KR20210106215A - Artificial intelligence heat detector by thermal imgae, thermal image heat detection system with the same and method for detecting the heat - Google Patents

Abstract

Disclosed are an artificial intelligence thermal imaging thermal detection system comprising a long-wavelength infrared high-density array sensor, and a detection method thereof. An artificial intelligence thermal imaging thermal detection system according to one embodiment of the present disclosure comprises: one or more artificial intelligence object to be measured detection parts composed of a convolutional neural network using a convolution filter; and one or more object to be measured heating output parts that outputs heating of the object to be measured. Therefore, the present invention is capable of allowing heating to be easily measured.

Description

Artificial intelligence thermal imager, thermal imaging system including same, and detection method { ARTIFICIAL INTELLIGENCE HEAT DETECTOR BY THERMAL IMGAE, THERMAL IMAGE HEAT DETECTION SYSTEM WITH THE SAME AND METHOD FOR DETECTING THE HEAT }

Embodiments of the present invention relate to thermal sensing technology in thermal sensing systems.

Long-wavelength infrared in the 10um band has temperature information of the object emitting it.

A single sensor or array sensor is used to measure the amount of long-wavelength infrared light. The single sensor method uses one infrared receiving sensor and outputs the average amount of infrared rays on the measurement surface of the object as an electric signal.

The array sensor method uses an array sensor composed of two or more single sensor elements, and in the case of a high-density array sensor, the measurement area occupied by each sensor element is very small, so it also serves as an image sensor. The image output of a long-wavelength infrared meter using such a device looks like a color or black-and-white video, so it is called a thermal imaging camera.

The heat detector uses the long-wavelength infrared amount meter technology and has the advantage of being able to easily measure the heat of the subject in a non-contact manner.

When measuring the heat of a subject with a heat sensor, the subject must be located in the measurement area of the heat sensor. To this end, the object to be measured may be moved to move the thermal sensor, or the object to be measured may be moved to enter the measurement area.

On the other hand, when the subject enters the measurement area, the measurer finds the subject in the thermal image and grasps its exothermic temperature value. The exothermic temperature value of the subject is output as a number or a specific color. This measurement method causes fatigue in the measurer, and there is a problem in that it is difficult to recognize heat when a plurality of subjects move quickly.

Korean Patent Publication No. 1017121360000 (2017.02.24) A method and apparatus for detecting fainting of an object using a thermal imaging camera is a technology related to a method and apparatus for detecting fainting of an object using a thermal imaging camera. The technique is characterized in that the object is extracted by the difference between two consecutive frames, the object is detected and tracked using a Kalman filter, and the trajectory and size change of the object are detected.

SUMMARY OF THE INVENTION An embodiment of the present invention is to provide a technique for automatically finding a measurement target in a thermal image.

An artificial intelligence thermal imaging thermal sensing system according to an embodiment of the present disclosure includes a thermal imaging thermal sensing unit for measuring the amount of radiated infrared rays of a target object with a long-wavelength infrared high-density array sensor; and an artificial intelligence object detection unit for detecting the object from the measured thermal image and a heat output unit for outputting the sensed object on the screen.

The object sensing unit may detect the object to be measured in the thermal image and output coordinates thereof.

The object heat output unit may output heat information of the object as an image using coordinates of the object received from the sensor.

According to the disclosed embodiment, by automatically detecting a target object and outputting heat information as an image, it is possible to reduce fatigue of a person and facilitate measurement of heat when a plurality of target objects move quickly.

1 is a view showing the configuration of an artificial intelligence thermal imaging thermal sensing system according to an embodiment of the present invention;
2 is a view showing the configuration of a thermal imaging thermal sensing unit;
3 is a diagram showing the configuration of an artificial intelligence target detection unit;
4 is a view showing the configuration of a heat generating output unit to be measured;
5 is a heat output image of an artificial intelligence thermal imaging thermal sensing system according to an embodiment of the present invention;
6 is a heat output image of a thermal imaging thermal sensing system in which the present invention is not implemented

2 is a diagram illustrating the configuration of a thermal imager. Referring to FIG. 2 , the thermal imaging unit measures the amount of radiated infrared rays of the object to be measured with a long-wavelength infrared high-density array sensor and outputs it in the form of an image. The output format outputs a digitized value for each pixel constituting the array sensor.

3 is a diagram illustrating the configuration of an artificial intelligence target sensing unit. The shape of the object to be learned previously learned from the input image is detected and the coordinates of a rectangle in contact with the outermost part of the object are output. In the embodiment of the present invention, facial heat detection is performed, but any surrounding object whose external shape does not change after AI prior learning can be the measured object. The artificial intelligence unit for detecting the shape of the object uses a convolutional neural network using a convolution filter. It is determined that a detailed description of the convolutional neural network may unnecessarily obscure the gist of the present invention, so a detailed description thereof will be omitted.

4 shows the configuration of a heat generating output unit to be measured. It receives the coordinates of the rectangle in contact with the outermost part of the object in the image, and outputs pixels above the specified heating temperature in color and the remaining pixels in black and white. Also, a rectangle that is in contact with the outermost part of the object is output in color.

In the above, the face heat detection embodiment has been described in detail as a representative embodiment of the present invention. It will be appreciated that various modifications are possible. Therefore, the scope of the present invention should not be limited to the described embodiments and should be defined by the claims described in this document as well as the claims and equivalents.

Claims (3)

A thermal imaging thermal sensing system comprising a thermal imaging thermal sensing unit comprising one or more long wavelength infrared high-density array sensors.
A thermal imaging thermal sensing system comprising one or more artificial intelligence subject detection units configured with a convolutional neural network using a convolution filter.
A thermal imaging thermal sensing system comprising one or more object heating output units that output pixels above the specified heating temperature in color and black and white for pixels below the specified heating temperature using the artificial intelligence target detection unit output.

Images