Particulate matter monitoring method and device, and system using same

Abstract

A particulate matter monitoring method and device, and a system using the same are disclosed. The disclosed particulate matter monitoring method according to an embodiment, which is a method performed by a monitoring device installed in a place for providing a particulate matter monitoring service, comprises the steps of: identifying whether a predetermined first measurement event occurs in an area covered by the monitoring device; if the first measurement event occurs, capturing an image of a point where the first measurement event occurs; and inputting the captured image according to the first measurement event into a pretrained first deep learning model so that the first deep learning model outputs a particulate matter concentration at the point where the first measurement event occurs.

Description

Fine dust monitoring method and device and system using the same

[0001] An embodiment of the present invention relates to a fine dust monitoring technology.

[0002] Fine dust is a type of dust whose particles are so small they are invisible to the naked eye. Research suggests that this fine dust overrides the body's defense system, causing various health problems, including cardiovascular, respiratory, and cerebrovascular diseases. The International Agency for Research on Cancer (IARC), a division of the World Health Organization (WHO), has designated fine dust as a Group 1 carcinogen. Due to these risks, fine dust is being analyzed as a contributing factor to the decline in economic activity among members of society.

[0003] In particular, construction site dust, due to its nature, generates high concentrations of fine dust at once, then rapidly decreases after work is completed. Therefore, workers moving around the area near the point where dust is generated are exposed to situations that can have adverse health effects.

[0004] An embodiment of the present invention provides a fine dust monitoring method and device capable of monitoring fine dust generated in a field, measuring its concentration, and providing a notification service, and a system using the same.

[0005] A method for monitoring fine dust according to one embodiment disclosed is a method performed in a monitoring device installed in a place for providing a fine dust monitoring service, the method comprising: a step of checking whether a preset first measurement event occurs in an area covered by the monitoring device; a step of photographing a point where the first measurement event occurs when the first measurement event occurs; and a step of inputting a photographed image according to the first measurement event into a pre-trained first deep learning model to output a fine dust concentration at a point where the first measurement event occurs.

[0006] The first measurement event may include one or more of: detection of a moving vehicle exceeding a preset speed, detection of noise generation exceeding a preset sound level, and detection of work motion of preset equipment.

[0007] The step of outputting the above fine dust concentration may cause the first deep learning model to extract a target area from the captured image and output the fine dust concentration of the extracted target area.

[0008] The step of outputting the above fine dust concentration may be performed by having the first deep learning model track the target area in the captured image and process the bounding box according to changes in the position and size of the target area, and when multiple target areas are extracted from the captured image, may be performed by processing the bounding box for each target area.

[0009] The above fine dust monitoring method may further include a step of generating a notification signal based on a target area of ​​the captured image and a fine dust concentration in the target area; and a step of transmitting the notification signal to a user terminal.

[0010] The step of generating the above notification signal can generate different types of notification signals depending on the range of the target area and the size of the fine dust concentration value of the target area.

[0011] The step of transmitting to the user terminal may include the step of determining a user terminal to which the notification signal is to be transmitted based on the field location corresponding to the target area and the location of the user terminal; and the step of transmitting the notification signal to the determined user terminal.

[0012] The step of generating the above notification signal may generate a first passive notification signal when the range of the target area is less than a preset reference range and the fine dust concentration value of the target area is less than a preset reference value, generate a first active notification signal when the range of the target area is less than a preset reference range and the fine dust concentration value of the target area is greater than or equal to the preset reference value, generate a second passive notification signal when the range of the target area is greater than or equal to the preset reference range and the fine dust concentration value of the target area is less than or equal to the preset reference value, and generate a second active notification signal when the range of the target area is greater than or equal to the preset reference range and the fine dust concentration value of the target area is greater than or equal to the preset reference value.

[0013] The first passive notification signal or the first active notification signal may be transmitted to a user terminal located within a first distance preset based on a field location corresponding to the target area, and the second passive notification signal or the second active notification signal may be transmitted to a user terminal located within a second distance set further than the first distance based on a field location corresponding to the target area.

[0014] The above fine dust monitoring method may further include a step of checking whether a preset second measurement event occurs; a step of photographing a front or preset location of the monitoring device when the second measurement event occurs; and a step of measuring the fine dust concentration at a shooting target location based on a captured image according to the second measurement event.

[0015] The step of measuring the concentration of fine dust may include: extracting a still image from the captured image; converting the extracted still image into an image having characteristics sensitive to fine dust to generate a converted image; and inputting the converted image into a pre-trained second deep learning model to output the concentration of fine dust at the captured point.

