Liquid type identification method based on convolutional neural network model

Abstract

The invention provides a liquid type identification method based on a convolutional neural network model. The method comprises the steps of obtaining a target microwave signal of a target container containing to-be-detected liquid; acquiring a target empty measurement microwave signal of the target container which does not contain the liquid; the difference value of the target microwave signal and the target space measurement microwave signal is input into a convolutional neural network model, target information, output by the convolutional neural network model, of the to-be-measured liquid is obtained, and the target information comprises the liquid type of the to-be-measured liquid; wherein the convolutional neural network model is obtained by training based on a sample difference value corresponding to a sample liquid and a liquid type label of the sample liquid, and the sample difference value is a difference value between a sample microwave signal of a sample container containing the sample liquid and a sample target empty measurement microwave signal of the sample container not containing the liquid. The liquid type identification accuracy can be improved, and the false alarm rate can be reduced.

Description

Technical Field

[0001] The present invention relates to the field of safety detection technology, and in particular to a liquid type recognition method based on a convolutional neural network model. Background Art

[0002] In the field of modern safety inspection, liquid detection is an important task, especially in public places such as airports and stations, where rapid and accurate identification of liquid dangerous goods is particularly important. Traditional methods for detecting liquid dangerous goods mainly rely on manual detection or detection equipment based on specific physical properties, such as measuring the dielectric constant and density of liquids. These methods have certain limitations in practical applications, such as slow detection speed and inability to adapt to a variety of liquid types.

[0003] With the development of science and technology, microwave signal detection technology has been widely used in the field of liquid detection due to its non-invasive and fast response characteristics. This technology determines the type of liquid by analyzing the liquid's absorption and reflection characteristics of microwave signals.

[0004] However, the existing technology still has the problem of low accuracy in liquid type recognition and high false alarm rate when using microwave signals for liquid detection. Summary of the invention

[0005] The present invention provides a liquid type recognition method based on a convolutional neural network model, which is used to solve the defects of low accuracy of liquid type recognition during liquid detection in the prior art, resulting in a high false alarm rate, and achieves the purpose of improving the accuracy of liquid type recognition and reducing the false alarm rate.

[0006] The present invention provides a liquid type recognition method based on a convolutional neural network model, comprising: Acquire a target microwave signal of a target container containing a liquid to be tested; Acquire a target empty microwave signal of the target container without liquid; Inputting the difference between the target microwave signal and the target air-measured microwave signal into a convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, wherein the target information includes the liquid type of the liquid to be tested; Among them, the convolutional neural network model is trained based on the sample difference corresponding to the sample liquid and the liquid type label of the sample liquid, and the sample difference is the difference between the sample microwave signal of the sample container containing the sample liquid and the sample target empty microwave signal of the sample container without liquid.

[0007] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the step of obtaining a target empty microwave signal of the target container without liquid includes: Acquire a target image of the target container; determining target attribute information of the target container based on the target image; Based on the correspondence between the attribute information and the air-detected microwave signal, the target air-detected microwave signal corresponding to the target attribute information is determined.

[0008] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the target information also includes the capacity of the liquid to be tested and the attributes of the target container, and the attributes of the target container include shape and / or material.

[0009] According to a liquid type recognition method based on a convolutional neural network model provided by the present invention, the convolutional neural network model is trained based on the following method: For each sample liquid, obtaining sample microwave signals of different sample containers containing the sample liquid; Acquire a sample empty microwave signal of each sample container that does not contain liquid; Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model; determining loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid; Based on the loss information, the model parameters of the initial convolutional neural network model are adjusted to obtain the convolutional neural network model.

[0010] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container is input into an initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model, including: Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model, and obtaining the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model; The determining of loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid includes: determining first loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid; determining second loss information based on the predicted volume of the sample liquid and the volume label of the sample liquid; The loss information is determined based on the first loss information and the second loss information.

[0011] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container is input into an initial convolutional neural network model to obtain the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model, including: Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model, and obtaining the predicted liquid type, predicted capacity and predicted properties of the sample liquid output by the initial convolutional neural network model; The determining the loss information based on the first loss information and the second loss information includes: determining third loss information based on the predicted attribute of the sample container and the attribute label of the sample container; The loss information is determined based on the first loss information, the second loss information, and the third loss information.

[0012] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the loss information is determined based on the first loss information, the second loss information and the third loss information, including: Respectively obtain a first weight corresponding to the liquid type, a second weight corresponding to the capacity of the liquid, and a third weight corresponding to the attribute of the container; The loss information is determined based on the first loss information, the second loss information, the third loss information, the first weight, the second weight, and the third weight.

[0013] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the step of obtaining a target microwave signal of a target container containing a liquid to be tested comprises: Acquire an initial microwave signal of a target container containing a liquid to be tested; The target microwave signal is determined based on a signal between a first trough and a preset number of peaks in the initial microwave signal.

[0014] According to a liquid type identification method based on a convolutional neural network model provided by the present invention, the target microwave signal is determined based on a signal between a first trough and a preset number of peaks in the initial microwave signal, comprising: Performing a preset number of cubic spline interpolations on the signal between the first trough and the preset number of peaks in the initial microwave signal to obtain each spline difference data; The target microwave signal is determined based on an average value of corresponding data in each of the spline difference data.

