WO2022061899A1

WO2022061899A1 - Infrared image processing method and device, and infrared camera

Info

Publication number: WO2022061899A1
Application number: PCT/CN2020/118439
Authority: WO
Inventors: 张青涛; 庹伟; 陈星�
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2020-09-28
Filing date: 2020-09-28
Publication date: 2022-03-31
Also published as: CN113692601A

Abstract

Disclosed in embodiments of the present application is an infrared image processing method, comprising: obtaining at least two infrared images, the at least two infrared images corresponding to a same target, and a grayscale value and the temperature of each infrared image having a preset mapping relationship; and stitching the at least two infrared images to obtain a first target infrared image, before the stitching, the mapping relationships of the at least two infrared images being the same, and the first target infrared image being used for temperature measurement. The method provided by the embodiments of the present application solves the problems that an infrared image obtained by stitching cannot be used for temperature measurement and has obvious visual jump.

Description

Infrared image processing method, device and infrared camera

technical field

The present application relates to the technical field of infrared image processing, and in particular, to an infrared image processing method, a device, an infrared camera, and a computer-readable storage medium.

Background technique

Infrared imaging technology is widely used to measure the temperature of objects. In some cases, because the target is too large, it is necessary to take multiple infrared images of the target to cover the target. In order to facilitate the user to locate the corresponding position of each infrared image in practice, multiple infrared images can be stitched together. However, at present, the large field of view infrared images obtained by splicing cannot be used for temperature measurement, and there are obvious visual jumps.

SUMMARY OF THE INVENTION

In view of this, embodiments of the present application provide an infrared image processing method, device, infrared camera, and computer-readable storage medium, so as to solve the problem that the infrared image obtained by splicing cannot be temperature-measured and has obvious visual jumps.

A first aspect of the embodiments of the present application provides an infrared image processing method, including:

acquiring at least two infrared images, wherein the at least two infrared images correspond to the same target, and the grayscale values of the infrared images have a preset mapping relationship with the temperature;

The at least two infrared images are stitched to obtain a first target infrared image, wherein, before the stitching is performed, the mapping relationships of the at least two infrared images are the same, and the first target infrared image is used for temperature.

A second aspect of an embodiment of the present application provides an infrared image processing apparatus, including: a processor and a memory storing instructions; the processor implements the following operations when executing the instructions:

A third aspect of the embodiments of the present application provides an infrared camera, including: an infrared lens, an infrared sensor, a processor, and a memory storing instructions;

The infrared sensor is used for collecting the original infrared image through the infrared lens;

The processor, when executing the instructions, performs the following operations:

A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores instructions, and when the instructions are executed by a processor, implement the infrared image processing method provided in the first aspect.

In the related art, since the multiple infrared images used for splicing have undergone different degrees of contrast stretching, these infrared images no longer have the same mapping relationship between gray value and temperature. Infrared images with a large field of view cannot be used for temperature measurement or the temperature measurement accuracy is very low. However, in the method provided by the embodiment of the present application, since the mapping relationship between the grayscale value and the temperature between the infrared images is the same before the splicing, the first target infrared image obtained after splicing can also apply a unified mapping relationship. , that is, the gray value of each pixel in the first target infrared image can be converted into a corresponding temperature by using the unified mapping relationship, so as to realize temperature measurement.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative labor.

FIG. 1 is a flowchart of an infrared image processing method provided by an embodiment of the present application.

FIG. 2 is another flowchart of the infrared image processing method provided by the embodiment of the present application.

FIG. 3 is a structural diagram of an infrared image processing apparatus provided by an embodiment of the present application.

FIG. 4 is a structural diagram of an infrared camera provided by an embodiment of the present application.

detailed description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of 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.

Infrared thermal imaging technology is a technology that converts infrared rays radiated by objects into infrared images through photoelectric conversion. It is based on the corresponding relationship/mapping relationship between the infrared radiation intensity of objects and the surface temperature of objects. Therefore, it can be measured by measuring the infrared rays of objects. Radiation intensity to measure the surface temperature of an object.

