CN113034503B - High-flux automatic cup separating method, device and system - Google Patents
High-flux automatic cup separating method, device and system Download PDFInfo
- Publication number
- CN113034503B CN113034503B CN202110591721.4A CN202110591721A CN113034503B CN 113034503 B CN113034503 B CN 113034503B CN 202110591721 A CN202110591721 A CN 202110591721A CN 113034503 B CN113034503 B CN 113034503B
- Authority
- CN
- China
- Prior art keywords
- sample tube
- state
- preset operation
- mechanical equipment
- target picture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000013145 classification model Methods 0.000 claims abstract description 72
- 230000008569 process Effects 0.000 claims abstract description 39
- 238000012544 monitoring process Methods 0.000 claims abstract description 31
- 238000000926 separation method Methods 0.000 claims abstract description 29
- 238000000605 extraction Methods 0.000 claims abstract description 22
- 239000007788 liquid Substances 0.000 claims description 64
- 238000012549 training Methods 0.000 claims description 18
- 238000012806 monitoring device Methods 0.000 claims description 15
- 238000012546 transfer Methods 0.000 claims description 15
- 238000010801 machine learning Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 6
- 230000007246 mechanism Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 abstract description 8
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000007635 classification algorithm Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 241000700605 Viruses Species 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 150000007523 nucleic acids Chemical class 0.000 description 1
- 102000039446 nucleic acids Human genes 0.000 description 1
- 108020004707 nucleic acids Proteins 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N35/00—Automatic analysis not limited to methods or materials provided for in any single one of groups G01N1/00 - G01N33/00; Handling materials therefor
- G01N35/0099—Automatic analysis not limited to methods or materials provided for in any single one of groups G01N1/00 - G01N33/00; Handling materials therefor comprising robots or similar manipulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Quality & Reliability (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Medical Informatics (AREA)
- Robotics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Automatic Analysis And Handling Materials Therefor (AREA)
Abstract
The embodiment of the invention discloses a high-flux automatic cup separating method, a device and a system, wherein the method comprises the following steps: in the process that the mechanical equipment automatically executes the cup separating process, a target picture after the mechanical equipment executes preset operation is obtained, state information corresponding to the target picture is obtained through a preset multi-class classification model, whether the first preset operation is executed successfully or not is determined according to the state information corresponding to the target picture, and the next operation after the first preset operation is executed under the condition that the first preset operation is executed successfully until the whole process of automatically separating cups is executed. Therefore, automatic cup separation operation on the sample is achieved, the extraction efficiency of the sample is improved, human participation is not needed, the biological hidden danger is reduced, and the probability of successful execution of the automatic cup separation process is improved by adopting a visual monitoring mode.
Description
Technical Field
The invention relates to the field of automatic processing, in particular to a high-throughput automatic cup separating method, device and system.
Background
Currently, for virus nucleic acid detection, a collected sample is usually placed in a sample tube, liquid needs to be taken out of the sample tube, and the taken liquid needs to be pipetted to a pore plate. However, the operation of taking the liquid from the sample tube by manual operation is not only inefficient but also easily causes a certain biological risk.
Disclosure of Invention
In view of this, the embodiment of the invention discloses a high-throughput automatic cup separation method, device and system, so that automatic cup separation operation on a sample is realized, the extraction efficiency of the sample is improved, human participation is not needed, the existing biological hidden danger is reduced, and the probability of successful execution of an automatic cup separation process is improved by adopting a visual monitoring mode.
A high-throughput automatic cup separation method is characterized by comprising the following steps:
in the process that mechanical equipment automatically executes a cup dividing process, acquiring a target picture after the mechanical equipment executes preset operation;
after a target picture after a first preset operation is obtained, inputting the target picture into a multi-class classification model trained in advance to obtain state information corresponding to the target picture; the state information corresponding to the target picture at least comprises: a sample tube transfer state, a sample tube switch cover state and a sample extraction state; the multi-class classification model is obtained by training a preset machine learning model through the corresponding relation between the historical pictures and the states after the preset operation is executed by the mechanical equipment;
determining whether a first preset operation is successfully executed or not according to the state information corresponding to the target picture;
and controlling the mechanical equipment to execute the next operation after the first preset operation under the condition that the first preset operation is executed successfully.
Alternatively to this, the first and second parts may,
the sample tube transfer state includes: the mechanical equipment grabs the sample tube, and the mechanical equipment does not grab the sample tube;
the sample tube switch cover state includes: the sample tube cover is in a screwed state, the sample tube cover is in a unscrewed state, and the sample tube cover is in a loose state;
the sample tube extraction state comprises: the mechanical equipment and the pipette head are in a connected state, the mechanical equipment and the pipette head are in an unconnected state, the liquid taking amount is in a standard state, and the liquid taking amount is in a non-standard state.
Optionally, after the target picture after the first preset operation is obtained, inputting the target picture into a multi-class classification model trained in advance, to obtain state information corresponding to the target picture, where the method includes:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state that whether the mechanical equipment grabs the sample tube or not;
when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment;
when the first preset operation is that the mechanical equipment executes the operation of taking liquid from the sample tube, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state whether the liquid taking amount reaches the standard or not;
when the first preset operation is that the mechanical equipment executes the operation of screwing the sample tube cover to the sample tube, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
and when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette tip, sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment.
Optionally, determining whether the first preset operation is successfully executed according to the state information corresponding to the picture includes:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, and the obtained state information is the state that the mechanical equipment grabs the sample tube, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover, and the obtained state information indicates that the sample tube cover is in the unscrewing state, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, the obtained state information is that the pipette tip is connected with the mechanical equipment, and the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of taking liquid from the sample tube, and the obtained state information comprises the state that the liquid taking amount reaches the standard, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment performs the operation of screwing the sample tube cover to the sample tube, and the obtained state information indicates that the sample tube cover is in a screwed state, the first preset operation is successfully performed;
and when the first preset operation is the operation of the mechanical equipment for unloading the pipette tip, and the obtained state information is that the pipette tip is not connected with the mechanical equipment, the first preset operation is successfully executed.
