Odor detection module and odor detection method

Abstract

To provide an odor detection module that can stably and accurately detect odors regardless of the passage of time or environmental factors. [Solution] The odor detection module 10 comprises a sensor element 11 for detecting odors and an odor presentation unit 12 for presenting a reference odor on the surface of the sensor element 11, with the sensor element 11 detecting the reference odor and the odor of the object to be measured. The odor detection module 10 may further include a processing unit 18 that receives the response output of the sensor element 11 and calculates intensity information and / or identification information of the odor of the object to be measured based on a trained model generated by machine learning.

Description

Technical Field

[0001] The present technology relates to an odor detection module and an odor detection method. More specifically, it relates to an odor detection module and an odor detection method that detect an odor to be measured using a sensor element that detects an odor and an odor presentation unit that presents a reference odor.

Background Art

[0002] Conventionally, an odor sensor that can objectively perform the identification or evaluation of a wide range of types of odors that have been subjectively performed using the human sense of smell has been proposed.

[0003] For example, in Patent Document 1, not all of the reference data obtained by measuring a plurality of types of standard odors are necessarily used. Depending on the type or classification of an unknown sample that is the measurement target, an appropriate standard odor for expressing the degree of strength of the unknown odor, etc., is appropriately selected, and an index value representing the degree of strength of the odor is calculated using only the reference data corresponding to the selected standard odor. An odor measuring device has been proposed.

[0004] Furthermore, Patent Document 2 proposes an odor evaluation method for evaluating changes in the odor quality of a target substance using an odor measuring device equipped with m (where m is an integer of 2 or more) odor sensors having different response characteristics from each other, the method comprising: a reference axis acquisition step in which a reference odor is obtained by mixing n (where n is an integer of 2 or more) types of additive odors, each of which have known odor quality, with the target odor, and the concentration of each of the n types of reference odors is adjusted to one or more levels, and measured with the odor measuring device, thereby positioning n evaluation reference axes corresponding to each reference odor in an m-dimensional space formed by the detection outputs from the m odor sensors; and an evaluation step in which a measurement point is located in the m-dimensional space by measuring the target odor, which is the target odor that has changed or may have changed, with the odor measuring device, and information reflecting the change in odor quality from the target odor to the target odor is created based on the positional relationship between the measurement point and the n evaluation reference axes. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2004-093447 [Patent Document 2] Japanese Patent Publication No. 2007-248377 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, it is known that odor sensors used in the technologies described in Patent Documents 1 and 2 suffer from the difficulty of reproducible measurements due to chemical or physical changes in the sensor surface over time or due to environmental factors. Such factors include, for example, instability due to humidity changes and changes in background values ​​during repeated use. Therefore, even if the sensitivity and discriminative ability are high, continuous use is not possible, which has hindered the practical application of odor sensors.

[0007] Therefore, the primary objective of this technology is to provide an odor detection module that can stably and accurately detect odors regardless of time-dependent or environmental factors. Here, "odor" refers to all kinds of smells, including pleasant and unpleasant scents. [Means for solving the problem]

[0008] This technology provides an odor detection module comprising a sensor element for detecting odors and an odor presentation unit that presents a reference odor on the surface of the sensor element, wherein the sensor element detects the reference odor and the odor of the object to be measured. The odor detection module may further include a processing unit that receives the response output of the sensor element and calculates intensity information and / or identification information of the odor to be measured based on a trained model generated by machine learning.

[0009] Furthermore, this technology provides an odor detection method that includes the steps of: presenting a reference odor; measuring the reference odor; measuring the odor of a target; acquiring the measured reference odor and the measured odor of the target; and calculating intensity information and / or identification information of the odor of the target based on a trained model generated by machine learning. [Brief explanation of the drawing]

