A method and system for identifying a dust area based on infrared channel infrared multispectral

Abstract

The present application relates to the technical field of monitoring dust area, and particularly relates to a method and system for identifying dust area based on infrared channel infrared multispectral. The method provided by the present application comprises the following steps: acquiring brightness temperature values of 8.5 mu m, 10.8 mu m and 12 mu m channels of a pixel in an infrared channel image of a to-be-identified area, and constructing an infrared multispectral dust index MIDI; performing cloud area pixel judgment and exclusion on the pixel in the to-be-identified area; if the MIDI value of the image of the to-be-identified area is higher than a dust identification threshold value after the cloud area pixel exclusion, it is determined that the pixel in the to-be-identified area is a dust pixel. The method and system for identifying dust area based on infrared channel infrared multispectral can be continuously monitored day and night to analyze the moving development path of dust, accurately monitor and warn dust weather, and more finely eliminate dust misjudgment information, thereby improving the accuracy of infrared channel identification of dust area.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of monitoring sand-dust area, and in particular to a method and system for identifying sand-dust area based on infrared channel. BACKGROUND

[0002] Sand-dust area is formed by strong wind that carries a large amount of loose sand or dust into the air. Sand-dust area makes the air turbid and reduces horizontal visibility, and has strong destructive power on traffic, crops, ecological environment and air quality.

[0003] Sand-dust area is a regional disastrous weather. There are many sand-dust prone areas in the world, including the deserts and desert areas in North Africa, East Asia, West Asia, South Asia and Australia, and the deserts in South America and North America. Satellite remote sensing technology can capture dynamic information of sand-dust area from space, and has high time resolution, and is one of the most effective means for monitoring, tracking and analyzing sand-dust area. The sand-dust observation method can be divided into ground-based and remote sensing. Ground-based observation is mainly single-point observation, which is difficult to detect sand-dust in a large area. In contrast, meteorological satellites have the characteristics of large imaging range and continuous spatial distribution, which is more advantageous for determining the spatial distribution range of sand-dust. With the development of remote sensing technology, satellite remote sensing sand-dust identification methods generally use the comprehensive application of reflection and emission spectrum information, such as visible light, short-wave infrared, medium infrared and far infrared, so the application period is limited to daytime. However, sand-dust area often occurs continuously for several or ten hours, and it is necessary to monitor it continuously day and night to analyze the moving development path of sand-dust, so as to more accurately predict and warn sand-dust area.

[0004] Therefore, it is urgent to overcome the defects of the prior art in the technical field. SUMMARY

[0005] The present application provides a method and system for identifying sand-dust area based on infrared channel, which solves the technical problem that the current satellite remote sensing sand-dust identification method is limited to the condition of visible light channel by analyzing the infrared multi-spectral channel image, excluding cloud area pixels first, and then identifying sand-dust.

[0006] In a first aspect, the present application provides a method for identifying sand-dust area based on infrared channel, which comprises:

[0007] Obtaining the brightness temperature values of 8.5 μm, 10.8 μm and 12 μm channels of the pixels in the infrared channel image of the area to be identified, and constructing an infrared multi-spectral sand-dust index MIDI;

[0008] Judging and excluding cloud area pixels from the pixels in the area to be identified;

[0009] If the image MIDI value of the to-be-identified region is higher than the dust identification threshold value after the cloud region pixels are excluded, it is determined that the to-be-identified region pixel is a dust pixel.

[0010] Preferably, the calculation method of the infrared multi-spectral dust index MIDI is as follows:

[0011] MIDI = (T 8.5 + T 12 ) / (2 x T 10.8 )

[0012] Wherein, T 8.5 , T 10.8 and T 12 are the brightness temperature values of the 8.5 μm, 10.8 μm and 12 μm channels of the pixel in the infrared channel image.

[0013] Preferably, the cloud region pixel judgment and exclusion of the to-be-identified region pixel specifically includes the following steps.

[0014] When T 10.8 is less than or equal to the first preset threshold value, it is determined that the type of the cloud region pixel is a high cloud region pixel.

[0015] Preferably, the cloud region pixel judgment and exclusion of the to-be-identified region pixel specifically includes the following steps.

[0016] When T 10.8 is greater than the high cloud preset threshold value, and the brightness temperature difference between T 10.8 and T 6.2 is less than or equal to the second preset threshold value, it is determined that the type of the cloud region pixel is a high cloud mixed region pixel containing high cloud information, wherein T 6.2 is the brightness temperature value of the 6.2 μm channel of the pixel in the infrared channel image.

[0017] Preferably, the cloud region pixel judgment and exclusion of the to-be-identified region pixel specifically includes the following steps.

[0018] When the brightness temperature difference between T 8.5 and T 10.8 is greater than the third preset threshold value, and the brightness temperature difference between T 12 and T 10.8 is less than the fourth preset threshold value, it is determined that the type of the cloud region pixel is a cloud band edge region pixel.

[0019] Preferably, the cloud region pixel judgment and exclusion of the to-be-identified region pixel specifically includes the following steps.

[0020] When the brightness temperature difference between T 10.8 and T 12 is greater than the fifth preset threshold value, it is determined that the type of the cloud region pixel is a common cloud region pixel.

