recognition model is used to perform secondary recognition on the image of the falling material, and based on the result of the secondary recognition, the sorting device 4 is used to send the unqualified products and the qualified products into different collection containers 9 respectively. The sorting device 4 includes an airflow sorter and a sorting machine, wherein the second acquisition device 3 is located on both sides of the sorting area at the bottom of the end of the conveyor belt 1, and is used to acquire images of the falling material, and the airflow sorter is located on one side of the sorting area, and is used to blow the unqualified products into the corresponding collection container 9.

[0057] In practical applications, the method described in this embodiment is used to screen the control modules in the device, such as Figure 4As shown, the screening device includes a main body 6, a feed port 5, a conveyor belt 1, a first collecting device 2, a second collecting device 3, a sorting device 4 and a control module, wherein the control module is electrically connected to the driving motor in the conveyor belt 1, the feed port 5 is fixedly connected to the main body 6, and is located above the conveyor belt 1, the conveyor belt 1 is arranged in the main body 6, a first collecting device 2 is arranged above the end of the conveyor belt 1, the first collecting device 2 is fixedly connected to the main body 6, the main body 6 is provided with a trough 7 at the bottom side of the end of the conveyor belt 1, the second collecting devices 3 are arranged on both sides of the trough 7, a sorting device 4 is arranged on one side of the trough 7, two material distribution ports 8 are arranged at the bottom of the trough 7, and a collecting container 9 is placed on the bottom side of the material distribution port 8.

[0058] When in use, the material is poured from the feed port 5 of the screening device, and the material falls on the conveyor belt 1 after entering the screening device. The conveyor belt 1 drives the material to move, and the image of the material in the detection area on the conveyor belt 1 is collected based on the first acquisition device 2, and the image is preliminarily recognized using the preset peony seed recognition model, and the material in the image is preliminarily divided into unqualified products and qualified products, and the result of the preliminary recognition is read. The transmission speed of the conveyor belt 1 is adjusted according to the proportion of unqualified products in the result of the preliminary recognition, and the speed is reduced when there are many unqualified products. The image of the material falling from the conveyor belt 1 is collected based on the second acquisition device 3, and the image of the falling material is secondary recognized using the preset peony seed recognition model. Based on the result of the secondary recognition, the unqualified products and qualified products are respectively sent into different collection containers 9 using the sorting device 4, and the sorting device 4 includes an airflow sorter and a sorting machine. In this embodiment, the sorting device 4 is usually an airflow sorter.

[0059] See also Figure 2 As a preferred embodiment of the present invention, the step of reading the preliminary identification result and adjusting the transmission speed of the conveyor belt 1 according to the proportion of unqualified products in the preliminary identification result includes:

[0060] Step S201, counting the number of materials identified as unqualified products and qualified products in the detection area;

[0061] Step S202, determining the proportion of unqualified products in the inspection area according to the number of unqualified products and qualified products in the inspection area, the proportion of unqualified products = the number of unqualified products / (the number of unqualified products + the number of qualified products);

[0062] Step S203, when the proportion of defective products in the detection area does not exceed the first preset threshold, adjusting the transmission speed of the conveyor belt 1 to the first preset speed and maintaining it;

[0063] Step S204, when the proportion of defective products in the detection area exceeds the first preset threshold, adjusting the transmission speed of the conveyor belt 1 to a second preset speed and maintaining it;

[0064] Among them, 0≤first preset threshold<1, first preset speed>second preset speed.

[0065] As a preferred embodiment of the present invention, the method further comprises: when the speed reduction is triggered, adjusting the speed reduction according to the position of the defective product.

[0066] The step of adjusting the speed reduction according to the position of the defective product when the speed reduction is triggered includes:

[0067] When deceleration is triggered (i.e., when adjusting from the first preset speed to the second preset speed), the initial speed when deceleration is triggered and the target speed after deceleration are obtained, and the target speed after deceleration is the second preset speed;

[0068] The detection area is divided into a number of unit grids of uniform size. During the transmission process of the conveyor belt 1, each unit grid is close to the end of the conveyor belt 1 in turn. The size of each grid unit can be adjusted according to the actual situation to ensure the accuracy and operability of the measurement;

