A DNA sequence printing device fault detection method, system and apparatus

Abstract

The application relates to the technical field of fault detection, and particularly discloses a DNA sequence printing equipment fault detection method, system and device, wherein the device comprises the following modules: a data synchronization module, which is used for acquiring printing information of a DNA sequence printer, and synchronously sending a trigger signal to a collection module according to the printing information; the collection module, which is used for collecting image data of a jet port of the DNA sequence printer after receiving the trigger signal; a data cache module, which is used for adding the image data to a data cache queue; and a detection module, which is used for inputting the image data into an AI detection model which has been trained, identifying the state of the jet port through the AI detection model, and outputting a detection result. The technical scheme of the application can quickly detect the fault of the injection device.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault detection, in particular to a DNA sequence printing equipment fault detection method, system and device. BACKGROUND

[0002] In biological research, it is necessary to frequently design biological materials with different DNA sequences, manufacture them for experiments, modify the design, manufacture the modified materials, and then experiment again. However, through a DNA sequence printer, biological materials can be printed according to the DNA sequence designed by researchers.

[0003] When the DNA sequence printer is working, the DNA medicine needs to be injected through the air pressure injection port of the injection device. When the injection device works for a long time, mechanical fatigue will cause the air pressure control valve of the injection device to malfunction or delay, and then cause the DNA printing sequence to be disordered. In addition, the solid crystals generated by the DNA medicine in the injection device after long-time work will also cause the injection port to be blocked, and then cause the DNA printing sequence to be missing.

[0004] In order to find the above-mentioned faults, manual detection is usually used at present, but the manual detection method has the following problems or shortcomings:

[0005] 1. The staff may miss detection after fatigue, or the staff may bring emotions to make the detection results subjective, so it is difficult to guarantee the reliability of the detection results;

[0006] 2. Due to the limited energy of the staff, the efficiency of manual detection has a fixed range, and cannot be greatly improved;

[0007] 3. Manual detection makes the DNA sequence printer unable to work in a closed environment, and is easy to cause medicine pollution;

[0008] 4. Manual detection cannot match the production speed of the high-speed printing of the DNA sequence printer production line.

[0009] Therefore, a method, system and device for quickly detecting the injection device of the DNA sequence printer are needed. SUMMARY

[0010] One of the purposes of the present application is to provide a DNA sequence printing equipment fault detection device capable of quickly detecting the fault of the injection device.

[0011] In order to solve the above technical problems, the present application provides the following technical solutions:

[0012] A DNA sequence printing equipment fault detection device, comprising:

[0013] The data synchronization module is configured to acquire printing information of the DNA sequence printer and synchronously send a trigger signal to the acquisition module according to the printing information.

[0014] The acquisition module is configured to acquire image data of a jet port of the DNA sequence printer after receiving the trigger signal.

[0015] The data caching module is configured to add the image data to a data caching queue.

[0016] The detection module is configured to input the image data into a trained AI detection model, identify a state of the jet port through the AI detection model, and output a detection result.

[0017] Preferably, the printing information includes a plurality of text data.

[0018] The data synchronization module is further configured to send the text data to the data caching module.

[0019] The data caching module is further configured to add the text data to the data caching queue.

[0020] The detection module is further configured to compare the detection result with a state of a nozzle in the text data and generate a comparison result.

[0021] Preferably, the method further includes a storage module and a visualization module.

[0022] The storage module is configured to store the image data.

[0023] The visualization module is configured to receive the comparison result and further configured to acquire corresponding image data from the storage module, generate a visual detection result based on the image data and the comparison result.

[0024] Preferably, the data synchronization module is further configured to verify the text data according to preset transmission encoding information, and send the text data to the data caching module after verification.

[0025] Preferably, the data caching module is further configured to convert a format of the image data into a preset format, and add the text data and the image data in the preset format to the data caching queue.

[0026] Preferably, the detection result includes a jetting state, a stopping state, an overflow state and a water drop state.