[0015] According to a liquid type recognition method based on a convolutional neural network model provided by the present invention, the method further includes: In the case that the probability values ​​of the two liquid types to be tested output by the convolutional neural network model are both greater than the preset values, and the difference between the probability values ​​of the two liquid types to be tested is less than the preset difference, for each of the probability values, based on the liquid type represented by the probability value, the capacity of the liquid to be tested and the properties of the target container, a preset microwave signal of the liquid type represented by the probability value is obtained; respectively determining the similarity between the target microwave signal and each of the preset microwave signals; The liquid type to which the preset microwave signal corresponding to the maximum similarity belongs is determined as the liquid type of the liquid to be tested.

[0016] The present invention also provides a liquid type identification device based on a convolutional neural network model, comprising: An acquisition module, used for acquiring a target microwave signal of a target container containing a liquid to be tested; The acquisition module is further used to acquire the target empty microwave signal of the target container without liquid; An input module, used for inputting the difference between the target microwave signal and the target air-measured microwave signal into a convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, wherein the target information includes the liquid type of the liquid to be tested; Among them, the convolutional neural network model is trained based on the sample difference corresponding to the sample liquid and the liquid type label of the sample liquid, and the sample difference is the difference between the sample microwave signal of the sample container containing the sample liquid and the sample target empty microwave signal of the sample container without liquid.

[0017] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for identifying the type of liquid based on the convolutional neural network model as described above is implemented.

[0018] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a liquid type identification method based on a convolutional neural network model as described in any one of the above.

[0019] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described methods for identifying liquid types based on a convolutional neural network model.

[0020] The liquid type identification method based on the convolutional neural network model provided by the present invention obtains the target microwave signal of the target container containing the liquid to be tested, and obtains the target empty-measured microwave signal of the target container without the liquid, and obtains the target information of the liquid to be tested output by the convolutional neural network model by determining the difference between the target microwave signal and the target empty-measured microwave signal, and inputting the difference into the convolutional neural network model, thereby obtaining the target information of the liquid to be tested output by the convolutional neural network model, and the target information includes the liquid type of the liquid to be tested. Since the determined difference between the target microwave signal and the target empty-measured microwave signal can characterize the microwave signal corresponding to the liquid to be tested, that is, the influence of the target container can be excluded, so that after the difference is input into the convolutional neural network model, the liquid type of the liquid to be tested is obtained to be more accurate, thereby improving the accuracy of the liquid type identification result and reducing the false alarm rate during the security inspection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic flow chart of a method for identifying liquid types based on a convolutional neural network model provided in an embodiment of the present invention.

[0023] Figure 2 A schematic diagram of an initial microwave signal provided by an embodiment of the present invention.

[0024] Figure 3 A schematic diagram of a target microwave signal after cubic spline interpolation provided in an embodiment of the present invention.

[0025] Figure 4 A schematic diagram of an initial sample empty-measured microwave signal provided by an embodiment of the present invention.

[0026] Figure 5 A schematic diagram of a sample space-measured microwave signal after cubic spline interpolation provided in an embodiment of the present invention.

[0027] Figure 6 A schematic diagram of the structure of a liquid type recognition device based on a convolutional neural network model provided in an embodiment of the present invention.

[0028] Figure 7A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] In the field of liquid detection, microwave signal analysis technology relies on the absorption and reflection characteristics of liquids to microwaves, and identifies the type of liquid by analyzing these characteristics. In the prior art, microwave signals are usually emitted to the target container containing the liquid to be detected, and then the microwave signals reflected by the liquid to be detected are received, so that the type of liquid is identified based on the emitted microwave signals and the received reflected microwave signals. However, the size, material and shape of the target container containing the liquid to be detected will affect the microwave signal. Therefore, if the liquid type is identified directly based on the above-mentioned reflected microwave signal, the identified liquid type will be wrong, which reduces the accuracy of liquid type identification and may cause false alarms.

[0031] In the embodiment of the present invention, taking the above problems into consideration, a liquid type recognition method based on a convolutional neural network model is proposed. In this method, since the dielectric constant of each liquid type is different, when monitoring is performed by multi-frequency microwave technology, the returned microwave signal will be different. Therefore, the liquid type can be detected based on the microwave signal. During the detection process, after obtaining the target microwave signal of the target container containing the liquid to be tested, it is also necessary to obtain the target empty-measured microwave signal of the target container without liquid. By determining the difference between the target microwave signal and the target empty-measured microwave signal, the microwave signal corresponding to the liquid to be tested can be obtained, that is, the influence of the target container can be eliminated. In this way, after the difference is input into the convolutional neural network model, the liquid type of the liquid to be tested is obtained to be more accurate, which improves the accuracy of the liquid type recognition result and reduces the false alarm rate during the security inspection process.

[0032] Combine the following Figures 1 to 5The liquid type recognition method based on the convolutional neural network model provided in an embodiment of the present invention is described. The execution subject of the method can be an electronic device such as a security inspection machine, a computer or a server, or a specially designed intelligent device, or a liquid type recognition device based on the convolutional neural network model arranged in the electronic device or the intelligent device. The liquid type recognition device based on the convolutional neural network model can be implemented by software, hardware or a combination of the two. The method can be applied to various scenarios where security checks are required, such as bus stations, railway stations, subway stations or high-speed railway stations.

[0033] Figure 1 A flow chart of a liquid type recognition method based on a convolutional neural network model provided in an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method includes: Step 101: Acquire a target microwave signal of a target container containing a liquid to be tested.