[0016] The step of measuring the concentration of fine dust may further include a step of acquiring at least one of environmental information and climate information during the shooting; and a step of determining what type of image to convert the still image into based on at least one of the environmental information and climate information.

[0017] The above fine dust monitoring method may further include a step of calculating a change in fine dust concentration value by at least one of weather, season, and month based on at least one of the fine dust concentration measured according to the first measurement event and the fine dust concentration measured according to the second measurement event, or a step of counting at least one of the number of occurrences, time, and occurrence frequency of fine dust exceeding a preset threshold value based on at least one of the fine dust concentration measured according to the first measurement event and the fine dust concentration measured according to the second measurement event.

[0018] A device according to one embodiment disclosed is a device installed in a place for providing a fine dust monitoring service, and includes an event detection module for checking whether a preset first measurement event occurs in an area covered by the device; a photographing module for photographing a point where the first measurement event occurs when the first measurement event occurs; and a first fine dust measurement module for inputting a photographed image according to the first measurement event into a pre-trained first deep learning model to output a fine dust concentration at a point where the first measurement event occurs.

[0019] A fine dust monitoring system according to one embodiment disclosed includes a monitoring device installed in a place for providing a fine dust monitoring service, which photographs a corresponding point upon occurrence of a preset first measurement event, inputs a photographed image according to the first measurement event into a pre-trained first deep learning model to extract a target area of ​​the photographed image, outputs a fine dust concentration in the target area, and generates a notification signal based on the target area and the fine dust concentration in the target area; and one or more user terminals which receive the notification signal from the monitoring device.

[0020] According to the disclosed embodiment, when dust is generated at a construction site or other construction site, it is possible to quickly detect it, measure the concentration of dust, and generate a notification signal to nearby users, thereby preventing users at the site from being exposed to danger due to the generation of dust.

[0021] Figure 1 is a drawing showing the configuration of a fine dust monitoring system according to one embodiment of the present invention.

[0022] Figure 2 is a block diagram showing the configuration of a monitoring device according to one embodiment of the present invention.

[0023] FIG. 3 is a diagram showing a state in which a first deep learning model extracts a target area from a captured image and outputs a fine dust concentration value of the extracted target area in one embodiment of the present invention.

[0024] FIG. 4 is a diagram showing a state in which a first deep learning model extracts multiple target areas from a captured image and outputs a fine dust concentration value of each extracted target area in one embodiment of the present invention.

[0025] FIG. 5 is a diagram showing various embodiments of extracting a target area and outputting a fine dust concentration value from a captured image according to one embodiment of the present invention.

[0026] FIG. 6 is a diagram showing first to sixth transformed images generated by performing a first type of transformation on a still image in one embodiment of the present invention.

[0027] Figure 7 is a flowchart for explaining a fine dust monitoring method according to one embodiment of the present invention.

[0028] Figure 8 is a flowchart for explaining a fine dust monitoring method according to another embodiment of the present invention.

[0029] FIG. 9 is a block diagram illustrating a computing environment including a computing device suitable for use in exemplary embodiments.

[0030] Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to facilitate a comprehensive understanding of the methods, devices, and / or systems described herein. However, these are merely examples and the present invention is not limited thereto.

[0031] In describing embodiments of the present invention, if a detailed description of a known technology related to the present invention is judged to unnecessarily obscure the gist of the present invention, the detailed description will be omitted. In addition, the terms described below are terms defined in consideration of their functions in the present invention, and this may vary depending on the intention or custom of the user or operator. Therefore, the definitions should be made based on the contents throughout this specification. The terminology used in the detailed description is only for the purpose of describing embodiments of the present invention and should not be limited in any way. Unless clearly used otherwise, the singular form includes the plural form. In this description, expressions such as "comprises" or "having" are intended to indicate certain features, numbers, steps, operations, elements, parts or combinations thereof, and should not be construed to exclude the presence or possibility of one or more other features, numbers, steps, operations, elements, parts or combinations thereof other than those described.

[0032] Additionally, while terms such as "first" and "second" may be used to describe various components, these components should not be limited by these terms. These terms may be used to distinguish one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a "second component," and similarly, a second component may also be referred to as a "first component."

[0033] FIG. 1 is a diagram showing the configuration of a fine dust monitoring system according to one embodiment of the present invention.