[0027] The second object of the present application is to provide a DNA sequence printing equipment fault detection method, including the following steps:

[0028] Synchronous acquisition step: acquire printing information of the DNA sequence printer; the printing information includes a plurality of text data, the opening time of the ejection port is determined through the text data, a trigger signal is synchronously sent to the acquisition module through the preset transmission encoding information based on the opening time of the ejection port, and the acquisition module is controlled to acquire image data of the ejection port;

[0029] Text verification step: verifying the text data according to the transmission encoding information, and adding the text data to a data cache queue after verification;

[0030] Cache step: converting the format of the image data into a preset format, and adding the image data after format conversion to the data cache queue;

[0031] AI inference step: inputting the image data after format conversion into the AI detection model after training to obtain a detection result from the AI detection model;

[0032] Comparison step: comparing the detection result with the nozzle state in the text data to generate a comparison result.

[0033] Preferably, the method further comprises a storage step of storing the image data after format conversion;

[0034] Display step: receiving the comparison result, acquiring corresponding stored image data, and generating a visual detection result based on the image data and the comparison result.

[0035] Preferably, the detection result includes an ejection state, a stop state, an overflow state and a water droplet state.

[0036] The third object of the present application is to provide a DNA sequence printing equipment fault detection system using the DNA sequence printing equipment fault detection device.

[0037] Compared with the prior art, the present application has the following beneficial effects:

[0038] In the present scheme, the trigger signal is synchronously sent to the acquisition module according to the printing information of the DNA sequence printer, which can ensure that the ejection of the DNA medicine and the photographing are performed at the same time, and the accuracy of the acquisition is ensured. The state of the ejection port is identified by the AI detection model and the detection result is output, which can efficiently and quickly monitor the state of the ejection port compared with manual detection; the fault detection of the DNA sequence printing device is realized. By acquiring the comparison result and the image data, the visual detection result is generated based on the image data and the comparison result, which can achieve remote visual detection alarm monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 For example one, a logic block diagram of a DNA sequence printing equipment fault detection state;

[0040] Figure 2 Figure 1 is a schematic diagram of the injection state of the DNA sequence printer fault detection device of the embodiment one;

[0041] Figure 3 Figure 2 is a schematic diagram of the stop state of the DNA sequence printer fault detection device of the embodiment one;

[0042] Figure 4 Figure 3 is a schematic diagram of the overflow state of the DNA sequence printer fault detection device of the embodiment one;

[0043] Figure 5 Figure 4 is a schematic diagram of the water drop state of the DNA sequence printer fault detection device of the embodiment one. DETAILED DESCRIPTION

[0044] The following will be further described in detail through specific embodiments:

[0045] Embodiment one

[0046] The embodiment provides a DNA sequence printer, which comprises a circular conveying belt for conveying test tubes, and further comprises an injection device, the injection port of the injection device being used for injecting DNA medicine to the test tubes. The number of injection ports is set according to actual conditions, and in the embodiment, the injection port is a pneumatic injection port.

[0047] As shown in Figure 1 The DNA sequence printer fault detection device of the embodiment comprises a collection module, a data synchronization module, a data cache module and a detection module.

[0048] The number of collection modules is several, and in the embodiment, the collection module is a CCD black-and-white industrial camera, which is arranged in a circular array around the conveying belt of the DNA sequence printer and is aligned with the injection port of the injection device.

[0049] In the embodiment, a light source is further included, which specifically adopts an LED strip light and is arranged above and below the CCD black-and-white industrial camera respectively, and illuminates the injection port through a bright field illumination mode.

[0050] The data synchronization module is used for acquiring printing information of the DNA sequence printer, and the printing information comprises a plurality of text data. The data synchronization module is further used for determining the opening time of the injection port through the text data, and based on the opening time of the injection port, a trigger signal is synchronously sent to the collection module through preset transmission encoding information, so that the photographing and the injection of DNA medicine by the injection port are synchronously performed. In the embodiment, the printing information is input by a staff, and is used for controlling the printing work of the DNA sequence printer. The transmission encoding information is specified in advance by both parties of communication, and in the embodiment, the transmission encoding information is a CRC16 communication check code, which can be used for checking whether the data transmission is wrong.