[0034] In this step, in the security inspection scene, passengers often carry liquids, so it is necessary to identify the liquid type of the liquid to prevent the situation of carrying dangerous liquids on board. In order to identify the liquid type of the liquid to be detected, it is first necessary to obtain the target microwave signal of the target container containing the liquid to be detected. Exemplarily, the target microwave signal can be obtained by a microwave sensor module provided in an electronic device, wherein the microwave sensor module can emit microwave signals to the target container containing the liquid to be detected, and these microwave signals will be absorbed and reflected after passing through the liquid to generate a specific waveform. It should be understood that the obtained target microwave signal includes physical property information such as the dielectric constant of the liquid. The dielectric constants of different types of liquids are different, and the frequency of the reflected microwave signal will change. In addition, the size, material and shape of the target container will also have a certain influence on the reflected microwave signal.

[0035] Step 102: Acquire a target empty microwave signal of a target container without liquid.

[0036] In this step, during the liquid type identification process, the material, size and shape of the target container itself will affect the propagation and reflection of the microwave signal. In order to accurately identify the liquid type, it is necessary to first obtain the effect of the target container on the microwave signal when there is no liquid, that is, the target empty microwave signal.

[0037] Among them, an air-measurement microwave signal library can be established in advance, which stores air-measurement microwave signals corresponding to containers of different materials, shapes and sizes. These air-measurement microwave signals are obtained by placing an empty container within the detection range of the microwave sensor module, and after the microwave sensor module transmits microwave signals to the container, these microwave signals are reflected back after passing through the empty container.

[0038] During security inspection, an image of a target container containing the liquid to be tested can be collected at the same time. The material, size and shape of the target container can be obtained by analyzing the image, and then the target air-detection microwave signal corresponding to the target container without liquid can be obtained by searching in a pre-established air-detection microwave signal library.

[0039] Step 103: input the difference between the target microwave signal and the target air-measured microwave signal into the convolutional neural network model to obtain the target information of the liquid to be measured output by the convolutional neural network model, where the target information includes the liquid type of the liquid to be measured.

[0040] Among them, the convolutional neural network model is trained based on the sample difference corresponding to the sample liquid and the liquid type label of the sample liquid. The sample difference is the difference between the sample microwave signal of the sample container containing the sample liquid and the sample target empty microwave signal of the sample container without liquid.

[0041] Specifically, after obtaining the target microwave signal and the target air-detected microwave signal, the difference between the two signals can be determined, and the target microwave signal can be calibrated by the difference. The difference can reflect the influence of the liquid to be detected on the microwave signal, while eliminating the interference of the target container itself on the microwave signal.

[0042] In order to make the obtained difference suitable for the convolutional neural network model, the difference can be further preprocessed, such as data normalization and standardization, and the preprocessed difference is input into the convolutional neural network model, so as to obtain the target information of the liquid to be tested output by the convolutional neural network model, and the target information includes the liquid type of the liquid to be tested. The liquid type may include alcohol, gasoline, acetone, rosin, potassium hydroxide, sulfuric acid or water.

[0043] The above-mentioned Convolutional Neural Network (CNN) model is a deep learning model, which may contain multiple convolutional layers, pooling layers, fully connected layers, and activation functions. The first convolutional layer has 32 filters, each with a filter size of 3. The ReLU (Rectified Linear Unit) activation function is used. The first maximum pooling layer has a pooling size of 2. The function of the pooling layer is to reduce the dimension of the data and extract features. This layer receives the output of the first convolutional layer. The second convolutional layer has 64 filters, each with a filter size of 3, and also uses the ReLU activation function. This layer receives input from the previous pooling layer, and the second maximum pooling layer has a pooling size of 2. This layer receives the output of the second convolutional layer. The third convolutional layer has 128 filters, each with a filter size of 3, and also uses the ReLU activation function. This layer receives the output of the previous pooling layer. The pooling size of the third maximum pooling layer is 2. This layer receives the output of the third convolutional layer. The convolutional neural network model is obtained by training the initial convolutional neural network model by obtaining sample differences corresponding to different sample liquids and pre-labeled liquid type labels of each sample liquid. The sample difference is the difference between the sample microwave signal of the sample container containing the sample liquid and the sample target empty microwave signal of the sample container without liquid.

[0044] The liquid type identification method based on the convolutional neural network model provided in the embodiment of the present invention obtains the target microwave signal of the target container containing the liquid to be tested, and obtains the target empty-measured microwave signal of the target container without the liquid, and obtains the target information of the liquid to be tested output by the convolutional neural network model by determining the difference between the target microwave signal and the target empty-measured microwave signal, and inputting the difference into the convolutional neural network model, thereby obtaining the target information of the liquid to be tested output by the convolutional neural network model, and the target information includes the liquid type of the liquid to be tested. Since the difference between the determined target microwave signal and the target empty-measured microwave signal can characterize the microwave signal corresponding to the liquid to be tested, that is, the influence of the target container can be excluded, so that after the difference is input into the convolutional neural network model, the liquid type of the liquid to be tested is obtained to be more accurate, thereby improving the accuracy of the liquid type identification result and reducing the false alarm rate during the security inspection process.

[0045] Illustratively, based on the above embodiments, the target information further includes the capacity of the liquid to be tested and the properties of the target container, and the properties of the target container include shape and / or material.