An infrared imaging system, such as an infrared thermal imager, an infrared camera and other equipment, may include an infrared lens and an infrared sensor (sensor). The infrared sensor can receive infrared signals through an infrared lens, and by photoelectric conversion of the infrared signals, an infrared image can be obtained. The infrared image is an image directly output by the sensor, which can be called a raw infrared image or a raw image.

Due to the insufficient response rate of the infrared sensor, when imaging a normal temperature scene, the obtained infrared image will have the problem of "high background and low contrast", that is, the background occupies a large grayscale range, while the target occupies a small grayscale. Therefore, if the original infrared image is directly provided to the user for viewing, it will be difficult for the user to recognize the target in the image. Therefore, the original infrared image collected by the sensor can usually be input to the processor, and the processor performs image enhancement processing on the original infrared image, so that the infrared image has better viewing quality. Here, the image enhancement processing performed by the processor may include contrast stretching. Through the contrast stretching, the grayscale range corresponding to the target in the infrared image can be enlarged, so that the user can clearly see and identify the target.

When using infrared imaging technology to measure the temperature of a target, in some cases, because the target is too large, taking only one image cannot cover the entire target, so it is necessary to take multiple infrared images of the target. However, in this case, since each infrared image only corresponds to a part of the target, the user cannot easily locate the part of the target that each infrared image corresponds to, especially the part and part of the target. When the recognition is not high.

For example, when performing temperature detection on photovoltaic panels, there are usually multiple photovoltaic panels, and these photovoltaic panels are regularly arranged in the field. If all photovoltaic panels are used as the temperature measurement target, multiple infrared images need to be taken to cover the entire photovoltaic panel site. At this time, since each photovoltaic panel is similar, it will be easy to distinguish without special markings. Which photovoltaic panel or panels each infrared image corresponds to, which may cause some photovoltaic panels to be missed, or cause some photovoltaic panels to be repeatedly temperature-measured.

In one embodiment, in response to the above problems, each captured infrared image can be spliced, so that each corresponding local infrared image can be merged into an infrared image with a large field of view, and based on the infrared image of the large field of view, Users can easily find out whether the target has an abnormal temperature and the specific location of the target.

But the above splicing also brings some problems. Since the infrared images used for splicing have undergone contrast stretching respectively during image processing, and the degree of stretching of each infrared image is usually different, the infrared image with a large field of view obtained after splicing/fusion will have Obvious visual jump, obvious splicing traces. In addition, due to the different stretching degree of each infrared image, the mapping relationship between the gray value and temperature of each infrared image is also different. Therefore, the large field of view infrared image obtained after splicing/fusion cannot be applied to a unified The mapping relationship cannot be used for temperature measurement.

Based on the above problems, an embodiment of the present application provides an infrared image processing method, which can be applied to an infrared image processing device, an infrared imaging device such as an infrared camera, an infrared thermal imager, or an infrared imaging system. , can also be applied to mobile platforms equipped with infrared sensors, such as drones equipped with infrared sensors.

Referring to FIG. 1, FIG. 1 is a flowchart of an infrared image processing method provided by an embodiment of the present application, and the method may include the following operations:

S110: Acquire at least two infrared images whose grayscale values and temperature have a preset mapping relationship.

Wherein, the at least two infrared images correspond to the same target.

S120, stitching the at least two infrared images to obtain a first target infrared image for temperature measurement.

Wherein, before performing the splicing, the mapping relationships of the at least two infrared images are the same.

In an embodiment, the preset mapping relationship of the at least two infrared images may be calibrated in advance.

For the acquired at least two infrared images, they may be infrared images corresponding to the same target. The target is the object to be temperature-measured, which can be an object, such as any object such as a transformer, a photovoltaic panel, and a power transmission line. When the target is an object, each acquired infrared image may correspond to a part of the object. Correspondingly, in an example, the first target infrared image obtained after splicing may correspond to the entire object. The target can also be a scene or site, such as the photovoltaic panel site example cited above, such as an entire substation. When the target is a scene or field, each acquired infrared image may correspond to a part of the scene or field. Correspondingly, in an example, the first infrared image of the target may correspond to the entire scene or field.

