An indirect temperature measurement method, device and medium based on visual intelligent recognition

Abstract

The present application provides an indirect temperature measurement method, device and medium based on visual intelligent recognition. The method includes: taking an image of a temperature sensing device installed on an object to be measured, and simultaneously taking an infrared image of the object to be measured; wherein, the image of the temperature sensing device includes the image information corresponding to any one or more of the following: the temperature region of the object to be measured, the ambient temperature region, the ultraviolet intensity region, and the deformation region of the temperature sensing device; performing color recognition on the image of the temperature sensing device to obtain the first recognized temperature of the object to be measured; performing temperature recognition on the infrared image of the object to be measured to obtain the second recognized temperature of the object to be measured; and determining the actual temperature of the object to be measured based on the relationship between the first recognized temperature and the second recognized temperature. The method provided by the present application can greatly improve the accuracy of the test results of the long-distance temperature measurement method.

Description

Technical Field

[0001] This application relates to the field of intelligent temperature measurement technology, and in particular, to an indirect temperature measurement method, device, and medium based on visual intelligent recognition. Background Art

[0002] In dangerous places such as high-voltage areas and underground tunnels, it is necessary to measure the temperature of equipment at a distance. Inspection personnel need to respond in a timely manner according to the temperature changes of the equipment to avoid the occurrence of dangerous accidents.

[0003] At present, there are many temperature measurement technologies that can measure temperature at a distance. For example, using an infrared thermometer for temperature measurement, but this method has a high cost and is easily affected by environmental factors (ambient temperature, dust in the air, etc.). The temperature measurement accuracy for bright or polished metal surfaces is low, requiring manual inspection operations and lacking real-time performance, making it very easy to have a danger during the inspection interval.

[0004] Using an ultraviolet detection instrument to measure the ultraviolet intensity, this method requires the object to emit light for testing, and a special ultraviolet test instrument is needed, with a high cost. It also requires manual inspection and testing, with low efficiency and no real-time performance.

[0005] Traditional temperature-indicating wax piece contact temperature measurement technology has low efficiency and no real-time performance. It requires manual inspection and cannot detect abnormalities in a timely manner. Moreover, after the temperature-indicating wax piece melts and falls off due to heat, it is easy to cause accidents and has potential safety hazards. Temperature measurement patch contact temperature measurement remains a single color without fading after exceeding the temperature threshold, but it also has low efficiency and no real-time performance, requiring manual inspection and unable to detect abnormalities in a timely manner.

[0006] Measuring temperature through a sensor can detect the temperature of the device in real time and has the ability to transmit data. However, the sensor needs to be powered by a battery, and the battery has a lifespan problem. In addition, when a fire risk occurs, the battery will catch fire or explode, which is more likely to exacerbate the severity of the risk.

[0007] Measuring temperature through infrared imaging by a drone is greatly affected by the weather, affecting the accuracy of the test data. Factors such as flight distance, battery power, and remote control communication methods restrict the benefits of drone technology, and the cost is too high to be popularized.

[0008] Therefore, researching a long-distance non-contact temperature measurement method with low cost, safety, effectiveness, and high accuracy has become an urgent problem to be solved. Summary of the Invention

[0009] Aiming at the problems existing in the above-mentioned prior art, the embodiments of this application propose an indirect temperature measurement method, device, and medium based on visual intelligent recognition, which solves the technical problem of low accuracy of the test results of the existing long-distance temperature measurement methods.

[0010] On the one hand, an embodiment of the present application provides an indirect temperature measurement method based on visual intelligent recognition. The method includes: capturing an image of a temperature sensing device installed on a measured object, and simultaneously capturing an infrared image of the measured object; wherein, the image of the temperature sensing device includes image information corresponding to any one or more of the following: the temperature region of the measured object, the ambient temperature region, the ultraviolet intensity region, and the deformation region of the temperature sensing device; performing color recognition on the image of the temperature sensing device to obtain a first recognized temperature of the measured object; performing temperature recognition on the infrared image of the measured object to obtain a second recognized temperature of the measured object; and determining the actual temperature of the measured object based on the relationship between the first recognized temperature and the second recognized temperature.

[0011] In the embodiment of the present application, by simultaneously capturing an image of a temperature sensing device installed on a measured object and an infrared image of the measured object, and respectively recognizing two temperatures on the two images, and determining the temperature of the measured object based on these two temperatures, the accuracy of remote temperature measurement can be improved.