[0010] [Figure 1] This is a schematic diagram showing an example configuration of an odor detection module according to the first embodiment of this technology. [Figure 2] This is a schematic diagram illustrating an example of operation of the odor detection module according to the first embodiment of this technology. [Figure 3] This is a schematic diagram illustrating an example of operation of the odor detection module according to the first embodiment of this technology. [Figure 4] This flowchart shows an example of an odor detection method according to the first embodiment of this technology. [Figure 5]It is a sequence diagram of smells presented by an olfactory display according to the first embodiment of the present technology. [Figure 6] It is a sequence diagram of smells presented by an olfactory display according to the first embodiment of the present technology. [Figure 7] It is a sequence diagram of smells presented by an olfactory display according to the first embodiment of the present technology. [Figure 8] It is a graph showing a sensorgram from the sequence presented by the olfactory display according to the first embodiment of the present technology. [Figure 9] It is a graph showing a sensorgram from the sequence presented by the olfactory display according to the first embodiment of the present technology. [Figure 10] It is a graph showing a sensorgram from the sequence presented by the olfactory display according to the first embodiment of the present technology. [Figure 11] It is a schematic configuration diagram showing a configuration example of an odor detection module according to the second embodiment of the present technology.

Embodiments for Carrying Out the Invention

[0011] Hereinafter, preferred embodiments for carrying out the present technology will be described with reference to the drawings. The embodiments described below show examples of representative embodiments of the present technology, and any of the embodiments can be combined. Also, the scope of the present technology is not construed narrowly by these. The description will be made in the following order. 1. First Embodiment (1) Configuration Example of Odor Detection Module (2) Operation Example of Odor Detection Module (3) Example (3-1) Sequence of Smells (3-2) Sensorgram 2. Second Embodiment (1) Configuration Example of Odor Detection Module (2) Operation Example of Odor Detection Module

[0012] According to the present technology, it is possible to stably and highly accurately detect odors regardless of time-dependent or environmental factors. Note that the above effects are not necessarily limited, and any of the effects shown in this specification or other effects that can be grasped from this specification may be achieved together with or instead of the above effects.

[0013] 1. First Embodiment (1) Configuration Example of Odor Detection Module Referring to FIG. 1, a configuration example of an odor detection module 10 according to the first embodiment of the present technology will be described. FIG. 1 is a schematic configuration diagram showing a configuration example of the odor detection module 10 according to the present embodiment.

[0014] As shown in FIG. 1, the odor detection module 10 according to the present embodiment includes a multiarray sensor 11, an olfactory display 12 as an odor presentation unit, exhaust mechanisms 13 and 14, a reference air flow inlet 15, and a measurement target air flow inlet 16. Further, the odor detection module 10 includes an AI processing unit 18 connected to the multiarray sensor 11.

[0015] The multiarray sensor 11 has a plurality of sensor elements for detecting odors arranged in an array, and each sensor element can detect different reference odors and the odor of a measurement target (object and / or space). Thus, having a plurality of sensor elements with different response characteristics enables various sensor responses. For example, if the sensor response is only ON / OFF and there is 1 sensor, only 2 patterns can be discriminated, but if there are 10 sensors, 2 to the 1,024th power enables 1,024 patterns to be discriminated.

[0016] Also, the surface of the multiarray sensor 11 is covered during normal times and is exposed during odor measurement. Thereby, contamination of the sensor surface can be prevented. Further, the surface of the unused part of the sensor element of the multiarray sensor 11 is exposed during odor measurement. By exposing a new sensor surface for each measurement, a stable sensor response can be obtained.

[0017] The surface of each sensor element in the multi-array sensor 11 is formed from one of the following materials: polymer materials, inorganic materials, or metallic materials, each having a different nanostructure or chemical properties. These sensitive films enable the identification of a variety of odors. Examples of sensor elements include metal oxide semiconductors, CMOS (Complementary Metal-Oxide-Semiconductor), SAW (Surface Acoustic Wave), SPR (Surface Plasmon Resonance), QCM (Quartz Crystal Microbalance), cantilevers, and piezoelectric elements; there are no particular limitations. Furthermore, the surface sensitive film of the sensor element can include, for example, various organic polymer films, SAM (Self-Assembled Monolayer) films, LB (Langmuir-Blodgett) films, and films containing biopolymers such as GPCR (GTP-binding Protein-Coupled Receptor).

[0018] The olfactory display 12 presents a reference scent on the surface of the multi-array sensor 11. The olfactory display 12 has multiple scent-holding units 17 that hold fragrances that constitute the reference scent. Each scent-holding unit 17 contains multiple fragrances in an interchangeable manner. Furthermore, the olfactory display 12 can independently control the multiple fragrances and change them over time to present the reference scent.

