CN-114694249-B - Thermal energy-kinetic energy image data generation method and human body state detection method

CN114694249BCN 114694249 BCN114694249 BCN 114694249BCN-114694249-B

Abstract

The invention discloses a heat energy-kinetic energy image data generation method and a human body state detection method, wherein the heat energy-kinetic energy image data generation method comprises the steps of respectively and synchronously acquiring visible light images and infrared images with temperature values of the head and neck of a person to be detected in a quasi-static state at three front and rear moments with the same time interval, carrying out homonymous point registration on the two images at the same moment, acquiring heat energy change index matrixes representing heat energy change spatial distribution of all position points of the head and neck according to the infrared images in all registration image groups, respectively acquiring kinetic energy change index matrixes representing kinetic energy change spatial distribution of all position points of the head and neck according to the two images, acquiring target kinetic energy change index matrixes by utilizing the two acquired kinetic energy change index matrixes, and acquiring heat energy-kinetic energy image data according to the heat energy change index matrixes to generate heat energy-kinetic energy images. The invention does not contact human body, does not depend on operation specialty, can describe human body characteristics more comprehensively, and can detect human body state accurately and quickly.

Inventors

ZHOU CHONG
GAO FANQIN

Assignees

脑意(北京)智能生物技术有限公司

Dates

Publication Date: 20260505
Application Date: 20220309

Claims (7)

1. A thermal energy-kinetic energy image data generating method, characterized by comprising: Synchronously acquiring visible light images and infrared images of the head and neck of a person to be tested in a quasi-static state at front and back three moments with the same time interval, wherein each infrared image carries a temperature value of each position point of the head and neck; Carrying out homonymy point registration on the visible light image and the infrared image at the same moment to obtain registration image groups respectively corresponding to the moments; The method comprises the steps of respectively obtaining temperature values of all position points of the head and neck part carried by all infrared images in a registration image group corresponding to all moments, calculating the thermal energy change index of each position point based on thermal energy by utilizing the temperature values of the position point at the three moments for each position point, and obtaining a thermal energy change index matrix representing the thermal energy change spatial distribution of all position points of the head and neck part according to the spatial arrangement of the position points; The method comprises the steps of obtaining a dynamic change index of a position point corresponding to each position point of a head and neck part according to the spatial arrangement of the position point, obtaining the dynamic change index matrix of the head and neck part under the image type, representing the dynamic change spatial distribution of the position point by the dynamic change index matrix of the head and neck part under the image type, and obtaining the dynamic change index matrix of the two image types based on fusion processing; and obtaining thermal energy-kinetic energy image data according to the thermal energy change index matrix and the target kinetic energy change index matrix so as to generate a thermal energy-kinetic energy image by using the thermal energy-kinetic energy image data.
2. The thermal energy-kinetic energy image data generating method according to claim 1, wherein before the homonymous point registration of the visible light image and the infrared image at the same time, the method further comprises: And performing homonymous point registration on the infrared images of the head and neck of the to-be-detected person, which are acquired at every two adjacent moments and are in a quasi-static state, in the three front and rear moments.
3. A thermal energy-kinetic energy image data generation method according to claim 1 or 2, characterized in that the process of homonymous point registration comprises: And respectively inputting a first image serving as a reference image and a second image serving as an image to be registered into a first LIFT network and a second LIFT network which are trained in advance, obtaining a registered image which is output by the second LIFT network and is obtained by registering the second image by taking the first image as a reference, and taking the first image and the registered image as a registration result of the same name point, wherein the first LIFT network and the second LIFT network are trained by utilizing a pre-constructed improved MatchNet network according to a plurality of sample reference images and sample images to be registered.
4. A thermal energy-kinetic energy image data generating method according to claim 3, characterized in that the modified MatchNet network comprises: The device comprises a first LIFT network, a second LIFT network and a metric network, wherein the first LIFT network and the second LIFT network are connected in parallel, the output ends of the first LIFT network and the second LIFT network are connected with the metric network, the first LIFT network and the second LIFT network are respectively used for extracting feature vectors from an input image in a training stage, and the metric network is used for measuring similarity of the two feature vectors.
5. The thermal energy-kinetic energy image data generating method according to claim 1, wherein the generating a thermal energy-kinetic energy image using the thermal energy-kinetic energy image data includes: generating a head and neck heat energy change image by using the heat energy change index matrix; and displaying the numerical value in the target kinetic energy change index matrix around the head and neck of the head and neck heat energy change image in a visual display mode to obtain the heat energy-kinetic energy image.
6. A human body state detection method, comprising: obtaining thermal energy-kinetic energy image data and/or thermal energy-kinetic energy images of the head and neck of the person to be tested by using the thermal energy-kinetic energy image data generation method according to any one of claims 1-5; And detecting the physiological state or psychological state of the person to be detected by using the thermal energy-kinetic energy image data and/or the thermal energy-kinetic energy image.
7. The human body state detection device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; a memory for storing a computer program; A processor for implementing the method steps of claim 6 when executing a program stored on a memory.