[0034] Referring to FIG. 1, a fine dust monitoring system (100) may include a monitoring device (102) and a user terminal (104). The monitoring device (102) and the user terminal (104) are communicatively connected to a monitoring server (104) via a communication network (150).

[0035] Here, the communication network (150) may include the Internet, one or more local area networks, wide area networks, a cellular network, a mobile network, other types of networks, or a combination of these networks.

[0036] In one embodiment, a fine dust monitoring system (100) may be installed at a construction site to measure fine dust and, based on the measurement results, issue notifications to workers or nearby residents. However, this is not a limitation, and the fine dust monitoring system (100) can be applied to various fields, including fire surveillance.

[0037] One or more monitoring devices (102) may be installed at a location (e.g., a construction site) for providing fine dust monitoring services. For example, multiple monitoring devices (102) may be installed spaced apart from each other to cover different areas depending on the area of ​​the location for providing fine dust monitoring services. The monitoring devices (102) may measure fine dust generated in the area of ​​their responsibility, generate a notification signal, and transmit the notification signal to a user terminal (104).

[0038] The user terminal (104) may be a terminal of a user receiving a fine dust monitoring service. The user terminal (104) may transmit location information of the user terminal (104) to the monitoring device (102). The user terminal (104) may receive a notification signal related to fine dust from the monitoring device (102). Here, the user may be a worker or manager at a construction site, but is not limited thereto, and may include residents near the construction site, etc.

[0039] In one embodiment, a user terminal (104) may be installed with an application for receiving fine dust monitoring services. The application may be stored on a computer-readable storage medium of the user terminal (104). The application includes a predetermined set of instructions executable by the processor of the user terminal (104).

[0040] FIG. 2 is a block diagram showing the configuration of a monitoring device (102) according to one embodiment of the present invention. Referring to FIG. 2, the monitoring device (102) may include an event detection module (111), a photographing module (113), a first fine dust measurement module (115), a notification module (117), a second fine dust measurement module (119), and a data management module (121).

[0041] The event detection module (111) can detect whether a preset first measurement event occurs in the area managed by the monitoring device (102). Here, the preset first measurement event is an event with a high probability of generating fine dust depending on the work situation at the site, and may include, but is not limited to, detection of a moving vehicle (e.g., a vehicle, a truck, etc.) exceeding a preset speed, detection of noise generation exceeding a preset sound level, and detection of work operation of preset equipment.

[0042] The photographing module (113) may be configured to photograph the point where the first measurement event occurs when a preset first measurement event occurs. In one embodiment, the photographing module (113) may include a camera capable of adjusting rotation and tilt. The photographing module (113) may adjust one or more of the rotation and tilt of the camera to photograph the point where the first measurement event occurs. If the occurrence of the first measurement event is caused by a moving object (e.g., a truck or a forklift), the photographing module (113) may track and photograph the object.

[0043] Additionally, the photographing module (113) can photograph the area covered by the monitoring device (102) according to a preset second measurement event. The second measurement event is an event for measuring the concentration of fine dust under normal conditions, and can occur according to a preset cycle or command. In one embodiment, the photographing module (113) can photograph the area in front of the photographing module (113) when the second measurement event occurs, but is not limited thereto, and can also photograph a preset location.

[0044] The first fine dust measurement module (115) can measure the concentration of fine dust at the shooting target point based on the image (i.e., the shooting image) captured by the shooting module (113) according to the first measurement event. The first fine dust measurement module (115) can measure the concentration of fine dust at the shooting target point from the shooting image based on deep learning. Accordingly, the first fine dust measurement module (115) can include a first deep learning model (115a). The first fine dust measurement module (115) can train the first deep learning model (115a) to predict the concentration of fine dust at the shooting target point from the shooting image according to the first measurement event.

[0045] Specifically, the first fine dust measurement module (115) can train the first deep learning model (115a) to extract a target area from the captured image. Here, the target area may be an area in the captured image where fine dust is generated upon the occurrence of the first measurement event.

[0046] The first fine dust measurement module (115) inputs a captured image as learning data into the first deep learning model (115a), and compares the target area predicted by the first deep learning model (115a) with the actual target area (i.e., the correct answer value) to train the first deep learning model (115a) so that the difference is minimized.

[0047] The first fine dust measurement module (115) can train the first deep learning model (115a) to process a target area in a captured image as a bounding box. The first fine dust measurement module (115) can train the first deep learning model (115a) to process a bounding box while tracking a target area according to changes in the position and size of the target area in the captured image. If multiple target areas are extracted from the captured image, the first fine dust measurement module (115) can process a bounding box for each target area.