[0021] Preferably, the method further comprises:

[0022] In the area determined as the dust area, the dust area with an area less than a preset area is filtered, and the dust area with a number of dust end member pixels exceeding a preset threshold is left as a high-confidence dust area;

[0023] It is determined whether the non-high-confidence dust area is within a preset distance of the high-confidence dust area, and if not, the dust area is considered to be misjudged.

[0024] In a second aspect, the present application provides an infrared multispectral dust area identification system based on an infrared channel, comprising a construction index module, a cloud removal module and a judgment module, wherein:

[0025] The construction index module obtains the channel brightness temperature values of 8.5 μm, 10.8 μm and 12 μm of the pixels in the infrared channel image of the area to be identified, and constructs an infrared multispectral dust index MIDI;

[0026] The cloud removal module excludes the cloud area pixels from the area to be identified;

[0027] After the cloud area pixels are excluded, if the judgment module determines that the MIDI value of the image of the area to be identified is higher than a dust identification threshold, the area to be identified is determined to be a dust area.

[0028] Preferably, the system further comprises a dust misjudgment area detection module, which judges the area of the dust area, filters the dust area with an area less than a preset area, and leaves the dust area with a number of dust end member pixels exceeding a preset threshold as a high-confidence dust area. The dust misjudgment area detection module judges whether the non-high-confidence dust area is within a preset distance of the high-confidence dust area, and if not, the dust area is considered to be misjudged.

[0029] In a third aspect, the present application provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the method for identifying a dust area based on an infrared channel according to the first aspect and / or the second aspect.

[0030] The method and system for identifying a dust area based on an infrared channel provided by the present application can continuously monitor day and night to analyze the moving development path of dust, accurately monitor and warn dust weather, and more finely eliminate dust misjudgment information, thereby improving the accuracy of dust area identification by the infrared spectral channel. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced as follows. Obviously, the drawings described below are only some of the embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0032] Figure 1 is a method step schematic diagram of identifying a sand and dust area based on an infrared channel provided by the first embodiment of the present application;

[0033] Figure 2 is a method step schematic diagram of detecting a sand and dust misjudgment area based on an infrared channel provided by the second embodiment of the present application;

[0034] Figure 3 is a system structure block diagram of identifying a sand and dust area based on an infrared channel provided by the third embodiment of the present application;

[0035] Figure 4 is a system structure block diagram of identifying a sand and dust area based on an infrared channel provided by the fourth embodiment of the present application, which adds a sand and dust misjudgment area detection to the third embodiment;

[0036] Figure 5 is a system structure block diagram of an electronic device provided by the fifth embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0038] In the description of the present application, the terms "inner", "outer", "longitudinal", "transverse", "upper", "lower", "top", "bottom", "left", "right", "front", "back" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and do not require the present application to be constructed and operated in a particular orientation, therefore should not be understood as a limitation of the present application.

[0039] In addition, the technical features involved in each of the embodiments of the present application described below can be combined with each other as long as there is no conflict. The present application will be described in detail below in combination with the drawings and embodiments.

[0040] Embodiment 1:

[0041] The application research of satellite remote sensing dust identification method using only infrared channel is less, and mainly using Fengyun 2 geostationary meteorological satellite and Japanese GMS satellite data. The new generation of Fengyun meteorological satellite has rich observation information and global observation capability, which provides data support for global sandstorm monitoring. Especially Fengyun 4 has a minute-level observation frequency, overlooking the concentrated sandstorm prone areas, including East Asia, West Asia, South Asia, Australia and other desert and desert areas, which can monitor the sandstorm information of this area in real time and dynamically. In this embodiment, the infrared channel data obtained by AGRI of Fengyun 4 satellite instrument is used as the infrared multispectral data mentioned in the method.

[0042] This embodiment provides a method for identifying sand dust area based on infrared channel, which comprises the following steps, as shown in the figure. Figure 1

[0043] S101: Obtain the brightness temperature values of 8.5 μm, 10.8 μm and 12 μm channels of the pixels in the infrared channel image of the area to be identified, and construct the infrared multispectral dust index MIDI.

[0044] When the sand dust area appears, the brightness temperature difference values of 8.5 μm channel, 12 μm channel and 10.8 μm channel of the underlying surface pixels in the desert area, desert area and vegetation area and other areas are positively increased, that is, the brightness temperature value of 8.5 μm of the pixel and the brightness temperature value of 12 μm of the pixel are close to or higher than the brightness temperature value of 10.8 μm of the pixel, so the infrared multispectral dust index (Multiple Infrared Dust Index, abbreviated as MIDI) based on Fengyun 4 satellite and AGRI instrument can be constructed.

[0045] S102: Cloud pixel judgment and exclusion is performed on the pixels in the area to be identified.

[0046] At present, it is known that the MIDI value of the sand dust area is obviously higher than the MIDI value under the local clear sky condition, but the MIDI value of some cloud areas may also be higher, reaching or approaching the MIDI value of the sand dust area, so if the image MIDI value of the area to be identified is higher than the image MIDI value of the local clear sky area, the cloud pixel in the area to be identified is excluded. Through analysis of 6.2 μm, 8.5 μm, 10.8 μm and 12 μm channel images of AGRI of Fengyun 4 satellite instrument, there are four cases that cause high MIDI value of cloud area:

[0047] ① High cloud: the 8.5 μm brightness temperature of high cloud pixel reaches or approaches the 10.8 μm brightness temperature of the pixel, so the MIDI value of high cloud is usually high, which can be used to identify high cloud by using the 10.8 μm channel brightness temperature value of the pixel.