[0046] Specifically, in the security inspection scenario, when the liquid to be tested is a dangerous liquid, it is very important to further determine the capacity of the liquid to be tested. Therefore, when training the convolutional neural network model, the capacity label of the sample liquid and the attribute label of the sample container can also be added. Therefore, after the difference between the target microwave signal and the target air-measured microwave signal is input into the trained convolutional neural network model, the convolutional neural network model can perform multi-task processing, thereby obtaining the capacity of the liquid to be tested and the attributes of the target container output by the convolutional neural network model, wherein the attributes of the target container include the shape and / or material of the target container.

[0047] Through the above method, the liquid to be tested can be identified from multiple dimensions, thereby improving the comprehensiveness and accuracy of the identification of the liquid to be tested.

[0048] Exemplarily, based on the above embodiment, when obtaining a target air-detected microwave signal of a target container without liquid, a target image of the target container can be obtained, and target attribute information of the target container can be determined based on the target image. Based on the correspondence between the attribute information and the air-detected microwave signal, the target air-detected microwave signal corresponding to the target attribute information can be determined.

[0049] Specifically, the correspondence between attribute information and air-measured microwave signals can be established in advance. For example, the air-measured microwave signals corresponding to containers of different shapes, sizes and materials can be obtained in advance, and the correspondence between the attribute information of different containers and the corresponding air-measured microwave signals can be stored.

[0050] During security inspection, the target image of the target container can be obtained through the image acquisition device, and the target attribute information of the target container can be determined from the target image through the computer vision algorithm. The target attribute information may include the size, material and shape of the target container, which may affect the microwave signal. For example, the target attribute information of the target container can be determined from the target image as a plastic bottle with threads.

[0051] After the target attribute information of the target container is obtained, based on the pre-established correspondence between the attribute information and the air-detection microwave signal, the target air-detection microwave signal corresponding to the target attribute information can be obtained.

[0052] In this embodiment, by pre-establishing the corresponding relationship between the attribute information and the air-detected microwave signal, after obtaining the target attribute information of the target container by image recognition, the target air-detected microwave signal can be accurately determined based on the corresponding relationship. In addition, in this way, the target air-detected microwave signal can be obtained without pouring out the liquid to be detected in the target container, thereby improving the efficiency and convenience of determining the target air-detected microwave signal.

[0053] Exemplarily, based on the above embodiment, the convolutional neural network model is trained based on the following method: For each sample liquid, sample microwave signals of different sample containers containing sample liquid are obtained, and sample empty-measured microwave signals of each sample container without liquid are obtained, and the sample difference between the sample microwave signal corresponding to each sample container and the sample empty-measured microwave signal is input into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model, and based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid, the loss information is determined, thereby adjusting the model parameters of the initial convolutional neural network model based on the loss information to obtain the convolutional neural network model.

[0054] Specifically, the sample liquid may include water, alcohol, acetone, rosin, acetonitrile, various fuels, and inorganic substances with strong ionization properties, etc. For each sample liquid, different sample containers may be used to hold it, and sample microwave signals of different sample containers containing the sample liquid are obtained. In addition, for each sample container, it is also necessary to obtain a sample empty microwave signal of the sample container when no liquid is contained in it.

[0055] For each sample liquid, it is necessary to calculate the difference between the sample microwave signal corresponding to the sample liquid and the sample empty microwave signal corresponding to each sample container. By calculating the difference, the interference of the sample container on the microwave signal can be eliminated, and the obtained difference is input into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model.

[0056] The predicted liquid type of the sample liquid is compared with the pre-labeled liquid type label of the sample liquid. After obtaining the loss information, the model parameters of the initial convolutional neural network model are adjusted based on the loss information. The above process is repeated until the obtained model converges or the number of iterations exceeds the preset number, and the final model is determined as the convolutional neural network model.

[0057] In this embodiment, by determining the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container, the interference of the sample container itself on the microwave signal can be eliminated. Therefore, when the model is trained based on the sample difference, the liquid type predicted by the final convolutional neural network model is more accurate.

[0058] Exemplarily, on the basis of the above-mentioned embodiments, when the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container is input into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model, the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container is input into the initial convolutional neural network model to obtain the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model. Correspondingly, the loss information is determined based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid, and the first loss information can be determined based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid, and the second loss information can be determined based on the predicted capacity of the sample liquid and the capacity label of the sample liquid, so that the loss information can be determined based on the first loss information and the second loss information.

[0059] Specifically, the initial convolutional neural network model can be a multi-task model. After the sample difference is input into the initial convolutional neural network model of the multi-task, the initial convolutional neural network model can simultaneously output the predicted liquid type and predicted capacity of the sample liquid. When the model is trained in this scenario, the loss information involved in the model training process can include two parts, namely, the first loss information obtained by comparing the predicted liquid type with the liquid type label of the pre-labeled sample liquid, and the second loss information obtained by comparing the predicted capacity with the capacity label of the pre-labeled sample liquid. Among them, the first loss information can be used to measure the accuracy of the model in liquid type prediction, which is the loss information determined when considering the liquid type prediction alone, and the second loss information can be used to measure the accuracy of the model in capacity prediction, which is the loss information determined when considering the capacity prediction alone.

[0060] The first loss information and the second loss information are combined as loss information, which is the target that needs to be minimized during the model training process. It can comprehensively consider the performance of the convolutional neural network model in liquid type prediction and capacity prediction.

[0061] In this embodiment, in addition to outputting the predicted liquid type, the initial convolutional neural network model can also output the predicted capacity. The total loss information is jointly determined by the first loss information determined based on the predicted liquid type and the liquid type label and the second loss information determined based on the predicted capacity and the capacity label. This allows the trained convolutional neural network model to predict both the liquid type and capacity at the same time, thereby improving the applicability and flexibility of the convolutional neural network model.