At least two infrared images corresponding to the same target can be stitched together. Here, in one embodiment, at least two infrared images may have overlapping parts or overlapping pixels, so that the stitching between infrared images can be achieved by performing region matching on the at least two infrared images /Fusion to get the first target infrared image.

For the pixels overlapping between infrared images, there are many ways to fuse them. If the first pixel point is any overlapping pixel point between infrared images, for the gray value of the first pixel point after fusion, in one embodiment, the first pixel point may be in the at least one The maximum gray value between the two infrared images is used as the gray value after fusion. For example, for example, the first pixel is an overlapping pixel between infrared image A and infrared image B. In infrared image A, the gray value corresponding to the first pixel is 255. The gray value corresponding to the pixel point is 155, then the maximum gray value of 255 can be taken as the gray value of the first pixel point after fusion, that is, the gray value of the first pixel point in the first target infrared image.

The above implementation of taking the maximum gray value as the gray value of the first pixel point after fusion can reflect the highest temperature of the actual position corresponding to the first pixel point, compared with the calculation of the average value in the prior art. The method is more realistic and more conducive to the early warning of temperature.

For the overlapping pixels between infrared images, in an implementation manner, weighted fusion may also be performed on the overlapping pixels. The above example of infrared image A and infrared image B can continue to be used for description. In this embodiment, if the fusion weight corresponding to infrared image A is 0.6, and the fusion weight corresponding to infrared image B is 0.4, the infrared image A and infrared image B can be used. For the respective fusion weights of the infrared images B, the gray value of the first pixel point after fusion is calculated as 0.6*255+0.4*155=215.

For the fusion weights corresponding to different infrared images, there are many ways to determine them. In one embodiment, the fusion weights corresponding to different infrared images may be preset. In an implementation manner, the corresponding fusion weight may be determined according to the content corresponding to the infrared image. For example, infrared image A and infrared image B are obtained by photographing a transformer, wherein the main body in infrared image A corresponds to the bushing of the transformer, and the main body in infrared image B corresponds to the oil tank of the transformer. The probability is higher, so a higher fusion weight can be determined for the infrared image A. In one embodiment, a fusion weight corresponding to the infrared image may also be determined according to the incident angle of the thermal radiation corresponding to the infrared image, and the overlapping pixels between at least two infrared images are fused using the fusion weight.

Different incident angles of thermal radiation will affect the accuracy of the infrared sensor's sensing radiation intensity. For example, in the example of shooting a photovoltaic panel, if the shooting angle is opposite to the front of the photovoltaic panel, the thermal radiation of the photovoltaic panel will be orthographically projected onto the infrared sensor through the infrared lens. According to the radiation angle, the radiation intensity sensed by the infrared sensor is the closest to the real value, and the gray value of the generated infrared image is also the closest to the gray value corresponding to the actual temperature. If the shooting angle is not facing the front of the photovoltaic panel, for example, it is opposite to the side of the photovoltaic panel, the incident angle of the thermal radiation will deviate from the orthographic angle, so that the radiation intensity sensed by the infrared sensor has a relatively large error, thus affecting the infrared The accuracy of the grayscale values of the image. Therefore, in one embodiment, the fusion weight corresponding to the infrared image may be negatively correlated with the angle difference, where the angle difference refers to the difference between the incident angle of the thermal radiation and the orthophoto angle, in other words, the infrared image The smaller the difference between the incident angle and the ortho-radiation angle of the corresponding thermal radiation, the larger the fusion weight corresponding to the infrared image can be.

In an implementation manner, the at least two infrared images may also have no overlapping pixels, for example, the edges to be spliced corresponding to the at least two infrared images may be exactly continuous (that is, the at least two infrared images There are no gaps between pixels), thus, splicing/fusion can also be performed according to the shooting angle of each infrared image to obtain the first target infrared image. However, if there is a gap of pixels between the at least two infrared images, it cannot be directly spliced. However, in an embodiment, if the size of the gap is within a preset range, the gap can be filled by an interpolation algorithm. Splicing again. Of course, there are other splicing methods, for example, the splicing/fusion of the at least two infrared images can also be realized based on the SIFT (Scale Invariant Feature Transform) algorithm.