[0048] In addition, the first fine dust measurement module (115) can train the first deep learning model (115a) to predict the fine dust concentration for the target area of ​​the captured image. The first fine dust measurement module (115) can train the first deep learning model (115a) to compare the fine dust concentration predicted value for the target area of ​​the first deep learning model (115a) with the actually measured fine dust concentration value (i.e., the correct answer value) so that the difference is minimized.

[0049] When the first fine dust measurement module (115) receives a captured image according to the first measurement event from the capture module (113) in a state where the learning of the first deep learning model (115a) is completed, the first fine dust measurement module (115) can input the captured image into the first deep learning model (115a) to extract a target area and output a fine dust concentration value of the extracted target area.

[0050] FIG. 3 is a diagram showing a state in which a first deep learning model (115a) extracts one target area from a captured image and outputs a fine dust concentration value of the extracted target area in one embodiment of the present invention, and FIG. 4 is a diagram showing a state in which a first deep learning model (115a) extracts multiple target areas from a captured image and outputs a fine dust concentration value of each extracted target area in one embodiment of the present invention. Here, a situation in which a building is demolished using a forklift at a construction site is shown as an example. FIG. 5 is a diagram showing various embodiments of extracting a target area from a captured image and outputting a fine dust concentration value according to one embodiment of the present invention.

[0051] The notification module (117) can receive the target area of ​​the captured image and the fine dust concentration value of the target area according to the first measurement event from the first fine dust measurement module (115). In addition, the notification module (117) can receive the location information of each user terminal (104). The notification module (117) can generate a notification signal based on the target area of ​​the captured image and the fine dust concentration value of the target area according to the first measurement event.

[0052] The notification module (117) can generate different types of notification signals depending on the range of the target area and the size of the fine dust concentration value in the target area. The notification module (117) can select the user terminal (104) to which the notification signal will be transmitted based on the on-site location corresponding to the target area and the location of the user terminal (104).

[0053] Specifically, the notification module (117) may generate a first passive notification signal when the range of the target area is less than a preset reference range and the fine dust concentration value of the target area is less than a preset reference value. The notification module (117) may transmit the first passive notification signal to a user terminal (104) located within a preset first distance based on a field location corresponding to the target area. The first passive notification signal may be a notification signal that prohibits users located within the first distance from approaching the target area.

[0054] The notification module (117) may generate a first active notification signal when the range of the target area is less than a preset reference range and the fine dust concentration value of the target area is greater than or equal to the preset reference value. The notification module (117) may transmit the first active notification signal to a user terminal (104) located within a preset first distance based on a field location corresponding to the target area. The first active notification signal may be a notification signal that induces users located within the first distance to move away from the vicinity of the corresponding target area.

[0055] The notification module (117) may generate a second passive notification signal when the range of the target area is greater than or equal to a preset reference range and the fine dust concentration value of the target area is less than the preset reference value. The notification module (117) may transmit the second passive notification signal to a user terminal (104) located within a second distance set further than the first distance based on a field location corresponding to the target area. The second passive notification signal may be a notification signal that prohibits users located within the second distance from approaching the target area.

[0056] The notification module (117) may generate a second active notification signal when the range of the target area is greater than or equal to a preset reference range and the fine dust concentration value of the target area is greater than or equal to a preset reference value. The notification module (117) may transmit the second active notification signal to a user terminal (104) located within a preset second distance based on a field location corresponding to the target area. The second active notification signal may be a notification signal that induces users located within the second distance to move away from the vicinity of the corresponding target area.

[0057] The second fine dust measurement module (119) can measure the fine dust concentration at the shooting target point based on the image (i.e., the shooting image) captured by the shooting module (113) according to the second measurement event. To measure the fine dust concentration based on deep learning, the second fine dust measurement module (119) can include a second deep learning model (119a).

[0058] Specifically, the second fine dust measurement module (119) can extract still images from the captured video of the second measurement event. That is, since the captured video of the second measurement event is composed of consecutive images, a plurality of still images can be extracted from the captured video. Here, the still images can be RGB (Red, Green, Blue) images.

[0059] The second fine dust measurement module (119) can convert the extracted still image into an image with different characteristics. The second fine dust measurement module (119) can convert the extracted still image into one or more images with characteristics sensitive to fine dust. In other words, the second fine dust measurement module (119) can convert the extracted still image into one or more images suitable for measuring the concentration of fine dust.