[0012] In a feasible implementation manner, the performing color recognition on the image of the temperature sensing device to obtain a first recognized temperature of the measured object specifically includes: performing image enhancement processing on the image of the temperature sensing device; determining the color information corresponding to the temperature region of the measured object of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the temperature region of the measured object of the temperature sensing device displays a preset color based on the temperature of the measured object; and determining the first recognized temperature of the measured object through a pre-stored first color correspondence relationship based on the color information corresponding to the temperature region of the measured object of the temperature sensing device; wherein the first color correspondence relationship is used to indicate the correspondence relationship between the temperature of the measured object and the color displayed by the temperature region of the measured object of the temperature sensing device.

[0013] In a feasible implementation manner, the determining the actual temperature of the measured object based on the first recognized temperature and the second recognized temperature specifically includes: calculating the mean and variance between the first recognized temperature and the second recognized temperature; in the case where the variance between the first recognized temperature and the second recognized temperature is less than a first preset threshold, determining the actual temperature of the measured object as the mean between the first recognized temperature and the second recognized temperature; or, in the case where the variance between the first recognized temperature and the second recognized temperature is greater than or equal to the first preset threshold, determining the actual temperature of the measured object as the first recognized temperature.

[0014] In the embodiment of the present application, the variances of two temperatures are calculated to balance the two identified temperatures. If the variances of the two temperatures are small, it indicates that the two temperatures are accurately identified, and then the average value of the two temperatures is directly used as the final actual temperature. If the variances of the two temperatures are large, it indicates that one of the temperatures is not accurately identified, and then the temperature value with higher identification accuracy is used as the final actual temperature. Thereby, the accuracy rate of temperature identification is improved.

[0015] In a feasible embodiment, after performing image enhancement processing on the image of the temperature sensing device, the method further includes: determining the color information corresponding to the ambient temperature region of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the ambient temperature region of the temperature sensing device displays a preset color based on the ambient temperature corresponding to the object to be measured; determining the ambient temperature corresponding to the object to be measured through a first color correspondence relationship based on the color information corresponding to the ambient temperature region of the temperature sensing device; wherein, the ambient temperature corresponding to the object to be measured is the temperature information within a region centered on the object to be measured and with a second preset threshold as the radius; in the case where the actual temperature of the object to be measured is greater than or equal to the ambient temperature corresponding to the object to be measured and less than a third preset threshold, determining that the object to be measured is in a normal working state; or, in the case where the actual temperature of the object to be measured is less than the ambient temperature corresponding to the object to be measured, determining that the object to be measured is in a non-working state; or, in the case where the actual temperature of the object to be measured is greater than or equal to the third preset threshold, determining that the object to be measured is in a dangerous state.

[0016] In a feasible embodiment, after performing image enhancement processing on the image of the temperature sensing device, the method further includes: determining the color information corresponding to the ultraviolet intensity region of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the ultraviolet intensity region of the temperature sensing device displays a preset color based on the ultraviolet intensity on the surface of the object to be measured; determining the ultraviolet intensity on the surface of the object to be measured through a pre-stored second color correspondence relationship based on the color information corresponding to the ultraviolet intensity region of the temperature sensing device; wherein, the second color correspondence relationship is used to indicate the correspondence relationship between the ultraviolet intensity on the surface of the object to be measured and the color displayed by the ultraviolet intensity region of the temperature sensing device.

[0017] In a feasible implementation, after determining the ultraviolet intensity on the surface of the object to be measured based on the color information corresponding to the ultraviolet intensity region of the temperature sensing device through a pre-stored second color correspondence relationship, the method further includes: positioning the temperature sensing device through GPS and Beidou dual-mode positioning technology to obtain the longitude and latitude information corresponding to the temperature sensing device; selecting a preset transmission method corresponding to the cloud platform; wherein, the preset transmission method includes at least any one or more of the following: WIFI transmission, ZIGBEE transmission, Bluetooth transmission, LORA transmission, 4G transmission, 5G transmission, Beidou short message transmission; in the case where the object to be measured is in a dangerous state and / or the ultraviolet intensity on the surface of the object to be measured is greater than or equal to a fourth preset threshold, sending the image of the temperature sensing device, the infrared image of the object to be measured, and the longitude and latitude information of the temperature sensing device to the cloud platform.

[0018] In the embodiment of the present application, by performing front-end analysis on the image and infrared image of the temperature sensing device, the analysis efficiency is improved, and a large amount of image transmission is not generated, saving traffic costs.