[0048] ​② High cloud mixed pixel: some mixed pixels containing high cloud information near high cloud area, but due to the influence of mixed pixel radiance average, the 8.5 μm channel brightness temperature of the pixel is still high, so the MIDI value is also high. The 6.2 μm and 10.8 μm channel spectra of the pixel can be used to identify high cloud mixed pixels. For pure high cloud pixels, the 6.2 μm channel brightness temperature of the pixel is similar to the 10.8 μm channel brightness temperature of the pixel, while for land or low cloud pixels, the 10.8 μm channel brightness temperature of the pixel is greater than the 6.2 μm channel brightness temperature of the pixel. The difference between the 10.8 μm channel brightness temperature of the pixel and the 6.2 μm channel brightness temperature of the pixel can be used to identify high cloud mixed pixels.

[0049] ③ Cloud band edge: in some cloud bands, the cloud band width of the 8.5 μm channel of the pixel is narrower than the 10.8 μm channel of the pixel, so the 8.5 μm channel brightness temperature of the pixel around the cloud band is significantly higher than the 10.8 μm channel brightness temperature of the pixel, causing the MIDI value around the cloud band to increase significantly. Since in the sand area, if the 8.5 μm channel brightness temperature of the pixel is higher than the 10.8 μm channel brightness temperature of the pixel by more than 3K, the 12 μm channel brightness temperature of the pixel will also be higher than the 10.8 μm channel brightness temperature of the pixel, so the difference between the 8.5 μm channel, 10.8 μm and 12 μm channel brightness temperatures of the pixel can be used to identify the cloud band edge.

[0050] ④ Normal cloud area identification: the 10.8 μm channel brightness temperature of the pixel in the normal cloud area is greater than the 12 μm channel brightness temperature of the pixel, and a suitable threshold can be set to identify the normal cloud area.

[0051] In this embodiment, the infrared multispectral sand dust index MIDI is used to identify the sand area to improve the recognition accuracy, and the infrared multispectral sand dust index MIDI is constructed as follows:

[0052] MIDI=(T 8.5 +T 12 ) / (2×T 10.8 )

[0053] Wherein: T 8.5 , T 10.8 , T 12 are the brightness temperatures of the 8.5 μm, 10.8 μm and 12 μm channels of the pixel in the infrared channel image.

[0054] In this embodiment, the cloud area pixel judgment and exclusion of the pixel in the area to be identified are performed, and the specific method includes:

[0055] When T 10.8 is less than or equal to the first preset threshold, the type of the cloud area pixel is judged to be a high cloud area pixel.

[0056] The 8.5 μm brightness temperature value of the high cloud area pixel is close to or equal to the 10.8 μm brightness temperature value, so the MIDI value of the high cloud area is usually high, and the 10.8 μm channel brightness temperature value of the pixel can be used to identify the high cloud area. The first preset threshold is a high cloud preset threshold, and is preferably 240 K, so when T 10.8 ≤240 K, the type of the cloud area pixel is determined to be a high cloud area pixel.

[0057] In the first embodiment, the cloud area pixel judgment and exclusion are performed on the pixels in the to-be-identified area, and the specific method further includes:

[0058] When T 10.8 is greater than the high cloud preset threshold, and the brightness temperature difference between T 10.8 and T 6.2 is less than or equal to the second preset threshold, the type of the cloud area pixel is determined to be a high cloud mixed area pixel containing high cloud information, wherein T 6.2 is the brightness temperature value of the 6.2 μm channel of the pixel in the infrared channel image.

[0059] Some mixed pixels containing high cloud information near the high cloud area do not satisfy the condition that T 10.8 ≤the first preset threshold, but due to the average influence of the mixed pixel radiance, the 8.5 μm channel brightness temperature value of the high cloud mixed area pixel is still high, so the MIDI value of the high cloud mixed area pixel is also high. The high cloud mixed pixel is identified by using the different response characteristics of the 6.2 μm channel spectrum of the pixel and the 10.8 μm channel spectrum of the pixel in the cloud area.

[0060] For a pure high cloud pixel, the brightness temperature values of the 6.2 μm channel of the pixel and the 10.8 μm channel of the pixel are similar, while in a land or a medium-low cloud pixel, the 10.8 μm channel brightness temperature value of the pixel is greater than the 6.2 μm channel brightness temperature value of the pixel by about 30 K, so the difference between the 10.8 μm channel brightness temperature value of the pixel and the 6.2 μm channel brightness temperature value of the pixel in the high cloud mixed pixel containing high cloud information is less than 30 K. The high cloud mixed pixel can be identified by using the brightness temperature difference between the 10.8 μm channel of the pixel and the 6.2 μm channel of the pixel, that is, when T 10.8 >240 K and T 10.8 -T 6.2 ≤the second preset threshold, the type of the cloud area pixel is determined to be a high cloud mixed area pixel containing high cloud information, and the second preset threshold is a high cloud mixed preset threshold, and is preferably 20 K.

