Deep learning-based thermal imaging camera image stabilization method utilizing environmental information for equipment overheating judgment and thermal imaging camera image stabilization system using the same

Abstract

The thermal imaging camera image stabilization system of the present invention is characterized by comprising a thermal imaging camera that acquires a thermal image of equipment driven by power, a current meter that measures the amount of current of the equipment, a sensor that acquires temperature data and humidity data of an area where the equipment is operated, and an image correction unit that generates a corrected thermal image by receiving a correction factor including the amount of current, temperature data, and humidity data from a pre-set thermal image correction deep learning model that receives a thermal image and outputs a corrected thermal image.

Description

Technology Field

[0001] The present invention relates to a thermal imaging camera image correction method, and more specifically, to a deep learning-based thermal imaging camera image stabilization method utilizing environmental information for determining equipment overheating and a thermal imaging camera image stabilization system using the same. Background Technology

[0003] Infrared cameras are being introduced to prevent fires in power relays of equipment installed in factories, but frequent malfunctions occur due to the low precision of the infrared cameras.

[0005] Figure 1 is an example of an image captured by a typical thermal imaging camera.

[0006] In other words, a thermal imaging camera is a device that generates an image by emitting infrared radiation, measuring the amount of infrared radiation emitted by an object. Based on this measured amount of infrared radiation, it estimates and displays the temperature.

[0007] However, these thermal imaging cameras have the disadvantage that absolute temperature measurement is difficult as they detect temperature differences between objects. Additionally, they are vulnerable to changes in ambient temperature and humidity. In other words, measurements can be distorted depending on changes in temperature and humidity. That is, even when photographing the same equipment, changes in the thermal image occur if the amount of change in ambient temperature or humidity is large. Furthermore, they have the disadvantages of being expensive per unit and having low resolution (120 x 160 to 320 x 240). Prior art literature

[0009] KR 10-2110399 B The problem to be solved

[0010] The present invention is proposed to solve the technical problem described above and provides a deep learning-based thermal imaging camera image stabilization method utilizing environmental information for determining equipment overheating that can improve the accuracy of infrared camera images by utilizing external environment data, and a thermal imaging camera image stabilization system using the same. means of solving the problem

[0012] According to one embodiment of the present invention for solving the above problem, a method for stabilizing a thermal image camera image based on environmental information utilization for determining equipment overheating is provided, comprising: a step of acquiring a thermal image of equipment driven by power from a thermal image camera; a step of measuring the amount of current of the equipment through a current meter; a step of acquiring temperature data and humidity data of an area where the equipment is operating; and a correction step in which a pre-set thermal image correction deep learning model receives the thermal image and outputs a corrected thermal image, wherein the thermal image correction deep learning model receives a correction factor including the amount of current, the temperature data, and the humidity data to generate a corrected thermal image.

[0013] In addition, according to another embodiment of the present invention, a thermal imaging camera image stabilization system is provided, comprising: a thermal imaging camera for acquiring a thermal image of equipment driven by power; a current meter for measuring the amount of current of the equipment; a sensor for acquiring temperature data and humidity data of an area where the equipment is operated; and an image correction unit for generating a corrected thermal image by receiving a correction factor including the amount of current, the temperature data, and the humidity data from a pre-set thermal image correction deep learning model, wherein the thermal image correction deep learning model receives the thermal image and outputs a corrected thermal image.

[0014] In addition, the correction deep learning model of the image correction unit in the present invention is characterized by additionally receiving date and time information and utilizing it as a correction factor. Effects of the invention

[0016] The deep learning-based thermal imaging camera image stabilization method utilizing environmental information for determining equipment overheating according to the present invention and the thermal imaging camera image stabilization system using the same can improve the accuracy of infrared camera images by utilizing correction factors including external environmental data, namely current amount, temperature data, and humidity data. Brief explanation of the drawing

[0018] Figure 1 is an example of an image captured by a typical thermal imaging camera. FIG. 2 is a conceptual diagram of the thermal imaging camera image stabilization system (1) of the present invention. FIG. 3 is a configuration diagram of a thermal imaging camera image stabilization system (1) according to an embodiment of the present invention. Specific details for implementing the invention

[0019] Hereinafter, in order to explain in detail enough for a person skilled in the art to easily implement the technical concept of the present invention, embodiments of the present invention will be described with reference to the attached drawings.

[0021] FIG. 2 is a conceptual diagram of the thermal imaging camera image stabilization system (1) of the present invention, and FIG. 3 is a configuration diagram of the thermal imaging camera image stabilization system (1) according to an embodiment of the present invention.

[0023] The thermal imaging camera image stabilization system (1) according to the present embodiment includes only a brief configuration to clearly explain the technical concept to be proposed.