It should be noted that the first target infrared image obtained by splicing can be a large field of view image of a two-dimensional plane, or a large field of view image with a spatial structure, such as a curved surface image, such as a spherical image, a cylindrical image, etc. , the embodiments of the present application are not limited.

In the related art, since the multiple infrared images used for splicing have undergone different degrees of contrast stretching, these infrared images no longer have the same mapping relationship between gray value and temperature. Infrared images with a large field of view cannot apply a unified mapping relationship and cannot be used for temperature measurement. However, in the method provided by the embodiment of the present application, since the mapping relationship between the grayscale value and the temperature between the infrared images is the same before the splicing, the first target infrared image obtained after the splicing can apply a unified mapping relationship, That is, the gray value of each pixel in the first target infrared image can be converted into a corresponding temperature by using the unified mapping relationship, so as to realize temperature measurement.

The mapping relationship between the gray value and the temperature between the infrared images used for splicing is the same, which can correspond to various situations. In one example, the infrared image used for stitching may be the image data directly output (direct output) from the infrared sensor, that is, the original infrared image (raw image). Since each infrared image used for splicing is a raw image, the mapping relationship between the infrared images is the same.

In an example, the infrared image used for stitching can be an infrared image obtained after performing specific preprocessing on the original infrared image (raw image after preprocessing). Of course, the specific preprocessing here means that the grayscale does not change. The preprocessing of the mapping relationship between the degree value and the temperature, for example, the preprocessing of correcting the original infrared image and/or the preprocessing of removing blemishes. The preprocessing for correction may include correcting the responsivity of the infrared sensor, and/or correcting the offset of the infrared sensor, and so on. The preprocessing for defect removal may include performing dead pixel removal on the original infrared image, and/or performing noise removal on the original infrared image.

In one example, each infrared image used for stitching may be an infrared image obtained after undergoing the same contrast stretching, where the same contrast stretching includes the same linear or non-linear contrast stretching. Since the contrast stretching performed on each infrared image is the same, the mapping relationship between the gray value and the temperature between the infrared images is still the same. Therefore, the first target infrared image obtained by stitching these infrared images A unified mapping relationship can still be applied, and the grayscale value of each pixel point of the first target infrared image can be converted into a corresponding temperature by using the mapping relationship to realize temperature measurement. Moreover, since the contrast stretching of each infrared image is the same, the contrast of the first target infrared image is still on the unified benchmark, and there will be no obvious visual jump.

For the convenience of description, the infrared image after the same contrast stretching is referred to as the first infrared image below. When determining the temperature corresponding to the grayscale value of the first infrared image, in one embodiment, the degree of stretching that the first infrared image has undergone before can be characterized by a stretching coefficient, so that the stretching coefficient to restore the first infrared image back to the infrared image before stretching, and the temperature corresponding to each pixel in the first infrared image can be determined according to the preset mapping relationship between the grayscale value and the temperature. In an embodiment, the preset mapping relationship can also be adjusted according to the stretching coefficient, and the temperatures corresponding to different grayscale values can be changed to obtain a new mapping relationship, so that the new mapping relationship can be directly Apply to the first infrared image.

No matter which infrared images used for stitching correspond to any of the above examples or embodiments, the mapping relationship between grayscale values and temperatures is the same among the infrared images used for stitching.

After the infrared image of the first target is obtained, the temperature corresponding to each pixel in the infrared image of the first target can be calculated according to the uniform mapping relationship between the gray value and the temperature, so that the temperature corresponding to each pixel can be used to determine the first target The temperature of the target object in the infrared image.

It should be noted that although the first target infrared image can be used for temperature measurement, the first target infrared image has not undergone image enhancement processing, so it still has the problem of "high background and low contrast", which is not suitable for users to watch. Therefore, in one embodiment, the infrared image of the first target can be used for temperature measurement, but not for display, and image enhancement processing can be performed on the infrared image of the first target to obtain the infrared image of the second target. Images can be used for display.

