Unmanned aerial vehicle based abnormal noise detection device and operating method thereof

An unmanned aerial vehicle based abnormal noise detection device comprising:

• a drone;
• an acoustic sensor attached to the drone and configured to collect acoustic data using a beam-forming technology;
• a camera attached to the acoustic sensor, the camera configured generate image data that includes an image of a measurement target;
• a controller which generates a visualized image, using the image data and the acoustic data;
• and
• a display which displays the visualized image,
• wherein the controller is configured to:
• generate a matrix including a plurality of cells on the image data and larger than or equal to the size of the image data,
• generate object data of an object corresponding to the measurement target included in the image data, the object data indicating a size and a position of the object that is a target of an abnormal noise detection,
• classify acoustic data that is not included in a preset first frequency band range from the acoustic data corresponding to each cell to generate abnormal acoustic data,
• generate, using the abnormal acoustic data and the object data, filtered abnormal acoustic data, and
• generate a visualized image in which positions of cells including the filtered abnormal acoustic data are displayed on the image data,
• wherein the first frequency band range is a range of the acoustic data measured during normal operation of the object.

A method for operating an unmanned aerial vehicle based abnormal noise detection device, the method comprising:

• capturing, to generate image data with a camera attached to a drone, an image of a measurement target included in the image data ;
• generating, based on the captured image, a matrix including a plurality of cells on the image data and corresponding to a size of the image data;
• generating object data of an object corresponding to the measurement target included in the image data, the object data indicating a size and a position of the object that is a target of an abnormal noise detection;
• collecting acoustic data corresponding to each of the plurality of cells by an acoustic sensor attached to the drone;
• classifying, using the collected acoustic data, acoustic data that is not included in a preset first frequency band range from the acoustic data to generate abnormal acoustic data;
• removing noise that is not abnormal acoustic data due to an abnormal state of the object from the abnormal acoustic data to generate filtered abnormal acoustic data;
• generating visualized image in which positions of cells that include the filtered abnormal acoustic data are displayed in the image data;
• providing the visualized image to a display by a radio communication module;
• and
• displaying the visualized image by the display.