Description

Thermal energy-kinetic energy image data generation method and human body state detection method Technical Field The invention belongs to the technical field of data processing, and particularly relates to a thermal energy-kinetic energy image data generation method and a human body state detection method. Background In medical research and in the auxiliary diagnosis of related diseases, some instruments are often used to detect the state of a human body by measuring physiological related parameters of the human body. For example, measuring an electrocardiogram, an electroencephalogram, etc. of a human body to detect a physiological state of the human body, and determining whether the subject is lying or not by recording a change in a physiological related parameter of stress when the subject is asked by a lie detector, etc. to realize a psychological state detection. However, the existing method generally needs to make human body contact when detecting human body state, and the detection process is often finished by relying on the professional knowledge of operators, and meanwhile, the obtained data information describing human body features is relatively one-sided, so that the method has certain limitation. Disclosure of Invention In order to solve the above problems in the prior art, the present invention provides a thermal energy-kinetic energy image data generating method, a human body state detecting method, and a human body state detecting apparatus. The technical problems to be solved by the invention are realized by the following technical scheme: In a first aspect, an embodiment of the present invention provides a thermal energy-kinetic energy image data generating method, including: Synchronously acquiring visible light images and infrared images of the head and neck of a person to be tested in a quasi-static state at front and back three moments with the same time interval, wherein each infrared image carries a temperature value of each position point of the head and neck; Carrying out homonymy point registration on the visible light image and the infrared image at the same moment to obtain registration image groups respectively corresponding to the moments; in the registration image group corresponding to all moments, a thermal energy change index matrix representing thermal energy change spatial distribution of each position point of the head and neck is obtained aiming at the infrared image type; in the registration image group corresponding to all moments, respectively obtaining a kinetic energy change index matrix representing the kinetic energy change spatial distribution of each position point of the head and neck part aiming at two image types, and obtaining a target kinetic energy change index matrix by utilizing the obtained two kinetic energy change index matrices; and obtaining thermal energy-kinetic energy image data according to the thermal energy change index matrix and the target kinetic energy change index matrix so as to generate a thermal energy-kinetic energy image by using the thermal energy-kinetic energy image data. In one embodiment of the present invention, before the co-name point registration is performed on the visible light image and the infrared image at the same time, the method further includes: And performing homonymous point registration on the infrared images of the head and neck of the to-be-detected person, which are acquired at every two adjacent moments and are in a quasi-static state, in the three front and rear moments. In one embodiment of the present invention, the process of homonymous point registration includes: And respectively inputting a first image serving as a reference image and a second image serving as an image to be registered into a first LIFT network and a second LIFT network which are trained in advance, obtaining a registered image which is output by the second LIFT network and is obtained by registering the second image by taking the first image as a reference, and taking the first image and the registered image as a registration result of the same name point, wherein the first LIFT network and the second LIFT network are trained by utilizing a pre-constructed improved MatchNet network according to a plurality of sample reference images and sample images to be registered. In one embodiment of the present invention, the improved MatchNet network comprises: The device comprises a first LIFT network, a second LIFT network and a metric network, wherein the first LIFT network and the second LIFT network are connected in parallel, the output ends of the first LIFT network and the second LIFT network are connected with the metric network, the first LIFT network and the second LIFT network are respectively used for extracting feature vectors from an input image in a training stage, and the metric network is used for measuring similarity of the two feature vectors. In one embodiment of the present invention, in the set of registration images cor