[0019] Exhaust mechanisms 13 and 14 exhaust air or airflow flowing near the surface of the multi-array sensor 11 to the outside. By exhausting the air in this way, odor mixing between the reference odor and the odor being measured can be eliminated. Exhaust mechanisms 13 and 14 may include, for example, blowers, fans, or compressed air dischargers.

[0020] The reference airflow inlet 15 sends the airflow of a reference odor emitted from the olfactory display 12 to the multi-array sensor 11.

[0021] The airflow inlet 16 to be measured is formed to be openable and closable, and when opened, it sends the odor of the target to be measured to the multi-array sensor 11.

[0022] The AI ​​processing unit 18 receives a reference measurement value, which is a sensorgram measuring a reference odor, and a target measurement value, which is a sensorgram measuring the odor of the target, and outputs intensity information and / or type of the odor of the target. For example, the AI ​​processing unit 18 receives a sensorgram obtained from a measurement, which is the response output from the multi-array sensor 11, and, based on a trained model generated by machine learning, calculates intensity information and / or identification information of the odor of the target using artificial intelligence (AI) processing and outputs it externally.

[0023] The AI ​​processing unit 18 can, for example, use AI software for intensity and discrimination calculations. The AI ​​algorithms of the AI ​​processing unit 18 can include, for example, various neural networks such as Deep Learning and GNN (graph neural network), SVM (support vector machine), SOM (self-organizing maps), KNN (k-nearest neighbor), Random Forest, and PCA (principal component analysis).

[0024] (2) Example of operation of the odor detection module Next, with reference to Figures 2 to 4, examples of odor detection operation by the odor detection module 10 will be described. Figure 2 is a schematic configuration diagram illustrating an example of operation in which the olfactory display 12 of the odor detection module 10 presents a reference odor to the multi-array sensor 11. Figure 3 is a schematic configuration diagram illustrating an example of operation in which the multi-array sensor 11 of the odor detection module 10 measures the odor of the target. Figure 4 is a flowchart illustrating an example of an odor detection method by the odor detection module 10.

[0025] First, an example of odor detection operation by the odor detection module 10 will be explained using Figures 2 and 3. As shown in Figure 2, as an example, the olfactory display 12 is initially placed above the reference airflow inlet 15, and the pre-measurement mode is started. When the pre-measurement mode is started, the odor retention section 17 of the mounted olfactory display 12 sequentially presents multiple patterns of reference odors to the multi-array sensor 11.

[0026] Next, as shown in Figure 3, immediately after the pre-measurement mode measurement, the system switches to the target measurement mode and opens the target airflow inlet 16. When the target airflow inlet 16 is opened, the odor of the target object and / or the target space is taken in from the target airflow inlet 16 to the multi-array sensor 11, and the multi-array sensor 11 measures the odor of the target. Once the multi-array sensor 11 measures the odor of the target, the sensorgram is input to the AI ​​processing unit 18.

[0027] Here, the artificial intelligence (AI) in the AI ​​processing unit 18 is trained in advance by presenting a large number of odor sequence patterns and known intensity and type odors. Subsequently, when a sensorgram is input to the AI ​​processing unit 18, the AI, which has been trained in advance by presenting a large number of odor sequence patterns and known intensity and type odors, is input the sensorgrams of each sequence during the pre-reference odor measurement and the sensorgram of the measurement target, and the intensity information and / or odor identification information (odor identification) of the measurement target is calculated.

[0028] Next, an example of an odor detection method (sensing) using the odor detection module 10 will be explained using Figure 4.

[0029] In step S1, the olfactory display 12 creates a presentation sequence of the odor library. In this step, the pre-measurement of known reference odors involves combining multiple reference odors and changing them over time.

[0030] In step S2, the olfactory display 12 sequences the presentation sequence of the odor library.

[0031] In step S3, the multi-array sensor 11 detects a reference odor.

[0032] In step S4, the multi-array sensor 11 exhausts a reference odor airflow from the exhaust mechanisms 13 and 14 and outputs a reference odor sensorgram to the AI ​​processing unit 18. The AI ​​processing unit 18 inputs the output sensorgram. Then, the process returns to step S1 and repeats steps S1 to S4 as many times as necessary. After that, the process proceeds to step S5.

[0033] Next, in step S5, the odor detection module 10 switches the inlet to the multi-array sensor 11 from the olfactory display 12 to the airflow inlet 16 to be measured.

[0034] In step S6, the multi-array sensor 11 detects the odor of the object to be measured.