[0061] In the first embodiment, the cloud area pixel judgment and exclusion are performed on the pixels in the to-be-identified area, and the specific method further includes:

[0062] When the brightness temperature difference between T 8.5 and T 10.8 is greater than the third preset threshold, and T 12 is greater than T 10.8the difference between the brightness temperature of the 8.5 μm channel and the 10.8 μm channel is less than the fourth preset threshold value, the cloud area pixel is determined as a cloud band edge area pixel.

[0063] In some cloud bands, the cloud band width of the 8.5 μm channel is narrower than that of the 10.8 μm channel, so the brightness temperature of the 8.5 μm channel is obviously higher than that of the 10.8 μm channel, more than 3 K, which causes the MIDI value of the cloud band edge area pixel to increase obviously. Since in the sand-dust weather, if the brightness temperature of the 8.5 μm channel is more than 3 K higher than that of the 10.8 μm channel, the brightness temperature of the 12 μm channel will also be higher than that of the 10.8 μm channel, so the cloud band edge information can be identified by the following condition: T 8.5 -T 10.8 ≥ the third preset threshold value and T 12 -T 10.8 ≤ the fourth preset threshold value, the cloud area pixel is determined as a cloud band edge area pixel, the third preset threshold value is preferably 3 K, and the fourth preset threshold value is preferably 1 K.

[0064] In the first embodiment, the cloud area pixel judgment and exclusion are performed on the pixels in the area to be identified, and the specific method further includes:

[0065] T 10.8 and T 12 The difference between the brightness temperature of the 8.5 μm channel and the 10.8 μm channel is greater than the fifth preset threshold value, the cloud area pixel is determined as a normal cloud area pixel.

[0066] The brightness temperature of the 10.8 μm channel of the normal cloud area pixel is greater than that of the 12 μm channel, and a suitable threshold value can be set to identify the normal cloud area, that is, when T10.8-T12≥ the fifth preset threshold value, it is considered as a normal cloud area pixel, and the fifth preset threshold value is preferably 1.5 K.

[0067] S103: If the image MIDI value of the area to be identified is higher than the sand-dust identification threshold value after the cloud area pixel exclusion, the pixels in the area to be identified are determined as sand-dust pixels.

[0068] In the first embodiment, the image MIDI value of the area to be identified is the MIDI value of the current time image. The current time image is an image without cloud exclusion.

[0069] The method for identifying the sand-dust area based on the infrared channel provided in the first embodiment can continuously monitor day and night to analyze the moving development path of the sand-dust, and accurately monitor and warn the sand-dust weather.

[0070] Embodiment 2:

[0071] In the first embodiment, the main basis for the dust identification of the Fengyun-4 meteorological satellite infrared multi-spectral dust index (MIDI) is that the MIDI value of the dust area is obviously higher than the MIDI value before the dust occurs. Although some cloud area identification conditions have been used in the dust identification, some cloud areas are not identified, resulting in some dust area misjudgment information, so it is necessary to remove the misjudged areas as dust areas. These misjudged dust areas are generally scattered in small patches.

[0072] Through the analysis of the MIDI ratio of some individual cases (i.e. the ratio of MIDI before and after the dust occurs), the MIDI ratio of the dust area, the cloud area and the desert clear sky area has the following characteristics: 1) The dust area generally has a large number of pixels with high MIDI ratio, i.e. dust endmember pixels; 2) The MIDI ratio of the desert clear sky area is obviously lower than that of the dust area; 3) High clouds (the 10.8 μm brightness temperature value of the pixel is lower than 240K) have a higher MIDI ratio; 4) The MIDI ratio of the ordinary cloud area is higher than that of the clear sky land and lower than that of the dust area; 5) Sometimes there are more small patches (several pixels) with high MIDI ratio, which is comparable to the MIDI ratio of the dust area, which may be related to the background map.

[0073] The second embodiment provides an infrared multi-spectral dust area misjudgment area detection method based on the infrared channel on the basis of the first embodiment, to solve the problem of reducing and eliminating dust misjudgment information. The method comprises the following steps, as shown in Figure 2

[0074] S201: In the area judged as the dust area, filter the dust area with an area smaller than a preset area, and leave the dust area with a number of dust endmember pixels exceeding a preset threshold as a high-confidence dust area.

[0075] The dust area with an area smaller than a preset area is removed from the dust area generated by the first embodiment. In the second embodiment, the preset area of the dust area is preferably 50 pixels. The dust area with a number of dust endmember pixels exceeding a preset number is regarded as a high-confidence dust area. In the second embodiment, the preset number of dust endmember pixels is preferably 300 pixels.

[0076] S202: Determine whether the non-high-confidence dust area is within a preset distance of the high-confidence dust area. If it is not within the preset distance, the area previously judged as the dust area is considered to be misjudged.

[0077] ​In the high-confidence dust area, all pixels are regarded as dust pixels, so it is possible to retain the dust pixels covered by thin clouds, and it is necessary to determine whether the remaining non-high-confidence dust area is in the vicinity of the high-confidence dust area. The method is to determine whether the non-high-confidence dust area is in the same geographical grid range as the high-confidence dust area. In this embodiment two, the size of the geographical grid range is set to 2x2 latitude and longitude. If it is not in the same geographical grid range, it is considered that the dust area identified in embodiment one is misjudged.

[0078] The method for detecting misjudgment of the infrared multispectral dust area provided in this embodiment two based on the infrared channel can more finely eliminate dust misjudgment information and improve the accuracy of identifying dust areas in the infrared channel.