[0051] The acquisition module is configured to acquire image data of the spray port after receiving the trigger signal. In this embodiment, the data synchronization module uses a plurality of data synchronizers, and each group of four CCD black-and-white industrial cameras is controlled by one data synchronizer.

[0052] The data synchronization module also checks the text data according to the transmission encoding information, and sends the text data to the data cache module after the check is passed. In this embodiment, the data cache module sends the text data in a serial communication mode. The text data includes the nozzle number and the nozzle state, and the nozzle state includes spraying and stopping spraying.

[0053] In this embodiment, the image data acquired by the acquisition module is transmitted to the data cache module through a gigabit Ethernet network.

[0054] The data cache module is also configured to convert the format of the image data to a preset format. In this embodiment, the preset format is a format supported by OpenCV. The data cache module is also configured to add the text data and the image data after format conversion to a data cache queue. In this embodiment, the data cache queue is a concurrent queue. Caching the text data and the image data in the form of a concurrent queue can solve the problem of high concurrency of data.

[0055] The detection module is configured to input the image data after format conversion into a trained AI detection model. The AI detection model identifies the state of the spray port and outputs a detection result. In this embodiment, the AI detection model uses a YOLOX target detection model, and the detection result includes a spraying state, a stopping state, an overflow state, and a water droplet state. The detection module is also configured to compare the detection result with the nozzle state in the text data to generate a comparison result and send the comparison result to the visualization module. In the comparison result of this embodiment, if the nozzle state is spraying and the detection result is a spraying state, or the nozzle state is stopping spraying and the detection result is a stopping state, it is normal; otherwise, it is abnormal, such as an overflow state, a water droplet state, or a nozzle state of spraying and a detection result of a stopping state. Figure 2 Figure 3 Figure 4 Figure 5

[0056] In other embodiments, a storage module and a visualization module can also be included. The storage module is configured to store the image data after format conversion. In this embodiment, the image data records the shooting time.

[0057] ​​​​The visualization module is configured to receive the comparison result and obtain corresponding image data from the storage module, and generate a visualization detection result based on the image data and the comparison result. In this embodiment, the visualization module receives the comparison result from the Ethernet in the form of a subscription. The visualization detection result is displayed in a GUI interface. In this embodiment, the visualization detection result includes a picture of the labeled injection port and log information. In this embodiment, the end of the injection port is framed, and the corresponding detection result is marked to achieve labeling. The log information is, for example, detection time: November 11, 2021 11:11:11; event: detection of overflow state; No. 1 nozzle overflow, No. 7 nozzle overflow, No. 18 nozzle overflow.

[0058] The embodiment also provides a DNA sequence printing device fault detection system using the DNA sequence printing device fault detection device.

[0059] Embodiment two

[0060] Based on the DNA sequence printing device fault detection device, the embodiment also provides a DNA sequence printing device fault detection method, including the following steps:

[0061] Synchronous acquisition step: obtaining printing information of the DNA sequence printer; the printing information includes a plurality of text data, the opening time of the injection port is determined through the text data, a trigger signal is synchronously sent to the acquisition module through the preset transmission encoding information based on the opening time of the injection port, and the acquisition module is controlled to acquire image data of the injection port; to realize the synchronization of photographing and injection of DNA drugs.

[0062] Text verification step: verifying the text data according to the transmission encoding information, adding the text data to the data cache queue after verification, the text data including nozzle number and nozzle state, the nozzle state including injection and stop injection. One piece of text data can control multiple nozzles.

[0063] Buffering step: converting the format of the image data into a preset format. In this embodiment, the preset format is a format supported by OpenCV. The image data after format conversion is added to the data cache queue, and in this embodiment, the data cache queue is a concurrent queue. The text data and the image data are cached through the form of a concurrent queue, which can solve the problem of high concurrency of data.