[0069] Count the number of materials identified as defective products in each unit grid, and determine whether each unit grid is a hotspot grid; specifically: when the number of defective materials in a unit grid exceeds a preset aggregation number, which can be 1 or 2, or other positive integers, the unit grid is determined to be a hotspot grid; when the number of defective materials in a unit grid does not exceed the preset aggregation number, the unit grid is determined to be a normal grid;

[0070] Obtaining the position information of the hot spot grid, and determining the distance between the hot spot grid and the end of the conveyor belt 1 according to the position information of the hot spot grid;

[0071] The target acceleration value is calculated according to the distance value between the hot spot grid and the end of the conveyor belt 1, the initial speed when the deceleration is triggered, and the target speed after the deceleration. The calculation formula is as follows:

[0072]

[0073] Wherein, v is the target speed after deceleration, v0 is the initial speed when deceleration is triggered, a is the target acceleration, and x is the distance between the hotspot grid and the end of conveyor belt 1;

[0074] The target acceleration value is used to uniformly decelerate the conveyor belt 1, so that when the hot spot grid reaches the end of the conveyor belt 1, the conveyor belt 1 has a transport speed that is the target speed after deceleration. When the acceleration is triggered, the acceleration value is usually a preset value and is constant, and can be set according to actual conditions.

[0075] In this embodiment, the detection area is divided and it is determined whether each unit grid is a hot spot grid. The target acceleration value is determined according to the position information of the hot spot grid, and the speed is reduced based on this, thereby achieving a fully effective speed reduction, which is beneficial to improving the accuracy of screening and classification.

[0076] See also Figure 3 Another object of the present invention is to provide a peony seed oil raw material screening and classification system based on image processing, the system comprising:

[0077] The material preliminary identification module 100 is used to collect images of materials in the detection area on the conveyor belt 1 based on the first collection device 2, and use the preset peony seed recognition model to perform preliminary identification on the image, and preliminarily classify the materials in the image into unqualified products and qualified products; wherein, in the present invention, the first collection device 2 is arranged above the conveyor belt 1, and is aimed at the detection area of ​​the conveyor belt 1, and is used to collect material images in the detection area of ​​the conveyor belt 1. In the present invention, the collection device can be an industrial camera, a camera or a scanner, etc. The detection area is usually determined according to the actual situation of the production line and the characteristics of the material flow. The selection of the detection area should ensure that it can cover the key position at the end of the conveyor belt 1 or before the material processing area;

[0078] The transmission speed adjustment module 200 is used to read the result of the preliminary identification and adjust the transmission speed of the conveyor belt 1 according to the proportion of unqualified products in the result of the preliminary identification. The proportion of unqualified products is negatively correlated with the transmission speed.

[0079] The identification and sorting execution module 300 is used to capture images of materials falling from the conveyor belt 1 based on the second acquisition device 3, and use a preset peony seed recognition model to perform secondary recognition on the images of the falling materials. Based on the results of the secondary recognition, the sorting device 4 is used to send unqualified products and qualified products into different collection containers 9 respectively, wherein the second acquisition device 3 is located on both sides of the sorting area at the bottom of the end of the conveyor belt 1, and is used to capture images of the falling materials. The airflow sorter is located on one side of the sorting area, and is used to blow unqualified products into the corresponding collection container 9.

[0080] As a preferred embodiment of the present invention, the transmission speed adjustment module 200 includes:

[0081] A quantity counting unit, used to count the number of materials identified as unqualified products and qualified products in the detection area;

[0082] A proportion calculation unit is used to determine the proportion of unqualified products in the inspection area according to the number of unqualified products and qualified products in the inspection area, and the proportion of unqualified products = the number of unqualified products / (the number of unqualified products + the number of qualified products);

[0083] A first speed adjustment unit, used for adjusting the transmission speed of the conveyor belt 1 to a first preset speed and maintaining it when the proportion of defective products in the detection area does not exceed a first preset threshold value;

[0084] A second speed adjustment unit, used to adjust the transmission speed of the conveyor belt 1 to a second preset speed and maintain it when the proportion of defective products in the detection area exceeds a first preset threshold value;

[0085] Among them, 0≤first preset threshold<1, first preset speed>second preset speed; when there are many defective products, the speed is reduced, which is beneficial to the accuracy of secondary identification, and also avoids the transmission speed of defective products being too high, resulting in too many qualified products being taken away when the defective products are eliminated, so that accurate elimination and accurate classification can be achieved.