[0079] Embodiment 3:

[0080] Based on embodiment one, this embodiment three provides a system for identifying dust areas based on infrared channels. The system includes a construction index module, a cloud removal module, and a judgment module, as shown in Figure 3

[0081] The construction index module obtains the channel brightness temperature values of 8.5 μm, 10.8 μm, and 12 μm of the pixels in the infrared channel image of the area to be identified, and constructs the infrared multispectral dust index MIDI.

[0082] When the dust area appears, the brightness temperature difference values of the 8.5 μm and 12 μm channels of the underlying surface pixels of the desert area, the desert area, and the vegetation area and the brightness temperature of the 10.8 μm channel of the pixels are all positively increased, that is, the brightness temperature value of the 8.5 μm of the pixel and the brightness temperature value of the 12 μm of the pixel are close to or higher than the brightness temperature value of the 10.8 μm of the pixel. Therefore, the construction index module is based on the infrared multispectral dust index (Multiple Infrared Dust Index, abbreviated as MIDI) of the Fengyun-4 satellite instrument AGRI.

[0083] The cloud removal module excludes the cloud area pixels in the area to be identified.

[0084] At present, it is known that the MIDI value of the dust area is significantly higher than the MIDI value under the local clear sky condition, but the MIDI value of some cloud areas may also be relatively high, reaching or approaching the MIDI value of the dust area. Therefore, if the image MIDI value of the area to be identified is higher than the image MIDI value of the local clear area, the cloud removal module excludes the cloud area pixels in the area to be identified.

[0085] The cloud removal module analyzes the 6.2 μm, 8.5 μm, 10.8 μm, and 12 μm channel images of the pixels of the Fengyun-4 satellite instrument AGRI to determine that there are four kinds of conditions in which the MIDI value of the cloud area is relatively high.

[0086] ​① High cloud: The 8.5 μm brightness temperature of the pixel of high cloud reaches or approaches the 10.8 μm brightness temperature of the pixel, so the MIDI value of high cloud is usually high, and the cloud removing module can identify high cloud by using the 10.8 μm channel brightness temperature value of the pixel.

[0087] ② High cloud mixed pixel: Some mixed pixels containing high cloud information near the high cloud area, but due to the influence of mixed pixel radiance average, the 8.5 μm channel brightness temperature of the pixel is still high, so the MIDI value is also high. The high cloud mixed pixel can be identified by using the different response characteristics of the 6.2 μm and 10.8 μm channels of the pixel in the cloud area. For pure high cloud pixels, the 6.2 μm channel and 10.8 μm channel brightness temperatures of the pixel are similar, while in land or low cloud pixels, the 10.8 μm channel brightness temperature of the pixel is higher than the 6.2 μm channel brightness temperature of the pixel, and the cloud removing module can identify the high cloud mixed pixel by using the brightness temperature difference value of the 10.8 μm channel and the 6.2 μm channel of the pixel.

[0088] ③ Cloud band edge: In some cloud bands, the cloud band width of the 8.5 μm channel of the pixel is narrower than that of the 10.8 μm channel of the pixel, so the 8.5 μm channel brightness temperature of the pixel around the cloud band is significantly higher than the 10.8 μm channel brightness temperature of the pixel, causing the MIDI value around the cloud band to increase significantly. Since in the dust area, if the 8.5 μm channel brightness temperature of the pixel is higher than the 10.8 μm channel brightness temperature of the pixel by more than 3K, the 12 μm channel brightness temperature of the pixel will also be higher than the 10.8 μm channel brightness temperature of the pixel, so the cloud removing module can identify the cloud band edge by using the brightness temperature difference of the 8.5 μm channel, the 10.8 μm channel and the 12 μm channel of the pixel.

[0089] ④ Ordinary cloud area identification: The 10.8 μm channel brightness temperature of the pixel in the ordinary cloud area is higher than the 12 μm channel brightness temperature of the pixel, and after setting an appropriate threshold, the cloud removing module can identify the ordinary cloud area.

[0090] In this embodiment three, the recognition accuracy is improved by identifying the dust area through the infrared multispectral dust index MIDI, and the infrared multispectral dust index MIDI is constructed in this embodiment three:

[0091] MIDI = (T 8.5 + T 12 ) / (2 × T 10.8 )

[0092] Wherein: T 8.5 , T 10.8 , T 12 are the brightness temperature values of the 8.5 μm, 10.8 μm and 12 μm channels of the pixel in the infrared channel image.

[0093] In the third embodiment, the cloud removing module excludes high cloud area pixels from the to-be-identified area. The specific method includes: when T 10.8 is less than or equal to the first preset threshold, the type of the cloud area pixel is determined to be a high cloud area pixel.

[0094] The 8.5 μm brightness temperature value of the pixel in the high cloud area reaches or approaches the 10.8 μm brightness temperature value of the pixel, and thus the MIDI value of the high cloud area is usually high. The high cloud area can be identified by using the 10.8 μm channel brightness temperature value of the pixel. The first preset threshold is a high cloud preset threshold, and is preferably 240 K. Thus, when T 10.8 ≤ 240 K, the type of the cloud area pixel is determined to be a high cloud area pixel.