[0019] In a feasible implementation, after sending the image of the temperature sensing device, the infrared image of the object to be measured, and the longitude and latitude information of the temperature sensing device to the cloud platform, the method further includes: inputting the image of the temperature sensing device and the infrared image of the object to be measured into an image recognition model on the cloud platform to obtain the rechecked temperature of the object to be measured and the rechecked ultraviolet intensity on the surface of the object to be measured; in the case where the rechecked temperature is greater than or equal to a third preset threshold, sending a temperature alarm message to the mobile terminal corresponding to the maintenance personnel; in the case where the rechecked ultraviolet intensity is greater than or equal to a fourth preset threshold, sending an ultraviolet intensity alarm message to the mobile terminal corresponding to the maintenance personnel.

[0020] In the embodiment of the present application, by sending some images that can prove that the object to be measured is in a dangerous state to the cloud platform for rechecking, the false alarm rate can be reduced, and thus the recognition accuracy of the temperature of the object to be measured can be improved.

[0021] In a feasible implementation, after determining the actual temperature of the object to be measured, the method further includes: in the case where the actual temperature of the object to be measured is greater than or equal to a third preset threshold, the deformation region of the temperature sensing device deforms, and the color information corresponding to the temperature region of the object to be measured by the temperature sensing device remains unchanged until the recovery instruction corresponding to the temperature sensing device is triggered.

[0022] In the temperature sensing device according to the embodiment of the present application, after it is detected that the temperature of the object to be measured exceeds the preset threshold or the ultraviolet intensity exceeds the preset threshold, the corresponding temperature region of the object to be measured presents a specific color, the ultraviolet intensity region also presents a specific color, and the deformation region deforms. This color and shape will no longer change with the temperature change of the object to be measured, so that the maintenance personnel can go to the site to check the temperature sensing device after receiving the alarm.

[0023] In a second aspect, an indirect temperature measurement device based on visual intelligent recognition according to an embodiment of the present application includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to: capture an image of a temperature sensing device installed on an object to be measured, and simultaneously capture an infrared image of the object to be measured; wherein, the image of the temperature sensing device includes any one or more of the following corresponding image information: the temperature region of the object to be measured, the ambient temperature region, the ultraviolet intensity region, and the deformation region of the temperature sensing device; perform color recognition on the image of the temperature sensing device to obtain a first recognized temperature of the object to be measured; perform temperature recognition on the infrared image of the object to be measured to obtain a second recognized temperature of the object to be measured; and determine the actual temperature of the object to be measured based on the relationship between the first recognized temperature and the second recognized temperature.

[0024] In a third aspect, an embodiment of the present application further provides a storage medium for indirect temperature measurement based on visual intelligent recognition, including: the storage medium is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program, and each program includes instructions, and when the instructions are executed by a terminal, the terminal is enabled to execute an indirect temperature measurement method based on visual intelligent recognition according to any one of the above embodiments.

[0025] In the embodiment of the present application, the temperature of the object to be measured is recognized by identifying the color and shape of the temperature sensing device, and the average value is calculated with the temperature obtained from the infrared image, so as to accurately and effectively identify the temperature of the object to be measured, generate an alarm signal in time and report it actively, with high efficiency and real-time performance. The recognition result is obtained through front-end AI intelligent analysis and edge computing, and the recognition result is reviewed by an image recognition model on the cloud platform side, with high accuracy, and can be widely applied to many industries such as home appliance industry, power companies (power transmission, power generation and power distribution), railway companies, electronic industry, medicine and health, etc., and is especially suitable for temperature monitoring in dangerous places such as high voltage and underground tunnels. Description of the Drawings

[0026] The accompanying drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0027] Figure 1 It is a flowchart of an indirect temperature measurement method based on visual intelligent recognition provided by an embodiment of the present application;

[0028] Figure 2 It is a schematic diagram of an indirect temperature measurement device based on visual intelligent recognition provided by an embodiment of the present application;

[0029] Figure 3 It is a schematic diagram of an indirect temperature measurement device based on visual intelligent recognition provided by an embodiment of the present application. Specific embodiments

[0030] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with the specific embodiments of the present application and the corresponding accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0031] The embodiments of the present application provide an indirect temperature measurement method, device, and medium based on visual intelligent recognition. By installing a temperature sensing device with multiple functions such as temperature measurement and ultraviolet intensity measurement on the object to be measured, the intelligent recognition sensor captures the image of the temperature sensing device and the infrared image of the object to be measured in real time, and performs front-end intelligent recognition on the two images. Based on the two recognized temperatures, the actual temperature of the object to be measured is determined, and a review is performed through the cloud platform, which greatly improves the accuracy of remote temperature measurement and solves the technical problem of low accuracy of test results in existing long-distance temperature measurement methods. The technical solutions proposed in the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0032] Figure 1 It is a flowchart of an indirect temperature measurement method based on visual intelligent recognition provided by an embodiment of the present application. As Figure 1 shown, the method may include steps S101-S107:

[0033] S101. The intelligent recognition sensor captures the image of the temperature sensing device installed on the object to be measured at a preset time interval, and simultaneously captures the infrared image of the object to be measured.