[0060] In one embodiment, the second fine dust measurement module (119) can generate a first converted image by converting a still image (i.e., an RGB image) into an HSV (Hue, Saturation, Value) image. Here, since the HSV image has pure color information in H (Hue), it can classify colors more easily than an RGB image, and can be less affected by changes in illuminance or shade by adjusting the V (Value) value.

[0061] The second fine dust measurement module (119) may use the HSV image itself as the first converted image when converting a still image into an HSV image, or may use the H (Hue) and S (Saturation) channels in the HSV image as the first converted image, or may use only the S (Saturation) channel in the HSV image as the first converted image. In this case, the H (Hue) or S (Saturation) value changes depending on the concentration of fine dust.

[0062] In addition, the second fine dust measurement module (119) can generate a second converted image by converting the still image by applying the DCP (Dark Channel Prior) technique. When the DCP (Dark Channel Prior) technique is applied to the still image, haze (a phenomenon in which the object appears hazy due to the propagation of light being obstructed by materials existing between the object and the camera) present in the still image can be removed. At this time, the second fine dust measurement module (119) can extract the transmittance feature of the still image based on the DCP (Dark Channel Prior) technique and use it as the second converted image. The second converted image has different turbidity depending on the concentration of fine dust.

[0063] In addition, the second fine dust measurement module (119) can generate a third transformed image by applying a Gabor filter to a still image and transforming the image. When the Gabor filter is applied to a still image, an edge can be extracted from the still image, and the clarity of the edge varies depending on the amount of fine dust.

[0064] In addition, the second fine dust measurement module (119) can generate a fourth transformed image by applying a Sobel filter to a still image to transform the image. The Sobel filter is a filter that filters according to the directionality of frequencies in an image, and can detect diagonal edges more sensitively than horizontal and vertical edges.

[0065] In addition, the second fine dust measurement module (119) can generate a fifth transformed image by applying the LBP (Local Binary Pattern) technique to a still image to transform the image. The LBP (Local Binary Pattern) technique extracts the features of the image by converting the pixel values ​​around each pixel of the image into binary numbers (0 or 1), and the binary numbers are generated based on the relative brightness difference between the central pixel and the neighboring pixels. In other words, if the neighboring pixel is larger than the central pixel, it is processed as binary 1, and if it is smaller, it is processed as binary 0.

[0066] In addition, the second fine dust measurement module (119) can generate a sixth transformed image by applying a Laplacian filter to a still image to transform the image. The Laplacian filter performs second differentiation in the horizontal and vertical directions of the image, thereby finding the center of the edge portion (i.e., the inflection point of the pixel value change).

[0067] FIG. 6 is a diagram illustrating first to sixth transformed images generated by performing a first type of transformation on a still image in one embodiment of the present invention. The second fine dust measurement module (119) can generate the first to sixth transformed images, respectively, based on the still image during the learning stage of the second deep learning model (119a).

[0068] Meanwhile, the second fine dust measurement module (119) can determine, based on input additional information, what image transformation to perform on a still image when performing image transformation in the inference step for measuring the concentration of fine dust according to the second measurement event after the learning of the second deep learning model (119a) described later is completed. Here, the additional information may be environmental information or climate information of the location for measuring the concentration of fine dust. In one embodiment, the additional information may be, but is not limited to, temperature, humidity, illuminance, and wind speed of the location for measuring the concentration of fine dust.

[0069] The second fine dust measurement module (119) can input one or more transformed images into the second deep learning model (119a) and train the second deep learning model (119a) to predict the fine dust concentration at the shooting target point. The second deep learning model (119a) can be trained so that the difference between the predicted fine dust concentration and the actually measured fine dust concentration (i.e., the correct answer value) is minimized.

[0070] Here, the correct answer value can be obtained from a device (e.g., a light scattering sensor) that actually measures the fine dust concentration at the target location. The fine dust concentration measuring device can measure the fine dust concentration at the location at the time of the shooting. However, this is not limited to this, and the correct answer value can also be obtained from an external organization by receiving fine dust concentration data for the area including the site.

[0071] The second fine dust measurement module (119) can input the first to sixth transformed images into the second deep learning model (119a), respectively. The first to sixth transformed images are not input simultaneously, but can be input sequentially for each learning epoch.