In the related art, before splicing, each infrared image used for splicing is stretched to different degrees of contrast, and then spliced, so that the spliced infrared image with a large field of view will have obvious visual jumps, and the picture will be spliced. The marks are obvious and the appearance is poor. However, in the embodiment of the present application, unified image enhancement processing is performed on the entire spliced first target infrared image. Therefore, in the obtained second target infrared image, the enhancement degree of each part is consistent, the overall picture is uniform and smooth, and there is no visual jump. Changes and obvious splicing marks. Moreover, after the image enhancement process, the gray scale range corresponding to the infrared image of the second target will be increased. Therefore, by displaying the infrared image of the second target, the user can clearly identify the content in the image.

For image enhancement processing, it may include contrast stretching (also referred to as contrast enhancement or grayscale stretching), where the contrast stretching may be linear or non-linear. In one embodiment, the image enhancement process may include detail enhancement, and/or pseudo-color mapping, in addition to contrast stretching. After these enhancement processes, the infrared image of the second target can better adapt to the viewing habits of the human eye, so that the user can accurately distinguish different objects in the image.

In the method provided in this embodiment of the present application, since the mapping relationship between the grayscale values and the temperature of each infrared image is the same before splicing, a unified mapping relationship can be applied to the first target infrared image obtained after splicing, which can be obtained by using for temperature measurement. Further, image enhancement processing is performed on the first target infrared image as a whole, and the obtained second target infrared image has the same enhancement degree of each part, the overall picture is uniform and smooth, there is no visual jump and obvious splicing traces, and it is suitable for display. Therefore, the infrared image of the first target can be used for temperature measurement, and the infrared image of the second target can be used for display. The user can not only watch the high-quality infrared image with a large field of view, but also accurately measure the infrared image with a large field of view. temperature of each target.

For the acquired at least two infrared images, when the method of the embodiment of the present application is applied to an infrared image processing device, since the infrared image processing device only has processing capability and does not have a component for collecting infrared images, the at least two infrared images The image may be acquired by the infrared image processing apparatus from other devices or apparatuses, for example, may be acquired from an external infrared camera or infrared sensor. When the method of the embodiment of the present application is applied to a device or device capable of collecting infrared images, such as an infrared camera and an infrared thermal imager, the at least two infrared images can be acquired by its own infrared sensor.

Since the method provided in the embodiment of the present application involves the splicing of multiple infrared images, the infrared images used for splicing may meet some requirements. As mentioned above, there may be a certain degree of overlap or just seamless connection. The realization of the requirements requires more accurate control of the shooting angle of the infrared image. In one embodiment, the at least two infrared images are obtained by separately photographing different regions of the target. Specifically, in one embodiment, an infrared camera that shoots infrared images can be connected to a gimbal, so that the infrared camera can shoot objects (scenes) from different shooting angles one by one under the control of the gimbal, and obtain multiple images. Infrared images available for stitching.

In one example, the above-mentioned pan/tilt can also be installed on a mobile platform such as a drone.

Referring to FIG. 2 below, FIG. 2 is another flowchart of the infrared image processing method provided by the embodiment of the present application. 2 corresponds to a relatively specific embodiment. In this embodiment, in S210, the target can be photographed from different directions or shooting angles (such as shooting angles 1-N) through the infrared lens and the infrared sensor. , to obtain multiple original infrared images, namely raw images 1-N. In S220, each raw image may be preprocessed separately to obtain preprocessed raw images 1-N. For the relevant content of the preprocessing, please refer to the corresponding description in the foregoing. In S230, the preprocessed raw images 1-N may be spliced to obtain a first target infrared image for temperature measurement. At S240, image enhancement processing may be performed on the first target infrared image to obtain a second target infrared image for display.

The above is a detailed description of the infrared image processing method provided by the embodiments of the present application.

Referring to FIG. 3 below, FIG. 3 is a structural diagram of an infrared image processing apparatus provided by an embodiment of the present application. The apparatus includes: a processor 310 and a memory 320 storing instructions; the processor implements the following operations when executing the instructions:

Optionally, the processor is further configured to perform image enhancement processing on the first target infrared image to obtain a second target infrared image, and the second target infrared image is used for display.

Optionally, the image enhancement processing includes: contrast stretching.

Optionally, the image enhancement processing further includes: detail enhancement and/or pseudo-color mapping.

Optionally, the infrared image is obtained by performing preprocessing of correction and/or defect removal on the original infrared image output by the infrared sensor.