[0034] Specifically, the ordinary camera of the intelligent recognition sensor captures images of the temperature sensing device installed on the object to be measured at preset time intervals. While the ordinary camera captures images of the temperature sensing device, the infrared camera of the intelligent recognition sensor captures infrared images of the object to be measured.

[0035] In one embodiment, the object to be measured can be equipment or objects that are not suitable for manual temperature measurement, such as dangerous equipment in a substation, high-voltage lines, and dangerous equipment in an underground tunnel. These dangerous equipment or objects require strict temperature control, otherwise dangerous situations such as fires are likely to occur. At the same time, excessive ultraviolet intensity can also harm objects made of plastic or rubber materials such as high-voltage lines, causing them to age rapidly. Therefore, it is necessary to monitor the ultraviolet intensity in real time and take remedial measures to slow down the aging speed of objects such as high-voltage lines.

[0036] S102. The intelligent recognition sensor performs image enhancement processing on the image of the temperature sensing device and the infrared image of the object to be measured.

[0037] Specifically, an interpolation algorithm based on the color ratio and color difference law is deployed on the intelligent recognition sensor, and the algorithm is used to perform image enhancement processing on the captured image of the temperature sensing device to improve the recognition accuracy of the image of the temperature sensing device. In addition, an artificial intelligence infrared image enhancement algorithm based on human vision is deployed on the intelligent recognition sensor, and the algorithm is used to perform image enhancement processing on the infrared image of the object to be measured to improve the recognition accuracy of the infrared image of the object to be measured.

[0038] S103. The intelligent recognition sensor recognizes the image of the temperature sensing device after image enhancement processing to obtain the actual temperature of the object to be measured, the ambient temperature corresponding to the object to be measured, the shape of the deformed area of the temperature sensing device, and the ultraviolet intensity on the surface of the object to be measured.

[0039] Specifically, the temperature sensing device includes four regions, namely the object temperature region, the ambient temperature region, the ultraviolet intensity region, and the deformation region. The temperature sensing device pre-stores a first color correspondence relationship between temperature and the display colors of the object temperature region and the ambient temperature region, and pre-stores a second color correspondence relationship between ultraviolet intensity and the display color of the ultraviolet intensity region. So that when the temperature sensing device monitors a specific temperature value, the object temperature region and the ambient temperature region of the object can display corresponding colors according to the first color correspondence relationship; or when the temperature sensing device monitors a specific ultraviolet intensity value, the ultraviolet intensity region can display the corresponding color according to the second color correspondence relationship. The intelligent recognition sensor also stores the above first color correspondence relationship and second color correspondence relationship.

[0040] The temperature region of the temperature sensing device for the object to be measured is used to display the color corresponding to the surface temperature of the object to be measured where the temperature sensing device is located. There is a heat insulation layer between the ambient temperature region and the object to be measured, which is used to display the color corresponding to the ambient temperature around the object to be measured, and the ambient temperature around the object to be measured is the temperature information within the region determined with the object to be measured as the center and the second preset threshold as the radius. The ultraviolet intensity region is used to display the color corresponding to the ultraviolet intensity on the surface of the object to be measured; the deformation region is of a preset shape, and after the temperature of the object exceeds the threshold, the preset shape will deform. For example, the preset shape of the deformation region is a circle, and after the temperature of the object exceeds 80 degrees, the circle of the deformation region will deform into other irregular shapes.

[0041] The intelligent recognition sensor performs color recognition on the image of the temperature sensing device after image enhancement processing to obtain the color of the temperature region of the object to be measured, the color of the ambient temperature region, and the color of the ultraviolet intensity region of the temperature sensing device. The intelligent recognition sensor obtains the first recognition temperature of the object to be measured according to the stored first color correspondence relationship and the color of the temperature region of the object to be measured; obtains the ambient temperature around the object to be measured according to the stored first color correspondence relationship and the color of the ambient temperature region; obtains the ultraviolet intensity on the surface of the object to be measured according to the stored second color relationship and the color of the ultraviolet intensity region. The intelligent recognition sensor performs shape recognition on the deformation region of the image of the temperature sensing device after image enhancement processing to obtain the shape of the deformation region.

[0042] Further, the intelligent recognition sensor performs temperature recognition on the infrared image of the object to be measured after image enhancement processing to obtain the second recognition temperature of the object to be measured.