Optionally, if the first preset operation fails to be executed, the first preset operation is executed again until the first preset operation is executed successfully.
Optionally, if the number of times of the execution failure of the first preset operation exceeds a preset threshold, an alarm is given.
The embodiment of the invention discloses a high-flux automatic cup separating device, which comprises:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a target picture after the mechanical equipment executes preset operation in the process of automatically executing a cup dividing process by the mechanical equipment;
the state information identification unit is used for inputting the target picture into a pre-trained multi-class classification model after the target picture after the first preset operation is acquired, and obtaining state information corresponding to the target picture; the state information corresponding to the target picture at least comprises: a sample tube transfer state, a sample tube switch cover state and a sample extraction state; the multi-class classification model is obtained by training a preset machine learning model through the corresponding relation between the historical pictures and the states after the preset operation is executed by the mechanical equipment;
the determining unit is used for determining whether the first preset operation is successfully executed or not according to the state information corresponding to the target picture;
and the control unit is used for controlling the mechanical equipment to execute the next operation after the first preset operation under the condition that the first preset operation is successfully executed.
Alternatively to this, the first and second parts may,
the sample tube transfer state includes: the mechanical equipment grabs the sample tube, and the mechanical equipment does not grab the sample tube;
the sample tube switch cover state includes: the sample tube cover is in a screwed state, the sample tube cover is in a unscrewed state, and the sample tube cover is in a loose state;
the sample tube extraction state comprises: the mechanical equipment and the pipette head are in a connected state, the mechanical equipment and the pipette head are in an unconnected state, the liquid taking amount is in a standard state, and the liquid taking amount is in a non-standard state.
Optionally, the state information identifying unit includes:
the first state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the mechanical equipment grabs the sample tube or not when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover;
the second state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover when the first preset operation is the operation of unscrewing the sample tube cover by the mechanical equipment;
the third state identification unit is used for sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment or not when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip;
the fourth state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state whether the liquid taking amount reaches the standard or not when the first preset operation is that the mechanical equipment executes the liquid taking operation from the sample tube;
the fifth state identification unit is used for executing the operation of screwing the sample tube cover to the sample tube when the first preset operation is the mechanical equipment, and sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
and the sixth state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment or not when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette tip.
The embodiment of the invention discloses a high-flux automatic cup separating system, which comprises:
a controller, a mechanical device and a monitoring device;
the controller is used for executing the high-flux automatic cup separating method;
the mechanical equipment is used for executing the operation in the cup separating process according to the instruction of the controller;
the monitoring equipment is used for shooting pictures after the mechanical equipment executes preset operation.
The embodiment of the invention discloses a high-flux automatic cup separating method, a device and a system, wherein the method comprises the following steps: in the process that mechanical equipment automatically executes a cup separating process, acquiring a target picture after the mechanical equipment executes preset operation, and acquiring state information corresponding to the target picture through a preset multi-class classification model, wherein the multi-class classification model is acquired after a preset machine learning model is trained through the corresponding relation between a history picture after the mechanical equipment executes the preset operation and the state; and determining whether the first preset operation is successfully executed according to the state information corresponding to the target picture, and executing the next operation after the first preset operation under the condition that the first preset operation is successfully executed until the whole process of automatic cup separation is completed. Therefore, automatic cup separation operation on the sample is achieved, the extraction efficiency of the sample is improved, human participation is not needed, the biological hidden danger is reduced, and the probability of successful execution of the automatic cup separation process is improved by adopting a visual monitoring mode.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic flow chart of a high throughput automatic cup-dispensing method according to an embodiment of the present invention;
FIG. 2 is another schematic flow chart of high throughput automatic cup dispensing provided by the embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a high-throughput automatic cup dispensing device provided by an embodiment of the invention;
fig. 4 shows a schematic structural diagram of a high-throughput automatic cup dispensing system provided by an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a schematic flow chart of a high throughput automatic cup dispensing method provided in an embodiment of the present invention is shown, in the embodiment, the method includes:
s101: in the process that mechanical equipment automatically executes a cup dividing process, acquiring a target picture after the mechanical equipment executes preset operation;
in this embodiment, the process of automatically executing cup separation by the mechanical device may include: grabbing the sample tube, taking down the sample tube cover from the sample tube, taking the pipette head, taking the liquid from the sample tube by the pipette head, detaching the pipette head, and screwing the sample tube cover onto the sample tube.
Wherein, in order to take off the sample tube lid from the sample tube, after having snatched the sample tube again, can also move mechanical equipment to the position of sample tube lid.
In addition, if a single pipetting operation is performed, the method may further comprise moving the sample tube back to the sample tube tray.
In this embodiment, in order to improve the execution efficiency of the automatic cup separation process, the operation on the sample tube and the operation on the pipette tip are processed in parallel, that is, when the sample tube is grabbed and the sample tube cover is taken down from the sample tube, the operation on taking the pipette tip is executed at the same time, and when the steps of taking the sample tube cover from the sample tube and taking the pipette tip are both successfully executed, the subsequent operations of taking the liquid from the sample tube, removing the pipette tip from the sample tube, screwing the sample tube cover onto the sample tube, and the like are executed by the pipette tip.
In this embodiment, in the process that the mechanical device automatically executes the cup dividing process, a picture of the mechanical device after executing the preset operation is photographed through the monitoring device.
The operation of the mechanical equipment, in which the monitoring equipment is arranged at a certain position, needs to be operated within the visual field of the monitoring equipment, or the monitoring equipment is arranged on a movable equipment and moves along with the operation of the mechanical equipment.