Optionally, the correction preprocessing includes responsivity correction and/or offset correction of the infrared sensor; the defect removal preprocessing includes dead pixel removal and/or noise removal.

Optionally, there are overlapping pixels between the at least two infrared images.

Optionally, the processor is configured to use the maximum grayscale value of the first pixel between the at least two infrared images as the first pixel when splicing the at least two infrared images. , wherein the first pixel is any pixel overlapping between the at least two infrared images.

Optionally, when stitching the at least two infrared images, the processor is configured to perform weighted fusion on overlapping pixels between the at least two infrared images.

Optionally, when the processor performs weighted fusion of the overlapping pixels between the at least two infrared images, the processor is configured to, according to the incident angle of the thermal radiation corresponding to the infrared images, determine the corresponding infrared images. fusion weight; the overlapping pixels between the at least two infrared images are fused by using the fusion weight.

Optionally, the difference between the incident angle and the orthophoto angle is negatively correlated with the fusion weight.

Optionally, the at least two infrared images are both subjected to the same contrast linear stretching.

Optionally, the first target infrared image is used to determine the temperature of the object in the first target infrared image according to the mapping relationship between the gray value and the temperature.

Optionally, the at least two infrared images are obtained by separately photographing different regions of the target.

It should be noted that the processor here can be a general-purpose processor or a specialized image processor, such as an ISP. In the specific implementation of hardware, the processor can be Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), Digital Signal Processor (DSP) , any of the Graphics Processing Unit (GPU). In one embodiment, the processor may be an FPGA processor, and the FPGA processor has the advantages of high speed, high real-time performance, and small size.

The specific implementation contents of the above-mentioned embodiments or implementation manners have been correspondingly described in the foregoing, and will not be repeated here.

In the device provided by the embodiment of the present application, since the mapping relationship between the grayscale value and the temperature between the infrared images is the same before the splicing, a unified mapping relationship can be applied to the first target infrared image obtained after the splicing. That is, the gray value of each pixel in the first target infrared image can be converted into a corresponding temperature by using the unified mapping relationship, so as to realize temperature measurement.

Referring to FIG. 4 below, FIG. 4 is a structural diagram of an infrared camera provided by an embodiment of the present application. The infrared camera includes: an infrared lens 410, an infrared sensor 420, a processor 430, and a memory 440 storing instructions;

Optionally, the image enhancement processing includes: contrast stretching.

Optionally, the infrared image is obtained by performing preprocessing on the original infrared image for correction and/or defect removal.

Optionally, the infrared camera is connected to a pan/tilt, and the at least two infrared images are captured by the infrared camera at different angles under the control of the pan/tilt.

Optionally, the processor includes any one of the following: FPGA, CPU, DSP, and GPU.

In the infrared camera provided by the embodiment of the present application, since before splicing, the mapping relationship between the gray value and temperature between the infrared images is the same, therefore, the first target infrared image obtained after splicing can be applied to a unified mapping relationship , that is, the gray value of each pixel in the first target infrared image can be converted into a corresponding temperature by using the unified mapping relationship, so as to realize temperature measurement.

The embodiments of the present application further provide a computer-readable storage medium, where the computer-readable storage medium stores instructions, and when the instructions are executed by the processor, implement any infrared image processing method provided by the embodiments of the present application.

In the above embodiments, various implementations are provided for each operation. As for which implementation is used for each operation, on the basis of no conflict or contradiction, those skilled in the art can freely choose or combine them according to the actual situation. Various embodiments are thus constituted. However, this application document is limited in space and does not describe various embodiments, but it is understood that various embodiments also belong to the scope disclosed by the embodiments of this application.

Embodiments of the present application may take the form of computer instruction products implemented on one or more storage media having program code embodied therein, including but not limited to disk storage, CD-ROM, optical storage, and the like. Computer-usable storage media includes both persistent and non-permanent, removable and non-removable media, and storage of information can be implemented by any method or technology. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash Memory or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists. The terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion such that a process, method, article or device comprising a list of elements includes not only those elements, but also other not expressly listed elements, or also include elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The methods, devices, equipment, etc. provided by the embodiments of the present application have been described in detail above. The principles and implementations of the present application are described in this paper by using specific examples. method and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation and application scope. Application restrictions.