[0035] In step S7, the multi-array sensor 11 exhausts the airflow of the odor to be measured from the exhaust mechanisms 13 and 14 and outputs a sensorgram of the odor to be measured to the AI ​​processing unit 18. The AI ​​processing unit 18 receives the output sensorgram as input.

[0036] In step S8, the AI ​​processing unit 18 performs AI-based pattern analysis from the sensorgrams of the pre-measured reference odor and the target odor. At this time, the built-in pattern matching AI calculates intensity information and identification information using the pre-measurement results of the reference odor and the target odor as input.

[0037] In step S9, the AI ​​processing unit 18 outputs the intensity information of the analyzed measurement target and identification information such as the type of odor to the outside.

[0038] As described above, the odor detection module 10 according to this embodiment is a module that integrates a multi-array sensor 11, which is an array of multiple sensor elements, and an olfactory display 12 that holds multiple reference odors. With this configuration, the odor detection module 10 can change the configuration of the multiple reference odors held in the mounted olfactory display 12 over time and present the aroma on the surface of the multi-array sensor 11, thereby calibrating the intensity and identification of the sensor signal of the object to be measured immediately afterward. As a result, the odor detection module 10 can stably and accurately detect and identify odors in the air regardless of temporal or environmental factors, including deterioration and climate.

[0039] Furthermore, since the odor detection module 10 compactly integrates the multi-array sensor 11 and the olfactory display 12, it can be mounted on portable devices such as mobile devices and wearable devices. Therefore, the odor detection module 10 is easy to carry and can measure odors in locations that suit the user's needs.

[0040] In this embodiment, a reference odor is measured before measuring the odor of the target odor. However, the reference odor may be measured simultaneously with the odor of the target odor, or after the odor of the target odor. Furthermore, the olfactory display 12 can present a reference odor according to the surrounding environment, such as temperature and humidity, and can also present a reference odor corrected according to the number of times it has been used.

[0041] (3) Examples Next, with reference to Figures 5 to 10, an example of odor detection using the odor detection module 10 according to this embodiment will be described. Figures 5 to 7 are sequence diagrams of odors presented by the olfactory display 12. Figures 5 to 7 show that the mix of fragrances presented changes over time. Figures 8 to 10 are graphs showing sensorgrams from each sequence shown in Figures 5 to 7. Figures 8 to 10 represent the sensor's response.

[0042] (3-1) Odor sequences Figure 5 shows how the olfactory display 12 holds eight types of standard odors A through H and presents them to the multi-array sensor 11 over time (Sequence 1). As shown in Figure 5, in Sequence 1, initially only odor A is presented, and after a predetermined time has elapsed, odor H is presented while continuing to present odor A. Next, the presentation of odor A is stopped, and while continuing to present odor H, the odor of the target to be measured, odor C, and odor F are presented. After that, the presentation of the odor of the target to be measured, odor C, and odor F is stopped, and only the presentation of odor H is continued.

[0043] Figure 6, similar to Figure 5, shows how the olfactory display 12 presents aromas to the multi-array sensor 11 (Sequence 2). As shown in Figure 6, in Sequence 2, initially no odors are presented, and after a predetermined time, odors E and F are presented. Next, while continuing to present odors E and F, odors B and C are presented. After that, only the presentation of odor C is stopped, and the presentation of odors B, E, and F continues, and after a predetermined time, the presentation of odors B, E, and F is also stopped. After that, only the odor to be measured is presented.

[0044] Figure 7, like Figures 5 and 6, shows how the olfactory display 12 presents aromas to the multi-array sensor 11 (sequence 3). As shown in Figure 7, in sequence 3, first only the odor of the target to be measured is presented. Next, the presentation of the target odor is stopped, and odors A and E are presented. Next, the presentation of odors A and E is stopped, and odors C and G are presented. After that, the presentation of odor C is stopped, and while continuing to present odor G, odor E is presented. After a predetermined time has elapsed, the presentation of odor G is stopped, and only odor E is presented.