[0095] In the third embodiment, the cloud removing module excludes high cloud mixed area pixels from the to-be-identified area. The specific method includes: when T 10.8 is greater than the high cloud preset threshold, and the brightness temperature difference between T 10.8 and T 6.2 is less than or equal to the second preset threshold, the type of the cloud area pixel is determined to be a high cloud mixed area pixel containing high cloud information, wherein: T 6.2 is the brightness temperature value of the 6.2 μm channel of the pixel in the infrared channel image.

[0096] Some mixed pixels containing high cloud information near the high cloud area do not satisfy the condition that T 10.8 ≤ the first preset threshold, but due to the average influence of the mixed pixel radiance, the 8.5 μm channel brightness temperature value of the high cloud mixed area pixel is still high, and thus the MIDI value of the high cloud mixed area pixel is also high. The high cloud mixed pixel can be identified by using the different response characteristics of the 6.2 μm channel and the 10.8 μm channel of the pixel in the cloud area.

[0097] For a pure high cloud pixel, the brightness temperature values of the 6.2 μm channel and the 10.8 μm channel of the pixel are similar, while for a land or middle-low cloud pixel, the 10.8 μm channel brightness temperature value of the pixel is greater than the 6.2 μm channel brightness temperature value of the pixel by about 30 K. Thus, the difference between the 10.8 μm channel brightness temperature value of the pixel and the 6.2 μm channel brightness temperature value of the pixel in the high cloud mixed pixel containing high cloud information is less than 30 K. The cloud removing module can identify the high cloud mixed pixel by using the brightness temperature difference between the 10.8 μm channel of the pixel and the 6.2 μm channel of the pixel, that is, when: T 10.8 > 240 K and T 10.8 -T 6.2 ≤ the second preset threshold, the type of the cloud area pixel is determined to be a high cloud mixed area pixel containing high cloud information. The second preset threshold is a high cloud mixed preset threshold, and is preferably 20 K.

[0098] In the third embodiment, the cloud removing module excludes cloud band edge area pixels from the to-be-identified area. The specific method includes: when T 8.5T 10.8 -T 12 ≥ third preset threshold value, and T 10.8 -T 8.5 ≤ fourth preset threshold value, the cloud area pixel is determined as a cloud band edge area pixel by the cloud removing module, the third preset threshold value is preferably 3K, and the fourth preset threshold value is preferably 1K.

[0099] In some cloud bands, the cloud band width of the 8.5 μm channel of a pixel is narrower than that of the 10.8 μm channel of the pixel, so that the brightness temperature value of the 8.5 μm channel of the cloud band edge area pixel is obviously higher than that of the 10.8 μm channel of the pixel, reaching more than 3K, causing the MIDI value of the cloud band edge area pixel to be obviously increased. Since in the sand-dust weather, if the brightness temperature value of the 8.5 μm channel of a pixel is more than 3K higher than that of the 10.8 μm channel of the pixel, the brightness temperature value of the 12 μm channel of the pixel will also be higher than that of the 10.8 μm channel of the pixel, so that the cloud band edge information can be identified by using the following conditions: T 8.5 -T 10.8 ≥ third preset threshold value and T 12 -T 10.8 ≤ fourth preset threshold value, the cloud area pixel is determined as a cloud band edge area pixel by the cloud removing module, the third preset threshold value is preferably 3K, and the fourth preset threshold value is preferably 1K.

[0100] In the third embodiment, the ordinary cloud area pixels are excluded from the to-be-identified area, and the specific method includes: T 10.8 -T 12 ≥ fifth preset threshold value, the cloud area pixel is determined as an ordinary cloud area pixel.

[0101] The brightness temperature value of the 10.8 μm channel of the ordinary cloud area pixel is higher than that of the 12 μm channel of the pixel, and after setting an appropriate threshold value, the cloud removing module can identify the ordinary cloud area, that is, when T

[0102] After the cloud area pixels are excluded, if the image MIDI value of the to-be-identified area is higher than the sand-dust identification threshold value, it is determined that the to-be-identified area is a sand-dust area.

[0103] In the third embodiment, the MIDI value of the to-be-identified area pixel is the MIDI value of the current time pixel. The current time image is an image without cloud exclusion.

[0104] The system for identifying a sand-dust area based on infrared channels provided in the third embodiment can continuously monitor day and night to analyze the moving development path of the sand-dust, and accurately monitor and warn the sand-dust weather.

[0105] Embodiment 4:

[0106] In embodiment three, the main basis for the dust identification of the Fengyun-4 meteorological satellite infrared multi-spectral dust index (MIDI) is that the MIDI value of the dust area is obviously higher than the MIDI value before the dust occurs. Although some cloud area identification conditions have been used in dust identification, some cloud areas are not identified, resulting in some dust area misjudgment information, so it is necessary to remove the misjudged areas as dust areas. These misjudged dust areas are generally scattered in small pieces.

[0107] Through the analysis of the MIDI ratio of some examples of the identification target (i.e. the ratio of MIDI before and after the dust occurs), the MIDI ratio of the dust area, the cloud area and the desert clear sky area has the following characteristics: 1) The dust area generally has a large number of pixels with high MIDI ratio (i.e. dust endmember pixels); 2) The MIDI ratio of the desert clear sky area is obviously lower than that of the dust area; 3) High clouds (10.8 μm brightness temperature value of the pixel is lower than 240K) have a higher MIDI ratio; 4) The MIDI ratio of the ordinary cloud area is higher than that of the clear sky land and lower than that of the dust area; 5) Sometimes there are more small pieces (several pixels) with high MIDI ratio, which is comparable to the MIDI ratio of the dust area, which may be related to the background map.