[0086] As a preferred embodiment of the present invention, the system further includes a speed reduction execution module, and the speed reduction execution module includes:

[0087] A speed acquisition unit, used for acquiring an initial speed when the speed reduction is triggered and a target speed after the speed reduction when the speed reduction is triggered (when the speed reduction is adjusted from the first preset speed to the second preset speed), wherein the target speed after the speed reduction is the second preset speed;

[0088] The area division unit is used to divide the detection area into a number of unit grids of uniform size. During the transmission process of the conveyor belt 1, each unit grid is close to the end of the conveyor belt 1 in turn. The size of each grid unit can be adjusted according to the actual situation to ensure the accuracy and operability of the measurement;

[0089] The hotspot grid determination unit is used to count the number of materials identified as defective products in each unit grid and determine whether each unit grid is a hotspot grid;

[0090] A position acquisition unit, used to acquire position information of a hot spot grid, and determine the distance between the hot spot grid and the end of the conveyor belt 1 according to the position information of the hot spot grid;

[0091] The acceleration calculation unit is used to calculate the target acceleration value according to the distance value between the hot spot grid and the end of the conveyor belt 1, the initial speed when the deceleration is triggered, and the target speed after the deceleration. The calculation formula is as follows:

[0092]

[0093] Wherein, v is the target speed after deceleration, v0 is the initial speed when deceleration is triggered, a is the target acceleration, and x is the distance between the hotspot grid and the end of conveyor belt 1;

[0094] The deceleration execution unit is used to uniformly decelerate according to the value of the target acceleration, so that when the hot spot grid reaches the end of the conveyor belt 1, the transportation speed of the conveyor belt 1 is the target speed after deceleration. When the acceleration is triggered, the acceleration value is a preset value and is constant, which can be set according to actual conditions.

[0095] It should be noted that, in this article, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0096] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for screening and classifying peony seed oil raw materials based on image processing, characterized in that: The method comprises the following steps: Based on the first acquisition device, images of materials in the detection area on the conveyor belt are acquired, and the images are preliminarily recognized using a preset peony seed recognition model, and the materials in the images are preliminarily classified into unqualified products and qualified products; Read the results of the preliminary identification, and adjust the transmission speed of the conveyor belt according to the proportion of unqualified products in the preliminary identification results. The proportion of unqualified products is negatively correlated with the transmission speed; The second acquisition device is used to acquire images of the materials falling from the conveyor belt, and a preset peony seed recognition model is used to perform secondary recognition on the images of the falling materials. Based on the results of the secondary recognition, the sorting device is used to send the unqualified products and the qualified products into different collection containers respectively; The method further includes: when the speed reduction is triggered, adjusting the speed reduction according to the position of the defective product; The step of adjusting the speed reduction according to the position of the defective product when the speed reduction is triggered includes: When deceleration is triggered, the initial speed when deceleration is triggered and the target speed after deceleration are obtained, wherein the target speed after deceleration is the second preset speed; The detection area is divided into several unit grids of uniform size. During the conveyor belt transmission process, each unit grid approaches the end of the conveyor belt in turn; Count the number of materials identified as defective products in each unit grid, and determine whether each unit grid is a hot spot grid; Acquire the position information of the hot spot grid, and determine the distance between the hot spot grid and the end of the conveyor belt according to the position information of the hot spot grid; The target acceleration value is calculated according to the distance value between the hot spot grid and the end of the conveyor belt, the initial speed when the deceleration is triggered, and the target speed after the deceleration. The calculation formula is as follows: ； in, is the target speed after deceleration, is the initial speed when deceleration is triggered, is the target acceleration, is the distance between the hotspot grid and the end of the conveyor belt; The uniform deceleration is performed according to the value of the target acceleration, so that when the hot spot grid reaches the end of the conveyor belt, the transportation speed of the conveyor belt is the target speed after deceleration.