[0043] Further, after the intelligent recognition sensor determines the first recognition temperature and the second recognition temperature, it can determine the actual temperature corresponding to the object to be measured according to the first recognition temperature and the second recognition temperature.

[0044] Specifically, first calculate the variance and mean between the first recognition temperature and the second recognition temperature. Then compare the calculated variance with the first preset threshold to determine the actual temperature of the object to be measured. Specifically, when the variance between the first recognition temperature and the second recognition temperature is less than the first preset threshold, determine the actual temperature of the object to be measured as the mean between the first recognition temperature and the second recognition temperature; or, when the variance between the first recognition temperature and the second recognition temperature is greater than or equal to the first preset threshold, determine the actual temperature of the object to be measured as the first recognition temperature.

[0045] For example, assume the first preset threshold is 5. If the variance between the first recognition temperature C1 and the second recognition temperature C2 is 3, and the variance between the first recognition temperature and the second recognition temperature is less than the first preset threshold, then the actual temperature of the object to be measured at this time is If the variance between the first recognized temperature C1 and the second recognized temperature C2 is 7, and the variance between the first recognized temperature and the second recognized temperature is greater than the first preset threshold, then the actual temperature of the object under test is C1.

[0046] S104. The intelligent recognition sensor determines the state of the object under test according to the temperature of the object under test, the ambient temperature corresponding to the object under test, the shape of the deformation area of the temperature sensing device, and the ultraviolet intensity on the surface of the object under test.

[0047] In an embodiment of the present application, the state of the object under test is determined by the magnitude relationship between the actual temperature of the object under test and the surrounding ambient temperature corresponding to the object under test.

[0048] Specifically, when the actual temperature of the object under test is less than the ambient temperature around the object under test, the object under test is in an unoperated state. When the actual temperature of the object under test is greater than or equal to the ambient temperature around the object under test and less than the third preset threshold, the object under test is in a normal operating state. And, when the actual temperature of the object under test is greater than or equal to the third preset threshold, and / or when the ultraviolet intensity on the surface of the object under test is greater than or equal to the fourth preset threshold, and / or when the deformation area of the temperature sensing device is deformed, the object under test is in a dangerous state.

[0049] S105. When the object under test is in a dangerous state, the intelligent recognition sensor sends the image of the temperature sensing device and the infrared image of the object under test to the cloud platform, and sends the corresponding warning information to the cloud platform.

[0050] Specifically, if the intelligent recognition sensor recognizes that the object under test is in a dangerous state, the intelligent recognition sensor sends the image of the corresponding temperature sensing device and the infrared image of the object under test to the cloud platform.

[0051] In an embodiment, when the actual temperature of the object under test is greater than or equal to the third preset threshold, the intelligent recognition sensor sends temperature warning information, the image of the corresponding temperature sensing device, and the infrared image of the object under test to the cloud platform. When the ultraviolet intensity on the surface of the object under test is greater than or equal to the fourth preset threshold, the intelligent recognition sensor sends ultraviolet intensity warning information and the image of the corresponding temperature sensing device to the cloud platform. When the deformation area of the temperature sensing device is deformed, the intelligent recognition sensor sends deformation warning information and the image of the corresponding temperature sensing device to the cloud platform.

[0052] S106. The cloud platform re-recognizes the received image of the temperature sensing device and the infrared image of the object under test to confirm the received warning information.

[0053] Specifically, the cloud platform receives the warning information sent by the intelligent recognition sensor and the image of the temperature sensing device or the infrared image of the object under test.

[0054] In one embodiment, if the received deformation warning information is received, the cloud platform inputs the image of the temperature sensing device into the image recognition model for re-identification. If the recognition result also indicates that the deformation area of the temperature sensing device has deformed, the cloud platform confirms that the deformation warning information is correct warning information.

[0055] If the received is temperature warning information, the cloud platform inputs the image of the temperature sensing device and the infrared image of the object under test into the image recognition model for re-identification, obtaining a third recognized temperature and a fourth recognized temperature. Calculate the variance and mean between the third recognized temperature and the fourth recognized temperature. If the variance between the third recognized temperature and the fourth recognized temperature is less than the first preset threshold, the rechecked temperature of the object under test is the mean of the third recognized temperature and the fourth recognized temperature. If the variance between the third recognized temperature and the fourth recognized temperature is greater than or equal to the first preset threshold, the rechecked temperature of the object under test is the third recognized temperature. If the obtained rechecked temperature is also greater than or equal to the third preset threshold, the cloud platform confirms that the temperature warning information is correct warning information.