[0072] The second deep learning model (119a) can receive the first converted image as input and output the first fine dust concentration prediction value. The second deep learning model (119a) can receive the second converted image as input and output the second fine dust concentration prediction value. Similarly, the second deep learning model (119a) can receive the third to sixth converted images as input and output the third to sixth fine dust concentration prediction values, respectively. The second deep learning model (119a) can be trained to compare the first to sixth fine dust concentration prediction values ​​with the correct answer value (actually measured fine dust concentration value) and minimize the difference between them.

[0073] During the learning process, the second fine dust measurement module (119) extracts a predicted value that is closest to the correct answer value among the first fine dust concentration predicted value to the sixth fine dust concentration predicted value, and stores the type of the converted image corresponding to the extracted predicted value by matching it with at least one of environmental information (e.g., illuminance of the target location, etc.) and climate information (e.g., temperature, humidity, and wind speed of the target location, etc.) at the time of shooting the still image.

[0074] For example, if the sixth fine dust concentration prediction value is closest to the correct answer value, the second fine dust measurement module (119) can store the type of converted image corresponding to the sixth fine dust concentration prediction value by matching it with environmental information and climate information at the time of image capture. Through this, it is possible to establish a standard for which type of image (i.e., which image among the first converted image to the sixth converted image) the captured image will be converted into based on which environmental information and climate information.

[0075] The second fine dust measurement module (119) can generate converted images by converting still images taken at various locations and time periods when training the second deep learning model (119a), and train the second deep learning model (119a) using the generated converted images as training data.

[0076] Meanwhile, although one second deep learning model (119a) is described here as an example, it is not limited thereto, and a deep learning model may be provided for each converted image. For example, a corresponding deep learning model may be provided for each converted image, such as a 2-1 deep learning model that takes a first converted image as input and outputs a first fine dust concentration prediction value, a 2-2 deep learning model that takes a second converted image as input and outputs a second fine dust concentration prediction value, etc.

[0077] The data management module (121) can manage the fine dust concentration values ​​measured by the first fine dust measurement module (115) and the second fine dust measurement module (119). In one embodiment, the data management module (121) can aggregate the number of occurrences, time, and occurrence frequency of fine dust exceeding a preset threshold value based on the fine dust concentration values ​​measured by at least one of the first fine dust measurement module (115) and the second fine dust measurement module (119) during a preset period.

[0078] The data management module (119) can calculate changes in fine dust concentration values ​​by weather, season, month, etc. based on the fine dust concentration values ​​measured by at least one of the first fine dust measurement module (115) and the second fine dust measurement module (119) over a preset period. For example, if the monitoring device (102) is installed at a construction site, the data management module (119) can also calculate changes in fine dust concentration values ​​by air of the construction site.

[0079] Meanwhile, the monitoring device (102) may further include a display (not shown) that displays the fine dust concentration value measured by at least one of the first fine dust measurement module (115) and the second fine dust measurement module (119).

[0080] According to the disclosed embodiment, when dust is generated at a construction site or other construction site, it is possible to quickly detect it, measure the concentration of dust, and generate a notification signal to nearby users, thereby preventing users at the site from being exposed to danger due to the generation of dust.

[0081] As used herein, the term "module" may refer to a functional and structural combination of hardware for implementing the technical concepts of the present invention and software for operating the hardware. For example, the term "module" may refer to a logical unit of a given code and hardware resources for executing the given code, and does not necessarily refer to physically connected code or a single type of hardware.

[0082] Figure 7 is a flowchart illustrating a fine dust monitoring method according to one embodiment of the present invention. While the illustrated flowchart divides the method into multiple steps, at least some of the steps may be performed in reverse order, combined with other steps and performed together, omitted, divided into substeps, or performed with one or more additional steps not illustrated.

[0083] Referring to FIG. 7, the monitoring device (102) can check whether a preset first measurement event occurs in the area in charge of the monitoring device (102) (S 101). If the preset first measurement event occurs as a result of the check in S 101, the monitoring device (102) can photograph the point where the first measurement event occurs (S 103). At this time, if the occurrence of the first measurement event is caused by a moving object (e.g., a truck or a forklift), the monitoring device (102) can track and photograph the object.

[0084] Next, the monitoring device (102) can input the captured image according to the first measurement event into the first deep learning model (115a) that has been previously trained to extract a target area from the captured image and output the fine dust concentration value of the extracted target area (S 105).

[0085] Next, the monitoring device (102) can generate a notification signal based on the target area of ​​the captured image and the fine dust concentration value of the target area (S 107). For example, the monitoring device (102) can generate one notification signal among the first passive notification signal, the first active notification signal, the second passive notification signal, and the second active notification signal based on the target area of ​​the captured image and the fine dust concentration value of the target area.