[0045] (3-2) Sensorgram Figure 8 shows sensorgram 1 from sequence 1 shown in Figure 5. The horizontal axis in Figure 8 represents time (seconds), and the vertical axis represents response (current change %). As shown in Figure 8, in sensorgram 1, for example, the responses of Sensor 7 and Sensor 8 increase as time progresses, while the response of Sensor 6 remains relatively small. Overall, the variation in the responses of Sensors 1 through 10 is small. Thus, Figure 8 illustrates the response of each sensor as its sensorgram response subtly changes due to a change in the fragrance mix it is exposed to. In Figure 8, for example, the timing of an extreme change in the fragrance mix is ​​shown as a line graph.

[0046] Figure 9, like Figure 8, shows sensorgram 2 from sequence 2 shown in Figure 6. As shown in Figure 9, in sensorgram 2, for example, the responses of Sensor 8 and Sensor 9 increase over time, while the response of Sensor 1 remains relatively small. Overall, the variability in the responses of Sensors 1 through 10 is slightly larger than in sensorgram 1.

[0047] Figure 10, like Figures 8 and 9, shows sensorgram 3 from sequence 3 shown in Figure 7. As shown in Figure 10, in sensorgram 3, for example, the responses of Sensor 1 and Sensor 8 increase over time, while the response of Sensor 4 remains relatively small. Overall, the variation in the responses of Sensors 1 through 10 is not very large, but it can be said that they are responding slowly over time.

[0048] The sensorgrams 1 to 3 created by the multi-array sensor 11 are output to the AI ​​processing unit 18. The AI ​​processing unit 18 is an AI that has been trained by odor sequence pattern presentation and known intensity and type odor presentation, and uses pattern matching AI to calculate the intensity information and identification information of the odor to be measured based on the input sensorgrams 1 to 3 and the surrounding environment.

[0049] 2. Second Embodiment (1) Example of odor detection module configuration Next, with reference to Figure 11, an example configuration of the odor detection module 20 according to the second embodiment of this technology will be described. Figure 11 is a schematic diagram showing an example configuration of the odor detection module 20. The difference between the odor detection module 20 and the odor detection module 10 according to the first embodiment is that the odor detection module 20 uses a communication network to send and receive information.

[0050] As shown in Figure 11, the odor detection module 20 according to this embodiment, like the odor detection module 10 according to the first embodiment, includes a multi-array sensor 11, an olfactory display 12 which is an odor presentation unit, exhaust mechanisms 13 and 14, a reference airflow inlet 15, a measurement target airflow inlet 16, and an AI processing unit 18. Furthermore, the odor detection module 20 includes a transmission / reception unit 21 which transmits and receives information with the cloud server 30.

[0051] When online, the transmitting / receiving unit 21 transmits information on the reference measurement value of a reference odor measured by the multi-array sensor 11 and the target measurement value of the odor to be measured to the cloud server 30, and receives the latest trained model from the cloud server 30. The transmitting / receiving unit 21 also outputs the received information to the AI ​​processing unit 18. When calculating the latest trained model, for example, the calculation process can be divided into two parts: the first calculation is performed by the AI ​​processing unit 18 of the odor detection module 20, which is an edge device close to the field, and in the second calculation, the cloud server 30 calculates the latest trained model based on the information after the first calculation.

[0052] Subsequently, the AI ​​processing unit 18 calculates the intensity information and / or identification information of the odor to be measured based on the latest trained model received by the transmitting / receiving unit 21.

[0053] (2) Example of operation of the odor detection module Next, with reference to Figure 11, an example of odor detection operation by the odor detection module 20 will be described.

[0054] Similar to Figure 2, as an example, the olfactory display 12 is initially placed above the reference airflow inlet 15, and the pre-measurement mode is started. When the pre-measurement mode is started, the odor retention section 17 of the mounted olfactory display 12 sequentially presents multiple patterns of reference odors to the multi-array sensor 11.

[0055] Next, as in Figure 3, immediately after the pre-measurement mode measurement, the system switches to the target measurement mode and opens the target airflow inlet 16. When the target airflow inlet 16 is opened, the odor of the target object and / or the target space is taken in from the target airflow inlet 16 to the multi-array sensor 11, and the multi-array sensor 11 measures the odor of the target. Once the multi-array sensor 11 measures the odor of the target, the sensorgram is input to the AI ​​processing unit 18.

[0056] Here, when online, the transmitting / receiving unit 21 transmits information of a reference measurement value, which is a sensorgram of a reference odor measured by the multi-array sensor 11, and a target measurement value, which is a sensorgram of the odor of the target to be measured, to the cloud server 30, and receives the latest trained model from the cloud server 30. When the transmitting / receiving unit 21 receives the latest trained model, it outputs it to the AI ​​processing unit 18.