[0062] Exemplarily, on the basis of the above-mentioned embodiments, when the sample difference between the sample microwave signal and the sample empty-measured microwave signal corresponding to each sample container is input into the initial convolutional neural network model to obtain the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model, the sample difference between the sample microwave signal and the sample empty-measured microwave signal corresponding to each sample container can also be input into the initial convolutional neural network model to obtain the predicted liquid type, predicted capacity and predicted attributes of the sample liquid output by the initial convolutional neural network model; accordingly, when the loss information is determined based on the first loss information and the second loss information, the third loss information is determined based on the predicted attributes of the sample container and the attribute label of the sample container, and the loss information is determined based on the first loss information, the second loss information and the third loss information.

[0063] Specifically, after the sample difference is input into the initial convolutional neural network model of the multi-task, the initial convolutional neural network model can also simultaneously output the predicted liquid type, predicted capacity and predicted attributes of the sample liquid container. When the model is trained in this scenario, the loss information involved in the model training process can include three parts, namely, the first loss information obtained by comparing the predicted liquid type with the liquid type label of the pre-labeled sample liquid, the second loss information obtained by comparing the predicted capacity with the capacity label of the pre-labeled sample liquid, and the third loss information obtained by comparing the predicted attribute of the sample container with the attribute label of the pre-labeled sample container. Among them, the first loss information can be used to measure the accuracy of the model in liquid type prediction, which is the loss information determined when the liquid type prediction is considered alone, the second loss information can be used to measure the accuracy of the model in capacity prediction, which is the loss information determined when the capacity prediction is considered alone, and the third loss information can be used to measure the accuracy of the model in container attribute prediction, which is the loss information determined when the container attribute prediction is considered alone.

[0064] The first loss information, the second loss information and the third loss information are combined as the final loss information. The loss information is the target that needs to be minimized during the model training process, which can comprehensively consider the performance of the convolutional neural network model in three aspects: liquid type prediction, capacity prediction and container property prediction.

[0065] In this embodiment, in addition to outputting the predicted liquid type and predicted capacity, the initial convolutional neural network model can also output the predicted attributes of the target container, and jointly determine the total loss information based on the first loss information, the second loss information, and the third loss information, so that the trained convolutional neural network model can simultaneously predict the liquid type, capacity, and container attributes, thereby improving the applicability and flexibility of the convolutional neural network model.

[0066] Exemplarily, based on the above-mentioned embodiments, when determining the loss information based on the first loss information, the second loss information and the third loss information, the first weight corresponding to the liquid type, the second weight corresponding to the capacity of the liquid and the third weight corresponding to the attribute of the container may be respectively obtained, and the loss information may be determined based on the first loss information, the second loss information, the third loss information, the first weight, the second weight and the third weight.

[0067] Specifically, the first weight, the second weight, and the third weight can reflect the relative importance of different prediction tasks in the overall model training process, which can be set according to actual conditions or experience. For example, when the importance of liquid type recognition is higher than that of target container attribute recognition, the first weight will be increased.

[0068] The first loss information, the second loss information and the third loss information are multiplied by their respective corresponding weights to obtain weighted loss information, and the sum of the weighted loss information is determined as the final loss information.

[0069] In this embodiment, by setting weights corresponding to different loss information, the training process of the convolutional neural network model can be more flexibly adapted to the importance of different tasks, thereby improving the overall performance of the convolutional neural network model.

[0070] Illustratively, based on the above embodiments, when obtaining the target microwave signal of the target container containing the liquid to be tested, the target microwave signal can be determined by obtaining the initial microwave signal of the target container containing the liquid to be tested, and based on the signal between the first trough and a preset number of peaks in the initial microwave signal.

[0071] Specifically, Figure 2 A schematic diagram of an initial microwave signal provided by an embodiment of the present invention, such as Figure 2 As shown, the target container containing the liquid to be tested is placed within the detection range of the microwave sensing module. The microwave sensing module will receive microwave signals that have been affected by the liquid to be tested and the target container. These signals will be converted into electrical signals to form initial microwave signals. The initial microwave signal is processed to generate a detection data point image. Due to the nature of the microwave sensing module itself, there will be only some obvious peaks in the detection data point image, so the signal between the first trough and the preset number of peaks in the detection data point image is intercepted, and these signals are used as target microwave signals. Among them, the preset number can be set according to the actual situation, for example, it can be set to 6. According to experiments, 6 obvious peaks can reflect the characteristic information of the liquid to be tested.

[0072] In this embodiment, after the initial microwave signal of the target container containing the liquid to be tested is obtained, the signal between the first trough and a preset number of peaks in the initial microwave signal is determined as the target microwave signal. Since the signal between the first trough and the preset number of peaks can already characterize the characteristic information of the liquid to be tested, only the signal between the first trough and the preset number of peaks is used as the target microwave signal for subsequent processing, which can greatly reduce the amount of calculation and improve the efficiency of liquid type recognition.

[0073] Exemplarily, based on the above embodiments, when determining the target microwave signal based on the signal between the first trough and a preset number of peaks in the initial microwave signal, cubic spline interpolation can be performed a preset number of times on the signal between the first trough and a preset number of peaks in the initial microwave signal to obtain each spline difference data, and the target microwave signal is determined based on the average value of the corresponding data in each spline difference data.