The monitoring device may include one or more devices.
In this embodiment, as can be seen from the above description, the cup separation process includes the above several processes, and based on the above process, S101 includes:
when the mechanical equipment executes the operation of grabbing the sample tube, pictures are shot through the monitoring equipment;
in other embodiments, the monitoring device takes a picture after the mechanical device performs the operation of grabbing the sample tube and moves the mechanical device to the position of the sample tube cover;
when the mechanical equipment carries out the operation of removing the sample tube cover from the sample tube, pictures are taken through the monitoring equipment;
after the mechanical equipment executes the operation of taking the pipette tip, shooting a picture through the monitoring equipment;
when the mechanical equipment performs liquid taking operation from the sample tube through the liquid-transfering gun head, pictures are shot through the monitoring equipment;
when the mechanical equipment executes the operation of detaching the pipette tip, pictures are shot through the monitoring equipment;
after the mechanical device performs the operation of screwing the sample tube cover onto the sample tube, a picture is taken by the monitoring device.
S102: after a target picture after a first preset operation is obtained, inputting the target picture into a multi-classification model trained in advance to obtain state information corresponding to the target picture; the state information corresponding to the target picture at least comprises: a sample tube transfer state, a sample tube switch cover state and a sample extraction state;
the multi-class classification model is obtained by training a preset machine learning model through the correspondence between the historical pictures and the states after the preset operation is executed by the mechanical equipment;
in this embodiment, the cup separation process includes a plurality of operations, and the first preset operation may be represented as any one operation in the cup separation process.
In this embodiment, a preset machine learning model is trained in advance, and a multi-class classification model with different states is obtained.
The preset machine learning model may be any model with classification capability, including: deep learning classification algorithm models, semantic segmentation models, instance segmentation models, models combining segmentation and classification algorithms, or models with traditional image processing algorithms.
Taking a deep learning classification algorithm model as an example, preferably, the embodiment discloses a multi-class classification neural network, the multi-class classification neural network takes Resnet as a basic module, and a decentralized attention mechanism is added among multiple layers.
In the process of training the multi-class classification neural network, when the mechanical equipment executes the automatic cup separating process, the historical picture of the mechanical equipment after executing the preset operation is used as a training sample, the state corresponding to the picture is used as a label, and the multi-class classification neural network is trained.
For example, the following steps are carried out: during training, the resolution of the training sample image is defined as 224 × 224; performing a traversal experiment on the number of single training samples and the total iteration number, and setting the values as follows, wherein the number of the single training samples is [8, 16, 24, 32], and the total iteration number is [250, 200, 150, 100 ]; wherein the initial learning rate is set to 0.005, and the learning rate is attenuated by a certain step length; optimizing a cross entropy loss function by using an Adam optimizer, using softmax as an activation function, performing model parallel training on two Tesla V100 graphics card servers, and finally obtaining the optimal training effect by using the single training sample number of 24 and the iteration number of 150 and realizing the optimal classification in a test set.
The multi-class classification model obtained through training has the capacity of recognizing different states represented by pictures.
In this embodiment, the states that can be recognized by the multi-class classification model include: sample tube transfer state, sample tube switch cover state, sample extraction state.
Wherein, the sample tube transition state includes: the mechanical equipment grabs the sample tube, and the mechanical equipment does not grab the sample tube;
the sample tube switch cover state includes: the sample tube cover is in a screwed state, the sample tube cover is in a unscrewed state, and the sample tube cover is in a loose state;
the sample tube cover is in a screwed state, which means that the sample tube cover is completely connected with the sample tube, and if the sample tube cover has threads, the sample tube cover is in the screwed state if the threads are not exposed; the sample tube cover is in a loose state, which indicates that the sample tube is unscrewed but is not removed from the sample tube, and if the sample tube cover has threads, the threads are exposed; or the sample tube cover is in a loose state, which can also indicate that the sample tube is not screwed; the sample tube cap is in the unscrewed state, which means that the sample tube and the sample tube cap are in a separated state, i.e. the sample tube cap is not present on the sample tube.
The sample tube extraction state comprises: the mechanical equipment and the pipette head are in a connected state, the mechanical equipment and the pipette head are in an unconnected state, the liquid taking amount is in a standard state, and the liquid taking amount is in a non-standard state.
In this embodiment, when the mechanical device grabs the sample tube, there are two situations, that is, the sample tube is successfully grabbed, the sample tube is not successfully grabbed, and if the sample tube is successfully grabbed, it indicates that the mechanical device grabs the sample tube successfully. When the mechanical device is used for opening the sample tube, there are two states, namely, the sample tube cover is in an open state (i.e., the sample tube cover is in a unscrewed state), and the sample tube cover is in an unopened state (i.e., the sample tube cover is in a screwed state or the sample tube cover is in a loose state, both of which indicate that the sample tube cover is in the unopened state). After a mechanical device such as a liquid transfer pump performs an operation of sucking a pipette tip, there are two situations, namely, the pipette pump sucks the upper pipette tip (the mechanical device and the pipette tip are in a connected state), and the liquid transfer pump does not suck the upper pipette tip (the mechanical device and the pipette tip are in an unconnected state); when the pipette tip is controlled to extract liquid from the sample tube, there are two situations, the amount of the extracted liquid reaches a preset standard (the liquid extraction amount is in a state of reaching the standard), and the amount of the extracted liquid does not reach the preset standard (the liquid extraction amount is in a state of not reaching the standard). After the mechanical equipment executes the operation of detaching the pipette tip, two situations exist, namely the mechanical equipment detaches the pipette tip (the mechanical equipment and the pipette tip are in an unconnected state), and the mechanical equipment does not detach the pipette tip (the mechanical equipment and the pipette tip are in a connected state). When a mechanical device performs a capping operation on a sample tube, there are two cases, where the sample tube cap is screwed (the sample tube cap is in a screwed state) and the sample tube cap is not screwed (the sample tube cap is in a loose state).