2. The method for screening and classifying peony seed oil raw materials based on image processing according to claim 1, characterized in that: The step of reading the preliminary identification result and adjusting the transmission speed of the conveyor belt according to the proportion of unqualified products in the preliminary identification result includes: Counting the number of materials identified as defective and qualified products in the inspection area; Determine the proportion of unqualified products in the inspection area based on the number of unqualified and qualified materials in the inspection area; When the proportion of defective products in the detection area does not exceed the first preset threshold, adjusting the transmission speed of the conveyor belt to the first preset speed and maintaining it; When the proportion of defective products in the detection area exceeds a first preset threshold, adjusting the transmission speed of the conveyor belt to a second preset speed and maintaining it; Among them, 0≤first preset threshold<1, first preset speed>second preset speed.

3. The method for screening and classifying peony seed oil raw materials based on image processing according to claim 1, characterized in that: The method is used for a control module in a screening device, which includes a main body, a feed port, a conveyor belt, a first collecting device, a second collecting device, a sorting device and a control module, wherein the control module is electrically connected to the conveyor belt, the feed port is fixedly connected to the main body and is located above the conveyor belt, the conveyor belt is arranged in the main body, a first collecting device is arranged above the end of the conveyor belt, the first collecting device is fixedly connected to the main body, a material trough is arranged on the bottom side of the end of the conveyor belt, second collecting devices are arranged on both sides of the material trough, a sorting device is arranged on one side of the material trough, two material distribution ports are arranged at the bottom of the material trough, and a collection container is placed on the bottom side of the material distribution port.

4. A peony seed oil raw material screening and classification system based on image processing, characterized in that: The system comprises: A material preliminary identification module is used to collect images of materials in the detection area on the conveyor belt based on the first acquisition device, use a preset peony seed recognition model to perform preliminary identification on the images, and preliminarily classify the materials in the images into unqualified products and qualified products; The transmission speed adjustment module is used to read the preliminary identification result and adjust the transmission speed of the conveyor belt according to the proportion of unqualified products in the preliminary identification result. The proportion of unqualified products is negatively correlated with the transmission speed. An identification and sorting execution module is used to collect images of materials falling from the conveyor belt based on a second acquisition device, perform secondary recognition on the images of the falling materials using a preset peony seed recognition model, and use a sorting device to send unqualified products and qualified products into different collection containers based on the results of the secondary recognition; The system further includes a speed reduction execution module, and the speed reduction execution module includes: A speed acquisition unit, used for acquiring an initial speed when the speed reduction is triggered and a target speed after the speed reduction when the speed reduction is triggered, wherein the target speed after the speed reduction is a second preset speed; The area division unit is used to divide the detection area into a number of unit grids of uniform size. During the conveyor belt transmission process, each unit grid approaches the end of the conveyor belt in turn; The hotspot grid determination unit is used to count the number of materials identified as defective products in each unit grid and determine whether each unit grid is a hotspot grid; A position acquisition unit, used to acquire the position information of the hot spot grid, and determine the distance between the hot spot grid and the end of the conveyor belt according to the position information of the hot spot grid; The acceleration calculation unit is used to calculate the target acceleration value according to the distance value between the hot spot grid and the end of the conveyor belt, the initial speed when the deceleration is triggered, and the target speed after the deceleration. The calculation formula is as follows: ； in, is the target speed after deceleration, is the initial speed when deceleration is triggered, is the target acceleration, is the distance between the hotspot grid and the end of the conveyor belt; The deceleration execution unit is used to perform uniform deceleration according to the value of the target acceleration, so that when the hot spot grid reaches the end of the conveyor belt, the transportation speed of the conveyor belt is the target speed after deceleration.

5. The peony seed oil raw material screening and classification system based on image processing according to claim 4, characterized in that: The transmission speed adjustment module comprises: A quantity counting unit, used to count the number of materials identified as unqualified products and qualified products in the detection area; A proportion calculation unit, used to determine the proportion of unqualified products in the inspection area according to the number of unqualified products and qualified products in the inspection area; A first speed adjustment unit, used for adjusting the transmission speed of the conveyor belt to a first preset speed and maintaining it when the proportion of defective products in the detection area does not exceed a first preset threshold value; A second speed adjustment unit, used for adjusting the transmission speed of the conveyor belt to a second preset speed and maintaining it when the proportion of defective products in the detection area exceeds a first preset threshold value; Among them, 0≤first preset threshold<1, first preset speed>second preset speed.

Images