Claims

An infrared image processing method, comprising:

acquiring at least two infrared images, wherein the at least two infrared images correspond to the same target, and the grayscale values of the infrared images have a preset mapping relationship with the temperature;

The at least two infrared images are stitched to obtain a first target infrared image, wherein, before the stitching is performed, the mapping relationships of the at least two infrared images are the same, and the first target infrared image is used for temperature.
The method according to claim 1, wherein after obtaining the first target infrared image, further comprising:

Perform image enhancement processing on the first target infrared image to obtain a second target infrared image, and the second target infrared image is used for display.
The method according to claim 2, wherein the image enhancement processing comprises: contrast stretching.
The method according to claim 3, wherein the image enhancement processing further comprises: detail enhancement and/or pseudo-color mapping.
The method according to claim 1, wherein the infrared image is obtained by performing preprocessing of correction and/or defect removal on the original infrared image output by the infrared sensor.
The method according to claim 5, wherein the correction preprocessing comprises responsivity correction and/or offset correction of the infrared sensor; and the defect removal preprocessing comprises dead pixel removal and/or noise removal.
The method according to claim 1, wherein the at least two infrared images have overlapping pixels.
The method according to claim 7, wherein the stitching of the at least two infrared images comprises:

Taking the maximum gray value of the first pixel between the at least two infrared images as the gray value of the first pixel, where the first pixel is between the at least two infrared images Any pixel that overlaps.
The method according to claim 7, wherein the stitching of the at least two infrared images comprises:

Weighted fusion is performed on the overlapping pixel points between the at least two infrared images.
The method according to claim 9, wherein the weighted fusion of the overlapping pixels between the at least two infrared images comprises:

Determine the fusion weight corresponding to the infrared image according to the incident angle of the thermal radiation corresponding to the infrared image;

The overlapping pixels between the at least two infrared images are fused by using the fusion weight.
The method according to claim 10, wherein the difference between the incident angle and the orthophoto angle is negatively correlated with the fusion weight.
The method according to claim 1, wherein the at least two infrared images are both subjected to the same contrast linear stretching.
The method according to claim 1, wherein the temperature measurement using the first target infrared image comprises:

According to the mapping relationship between the gray value and the temperature, the temperature of the object in the infrared image of the first target is determined.
The method according to claim 1, wherein the at least two infrared images are obtained by separately photographing different regions of the target.
The method according to claim 14, wherein the method is applied to an infrared camera connected to a gimbal, and the at least two infrared images are captured by the infrared camera at different angles under the control of the gimbal owned.
An infrared image processing device, comprising: a processor and a memory storing instructions; the processor implements the following operations when executing the instructions:

acquiring at least two infrared images, wherein the at least two infrared images correspond to the same target, and the grayscale values of the infrared images have a preset mapping relationship with the temperature;

The at least two infrared images are stitched to obtain a first target infrared image, wherein, before the stitching is performed, the mapping relationships of the at least two infrared images are the same, and the first target infrared image is used for temperature.
The device according to claim 16, wherein the processor is further configured to perform image enhancement processing on the first target infrared image to obtain a second target infrared image, and the second target infrared image is used for show.
The apparatus of claim 17, wherein the image enhancement processing comprises: contrast stretching.
The apparatus according to claim 18, wherein the image enhancement processing further comprises: detail enhancement and/or pseudo-color mapping.
The device according to claim 16, wherein the infrared image is obtained by performing preprocessing of correction and/or defect removal on the original infrared image output by the infrared sensor.
The apparatus according to claim 20, wherein the correction preprocessing comprises responsivity correction and/or offset correction of the infrared sensor; and the defect removal preprocessing comprises dead pixel removal and/or noise removal.
The device according to claim 16, wherein the at least two infrared images have overlapping pixels.
The device according to claim 22, wherein the processor is configured to, when splicing the at least two infrared images, combine the maximum gray level of the first pixel between the at least two infrared images The degree value is used as the gray value of the first pixel point, wherein the first pixel point is any pixel point overlapping between the at least two infrared images.
The device according to claim 22, wherein the processor is configured to perform weighted fusion of overlapping pixels between the at least two infrared images when splicing the at least two infrared images.
The device according to claim 24, wherein when the processor performs weighted fusion on the pixel points overlapping between the at least two infrared images, the processor is configured to: according to the incidence of thermal radiation corresponding to the infrared images angle, and determine the fusion weight corresponding to the infrared image; and use the fusion weight to fuse the overlapping pixels between the at least two infrared images.
The apparatus according to claim 25, wherein the difference between the incident angle and the orthophoto angle is negatively correlated with the fusion weight.
The device according to claim 16, wherein the at least two infrared images are both subjected to the same linear stretching of contrast.
The apparatus according to claim 16, wherein the first target infrared image is used to determine the temperature of the object in the first target infrared image according to the mapping relationship between grayscale values and temperature.
The device according to claim 16, wherein the at least two infrared images are obtained by separately photographing different regions of the target.
The device according to claim 16, wherein the processor comprises any one of the following: a field programmable logic gate array, a central processing unit, a digital signal processor, and a graphics processor.
An infrared camera, comprising: an infrared lens, an infrared sensor, a processor, and a memory storing instructions;