As can be seen from the above description, when the mechanical device performs different operations of automatically separating cups, the multi-class classification model is extracted into different states based on pictures obtained after the different operations, specifically, S102 includes:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state that whether the mechanical equipment grabs the sample tube or not;
when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment;
when the first preset operation is that the mechanical equipment executes liquid taking operation from the sample tube, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain a state whether the liquid taking amount reaches the standard or not;
when the first preset operation is that the mechanical equipment executes the operation of screwing the sample tube cover to the sample tube, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
and when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette tip, sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment.
S103: determining whether a first preset operation is successfully executed or not according to the state information corresponding to the picture;
in this embodiment, as can be known from the above description, under different operations of the mechanical device for performing automatic cup sorting, the multi-class classification model may extract different states based on pictures obtained after the different operations, where the different states may represent whether the corresponding operations of the mechanical device for performing automatic cup sorting are successfully performed, and specifically, S103 includes:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, and the obtained state information is the state that the mechanical equipment does not grab the sample tube, the mechanical equipment is indicated to be successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover, and the obtained state information indicates that the sample tube cover is in the unscrewing state, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, the obtained state information is that the pipette tip is connected with the mechanical equipment, and the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the liquid taking operation from the sample tube, and the obtained state information comprises the state that the liquid taking amount reaches the standard, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment performs the operation of screwing the sample tube cover to the sample tube, and the obtained state information indicates that the sample tube cover is in a screwed state, the first preset operation is successfully performed;
and when the first preset operation is the operation of the mechanical equipment for unloading the pipette tip, and the obtained state information is that the pipette tip is not connected with the mechanical equipment, the first preset operation is successfully executed.
Alternatively, in another case:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, and the obtained state information is the state that the mechanical equipment grabs the sample tube, the first preset operation is failed to execute, namely the mechanical equipment does not grab the sample tube;
when the first preset operation is that the mechanical equipment performs an operation of unscrewing the sample tube cover, and the obtained state information is that the sample tube cover is in a screwed state or a loosened state, the first preset operation is failed to perform, namely that the sample tube cover is not unscrewed;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, the obtained state information is that the pipette tip and the mechanical equipment are in an unconnected state, and the first preset operation is failed to execute, namely the mechanical equipment does not take the pipette tip;
when the first preset operation is that the mechanical equipment executes the liquid taking operation from the sample tube, and the obtained state information is that the liquid taking amount does not reach the standard, the first preset operation is failed to execute, namely the liquid amount extracted by the liquid-transfer gun head does not reach the standard;
when the first preset operation is that the mechanical equipment executes the operation of screwing the sample tube cover to the sample tube, and the obtained state information is that the sample tube cover is in an unscrewing state or in a loosening state, the first preset operation is failed to execute, namely the sample tube cover is not screwed;
and when the first preset operation is the operation of the mechanical equipment for removing the pipette tip, and the obtained state information is the state that the pipette tip is connected with the mechanical equipment, the first preset operation fails to be performed, namely the pipette tip is not removed from the removing equipment.
S104: under the condition that the first preset operation is successfully executed, controlling the mechanical equipment to execute the next operation after the first preset operation;
in this embodiment, the process of automatically executing cup separation by the mechanical device may include: grabbing the sample tube, taking down the sample tube cover from the sample tube, taking the pipette head, taking the liquid from the sample tube by the pipette head, detaching the pipette head, and screwing the sample tube cover onto the sample tube.
Wherein, in order to take off the sample tube lid from the sample tube, after having snatched the sample tube again, can also move mechanical equipment to the position of sample tube lid.
Or the process of automatically performing cup separation through mechanical equipment can further comprise the following steps: when the sample tube is grabbed and the sample tube cover is taken down from the sample tube, the operation of taking the pipette head is executed simultaneously, and under the condition that the steps of taking the sample tube cover from the sample tube and taking the pipette head are successfully executed, the subsequent operations of taking liquid from the sample tube by the pipette head, detaching the pipette head, screwing the sample tube cover onto the sample tube and the like are executed.
In the process of executing the automatic cup separation process, after a certain preset operation is executed, pictures are shot through the monitoring equipment, the shot pictures are identified through the multi-class classification model, state information corresponding to the pictures is obtained, whether the preset operation is executed successfully or not is determined based on the state information, and if the preset operation is executed successfully, the next operation is executed until the automatic cup separation process is executed completely.
And under the condition that the execution of the first preset operation fails, controlling the mechanical equipment to execute the first preset operation again until the execution of the first preset operation succeeds.
Further, in order to avoid that the whole process of automatic cup separation cannot be completed due to abnormal operation or that the whole process of automatic cup separation is difficult to realize due to abnormal operation, the frequency of executing the first preset operation is detected, and if the frequency of executing the first preset operation failure exceeds a preset threshold value, an alarm is given.
In this embodiment, in the process of automatically executing the cup separating process by the mechanical device, a target picture after the mechanical device executes a preset operation is obtained, and state information corresponding to the target picture is obtained through a preset multi-class classification model, wherein the multi-class classification model is obtained by training a preset machine learning model through a corresponding relation between a history picture after the mechanical device executes the preset operation and a state; and determining whether the first preset operation is successfully executed according to the state information corresponding to the target picture, and executing the next operation after the first preset operation under the condition that the first preset operation is successfully executed until the whole process of automatic cup separation is completed. Therefore, automatic cup separation operation on the sample is achieved, the extraction efficiency of the sample is improved, human participation is not needed, the biological hidden danger is reduced, and the probability of successful execution of the automatic cup separation process is improved by adopting a visual monitoring mode.