[0086] Next, the monitoring device (102) can determine a user terminal (104) to which a notification signal will be transmitted based on the field location corresponding to the target area and the location of the user terminal (104), and transmit the notification signal to the determined user terminal (104) (S 109).

[0087] Figure 8 is a flowchart illustrating a fine dust monitoring method according to another embodiment of the present invention. While the illustrated flowchart divides the method into multiple steps, at least some of the steps may be performed in reverse order, combined with other steps and performed together, omitted, divided into substeps, or performed with one or more additional steps not illustrated.

[0088] Referring to FIG. 8, the monitoring device (102) can check whether a preset second measurement event occurs (S 201). If the second measurement event occurs as a result of the check in S 201, the monitoring device (102) can photograph the front or a preset location (S 203).

[0089] Next, the monitoring device (102) can obtain one or more of environmental information and climate information during shooting (S 205). Next, the monitoring device (102) can extract a still image from the shooting video and determine what type of image to convert the still image into based on one or more of the environmental information and climate information (S 207).

[0090] Next, the monitoring device (102) can generate a converted image by converting a still image into a predetermined type of image (S 209). Next, the monitoring device (102) can input the converted image into a pre-trained second deep learning model (119a) to output a fine dust concentration value of the shooting target location (S 211).

[0091] FIG. 9 is a block diagram illustrating a computing environment (10) including a computing device suitable for use in exemplary embodiments. In the illustrated embodiment, each component may have different functions and capabilities other than those described below, and may include additional components other than those described below.

[0092] The illustrated computing environment (10) includes a computing device (12). In one embodiment, the computing device (12) may be a monitoring device (102). Additionally, the computing device (12) may be a user terminal (104).

[0093] A computing device (12) includes at least one processor (14), a computer-readable storage medium (16), and a communication bus (18). The processor (14) may cause the computing device (12) to operate according to the exemplary embodiments mentioned above. For example, the processor (14) may execute one or more programs stored in the computer-readable storage medium (16). The one or more programs may include one or more computer-executable instructions, which, when executed by the processor (14), may be configured to cause the computing device (12) to perform operations according to the exemplary embodiments.

[0094] A computer-readable storage medium (16) is configured to store computer-executable instructions or program code, program data, and / or other suitable forms of information. A program (20) stored in the computer-readable storage medium (16) includes a set of instructions executable by the processor (14). In one embodiment, the computer-readable storage medium (16) may be a memory (volatile memory such as random access memory, non-volatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, any other form of storage medium that can be accessed by the computing device (12) and store desired information, or a suitable combination thereof.

[0095] A communication bus (18) interconnects various other components of the computing device (12), including the processor (14) and computer-readable storage media (16).

[0096] The computing device (12) may also include one or more input / output interfaces (22) that provide interfaces for one or more input / output devices (24) and one or more network communication interfaces (26). The input / output interfaces (22) and the network communication interfaces (26) are connected to the communication bus (18). The input / output devices (24) may be connected to other components of the computing device (12) via the input / output interfaces (22). Exemplary input / output devices (24) may include input devices such as pointing devices (such as a mouse or a trackpad), a keyboard, a touch input device (such as a touchpad or a touchscreen), a voice or sound input device, various types of sensor devices and / or photographing devices, and / or output devices such as display devices, printers, speakers and / or network cards. The exemplary input / output devices (24) may be included within the computing device (12) as a component constituting the computing device (12), or may be connected to the computing device (12) as a separate device distinct from the computing device (12).

[0097] While representative embodiments of the present invention have been described in detail above, those skilled in the art will appreciate that various modifications to the above-described embodiments are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the claims set forth below but also by equivalents thereof.

Claims

1. A method performed in a monitoring device installed in a place to provide fine dust monitoring service, A step of checking whether a preset first measurement event occurs in the area covered by the above monitoring device; When the first measurement event occurs, a step of photographing the point where the first measurement event occurs; and A method for monitoring fine dust, comprising the step of inputting a captured image according to the first measurement event into a first deep learning model that has been previously learned so as to output the fine dust concentration at a point where the first measurement event occurred.

2. In claim 1, The above first measurement event is, A method for monitoring fine dust, comprising at least one of: detecting a vehicle having a speed exceeding a preset speed, detecting noise generation exceeding a preset sound level, and detecting work operation of preset equipment.