[0074] Specifically, after selecting the signal between the first trough and the preset number of peaks from the initial microwave signal, since the convolutional neural network model requires the input feature points to be consistent when predicting the liquid type, and the number of signals between the first trough and the preset number of peaks in the initial microwave signal is not exactly the same, roughly 100-160 points, it is necessary to interpolate the signal between the first trough and the preset number of peaks. In order to keep the original waveform unchanged as much as possible, cubic spline interpolation can be selected.

[0075] Figure 3 A schematic diagram of a target microwave signal after cubic spline interpolation provided by an embodiment of the present invention, such as Figure 3 As shown, the signal between the first trough and the preset number of peaks is interpolated to 200 points by cubic spline for a preset number of times, where, assuming that the preset number of times is 3, the data obtained after the preset number of cubic spline interpolations are: b1=[c1, c2, ......., c200], b2=[d1, d2, ......., d200], b3=[e1, e2, ......., e200]. Since the waveform will fluctuate, in order to minimize this fluctuation, the average value of the interpolated data can be calculated, and the target microwave signal is finally obtained as a1: a1=[fn], fn=(cn+dn+e2) / 3.

[0076] In this embodiment, by performing a preset number of cubic spline interpolations on the signal between the first trough and a preset number of peaks in the initial microwave signal, and determining the target microwave signal based on the average value of the corresponding data in each spline interpolation data, the amount of data in the target microwave signal can be made consistent, which can adapt to the input of the convolutional neural network model and avoid the situation where the convolutional neural network model has erroneous prediction results due to inconsistent data amount.

[0077] It should be noted that, in the process of training the convolutional neural network model, the sample microwave signals of different sample containers containing sample liquids and the sample blank microwave signals of each sample container without liquid are obtained based on the same method.

[0078] Specifically, Figure 4 A schematic diagram of an initial sample empty microwave signal provided by an embodiment of the present invention, such as Figure 4 As shown, the signal from the first trough to the sixth peak in the initial sample empty-measured microwave signal is useful information. Therefore, the sample empty-measured microwave signal can be determined based on the signal from the first trough to the sixth peak intercepted from the initial sample empty-measured microwave signal.

[0079] Figure 5 A schematic diagram of a sample space-measured microwave signal after cubic spline interpolation provided by an embodiment of the present invention, such as Figure 5 As shown, the sample empty-measured microwave signal can be determined by performing a preset number of cubic spline interpolations on the signal between the first trough and the preset number of peaks in the initial sample empty-measured microwave signal, and based on the average value of the corresponding data in each spline interpolation data.

[0080] Similarly, after the initial sample microwave signal is acquired, the signal between the first trough and the sixth peak is intercepted from the initial sample microwave signal, and a preset number of cubic spline interpolations are performed to obtain the sample microwave signal.

[0081] Exemplarily, on the basis of the above embodiments, when the convolutional neural network model outputs probability values ​​of the liquid types of the two liquids to be tested that are greater than preset values, and the difference in the probability values ​​of the liquid types of the two liquids to be tested is less than the preset difference, for each probability value, based on the liquid type represented by the probability value, the capacity of the liquid to be tested and the properties of the target container, a preset microwave signal of the liquid type represented by the probability value is obtained, and the similarity between the target microwave signal and each preset microwave signal is determined respectively, and the liquid type to which the preset microwave signal corresponding to the maximum similarity belongs is determined as the liquid type of the liquid to be tested.

[0082] Specifically, preset microwave signals for different liquid types, different capacities, and different containers for holding liquids can be stored in the database in advance, wherein the preset microwave signals are correct signals, which can be used to characterize the characteristics exhibited by liquids of different types and capacities when held in different containers.

[0083] In practical applications, there may be a situation where the characteristics of two liquids are relatively similar. At this time, the probability values ​​of the two liquid types to be tested output by the convolutional neural network model may be greater than the preset values, and the difference between the probability values ​​of the two liquid types may be less than the preset difference. That is, when the probability values ​​are not much different, the liquid type of the liquid to be tested cannot be further determined based on the probability values.

[0084] For these two types of liquids, the preset microwave signals corresponding to the predicted liquid capacity and the properties of the target container for each type of liquid can be obtained from the aforementioned database.

[0085] Furthermore, the collected target microwave signal and the preset microwave signal can be compared to determine the similarity between the target microwave signal and each preset microwave signal, and the liquid type to which the preset microwave signal corresponding to the maximum similarity belongs can be determined as the final liquid type of the liquid to be tested.

[0086] In this embodiment, when the convolutional neural network model outputs probability values ​​of two liquid types that are greater than preset values ​​and the difference between the probability values ​​of the two liquid types is less than the preset value, the preset microwave signals corresponding to the two liquid types will be obtained respectively, and by determining the similarity between the target microwave signal and each preset microwave signal, the liquid type to which the preset microwave signal corresponding to the maximum similarity belongs is determined as the liquid type of the liquid to be tested, and by comparing the target microwave signal with each preset microwave signal, when the liquid type cannot be accurately distinguished, the liquid type of the liquid to be tested can be accurately determined with the help of the preset microwave signal, thereby improving the accuracy of the liquid type.

[0087] The liquid type identification device based on the convolutional neural network model provided by the present invention is described below. The liquid type identification device based on the convolutional neural network model described below and the liquid type identification method based on the convolutional neural network model described above can be referenced to each other.