[0108] This embodiment four is based on embodiment three, and provides an infrared multi-spectral dust misjudgment area detection system based on an infrared channel, to solve the problem of reducing and eliminating dust misjudgment information, as shown in Figure 4 The dust misjudgment area detection module judges the area of the dust area, and filters the dust area smaller than the preset area; the dust area with the number of dust endmember pixels exceeding the preset threshold is left as the high-confidence dust area, and the dust misjudgment area detection module judges whether the non-high-confidence dust area is within the preset distance of the high-confidence dust area. If it is not within the preset distance, it is considered that the dust area is misjudged.

[0109] The dust area generated in embodiment three is used to remove the dust area with an area smaller than the preset area. In this embodiment four, the dust area with a preset area of the dust misjudgment detection module is preferably 50 pixels. The dust area with a number of dust endmember pixels exceeding the preset number is considered as the high-confidence dust area. In this embodiment four, the number of dust endmember pixels of the dust misjudgment area detection module is preferably 300 pixels.

[0110] The dust misjudgment area detection module judges whether the non-high-confidence dust area is within the preset distance of the high-confidence dust area. If it is not within the preset distance, it is considered that the area previously determined as the dust area is misjudged.

[0111] In the high-confidence dust area, all pixels are regarded as dust pixels, so it is possible to retain the dust pixels covered by thin clouds, and then it is necessary to determine whether the remaining non-high-confidence dust area is in the vicinity of the high-confidence dust area. The method is that the dust misjudgment area detection module determines whether the non-high-confidence dust area is in the same geographical grid range as the high-confidence dust area. In this embodiment four, the size of the geographical grid is set to 2x2 latitude and longitude. If it is not in the same geographical grid range, it is considered that the dust area identified in embodiment three is misjudged.

[0112] The system for identifying dust areas based on infrared channels provided in this embodiment four can not only continuously monitor day and night to analyze the moving development path of dust, but also can more finely eliminate dust misjudgment information and improve the accuracy of infrared multispectral identification of dust areas.

[0113] Embodiment 5:

[0114] The electronic device provided in this embodiment five includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the method for identifying dust areas based on infrared channels of infrared multispectrum provided in any of the above embodiments.

[0115] As shown in Figure 5 The electronic device 10 can include a processor 100, a memory 101, a bus 102, and a communication interface 103. The processor 100, the communication interface 103, and the memory 101 are connected through the bus 102. The memory 101 stores a computer program executable on the processor 100. When the processor 100 executes the computer program, it executes the method for identifying dust areas based on infrared channels of infrared multispectrum provided in embodiment one and embodiment two.

[0116] The memory 101 can include a high-speed random access memory (RAM) and can also include a non-volatile memory such as at least one disk memory. The communication interface 103 can be wired or wireless, and can use the Internet, a wide area network, a local network, a metropolitan area network, etc.

[0117] The bus 102 can be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 101 is used to store programs, and the processor 100 executes the programs after receiving execution instructions. The method and system for identifying dust areas based on infrared channels of infrared multispectrum provided in embodiments one to four can be applied to the processor 100 or implemented by the processor 100.

[0118] The processor 100 can be an integrated circuit chip with signal processing capability. In implementation, each step of the above method can be completed by integrated logic circuits or instructions in the form of software in the processor 100. The processor 100 described above can be a general-purpose processor, which can include a central processing unit (CPU), a network processor (NP), etc.; or can be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a ready programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The method, steps and logic block diagram provided in Embodiments 1 to 4 for identifying a sand-dust area based on an infrared channel can be implemented or executed by the processor. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The method and system for identifying a sand-dust area based on an infrared channel provided in Embodiments 1 to 4 can be directly embodied as a hardware coding processor to execute, or be executed by a combination of hardware and software modules in the coding processor. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, or other mature storage medium in the art. The storage medium is located in the storage 101, and the processor 100 reads information in the storage 101 and combines the hardware to complete the steps of the above method.

[0119] The electronic device provided in Embodiment 5 has the same inventive concept as the method and system provided in Embodiments 1 to 4, and has the same beneficial effects as the method it adopts, runs or implements.

[0120] Embodiment 6:

[0121] The computer readable storage medium provided in Embodiment 6 has a computer program stored thereon, which is executed by a processor to implement the method for identifying a sand-dust area based on an infrared channel according to any of the above embodiments.

[0122] The computer readable storage medium is an optical disc 20, and the computer readable storage medium stores a computer program (i.e. a program product). When the computer program is run by a processor, the method provided in any of the above embodiments is executed.

[0123] It should be noted that examples of computer-readable storage media can include but are not limited to phase-change RAM (PRAM), static RAM (SRAM), dynamic RAM (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), flash memory or other optical, magnetic, or solid-state storage media. Details of such computer-readable storage media are not critical to the application.

[0124] The computer-readable storage medium provided in this embodiment six is based on the same inventive concept as the methods provided in the first embodiment and the second embodiment, and has the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0125] It should be noted that the term "module" is not intended to be limited to a particular physical form. Depending on the specific application, the module can be implemented as hardware, firmware, software and / or a combination thereof. In addition, different modules can share common components or even be implemented by the same components. There can or can not be a clear boundary between different modules.