[0025] Referring to FIGS. 2 and 3, the thermal imaging camera image stabilization system (1) is configured to include electrical / electronic equipment (10), a thermal imaging camera (100), a sensor (200), a current meter (300), and an image correction unit (400).

[0027] The main operation of the thermal imaging camera image stabilization system (1) configured as described above is as follows.

[0028] The electrical / electronic equipment (10) is an electrically powered device, and heat is generated during operation, so the overheating / fire status of the electrical / electronic equipment (10) can be determined through the thermal image of the thermal imaging camera (100).

[0030] A thermal imaging camera (100) acquires a thermal image of electrical / electronic equipment (10) that is powered by electricity.

[0031] The sensor (200) acquires temperature and humidity data in the area where the electrical / electronic equipment (10) is operating, and may be composed of a thermometer and a hygrometer. A sensor that measures temperature and humidity simultaneously may be used.

[0032] The current meter (300) measures the current and power of the electrical / electronic equipment (10).

[0034] In particular, the image correction unit (400) is equipped with a pre-set thermal image correction deep learning model, and the pre-set thermal image correction deep learning model receives a thermal image as input and outputs a corrected thermal image.

[0035] That is, the thermal image correction deep learning model of the image correction unit (400) receives a correction factor including current amount, temperature data and humidity data and generates a corrected thermal image.

[0037] At this time, the correction deep learning model of the image correction unit (400) can receive additional date and time information and use it as a correction factor. That is, a corrected thermal image can be generated by considering the current amount, temperature data, and humidity data for each date and time information.

[0038] That is, the correction deep learning model of the image correction unit (400) is pre-learned through thermal images corresponding to current amount, temperature data, and humidity data for the electrical / electronic equipment (10), so it can generate a corrected thermal image in real time by considering current amount, temperature data, and humidity data for each date and time based on the learned information.

[0040] As described above, the deep learning-based thermal imaging camera image stabilization method utilizing environmental information for determining equipment overheating of the thermal imaging camera image stabilization system (1) is,

[0041] A step of acquiring a thermal image of power-driven equipment from a thermal imaging camera is performed.

[0042] In addition, a step is performed to measure the current of the equipment using a current meter.

[0043] In addition, a step of acquiring temperature and humidity data of the area where the equipment is operating is performed.

[0044] Finally, a pre-configured thermal image correction deep learning model receives a thermal image as input and outputs a corrected thermal image, and a correction step is performed in which the thermal image correction deep learning model receives a correction factor including current amount, temperature data, and humidity data to generate a corrected thermal image.

[0046] The deep learning-based thermal imaging camera image stabilization method utilizing environmental information for determining equipment overheating according to the present invention and the thermal imaging camera image stabilization system using the same can improve the accuracy of infrared camera images by utilizing correction factors including external environmental data, namely current amount, temperature data, and humidity data.

[0048] As such, those skilled in the art to which the present invention pertains will understand that the present invention may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and their equivalents should be interpreted as being included within the scope of the present invention. Explanation of the symbols

[0050] 10 : Electrical / Electronic Equipment 100 : Thermal imaging camera 200 : Sensor 300 : Current meter 400 : Image correction unit

Claims

Claim 1 A method for deep learning-based thermal imaging camera image stabilization utilizing environmental information for determining equipment overheating, comprising: a step of acquiring a thermal image of equipment driven by power from a thermal imaging camera; a step of measuring the amount of current of the equipment through a current meter; a step of acquiring temperature data and humidity data of an area where the equipment is operating; and a correction step in which a pre-set thermal image correction deep learning model receives the thermal image and outputs a corrected thermal image, wherein the thermal image correction deep learning model receives a correction factor including the amount of current, the temperature data, and the humidity data to generate a corrected thermal image. Claim 2 A method for deep learning-based thermal imaging camera image stabilization utilizing environmental information for determining equipment overheating, characterized in that, in the correction step of claim 1, the correction deep learning model additionally receives date and time information and utilizes it as a correction factor. Claim 3 A thermal imaging camera for acquiring a thermal image of equipment driven by power; a current meter for measuring the current amount of the equipment; a sensor for acquiring temperature data and humidity data of an area where the equipment is operating; and an image correction unit for generating a corrected thermal image by receiving a correction factor including the current amount, the temperature data, and the humidity data from a pre-set thermal image correction deep learning model, wherein the thermal image correction deep learning model receives the thermal image and outputs a corrected thermal image. Claim 4 A thermal imaging camera image stabilization system according to claim 3, characterized in that the correction deep learning model of the image correction unit receives additional date and time information and utilizes it as a correction factor.

Images