[0056] If the received is ultraviolet intensity warning information, the cloud platform inputs the image of the temperature sensing device into the image recognition model for re-identification, obtaining a rechecked ultraviolet intensity. If the obtained rechecked ultraviolet intensity is also greater than or equal to the fourth preset threshold, the cloud platform confirms that the ultraviolet intensity warning information is correct warning information.

[0057] S107. The cloud platform sends the confirmed correct warning information to the mobile terminal of the maintenance personnel.

[0058] Specifically, after the cloud platform confirms that the warning information sent by the intelligent recognition sensor is correct warning information, it sends the correct warning information to the mobile terminal of the maintenance personnel to remind the maintenance personnel to go to the site to check the object under test and perform maintenance.

[0059] In one embodiment, when the actual temperature of the object under test is greater than or equal to the third preset threshold, the deformation area of the temperature sensing device deforms, and the temperature area of the object under test of the temperature sensing device shows a warning color. At this time, the shape of the deformation area of the temperature sensing device and the color of the temperature area of the object under test remain unchanged for the maintenance personnel to check the temperature sensing device on site. Until the maintenance personnel trigger the recovery button of the temperature sensing device at the site, the temperature sensing device returns to its original state.

[0060] The above is a method embodiment provided by the present application. Based on the same inventive concept, the present application embodiment also provides an indirect temperature measurement device based on visual intelligent recognition.

[0061] Figure 2 Schematic diagram of an indirect temperature measurement device based on visual intelligent recognition provided by the present application embodiment, asFigure 2 As shown in Figure 2 , the indirect temperature measurement device 200 based on visual intelligent recognition includes a temperature sensing device 210, an intelligent recognition sensor 220, and a cloud platform 230. The temperature sensing device 210 includes a temperature region 211 of the object to be measured, an ambient temperature region 212, an ultraviolet intensity region 213, and a deformation region 214.

[0062] Specifically, the temperature sensing device 210 is installed on the object to be measured, and the display interface of the temperature sensing device has four display regions. The temperature region 211 of the object to be measured is used to display the color corresponding to the real-time temperature of the object to be measured where the temperature sensing device is located. The ambient temperature region 212 is used to display the color corresponding to the real-time ambient temperature around the object to be measured. The ultraviolet intensity region 213 is used to display the color corresponding to the real-time ultraviolet intensity on the surface of the object to be measured. The deformation region 214 is a preset shape, and after the temperature of the object to be measured exceeds the threshold, the preset shape will deform.

[0063] The ordinary camera of the intelligent recognition sensor 220 is directly facing the display interface of the temperature sensing device 210, and takes images of the display interface of the temperature sensing device 210 based on a preset time interval. The infrared camera of the intelligent recognition sensor 220 is directly facing the object to be measured, and while the ordinary camera takes images of the display interface of the temperature sensing device 210, it takes infrared images of the object to be measured. The intelligent recognition sensor is also used to perform image enhancement and image recognition on the images of the display interface of the temperature sensing device 210 and the infrared images of the object to be measured, and determine the actual temperature of the object to be measured, the ultraviolet intensity on the surface of the object to be measured, and the shape of the deformation region of the temperature sensing device according to the recognition results. If the actual temperature of the object to be measured is greater than or equal to the third preset threshold, and / or the ultraviolet intensity on the surface of the object to be measured is greater than or equal to the fourth preset threshold, and / or the deformation region of the temperature sensing device deforms, then the corresponding images of the display interface of the temperature sensing device 210, the infrared images of the object to be measured, and the alarm information are sent to the cloud platform 230.

[0064] The cloud platform 230 is used to perform re-recognition on the images of the display interface of the temperature sensing device 210 and the infrared images of the object to be measured sent by the intelligent recognition sensor 220, confirm whether the alarm information is true, and if so, forward the alarm information to the mobile terminal of the maintenance personnel.

[0065] In addition, the embodiment of the present application also provides an indirect temperature measurement device based on visual intelligent recognition. Figure 3 The following is a schematic diagram of an indirect temperature measurement device based on visual intelligent recognition provided by the embodiment of the present application. As Figure 3 shown, the device includes:

[0066] At least one processor 301; and a memory 302 communicatively connected to the at least one processor; wherein,

[0067] The memory 302 stores instructions executable by at least one processor 301. The instructions are executed by at least one processor 301 to enable at least one processor 301 to:

[0068] Based on a preset time interval, capture an image of the temperature sensing device installed on the object to be measured, and at the same time capture an infrared image of the object to be measured, and perform image enhancement processing on the image of the temperature sensing device and the infrared image of the object to be measured respectively; wherein, the temperature sensing device includes a temperature region of the object to be measured, an ambient temperature region, an ultraviolet intensity region, and a deformation region;

[0069] Perform color recognition on the image of the temperature sensing device after image enhancement processing to obtain a first recognized temperature of the object to be measured; perform temperature recognition on the infrared image of the object to be measured after image enhancement processing to obtain a second recognized temperature of the object to be measured;

[0070] Based on the first recognized temperature and the second recognized temperature, determine the actual temperature of the object to be measured.