[0126] The algorithms and displays presented herein are not inherently related to any particular computer, virtual apparatus, or other device. Various general purpose devices can be used with these examples based on the teachings herein. The structure required to construct such devices will be apparent to those skilled in the art from the description above. In addition, the present embodiment four is not intended to be limited to any particular programming language. It will be appreciated that a variety of programming languages can be used to implement the present embodiment four described herein, and that the examples given above are intended to be exemplary only.

[0127] It should be understood that, although the steps in the flowcharts of the drawings are shown in a sequential order, these steps are not necessarily performed in the order shown by the arrows. Unless otherwise explicitly stated herein, the steps are not necessarily performed in the order shown, and can be performed in other orders. Moreover, at least some of the steps in the flowcharts of the drawings can include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but can be performed at different times, and the order of their execution is not necessarily sequential, but can be performed alternately or alternately with at least part of other steps or sub-steps or stages of other steps.

[0128] The above embodiments only express the implementation of the present application, and the description is more specific and detailed, but it should not be understood as limiting the scope of the patent of the present application. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for identifying a dust area based on infrared channel and infrared multispectral, characterized in that the method The method comprises the following steps: obtaining the brightness temperature values of the 8.5 µm, 10.8 µm and 12 µm channels of the pixels in the infrared channel image of the to-be-judged area, and constructing an infrared multispectral dust index MIDI; judging and excluding the cloud area pixels of the to-be-judged area pixels; if the image MIDI value of the to-be-judged area is higher than the dust judgment threshold value after the cloud area pixel exclusion, it is determined that the to-be-judged area pixel is a dust pixel; the calculation method of the constructed infrared multispectral dust index MIDI is: MIDI = (T 8.5 + T 12 ) / (2 x T 10.8 ) where: T 8.5 , T 10.8 , T 12 are the brightness temperature values of the 8.5 pm, 10.8 pm and 12 pm channels of the pixels in the infrared channel image, respectively.

2. The method for identifying sand-dust region based on infrared channel of infrared multispectral according to claim 1, characterized in that, the specific method for judging and excluding the cloud area pixels of the to-be-judged area pixels comprises: When T 10.8 If the value is less than or equal to a first preset threshold, the cloud region pixel is determined as a high cloud region pixel.

3. The method for identifying sand-dust region based on infrared channel of infrared multispectral according to claim 2, characterized in that, the specific method for judging and excluding the cloud area pixels of the to-be-judged area pixels further comprises: When T 10.8 When it exceeds the preset threshold of Gaoyun, and T 10.8 With T 6.2 When the brightness temperature difference is less than or equal to the second preset threshold, the type of cloud area pixel is determined to be a high-cloud mixed area pixel containing high cloud information, where: T 6.2 This represents the brightness temperature value of the 6.2µm channel of a pixel in the infrared image.

4. The method for identifying sand-dust region based on infrared channel of infrared multispectral according to claim 3, characterized in that, the specific method for judging and excluding the cloud area pixels of the to-be-judged area pixels further comprises: When T 8.5 The brightness temperature difference between T 10.8 When the brightness temperature difference between T 12 The brightness temperature difference between T 10.8 The cloud band edge region pixel is determined.

5. The method for identifying sand-dust region based on infrared channel of infrared multispectral according to claim 4, characterized in that, the specific method for judging and excluding the cloud area pixels of the to-be-judged area pixels further comprises: T 10.8 When the brightness temperature difference value of T 12 is greater than the fifth preset threshold, the type of the cloud region pixel is determined as a normal cloud region pixel.

6. The method for identifying dust area based on infrared channel of infrared multispectral according to any one of claims 1 to 5, characterized in that, the method further comprises: in the area judged as a dust area, filtering the dust area with an area smaller than a preset area, and leaving the dust area with a dust end member pixel number exceeding a preset threshold value as a high-confidence dust area; judging whether a non-high-confidence dust area is within a preset distance of the high-confidence dust area, and if not, considering that the dust area is misjudged.

7. A system for identifying dust regions based on the infrared channel according to the method of any one of claims 1 to 5, characterized in that, The system comprises a constructed index module, a cloud removal module and a judgment module, wherein: the constructed index module obtains the brightness temperature values of the 8.5 µm, 10.8 µm and 12 µm channels of the pixels in the infrared channel image of the to-be-judged area, and constructs an infrared multispectral dust index MIDI; the cloud removal module judges and excludes the cloud area pixels of the to-be-judged area; after the cloud area pixel exclusion is completed, if the judgment module judges that the image MIDI value of the to-be-judged area is higher than the dust judgment threshold value, it is determined that the to-be-judged area is a dust area.

8. The method for identifying sand-dust region based on infrared channel of infrared multispectral according to claim 7, characterized in that, The system further comprises a dust misjudgment area detection module, which judges the area of the dust area, filters the dust area smaller than the preset area, and leaves the dust area with a dust end member pixel number exceeding a preset threshold value as a high-confidence dust area. The dust misjudgment area detection module judges whether a non-high-confidence dust area is within a preset distance of the high-confidence dust area, and if not, considers that the dust area is misjudged.

9. An electronic device, comprising: The system comprises a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the program to realize the method for judging the dust area based on the infrared channel according to any one of claims 1-6.

Images