3. In claim 1, The step of outputting the above fine dust concentration is: A method for monitoring fine dust, which causes the first deep learning model to extract a target area from the captured image and output the concentration of fine dust in the extracted target area.

4. In claim 3, The step of outputting the above fine dust concentration is: A method for monitoring fine dust, wherein the first deep learning model is configured to process a bounding box while tracking a target area according to changes in the position and size of the target area in the captured image, and, when multiple target areas are extracted from the captured image, process a bounding box for each target area.

5. In claim 3, The above fine dust monitoring method is, A step of generating a notification signal based on the target area of ​​the above-described shooting image and the concentration of fine dust in the target area; and A method for monitoring fine dust, further comprising the step of transmitting the above notification signal to a user terminal.

6. In claim 5, The steps for generating the above notification signal are: A method for monitoring fine dust, which generates different types of notification signals depending on the range of the target area and the size of the fine dust concentration value in the target area.

7. In claim 6, The step of transmitting to the above user terminal is: A step of determining a user terminal to which the notification signal is to be transmitted based on the field location corresponding to the target area and the location of the user terminal; and A method for monitoring fine dust, comprising the step of transmitting the notification signal to the determined user terminal.

8. In claim 6, The steps for generating the above notification signal are: If the range of the target area is less than the preset standard range and the fine dust concentration value of the target area is less than the preset standard value, a first negative alarm signal is generated, If the range of the above target area is less than the preset standard range and the fine dust concentration value of the above target area is greater than the preset standard value, a first active notification signal is generated. If the range of the target area is greater than or equal to the preset standard range and the concentration value of fine dust in the target area is less than the preset standard value, a second negative alarm signal is generated. A fine dust monitoring method, wherein a second active notification signal is generated when the range of the target area is greater than or equal to a preset standard range and the fine dust concentration value of the target area is greater than or equal to a preset standard value.

9. In claim 8, The first passive notification signal or the first active notification signal is transmitted to a user terminal located within a first preset distance based on a field location corresponding to the target area, A method for monitoring fine dust, wherein the second passive notification signal or the second active notification signal is transmitted to a user terminal located within a second distance that is set further than the first distance based on a field location corresponding to the target area.

10. In claim 1, The above fine dust monitoring method is, A step of determining whether a preset second measurement event occurs; When the second measurement event occurs, the step of photographing the front or preset location of the monitoring device; and A method for monitoring fine dust, further comprising a step of measuring the concentration of fine dust at a shooting target point based on a shooting image according to the second measurement event.

11. In claim 10, The step of measuring the above fine dust concentration is: A step of extracting a still image from the above-mentioned captured video; A step of generating a converted image by converting the extracted still image into an image having a characteristic sensitive to fine dust; and A method for monitoring fine dust, comprising a step of inputting the above-mentioned converted image into a pre-trained second deep learning model to output the fine dust concentration at the photographing target point.

12. In claim 11, The step of measuring the above fine dust concentration is: A step of acquiring at least one of environmental information and climate information during the above shooting; and A method for monitoring fine dust, further comprising a step of determining what type of image to convert the still image into based on at least one of the environmental information and climate information.

13. In claim 10, The above fine dust monitoring method is, A method for monitoring fine dust, further comprising the step of calculating changes in fine dust concentration values ​​by at least one of weather, season, and month based on at least one of the fine dust concentration measured according to the first measurement event and the fine dust concentration measured according to the second measurement event, or calculating at least one of the number of times, time, and occurrence frequency of fine dust exceeding a preset threshold value based on at least one of the fine dust concentration measured according to the first measurement event and the fine dust concentration measured according to the second measurement event.

14. A device installed in a location to provide fine dust monitoring services. An event detection module for determining whether a preset first measurement event occurs in the area covered by the above device; A photographing module that photographs the point where the first measurement event occurs when the first measurement event occurs; and A device including a first fine dust measurement module that inputs a captured image according to the first measurement event into a first deep learning model that has been previously trained to output the fine dust concentration at the point where the first measurement event occurred.

15. A monitoring device installed in a place for providing a fine dust monitoring service, which photographs a corresponding point upon occurrence of a preset first measurement event, inputs the photographed image according to the first measurement event into a pre-learned first deep learning model to extract a target area of ​​the photographed image, outputs the fine dust concentration of the target area, and generates a notification signal based on the target area and the fine dust concentration of the target area; and A fine dust monitoring system comprising one or more user terminals receiving the notification signal from the monitoring device.

Images