Referring to fig. 2, another schematic flow chart of high throughput automatic cup dispensing provided by an embodiment of the present invention is shown, in this embodiment, the method includes:
s201: the mechanical equipment performs the operation of grabbing the sample tube, moves the mechanical equipment to the position of the sample tube cover, and acquires a first picture through the monitoring equipment;
the first picture is a picture of the mechanical device taken after the mechanical device performs the operation of grabbing the sample tube, for example, if the monitoring device is arranged at a certain fixed position, the first picture taken is a picture taken after the mechanical device grabs the sample tube and moves to the field range of the monitoring device.
Identifying the state information of the first picture through a pre-trained multi-class classification model, and if the state information of the first picture is that the mechanical equipment grabs the sample tube, indicating that the mechanical equipment grabs the sample tube is successfully executed;
if the state information of the first picture indicates that the mechanical equipment does not grab the sample tube, the operation of grabbing the sample tube by the mechanical equipment fails, the operation of grabbing the sample tube is executed again, and if the operation of grabbing the sample tube for the second time fails, an alarm is given;
s202: if the mechanical equipment successfully performs the operation of grabbing the sample tube, the mechanical equipment performs the uncovering operation and acquires a second picture through the monitoring equipment;
the second picture is a picture of the sample tube shot after the mechanical equipment is subjected to the uncovering operation.
Identifying the state of the second picture through a pre-trained multi-class classification model, and if the state information of the second picture is that the sample tube cover is in an unscrewing state, indicating that the uncapping operation of the mechanical equipment is successfully executed;
if the state information of the second picture is that the sample tube cover is in a screwed state or a loosened state, indicating that the mechanical equipment cover opening operation fails to be executed, executing the cover opening operation again, and if the secondary cover opening operation fails to be executed, alarming and reminding;
s203: the mechanical equipment executes the operation of taking the pipette tip, and a third picture is collected through the monitoring equipment;
the third picture is a picture of the pipette head taken after the mechanical device executes the pipette head, for example, if the monitoring device is set at a certain fixed position, the third picture is a picture taken after the mechanical device executes the operation of taking the pipette head and moves to the field range of the monitoring device.
Recognizing the state of the third picture through a pre-trained multi-class classification model, and if the state information of the third picture is that the mechanical equipment is in a connection state with the pipette head, indicating that the mechanical equipment takes the pipette head up;
if the state information of the third picture indicates that the mechanical equipment is not connected with the pipetting gun head, the mechanical equipment executes the operation of taking the pipetting gun head again, and if the operation of taking the pipetting gun head twice fails, an alarm is given;
s204: if the mechanical equipment successfully executes the operation of taking the pipette tips and the mechanical equipment successfully opens the cover, the pipette tips execute the operation of taking liquid from the sample tube and acquire a fourth picture through the monitoring equipment;
and the fourth picture is a picture of the pipette tip taken after the pipette tip performs the liquid taking operation from the sample tube.
Identifying the state of the fourth picture through a pre-trained multi-class classification model, and if the state information of the fourth picture is that the liquid taking amount is in a standard state, indicating that the liquid in the sample tube is successfully extracted by the pipette tip;
and if the state information of the fourth picture indicates that the liquid taking amount is not in the standard state, the liquid taking device indicates that the liquid taking gun head is not successfully taken into the sample tube, the liquid taking gun head is controlled again to perform liquid taking operation from the sample tube, and if the liquid taking operation fails to be performed again, alarm reminding is performed.
S205: if the liquid taking of the pipette tip from the sample tube is successful, the mechanical equipment performs the operation of screwing the sample tube cover to the sample tube, and a fifth picture is shot through the monitoring equipment;
identifying the state of a fifth picture through a pre-trained multi-class classification model, and if the state information of the fifth picture is that the sample tube cover is in a screwing state, indicating that the mechanical equipment successfully performs the operation of screwing the sample tube cover to the sample tube; if the state information of the fifth picture includes: if the sample tube cover is in an unscrewing state and the sample tube cover is in a loosening state, the operation of screwing the sample tube cover to the sample tube by mechanical equipment fails, the operation of screwing the sample tube cover is continuously executed, and if the operation of screwing the sample tube cover for the second time fails, an alarm is given;
s206: if the pipette tip successfully takes the liquid from the sample tube, the mechanical equipment executes the operation of detaching the pipette tip, and a sixth picture is shot by the monitoring equipment;
the sixth picture is a picture of the pipette tip taken after the mechanical equipment executes the operation of detaching the pipette tip;
identifying the state of the fifth picture through a pre-trained multi-class classification model, and if the state information of the fifth picture is that the mechanical equipment and the pipette head are in an unconnected state, indicating that the mechanical equipment successfully executes the operation of detaching the pipette head;
and if the state information of the sixth picture indicates that the mechanical equipment is in a connection state with the pipette tip, the mechanical equipment fails to execute the operation of detaching the pipette tip, the operation of detaching the pipette tip is executed again, and if the operation of secondarily executing the operation of detaching the pipette tip fails, an alarm is given.
S207: if the sample tube cap is screwed down and the pipette tips have been unloaded, the sample tubes are moved back to the tray.
In this embodiment, in the process of automatically executing the cup separating process by the mechanical device, a target picture after the mechanical device executes a preset operation is obtained, and state information corresponding to the target picture is obtained through a preset multi-class classification model, wherein the multi-class classification model is obtained by training a preset machine learning model through a corresponding relation between a history picture after the mechanical device executes the preset operation and a state; and determining whether the first preset operation is successfully executed according to the state information corresponding to the target picture, and executing the next operation after the first preset operation under the condition that the first preset operation is successfully executed until the whole process of automatic cup separation is completed. Therefore, automatic cup separation operation on the sample is achieved, the extraction efficiency of the sample is improved, human participation is not needed, the biological hidden danger is reduced, and the probability of successful execution of the automatic cup separation process is improved by adopting a visual monitoring mode.