[0064] AI inference step: inputting the image data after format conversion into the trained AI detection model to obtain a detection result from the AI detection model. In this embodiment, the AI detection model uses a YOLOX target detection model, and the detection result includes injection state( Figure 2 ), stop state( Figure 3 ), overflow state( Figure 4 ) and water droplet stateFigure 5 )。

[0065] Comparing step: comparing the detection result with the nozzle state in the text data to generate a comparison result; in the comparison result of the embodiment, the nozzle state is spraying, the detection result is the spraying state, or the nozzle state is stopping spraying, the detection result is the stopping state, which is normal; other cases are abnormal, such as the overflow state, the water droplet state, or the nozzle state is spraying, the detection result is the stopping state, etc.

[0066] Storage step: storing the image data after the format conversion, and the image data records the shooting time in the embodiment.

[0067] Display step: receiving the comparison result, obtaining the corresponding image data, and generating a visual detection result based on the image data and the comparison result; in the embodiment, the visual detection result is displayed in the GUI interface. The visual detection result includes the picture of the labeled spraying port and the log information. In the embodiment, the end of the spraying port is framed and labeled with the corresponding detection result to achieve labeling. The log information is, for example, detection time: November 11, 2021 11:11:11; event: overflow state detected; No. 1 nozzle is overflowing, No. 7 nozzle is overflowing, and No. 18 nozzle is overflowing.

[0068] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described by referring to the preferred embodiments of the present application, it should be understood by those skilled in the art that various changes can be made in form and details without departing from the spirit and scope of the present application as defined by the appended claims.

Claims

1. A fault detection device for DNA sequence printing equipment, characterized in that, include: The data synchronization module is used to acquire the printing information of the DNA sequencer and send trigger signals to the acquisition module synchronously based on the printing information. The printing information is entered by staff to control the printing operation of the DNA sequence printer; The acquisition module is used to acquire image data from the nozzle of the DNA sequencer printer after receiving a trigger signal; The data caching module is used to add image data to the data caching queue; The detection module is used to input image data into the trained AI detection model, identify the state of the jet nozzle through the AI ​​detection model, and output the detection results. The printed information includes several text data; The data synchronization module is also used to send text data to the data caching module; The data caching module is also used to add text data to the data cache queue; The detection module is also used to compare the detection results with the nozzle status in the text data to generate comparison results; It also includes a storage module and a visualization module; The storage module is used to store image data; The visualization module is used to receive comparison results and also to retrieve corresponding image data from the storage module. Based on the image data and comparison results, it generates visualized detection results. The data synchronization module also verifies the text data according to the preset transmission encoding information, and sends the text data to the data cache module after the verification is successful. The data caching module is also used to convert the image data format into a preset format; the data caching module is also used to add text data and format-converted image data to the data caching queue; The detection results include spraying state, stopped state, overflowing state, and water droplet state.

2. A method for detecting faults in a DNA sequence printing device, using the apparatus described in claim 1, characterized in that, Includes the following steps: Synchronous acquisition steps: Obtain the printing information of the DNA sequencer; the printing information includes several text data, the opening time of the ejector nozzle is determined by the text data, and based on the opening time of the ejector nozzle, a trigger signal is synchronously sent to the acquisition module through preset transmission encoding information to control the acquisition module to acquire image data of the ejector nozzle; Text verification steps: Verify the text data according to the transmission encoding information. After the verification is successful, add the text data to the data cache queue. Caching steps: Convert the image data to a preset format, and add the converted image data to the data cache queue; AI inference steps: Input the format-converted image data into the trained AI detection model, and obtain the detection results from the AI ​​detection model; Comparison steps: Compare the detection results with the nozzle status in the text data to generate comparison results.

3. The DNA sequence printing equipment fault detection method according to claim 2, characterized in that: It also includes a storage step: storing the format-converted image data; The process involves: receiving the comparison results, acquiring the corresponding stored image data, and generating a visual detection result based on the image data and the comparison results.

Images