[0071] The embodiment of the present application also provides a storage medium for indirect temperature measurement based on visual intelligent recognition, including: the storage medium is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program, and each program includes instructions, and when the instructions are executed by the terminal, the terminal is enabled to execute the above-mentioned method for indirect temperature measurement based on visual intelligent recognition.

[0072] By integrating various test functions on the temperature sensing device in the embodiment of the present application, information such as the temperature of the object to be measured, the ambient temperature of the object to be measured, and the ultraviolet intensity on the surface of the object to be measured, which are likely to cause harm to the object to be measured, can be identified in a picture of the temperature sensing device, improving the efficiency of hazard prevention. At the same time, by using the mutual corroboration of these two types of images, namely the ordinary image and the infrared image, the recognized temperature is more accurate, and through the review of the cloud platform, the accuracy of the test results of the long-distance temperature measurement method can be greatly improved.

[0073] The above is only a preferred embodiment of the present application and is not intended to limit the present application. It should be noted that for those of ordinary skill in the art, the present application may have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made without departing from the principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. An indirect temperature measurement method based on visual intelligent recognition, characterized in that The method includes: Taking an image of a temperature sensing device installed on the object to be measured, and simultaneously taking an infrared image of the object to be measured; wherein, the image of the temperature sensing device includes image information corresponding to any one or more of the following: the temperature region of the object to be measured, the ambient temperature region, the ultraviolet intensity region, and the deformation region of the temperature sensing device; Performing color recognition on the image of the temperature sensing device to obtain a first recognized temperature of the object to be measured, specifically including: Performing image enhancement processing on the image of the temperature sensing device; Determining the color information corresponding to the temperature region of the object to be measured of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the temperature region of the object to be measured of the temperature sensing device displays a preset color based on the temperature of the object to be measured; Based on the color information corresponding to the temperature region of the object to be measured of the temperature sensing device, determining the first recognized temperature of the object to be measured through a pre-stored first color correspondence; wherein the first color correspondence is used to indicate the correspondence between the temperature of the object to be measured and the color displayed by the temperature region of the object to be measured of the temperature sensing device; Performing temperature recognition on the infrared image of the object to be measured to obtain a second recognized temperature of the object to be measured; Determining the actual temperature of the object to be measured based on the relationship between the first recognized temperature and the second recognized temperature.

2. The indirect temperature measurement method based on visual intelligent recognition according to claim 1, wherein, The determining the actual temperature of the object to be measured based on the first recognized temperature and the second recognized temperature specifically includes: Calculating the mean and variance between the first recognized temperature and the second recognized temperature; When the variance between the first recognized temperature and the second recognized temperature is less than a first preset threshold, determining that the actual temperature of the object to be measured is the mean between the first recognized temperature and the second recognized temperature; or, When the variance between the first recognized temperature and the second recognized temperature is greater than or equal to the first preset threshold, determining that the actual temperature of the object to be measured is the first recognized temperature.

3. The indirect temperature measurement method based on visual intelligent recognition according to claim 1, characterized in that After performing image enhancement processing on the image of the temperature sensing device, the method further includes: Determining the color information corresponding to the ambient temperature region of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the ambient temperature region of the temperature sensing device displays a preset color based on the ambient temperature corresponding to the object to be measured; Based on the color information corresponding to the ambient temperature region of the temperature sensing device, determining the ambient temperature corresponding to the object to be measured through the first color correspondence; wherein, the ambient temperature corresponding to the object to be measured is the temperature information within a region determined with the object to be measured as the center and a second preset threshold as the radius; the first color correspondence is also used to indicate the correspondence between the ambient temperature of the object to be measured and the color displayed by the ambient temperature region of the temperature sensing device; When the actual temperature of the object to be measured is greater than or equal to the ambient temperature corresponding to the object to be measured and less than a third preset threshold, determining that the object to be measured is in a normal working state; or, When the actual temperature of the object to be measured is less than the ambient temperature corresponding to the object to be measured, determining that the object to be measured is in a non-working state; or, When the actual temperature of the object under test is greater than or equal to a third preset threshold, it is determined that the object under test is in a dangerous state.