Referring to fig. 3, a schematic structural diagram of a high throughput automatic cup dispensing device provided in an embodiment of the present invention is shown, in this embodiment, the device includes:
the acquiring unit 301 is configured to acquire a picture after a mechanical device performs a preset operation in a process that the mechanical device automatically performs a cup separating process;
the state information identification unit 302 is configured to, after a target picture after a first preset operation is acquired, input the target picture into a multi-class classification model trained in advance to obtain state information corresponding to the target picture; the state information corresponding to the target picture at least comprises: a sample tube transfer state, a sample tube switch cover state and a sample extraction state; the multi-class classification model is obtained by training a preset machine learning model through the corresponding relation between the historical pictures and the states after the preset operation is executed by the mechanical equipment;
a determining unit 303, configured to determine whether a first preset operation is successfully executed according to the state information corresponding to the target picture;
and the control unit 304 is configured to control the mechanical device to execute a next operation after the first preset operation if the first preset operation is successfully executed.
Alternatively to this, the first and second parts may,
the sample tube transfer state includes: the mechanical equipment grabs the sample tube, and the mechanical equipment does not grab the sample tube;
the sample tube switch cover state includes: the sample tube cover is in a screwed state, the sample tube cover is in a unscrewed state, and the sample tube cover is in a loose state;
the sample tube extraction state comprises: the mechanical equipment and the pipette head are in a connected state, the mechanical equipment and the pipette head are in an unconnected state, the liquid taking amount is in a standard state, and the liquid taking amount is in a non-standard state.
Optionally, the state information identifying unit includes:
the first state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the mechanical equipment grabs the sample tube or not when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover;
the second state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover when the first preset operation is the operation of unscrewing the sample tube cover by the mechanical equipment;
the third state identification unit is used for sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment or not when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip;
the fourth state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state whether the liquid taking amount reaches the standard or not when the first preset operation is that the mechanical equipment executes the liquid taking operation from the sample tube;
the fifth state identification unit is used for executing the operation of screwing the sample tube cover to the sample tube when the first preset operation is the mechanical equipment, and sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
and the sixth state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment or not when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette tip.
Optionally, the determining unit includes:
the first determining subunit is used for indicating that the first preset operation is successfully executed when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover and the obtained state information is the state that the mechanical equipment grabs the sample tube;
the second determining subunit is used for indicating that the first preset operation is successfully executed when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover and the obtained state information is that the sample tube cover is in the unscrewing state;
the third determining subunit is used for obtaining the state information that the pipette tip is in a connection state with the mechanical equipment when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, and the first preset operation is successfully executed;
the fourth determining subunit is configured to, when the first preset operation is that the mechanical device performs a liquid taking operation from the sample tube, and the obtained state information includes a state that the liquid taking amount reaches the standard, indicate that the first preset operation is successfully performed;
a fifth determining subunit, configured to indicate that the first preset operation is successfully executed when the first preset operation is an operation of the mechanical device to screw the sample tube cap to the sample tube, and the obtained state information indicates that the sample tube cap is in a screwed state;
and the sixth determining subunit is used for indicating that the first preset operation is successfully executed when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette head and the obtained state information is that the pipette head is not connected with the mechanical equipment.
Optionally, the method further includes:
and the repeated execution unit is used for executing the first preset operation again if the execution of the first preset operation fails until the execution of the first preset operation is successful.
Optionally, the method further includes:
and the early warning unit is used for alarming and reminding if the number of times of the execution failure of the first preset operation exceeds a preset threshold value.
Through the device of this embodiment, not only realized dividing the cup operation automatically to the sample, improved the extraction efficiency of sample to need not artificial participation, reduced the biological hidden danger that exists, and, through the mode that adopts visual monitoring, improved the probability that automatic branch cup flow was carried out successfully.
Referring to fig. 4, a schematic structural diagram of a high throughput automatic cup dispensing system according to an embodiment of the present invention is shown, in this embodiment, the system includes:
a controller 401, a mechanical device 402, a monitoring device 403;
the controller is used for executing the high-throughput automatic cup separation method, which is not described in detail in this embodiment;
the mechanical equipment is used for executing the operation in the cup separating process according to the instruction of the controller;
the monitoring equipment is used for shooting pictures after the mechanical equipment executes preset operation.
The monitoring device can be arranged at a certain position, the mechanical device executes certain operation and then reaches the field range of the monitoring device, and the monitoring device shoots corresponding pictures. Or the monitoring device is arranged on a movable device and moves along with the operation of the mechanical device.
After the mechanical equipment performs the operation of grabbing the sample tube, the monitoring equipment shoots a picture of the current state of the mechanical equipment; after the mechanical equipment performs the uncovering operation, the monitoring equipment shoots a picture of the current state of the sample tube; after the mechanical equipment executes the operation of taking the pipette tip, the monitoring equipment shoots a picture of the current state of the mechanical equipment; the pipette tip performs liquid taking operation from the sample tube, and the monitoring equipment shoots a picture of the current state of the pipette tip; the mechanical equipment executes the operation of screwing the sample tube cover to the sample tube, and the monitoring equipment shoots the picture of the current state of the sample tube; the mechanical equipment executes the operation of detaching the pipette head, and the monitoring equipment shoots the picture of the mechanical equipment.
The mechanical device may include one or more of, for example, a manipulator, and a pipetting pump, for example, a mechanical device for performing an operation of grasping a specimen tube, an operation of uncapping the specimen tube, and an operation of screwing a specimen tube cap onto the specimen tube, and a mechanical device for performing an operation of pipetting a pipette tip, an operation of pipetting a liquid from the specimen tube, and an operation of detaching the pipette tip from the specimen tube.