[0057] Subsequently, when the latest trained model and sensorgrams are input to the AI ​​processing unit 18, the AI, which has already learned the latest trained model through the presentation of a large number of odor sequence patterns and known intensity and type odors, is input the sensorgrams of each sequence during the pre-reference odor measurement and the sensorgrams of the measurement target, and the intensity information and / or odor identification information (odor identification) of the measurement target is calculated.

[0058] As described above, the odor detection module 20 according to this embodiment offers the same effects as the first embodiment, and because the AI ​​processing unit 18 performs learning based on the latest trained model, it can further improve the accuracy of identifying the measurement target compared to the odor detection module 10 according to the first embodiment. Furthermore, the odor detection module 20 can also be used as an odor detection system in combination with the cloud server 30.

[0059] The odor detection module related to this technology can accurately identify odors of measurement targets through objective evaluation, whereas conventional methods were difficult to adapt to the environment and relied on subjective identification. Therefore, this odor detection module can be utilized, for example, in the field of environmental assessment.

[0060] Furthermore, this technology can be configured as follows: (1) A sensor element that detects odors, The surface of the sensor element includes an odor-presenting unit that presents a reference odor, Equipped with, A scent detection module in which the sensor element detects the reference scent and the scent of the object to be measured. (2) The odor detection module according to (1), further comprising a processing unit that inputs the response output of the sensor element and calculates the intensity information and / or identification information of the odor to be measured based on a trained model generated by machine learning. (3) The odor detection module according to (1) or (2), wherein the odor presentation unit holds a plurality of fragrances. (4) The odor detection module according to (3), wherein the odor presentation unit independently controls the plurality of fragrances to change them over time and presents the standard odor. (5) The odor detection module according to any one of (1) to (4), wherein the odor presentation unit presents the standard odor according to the surrounding environment. (6) The odor detection module according to any one of (1) to (5), wherein the odor presentation unit presents the standard odor corrected according to the number of times it has been used. (7) An odor detection module according to any one of (1) to (6), comprising a plurality of the aforementioned sensor elements, wherein the plurality of the aforementioned sensor elements are arranged in an array. (8) The odor detection module according to any one of (1) to (7), further comprising an exhaust mechanism for discharging air near the surface of the sensor element. (9) The odor detection module according to any one of (1) to (8), wherein the sensor element's surface is covered under normal conditions and exposed when odor is measured. (10) The odor detection module according to any one of (1) to (9), wherein the surface of the unused portion of the sensor element is exposed when odor is measured. (11) The odor detection module according to any one of (1) to (10), comprising a plurality of the aforementioned sensor elements, wherein the surface of each of the sensor elements is formed of one of the following materials: a polymer material, an inorganic material, or a metallic material having a different nanostructure or chemical properties. (12) The odor detection module according to (2), wherein the processing unit receives a reference measurement value for measuring the standard odor and a target measurement value for measuring the odor of the target, and outputs intensity information and / or identification information for the odor of the target. (13) The system further includes a transmitting and receiving unit that, when online, transmits information on the reference measurement value of the aforementioned standard odor and the target measurement value of the aforementioned target odor to a cloud server, and receives the latest trained model from the cloud server. The odor detection module according to (2), wherein the processing unit calculates intensity information and / or identification information of the odor to be measured based on the latest learned model received by the transmitting / receiving unit. (14) The step of presenting a standard scent, The steps include measuring the odor of the aforementioned standard, The steps include measuring the odor of the object to be measured, The steps include: obtaining the measured reference odor and the measured odor of the target to be measured, and calculating intensity information and / or identification information of the odor of the target to be measured based on a trained model generated by machine learning; A method for detecting odors, including the following. [Explanation of Symbols]

[0061] 10, 20 Odor detection modules 11. Multi-array sensor (sensor element) 12. Olfactory display (odor presentation unit) 13, 14 Exhaust mechanism 15 Standard airflow inlet 16. Airflow inlet to be measured 17 Odor retention part 18 AI Processing Unit 21 Transmitter / Receiver 30 Cloud Servers

Claims

1. A sensor element that detects odors, The surface of the sensor element includes an odor-presenting unit that presents a reference odor, Equipped with, The odor presentation unit presents a standard odor that changes over time. The aforementioned sensor element is an odor detection module that detects the odor of the object to be measured based on the detection of the aforementioned reference odor.