4. The indirect temperature measurement method based on visual intelligent recognition according to claim 3, wherein After performing image enhancement processing on the image of the temperature sensing device, the method further includes: Determining the color information corresponding to the ultraviolet intensity region of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the ultraviolet intensity region of the temperature sensing device displays a preset color based on the ultraviolet intensity on the surface of the object under test. Based on the color information corresponding to the ultraviolet intensity region of the temperature sensing device, determining the ultraviolet intensity on the surface of the object under test through a pre-stored second color correspondence; wherein, the second color correspondence is used to indicate the correspondence between the ultraviolet intensity on the surface of the object under test and the color displayed by the ultraviolet intensity region of the temperature sensing device.

5. The indirect temperature measurement method based on visual intelligent recognition according to claim 4, wherein After determining the ultraviolet intensity on the surface of the object under test based on the color information corresponding to the ultraviolet intensity region of the temperature sensing device through a pre-stored second color correspondence, the method further includes: Positioning the temperature sensing device through GPS and Beidou dual-mode positioning technology to obtain the longitude and latitude information corresponding to the temperature sensing device. Selecting a preset transmission method corresponding to the cloud platform; wherein, the preset transmission method includes at least any one or more of the following: WIFI transmission, ZIGBEE transmission, Bluetooth transmission, LORA transmission, 4G transmission, 5G transmission, Beidou short message transmission. When the object under test is in a dangerous state, and / or when the ultraviolet intensity on the surface of the object under test is greater than or equal to a fourth preset threshold, sending the image of the temperature sensing device, the infrared image of the object under test, and the longitude and latitude information of the temperature sensing device to the cloud platform.

6. The indirect temperature measurement method based on visual intelligent recognition according to claim 5, wherein After sending the image of the temperature sensing device, the infrared image of the object under test, and the longitude and latitude information of the temperature sensing device to the cloud platform, the method further includes: Inputting the image of the temperature sensing device and the infrared image of the object under test into an image recognition model on the cloud platform to obtain the rechecked temperature of the object under test and the rechecked ultraviolet intensity on the surface of the object under test. When the rechecked temperature is greater than or equal to the third preset threshold, sending temperature alarm information to the mobile terminal corresponding to the maintenance personnel; and When the rechecked ultraviolet intensity is greater than or equal to the fourth preset threshold, sending ultraviolet intensity alarm information to the mobile terminal corresponding to the maintenance personnel.

7. The indirect temperature measurement method based on visual intelligence recognition according to claim 1, characterized in that After determining the actual temperature of the object under test, the method further includes: When the actual temperature of the object under test is greater than or equal to the third preset threshold, determining that the deformation region of the temperature sensing device deforms, and the color information corresponding to the temperature region of the object under test of the temperature sensing device remains unchanged until the recovery instruction corresponding to the temperature sensing device is triggered.

8. An indirect temperature measurement device based on visual intelligent recognition, characterized in that, Including: At least one processor; And a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute: Capture an image of the temperature sensing device installed on the object to be measured, and at the same time capture an infrared image of the object to be measured; wherein, the image of the temperature sensing device includes the image information corresponding to any one or more of the following: the temperature region of the object to be measured, the ambient temperature region, the ultraviolet intensity region, and the deformation region of the temperature sensing device; Perform color recognition on the image of the temperature sensing device to obtain the first recognized temperature of the object to be measured, specifically including: Perform image enhancement processing on the image of the temperature sensing device; Determine the color information corresponding to the temperature region of the object to be measured of the temperature sensing device in the image of the temperature sensing device after the image enhancement processing; wherein, the temperature region of the object to be measured of the temperature sensing device displays a preset color based on the temperature of the object to be measured; Based on the color information corresponding to the temperature region of the object to be measured of the temperature sensing device, determine the first recognized temperature of the object to be measured through a pre-stored first color correspondence relationship; wherein the first color correspondence relationship is used to indicate the correspondence relationship between the temperature of the object to be measured and the color displayed by the temperature region of the object to be measured of the temperature sensing device; Perform temperature recognition on the infrared image of the object to be measured to obtain the second recognized temperature of the object to be measured; Determine the actual temperature of the object to be measured based on the relationship between the first recognized temperature and the second recognized temperature.

9. A storage medium for indirect temperature measurement based on visual intelligent recognition, characterized in that, Include: The storage medium is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program. Each program includes instructions that, when executed by the terminal, cause the terminal to execute a method for indirectly measuring temperature based on visual intelligent recognition according to any one of claims 1-7.

Images