[0088] Figure 6 A schematic diagram of a liquid type recognition device based on a convolutional neural network model provided in an embodiment of the present invention, referring to Figure 6 As shown, the liquid type identification device 600 based on the convolutional neural network model includes: An acquisition module 11 is used to acquire a target microwave signal of a target container containing a liquid to be tested; The acquisition module 11 is further used to acquire the target empty microwave signal of the target container without liquid; An input module 12 is used to input the difference between the target microwave signal and the target air-measured microwave signal into a convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, wherein the target information includes the liquid type of the liquid to be tested; Among them, the convolutional neural network model is trained based on the sample difference corresponding to the sample liquid and the liquid type label of the sample liquid, and the sample difference is the difference between the sample microwave signal of the sample container containing the sample liquid and the sample target empty microwave signal of the sample container without liquid.

[0089] In an exemplary embodiment, the acquisition module 11 is specifically configured to: Acquire a target image of the target container; determining target attribute information of the target container based on the target image; Based on the correspondence between the attribute information and the air-detected microwave signal, the target air-detected microwave signal corresponding to the target attribute information is determined.

[0090] In an exemplary embodiment, the target information further includes the volume of the liquid to be tested and the properties of the target container, and the properties of the target container include shape and / or material.

[0091] In an exemplary embodiment, the convolutional neural network model is trained based on the following method: For each sample liquid, obtaining sample microwave signals of different sample containers containing the sample liquid; Acquire a sample empty microwave signal of each sample container that does not contain liquid; Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model; determining loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid; Based on the loss information, the model parameters of the initial convolutional neural network model are adjusted to obtain the convolutional neural network model.

[0092] In an exemplary embodiment, the input module 12 is further configured to: Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model, and obtaining the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model; The device further comprises: a determination module, wherein: A determination module is used to determine first loss information based on a predicted liquid type of the sample liquid and a liquid type label of the sample liquid; determine second loss information based on a predicted capacity of the sample liquid and a capacity label of the sample liquid; and determine the loss information based on the first loss information and the second loss information.

[0093] In an exemplary embodiment, the input module 12 is further configured to: Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model, and obtaining the predicted liquid type, predicted capacity and predicted properties of the sample liquid output by the initial convolutional neural network model; A determination module is used to determine the third loss information based on the predicted attribute of the sample container and the attribute label of the sample container; and to determine the loss information based on the first loss information, the second loss information and the third loss information.

[0094] In an exemplary embodiment, the determining module is specifically configured to: Respectively obtain a first weight corresponding to the liquid type, a second weight corresponding to the capacity of the liquid, and a third weight corresponding to the attribute of the container; The loss information is determined based on the first loss information, the second loss information, the third loss information, the first weight, the second weight, and the third weight.

[0095] In an exemplary embodiment, the acquisition module 11 is specifically configured to: Acquire an initial microwave signal of a target container containing a liquid to be tested; The target microwave signal is determined based on a signal between a first trough and a preset number of peaks in the initial microwave signal.

[0096] In an exemplary embodiment, the acquisition module 11 is specifically configured to: Performing a preset number of cubic spline interpolations on the signal between the first trough and the preset number of peaks in the initial microwave signal to obtain each spline difference data; The target microwave signal is determined based on an average value of corresponding data in each of the spline difference data.

[0097] In an exemplary embodiment, the acquisition module 11 is further used to acquire, for each of the probability values, a preset microwave signal of the liquid type represented by the probability value based on the liquid type represented by the probability value, the capacity of the liquid to be tested, and the property of the target container, when the probability values ​​of the liquid types of the two liquids to be tested output by the convolutional neural network model are both greater than a preset value and the difference between the probability values ​​of the liquid types of the two liquids to be tested is less than a preset difference; A determination module, used to respectively determine the similarity between the target microwave signal and each of the preset microwave signals; The determination module is used to determine the liquid type to which the preset microwave signal corresponding to the maximum similarity belongs as the liquid type of the liquid to be tested.

[0098] The liquid type identification device based on the convolutional neural network model of this embodiment can be used to execute the method of any embodiment in the liquid type identification method based on the convolutional neural network model. Its specific implementation process and technical effects are similar to those in the liquid type identification method based on the convolutional neural network model. For details, please refer to the detailed introduction in the liquid type identification method based on the convolutional neural network model, which will not be repeated here.

[0099] Figure 7 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention, such as Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute a liquid type recognition method based on a convolutional neural network model, the method comprising: obtaining a target microwave signal of a target container containing a liquid to be tested; obtaining a target air-measured microwave signal of the target container without a liquid; inputting the difference between the target microwave signal and the target air-measured microwave signal into the convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, the target information including the liquid type of the liquid to be tested; wherein the convolutional neural network model is obtained by training based on a sample difference corresponding to a sample liquid and a liquid type label of the sample liquid, the sample difference being the difference between a sample microwave signal of a sample container containing the sample liquid and a sample target air-measured microwave signal of a sample container without a liquid.

[0100] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0101] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the liquid type identification method based on the convolutional neural network model provided by the above-mentioned methods, and the method includes: obtaining a target microwave signal of a target container containing a liquid to be tested; obtaining a target empty-measured microwave signal of the target container without liquid; inputting the difference between the target microwave signal and the target empty-measured microwave signal into a convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, and the target information includes the liquid type of the liquid to be tested; wherein the convolutional neural network model is trained based on a sample difference corresponding to a sample liquid and a liquid type label of the sample liquid, and the sample difference is the difference between a sample microwave signal of a sample container containing the sample liquid and a sample target empty-measured microwave signal of the sample container without liquid.

