Image-based fine dust measurement method and system
A fine dust measurement system includes a video photographing device that photographs a target place, generates a converted image or converted data by converting all or a portion of the photographed image and transmits the converted image or converted data, and a fine dust measurement that receives the converted image or converted data from the video photographing device, and trains a deep learning model to output a fine dust concentration of the target place by inputting the received converted image or converted data into the deep learning model.
1 . A fine dust measurement system comprising:
a video photographing device configured to photograph a target place, generate a converted image or converted data by converting all or a portion of the photographed image and transmit the converted image or converted data; and
a fine dust measurement device configured to receive the converted image or converted data from the video photographing device, and train a deep learning model to output a fine dust concentration of the target place by inputting the received converted image or converted data into the deep learning model,
wherein the video photographing device is configured to generate a converted image by performing a first type of conversion for converting all or a portion of the photographed image into an image with different characteristics, or generate converted data by performing a second type of conversion for converting all or a portion of the photographed image into data in another form,
wherein the video photographing device is configured to determine which type of conversion to perform among the first type of conversion and the second type of conversion depending on a degree of network communication between the video photographing device and the fine dust measurement device.
2 . The fine dust measurement system of claim 1 , wherein the video photographing device is configured to determine to perform the first type of conversion when the degree of network communication is equal to or greater than a preset first reference level and determine to perform the second type of conversion when the degree of network communication is less than a second reference level set to be lower than the first reference level.
3 . The fine dust measurement system of claim 1 , wherein the video photographing device is configured to generate a plurality of converted images of different types according to preset conversion methods, respectively, during the first type of conversion and transmits the plurality of converted images of different types to the fine dust measurement device, and the fine dust measurement device is configured to input the plurality of converted images into the deep learning model, respectively, to train the deep learning model so that prediction values of fine dust concentration for the plurality of converted images are output, respectively, and a difference between each of the prediction values of fine dust concentration and a correct answer value is minimized.
4 . The fine dust measurement system of claim 3 , wherein the fine dust measurement device is configured to extract a prediction value which is closest to the correct answer value among the prediction values of fine dust concentration, store a type of converted image corresponding to the extracted prediction value by matching the type with one or more of environmental information and climate information during photographing, and share the matched and stored information with the video photographing device.
5 . The fine dust measurement system of claim 4 , wherein the video photographing device is configured to acquire one or more of the environmental information and climate information during photographing of the target place, determine what type of image to convert, all or a portion of the photographed image based on one or more of the acquired environmental information and climate information, generate a converted image by converting all or a portion of the photographed image into an image of the determined type, and transmit the converted image to the fine dust measurement device.
6 . The fine dust measurement system of claim 5 , wherein the fine dust measurement device is configured to output the fine dust concentration of the target place by inputting the converted image into a previously trained deep learning model.
7 . A fine dust measurement method performed on a computing device including one or more processors and a memory that stores one or more programs executed by the one or more processors, the method comprising: photographing a target place; generating a converted image or converted data by converting all or a portion of a photographed image; and transmitting the converted image or the converted data to a fine dust measurement device equipped with a deep learning model, wherein the fine dust measurement device trains a deep learning model to output a fine dust concentration of the target place by inputting the converted image or converted data into the deep learning model,
wherein, in the generating, a converted image is generated by performing a first type of conversion for converting all or a portion of the photographed image into an image with different characteristics, or converted data is generated by performing a second type of conversion for converting all or a portion of the photographed image into data in another form,
wherein the generating includes determining to perform the first type of conversion when a degree of network communication with the fine dust measurement device is equal to or greater than a preset first reference level and determining to perform the second type of conversion when the degree of network communication is less than a second reference level set to be lower than the first reference level.
8 . The fine dust measurement method of claim 7 , further comprising:
acquiring one or more of the environmental information and climate information during photographing of the target place; and
determining what type of image to convert all or a portion of the photographed image based on one or more of the acquired environmental information and climate information during the first type of conversion.