2. The odor detection module according to claim 1, wherein the odor presentation unit holds a plurality of fragrances.

3. The odor detection module according to claim 2, wherein the odor presentation unit independently controls and changes the plurality of fragrances over time.

4. The odor detection module according to claim 1, wherein the sensor element detects the odor of the target to be measured after first detecting the reference odor.

5. The odor detection module according to claim 1, wherein the sensor element is a multi-array sensor having a plurality of arrayed sensor elements.

6. The odor detection module according to claim 1, further comprising an exhaust mechanism for exhausting the airflow flowing near the surface of the sensor element to the outside.

7. The odor detection module according to claim 6, further comprising a processing unit that receives the response output of the sensor element and calculates the intensity information and / or identification information of the odor of the object to be measured based on a trained model generated by machine learning.

8. The odor detection module according to claim 7, wherein the sensor element detects the standard odor, then exhausts the airflow of the standard odor from the exhaust mechanism, and outputs a sensorgram of the standard odor to the processing unit.

9. The odor detection module according to claim 7, wherein the sensor element detects the odor of the object to be measured, then exhausts the airflow of the odor of the object to be measured from the exhaust mechanism, and outputs a sensorgram of the odor of the object to be measured to the processing unit.

10. The odor detection module according to claim 7, wherein the processing unit receives a reference measurement value, which is a sensorgram measuring the reference odor, and a target measurement value, which is a sensorgram measuring the odor of the target to be measured, and outputs the intensity and / or type of the odor of the target to be measured.

11. The odor detection module according to claim 10, wherein the processing unit performs AI-based pattern analysis from the reference odor sensorgram and the odor sensorgram of the target odor.

12. The odor detection module according to claim 11, wherein the processing unit performs calculations of the intensity information and / or the identification information using a pattern matching AI that takes the results of detecting the standard odor and detecting the odor of the object to be measured as inputs.

13. The odor detection module according to claim 7, wherein the AI ​​in the processing unit is based on a trained model generated in advance by presenting odor sequence patterns and known intensity and type odors.

14. The system further includes a transmitting / receiving unit that, when online, transmits information on the reference measurement value of the aforementioned standard odor and the target measurement value of the aforementioned target odor to a cloud server, and receives the latest trained model from the cloud server. The odor detection module according to claim 7, wherein the processing unit calculates intensity information and / or identification information of the odor to be measured based on the latest learned model received by the transmitting / receiving unit.

15. The step of presenting a standard scent, The steps include measuring the odor of the aforementioned standard, The steps include measuring the odor of the object to be measured, Includes, In the step of presenting the aforementioned standard scent, the aforementioned standard scent is presented while changing over time. A method for detecting odors, wherein the step of measuring the odor of the object to be measured is to measure the odor of the object to be measured based on the measurement of the reference odor in the step of measuring the reference odor.

16. The odor detection method according to claim 15, wherein in the step of presenting the aforementioned standard odor, multiple fragrances are independently controlled and changed over time.

17. The odor detection method according to claim 15, further comprising the step of measuring the odor of the target to be measured after the step of measuring the standard odor.

18. The odor detection method according to claim 15, further comprising, after the step of measuring the standard odor and / or the step of measuring the odor to be measured, an exhaust step of exhausting the airflow of the standard odor and / or the airflow of the odor to be measured to the outside.

19. The odor detection method according to claim 18, further comprising the steps of obtaining the measured reference odor and the measured odor of the target to be measured, and calculating intensity information and / or identification information of the odor of the target to be measured based on a trained model generated by machine learning.

20. The odor detection method according to claim 19, wherein the exhaust step is performed after the step of measuring the standard odor and / or the step of measuring the odor of the object to be measured.

21. In the step of measuring the aforementioned standard odor, a sensorgram of the standard odor is output. The odor detection method according to claim 20, wherein after the exhaust step, the calculation step is performed based on the outputted reference odor sensorgram.

22. In the step of measuring the odor of the object to be measured, a sensorgram of the odor of the object to be measured is output. The odor detection method according to claim 21, wherein after the exhaust step, the calculation step is performed based on the sensorgram of the odor to be measured that is output.

Images