The infrared sensor is used for collecting the original infrared image through the infrared lens;

The processor, when executing the instructions, performs the following operations:

acquiring at least two infrared images, wherein the at least two infrared images correspond to the same target, and the grayscale values of the infrared images have a preset mapping relationship with the temperature;

The at least two infrared images are stitched to obtain a first target infrared image, wherein, before the stitching is performed, the mapping relationships of the at least two infrared images are the same, and the first target infrared image is used for temperature.
The infrared camera of claim 31, wherein the processor is further configured to perform image enhancement processing on the first target infrared image to obtain a second target infrared image, the second target infrared image using on display.
The infrared camera of claim 32, wherein the image enhancement processing comprises: contrast stretching.
The infrared camera of claim 33, wherein the image enhancement processing further comprises: detail enhancement and/or pseudo-color mapping.
The infrared camera of claim 31, wherein the infrared image is obtained by performing preprocessing on the original infrared image for correction and/or defect removal.
The infrared camera according to claim 35, wherein the correction preprocessing comprises responsivity correction and/or offset correction of the infrared sensor; the defect removal preprocessing comprises dead pixel removal and/or noise removal .
The infrared camera of claim 31, wherein the at least two infrared images have overlapping pixels.
The infrared camera according to claim 37, wherein, when the processor stitches the at least two infrared images, the maximum value of the first pixel point between the at least two infrared images is The gray value is used as the gray value of the first pixel, wherein the first pixel is any pixel overlapping between the at least two infrared images.
The infrared camera according to claim 37, wherein the processor is configured to perform weighted fusion of overlapping pixels between the at least two infrared images when splicing the at least two infrared images .
The infrared camera according to claim 39, wherein when the processor performs weighted fusion on the pixels overlapping between the at least two infrared images, the processor is configured to: according to the thermal radiation corresponding to the infrared images The incident angle is determined, and the fusion weight corresponding to the infrared image is determined; the overlapping pixel points between the at least two infrared images are fused by using the fusion weight.
The infrared camera according to claim 40, wherein the difference between the incident angle and the orthophoto angle is negatively correlated with the fusion weight.
The infrared camera according to claim 31, wherein the at least two infrared images are linearly stretched with the same contrast.
The infrared camera of claim 31, wherein the first target infrared image is used to determine the temperature of the object in the first target infrared image according to the mapping relationship between grayscale values and temperature.
The infrared camera according to claim 31, wherein the at least two infrared images are obtained by separately photographing different areas of the target.
The infrared camera according to claim 31, wherein the infrared camera is connected to a pan/tilt, and the at least two infrared images are captured by the infrared camera at different angles under the control of the pan/tilt.
The infrared camera according to claim 31, wherein the processor comprises any one of the following: a field programmable logic gate array, a central processing unit, a digital signal processor, and a graphics processor.
A computer-readable storage medium, characterized in that, the computer-readable storage medium stores instructions, and when the instructions are executed by a processor, the infrared image processing method according to any one of claims 1-15 is implemented.