Through the system of this embodiment, not only realized dividing the cup operation automatically to the sample, improved the extraction efficiency of sample to need not artificial participation, reduced the biological hidden danger that exists, and, through the mode that adopts visual monitoring, improved the probability that automatic branch cup flow was carried out successfully.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. A high-throughput automatic cup separation method is characterized by comprising the following steps:
in the process that mechanical equipment automatically executes a cup dividing process, acquiring a target picture after the mechanical equipment executes preset operation;
after a target picture after a first preset operation is obtained, inputting the target picture into a multi-class classification model trained in advance to obtain state information corresponding to the target picture; the state information corresponding to the target picture at least comprises: a sample tube transfer state, a sample tube switch cover state and a sample extraction state; the multi-class classification model is obtained by training a preset machine learning model through the corresponding relation between the historical pictures and the state information after the preset operation is executed by the mechanical equipment; adding a decentralized attention mechanism between multiple layers of the multi-class classification model;
determining whether a first preset operation is successfully executed or not according to the state information corresponding to the target picture;
under the condition that the first preset operation is successfully executed, controlling the mechanical equipment to execute the next operation after the first preset operation;
and if the execution of the first preset operation fails, executing the first preset operation again until the execution of the first preset operation is successful.
2. The method of claim 1,
the sample tube transfer state includes: the mechanical equipment grabs the sample tube, and the mechanical equipment does not grab the sample tube;
the sample tube switch cover state includes: the sample tube cover is in a screwed state, the sample tube cover is in a unscrewed state, and the sample tube cover is in a loose state;
the sample tube extraction state comprises: the mechanical equipment and the pipette head are in a connected state, the mechanical equipment and the pipette head are in an unconnected state, the liquid taking amount is in a standard state, and the liquid taking amount is in a non-standard state.
3. The method according to claim 2, wherein the inputting the target picture into a multi-class classification model trained in advance after the target picture after the first preset operation is obtained to obtain the state information corresponding to the target picture comprises:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state that whether the mechanical equipment grabs the sample tube or not;
when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment;
when the first preset operation is that the mechanical equipment executes the operation of taking liquid from the sample tube, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state whether the liquid taking amount reaches the standard or not;
when the first preset operation is that the mechanical equipment executes the operation of screwing the sample tube cover to the sample tube, sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
and when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette tip, sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment.
4. The method according to claim 3, wherein the determining whether the first preset operation is successfully executed according to the state information corresponding to the picture comprises:
when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover, and the obtained state information is the state that the mechanical equipment grabs the sample tube, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of unscrewing the sample tube cover, and the obtained state information indicates that the sample tube cover is in the unscrewing state, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip, the obtained state information is that the pipette tip is connected with the mechanical equipment, and the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment executes the liquid taking operation from the sample tube, and the obtained state information comprises the state that the liquid taking amount reaches the standard, the first preset operation is successfully executed;
when the first preset operation is that the mechanical equipment performs the operation of screwing the sample tube cover to the sample tube, and the obtained state information indicates that the sample tube cover is in a screwed state, the first preset operation is successfully performed;
and when the first preset operation is the operation of the mechanical equipment for unloading the pipette tip, and the obtained state information is that the pipette tip is not connected with the mechanical equipment, the first preset operation is successfully executed.
5. The method of claim 1, wherein if the number of times of the first predetermined operation execution failure exceeds a predetermined threshold, an alarm is given.
6. A high throughput automatic cup dispensing device, comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a target picture after the mechanical equipment executes preset operation in the process of automatically executing a cup dividing process by the mechanical equipment;
the state information identification unit is used for inputting the target picture into a pre-trained multi-class classification model after the target picture after the first preset operation is acquired, and obtaining state information corresponding to the target picture; the state information corresponding to the target picture at least comprises: a sample tube transfer state, a sample tube switch cover state and a sample extraction state; the multi-class classification model is obtained by training a preset machine learning model through the corresponding relation between the historical pictures and the states after the preset operation is executed by the mechanical equipment; adding a decentralized attention mechanism between multiple layers of the multi-class classification model;
the determining unit is used for determining whether the first preset operation is successfully executed or not according to the state information corresponding to the target picture;
the control unit is used for controlling the mechanical equipment to execute the next operation after the first preset operation under the condition that the first preset operation is executed successfully; and if the execution of the first preset operation fails, executing the first preset operation again until the execution of the first preset operation is successful.
7. The apparatus of claim 6,
the sample tube transfer state includes: the mechanical equipment grabs the sample tube, and the mechanical equipment does not grab the sample tube;
the sample tube switch cover state includes: the sample tube cover is in a screwed state, the sample tube cover is in a unscrewed state, and the sample tube cover is in a loose state;
the sample tube extraction state comprises: the mechanical equipment and the pipette head are in a connected state, the mechanical equipment and the pipette head are in an unconnected state, the liquid taking amount is in a standard state, and the liquid taking amount is in a non-standard state.
8. The apparatus of claim 6, wherein the status information identifying unit comprises:
the first state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the mechanical equipment grabs the sample tube or not when the first preset operation is the operation that the mechanical equipment grabs the sample tube and moves to the position of the sample tube cover;
the second state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover when the first preset operation is the operation of unscrewing the sample tube cover by the mechanical equipment;
the third state identification unit is used for sending a target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment or not when the first preset operation is that the mechanical equipment executes the operation of taking the pipette tip;
the fourth state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state whether the liquid taking amount reaches the standard or not when the first preset operation is that the mechanical equipment executes the liquid taking operation from the sample tube;
the fifth state identification unit is used for executing the operation of screwing the sample tube cover to the sample tube when the first preset operation is the mechanical equipment, and sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the opening and closing state of the sample tube cover;
and the sixth state identification unit is used for sending the target picture obtained after the first preset operation to the multi-class classification model to obtain the state of whether the pipette tip is connected with the mechanical equipment or not when the first preset operation is that the mechanical equipment executes the operation of detaching the pipette tip.