[0102] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the liquid type identification method based on the convolutional neural network model provided by the above-mentioned methods, the method comprising: obtaining a target microwave signal of a target container containing the liquid to be tested; obtaining a target empty-measured microwave signal of the target container without liquid; inputting the difference between the target microwave signal and the target empty-measured microwave signal into a convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, the target information including the liquid type of the liquid to be tested; wherein the convolutional neural network model is trained based on a sample difference corresponding to a sample liquid and a liquid type label of the sample liquid, the sample difference being the difference between a sample microwave signal of a sample container containing the sample liquid and a sample target empty-measured microwave signal of the sample container without liquid.

[0103] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A liquid type recognition method based on a convolutional neural network model, characterized in that: include: Acquire a target microwave signal of a target container containing a liquid to be tested; Acquire a target empty microwave signal of the target container without liquid; Inputting the difference between the target microwave signal and the target air-measured microwave signal into a convolutional neural network model to obtain target information of the liquid to be tested output by the convolutional neural network model, wherein the target information includes the liquid type of the liquid to be tested; Among them, the convolutional neural network model is trained based on the sample difference corresponding to the sample liquid and the liquid type label of the sample liquid, and the sample difference is the difference between the sample microwave signal of the sample container containing the sample liquid and the sample target empty microwave signal of the sample container without liquid.

2. The liquid type recognition method based on the convolutional neural network model according to claim 1 is characterized in that: The step of obtaining a target empty microwave signal of the target container without liquid comprises: Acquire a target image of the target container; determining target attribute information of the target container based on the target image; Based on the correspondence between the attribute information and the air-detected microwave signal, the target air-detected microwave signal corresponding to the target attribute information is determined.

3. The liquid type recognition method based on the convolutional neural network model according to claim 1 is characterized in that: The target information also includes the capacity of the liquid to be tested and the properties of the target container, and the properties of the target container include shape and / or material.

4. The liquid type recognition method based on the convolutional neural network model according to any one of claims 1 to 3, characterized in that: The convolutional neural network model is trained based on the following method: For each sample liquid, obtaining sample microwave signals of different sample containers containing the sample liquid; Acquire a sample empty microwave signal of each sample container that does not contain liquid; Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model; determining loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid; Based on the loss information, the model parameters of the initial convolutional neural network model are adjusted to obtain the convolutional neural network model.

5. The liquid type recognition method based on the convolutional neural network model according to claim 4 is characterized in that: The step of inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container into the initial convolutional neural network model to obtain the predicted liquid type of the sample liquid output by the initial convolutional neural network model comprises: Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model, and obtaining the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model; The determining of loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid includes: determining first loss information based on the predicted liquid type of the sample liquid and the liquid type label of the sample liquid; determining second loss information based on the predicted volume of the sample liquid and the volume label of the sample liquid; The loss information is determined based on the first loss information and the second loss information.

6. The liquid type recognition method based on the convolutional neural network model according to claim 5 is characterized in that: The step of inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each sample container into the initial convolutional neural network model to obtain the predicted liquid type and predicted capacity of the sample liquid output by the initial convolutional neural network model comprises: Inputting the sample difference between the sample microwave signal and the sample empty microwave signal corresponding to each of the sample containers into the initial convolutional neural network model, and obtaining the predicted liquid type, predicted capacity and predicted properties of the sample liquid output by the initial convolutional neural network model; The determining the loss information based on the first loss information and the second loss information includes: determining third loss information based on the predicted attribute of the sample container and the attribute label of the sample container; The loss information is determined based on the first loss information, the second loss information, and the third loss information.

7. The liquid type recognition method based on the convolutional neural network model according to claim 6 is characterized in that: The determining the loss information based on the first loss information, the second loss information and the third loss information includes: Respectively obtain a first weight corresponding to the liquid type, a second weight corresponding to the capacity of the liquid, and a third weight corresponding to the attribute of the container; The loss information is determined based on the first loss information, the second loss information, the third loss information, the first weight, the second weight, and the third weight.

8. The liquid type recognition method based on the convolutional neural network model according to claim 1 is characterized in that: The step of obtaining a target microwave signal of a target container containing the liquid to be tested comprises: Acquire an initial microwave signal of a target container containing a liquid to be tested; The target microwave signal is determined based on a signal between a first trough and a preset number of peaks in the initial microwave signal.

9. The liquid type recognition method based on the convolutional neural network model according to claim 8 is characterized in that: The determining the target microwave signal based on a signal between a first trough and a preset number of peaks in the initial microwave signal comprises: Performing a preset number of cubic spline interpolations on the signal between the first trough and the preset number of peaks in the initial microwave signal to obtain each spline difference data; The target microwave signal is determined based on an average value of corresponding data in each of the spline difference data.

10. The liquid type recognition method based on the convolutional neural network model according to claim 3 is characterized in that: The method further comprises: In the case that the probability values ​​of the two liquid types of the liquids to be tested output by the convolutional neural network model are both greater than the preset values, and the difference between the probability values ​​of the two liquid types of the liquids to be tested is less than the preset difference, for each of the probability values, based on the liquid type represented by the probability value, the capacity of the liquid to be tested and the properties of the target container, a preset microwave signal of the liquid type represented by the probability value is obtained; respectively determining the similarity between the target microwave signal and each of the preset microwave signals; The liquid type to which the preset microwave signal corresponding to the maximum similarity belongs is determined as the liquid type of the liquid to be tested.

Images