9. A high throughput automated cup dispensing system, comprising:
a controller, a mechanical device and a monitoring device;
the controller is used for executing the high-throughput automatic cup separation method of any one of the claims 1 to 5;
the mechanical equipment is used for executing the operation in the cup separating process according to the instruction of the controller;
the monitoring equipment is used for shooting pictures after the mechanical equipment executes preset operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110591721.4A CN113034503B (en) | 2021-05-28 | 2021-05-28 | High-flux automatic cup separating method, device and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110591721.4A CN113034503B (en) | 2021-05-28 | 2021-05-28 | High-flux automatic cup separating method, device and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113034503A CN113034503A (en) | 2021-06-25 |
CN113034503B true CN113034503B (en) | 2021-08-24 |
Family
ID=76456215
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110591721.4A Active CN113034503B (en) | 2021-05-28 | 2021-05-28 | High-flux automatic cup separating method, device and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113034503B (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111191590A (en) * | 2019-12-30 | 2020-05-22 | Oppo(重庆)智能科技有限公司 | Model training method and device, storage medium and electronic equipment |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1558765A4 (en) * | 2002-11-08 | 2005-11-30 | Pharmacia Corp | High throughput automatic nucleic acid isolation and quantitation methods |
CN106056035A (en) * | 2016-04-06 | 2016-10-26 | 南京华捷艾米软件科技有限公司 | Motion-sensing technology based kindergarten intelligent monitoring method |
CN109685102B (en) * | 2018-11-13 | 2024-07-09 | 平安科技(深圳)有限公司 | Chest focus image classification method, device, computer equipment and storage medium |
CN110197166B (en) * | 2019-06-04 | 2022-09-09 | 西安建筑科技大学 | Vehicle body loading state recognition device and method based on image recognition |
CN112288345B (en) * | 2019-07-25 | 2024-08-20 | 顺丰科技有限公司 | Loading and unloading port state detection method and device, server and storage medium |
CN110490120B (en) * | 2019-08-14 | 2022-03-25 | 北京达佳互联信息技术有限公司 | Dangerous behavior detection method and device, server and storage medium |
CN111310590B (en) * | 2020-01-20 | 2023-07-11 | 北京西米兄弟未来科技有限公司 | Action recognition method and electronic equipment |
CN210481395U (en) * | 2020-04-04 | 2020-05-08 | 博奥生物集团有限公司 | High-throughput full-automatic nucleic acid detection system |
CN111860448A (en) * | 2020-07-30 | 2020-10-30 | 北京华捷艾米科技有限公司 | Hand washing action recognition method and system |
CN212780841U (en) * | 2021-02-08 | 2021-03-23 | 博奥生物集团有限公司 | Cup separating device and sample analysis equipment |
-
2021
- 2021-05-28 CN CN202110591721.4A patent/CN113034503B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111191590A (en) * | 2019-12-30 | 2020-05-22 | Oppo(重庆)智能科技有限公司 | Model training method and device, storage medium and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN113034503A (en) | 2021-06-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6018406B2 (en) | Operation method of automatic sample work cell | |
CN107257161B (en) | Transformer substation disconnecting link remote control auxiliary checking method and system based on state recognition algorithm | |
Liao et al. | An accurate segmentation method for white blood cell images | |
CN106288591B (en) | A kind of refrigerator and food management method with food materials management function | |
CN108010026B (en) | Protection pressing plate state identification method and device | |
CN112189170B (en) | Method for processing a workpiece by means of a removal tool and machine for carrying out the method | |
WO2006037941A1 (en) | Determination of the boundaries between fractions and extraction of selected fractions in a fractionated sample | |
JP2016028607A5 (en) | ||
EP3798924A1 (en) | System and method for classifying manufactured products | |
CN105095848B (en) | The method and system of target identification | |
CN101221623A (en) | Object type on-line training and recognizing method and system thereof | |
CN113034503B (en) | High-flux automatic cup separating method, device and system | |
CN111986195B (en) | Appearance defect detection method and system | |
CN114820598B (en) | PCB defect detection system and PCB defect detection method | |
WO2021201251A1 (en) | Method for analyzing composition of electronic/electrical apparatus component layer, method for processing electronic/electrical apparatus component layer, device for analyzing composition of electronic/electrical apparatus component layer, and device for processing electronic/electrical apparatus component layer | |
CN110877019A (en) | Traditional Chinese medicinal material impurity removing device and method | |
CN111210412A (en) | Package detection method and device, electronic equipment and storage medium | |
CN108038418B (en) | Garbage cleaning method and device | |
EP3513377A1 (en) | Computer-implemented apparatus and method for performing a genetic toxicity assay | |
CN113469137A (en) | Abnormal behavior recognition method and device, storage medium and electronic device | |
CN111646332A (en) | Method and system for identifying abnormal operation of elevator | |
US20230118305A1 (en) | Method and apparatus for identifying an article | |
CN118731385A (en) | Sample analysis system | |
CN113524184B (en) | Sample rack grabbing method, device and system | |
CN116310725A (en) | Method and device for detecting calcium carbide proofing state and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20240920 Address after: 102206 No. 18, life science Road, Beijing, Changping District Patentee after: CAPITALBIO Corp. Country or region after: China Patentee after: TSINGHUA University Address before: 102206 No. 18, life science Road, Beijing, Changping District Patentee before: CAPITALBIO Corp. Country or region before: China |