CN112001263A - A method and system for selecting reference detectors of a linear scan remote sensor - Google Patents

Abstract

Embodiments of the present invention provide a method and system for selecting reference detectors of a line scan remote sensor. The method includes: acquiring an original remote sensing image, correcting the original remote sensing image through relative calibration, and obtaining a corrected image; based on the original remote sensing image The average brightness change index, the image information entropy index and the image radiation resolution index are obtained by calculating the corrected image and the image average brightness change index, the image information entropy index and the image radiation resolution index. The detector with the highest comprehensive score is used as the benchmark detector for relative calibration. Aiming at the problem of how to select the reference detector in the relative calibration process of the linear scan remote sensor, the embodiment of the present invention provides a scoring method based on the distribution characteristics of the output code value of the detector, and selects the detector with the highest score as the reference detector. From the perspective of the statistical characteristics of the earth observation data, the most suitable reference detector is selected.

Description

A method and system for selecting reference detectors of a linear scan remote sensor

technical field

The invention relates to the technical field of remote sensing, in particular to a method and a system for selecting a reference detector element of a linear array scanning remote sensor.

Background technique

In line array scanning imaging remote sensing data, streaks are very common. For example, the Medium Resolution Spectral Imager (MERSI) of Fengyun-3 (FY-3) is a 10/40-detector line array instrument, up and down the star. The original image has streaks. This kind of fringe phenomenon is generally corrected by the relative calibration between the detectors to remove the fringe.

At present, there are mainly three types of relative calibration methods: one is to use on-board normalization, that is, to use electronic technology to convert the original sampled code value (Digital Number, DN) according to the relative difference of the radiation response between the detectors, linear Converted and rounded and downloaded. This method can greatly suppress streaks, but often not completely. The other type is based on computer image algorithm elimination, such as wavelet transform, median filter and so on. These algorithms can achieve a good stripe removal effect for a single image. The disadvantage is that the algorithm relies on the image itself to extract correction parameters, but these parameters are not stable enough to meet the requirements of automatic processing; at the same time, due to the nonlinear processing of the image, the reflectivity information contained in the processed image is irreversibly changed. Introduce errors that are difficult to quantitatively evaluate for subsequent quantitative applications. In terms of time complexity, different algorithms consume different time, but they often affect the processing time of the business system. There is another category, the method of empirical distribution function matching is discussed more, the method is stable and reliable, and the time complexity is low, which hardly affects the quantitative application.

In this kind of relative calibration method, the reference detector is the key technical parameter, and different detectors in the channel are selected as the reference detector, and the quality of the remote sensing image after relative calibration is different. Generally, the detector with the most sensitive radiation response is used as the reference detector, but this method using a single measurement standard cannot fully reflect the influence of the reference detector selection and correction on the image.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a method and system for selecting a reference detector of a linear scanning remote sensor, so as to solve the defects in the prior art and realize the selection of the most suitable reference detector.

In a first aspect, an embodiment of the present invention provides a method for selecting a reference detector of a line scan remote sensor, including:

Obtaining an original remote sensing image, and correcting the original remote sensing image through relative calibration to obtain a corrected image;

Calculate the average brightness change index of the image based on the original remote sensing image and the corrected image;

Calculate the image information entropy index based on the original remote sensing image and the corrected image;

The image radiation resolution index is calculated based on the original remote sensing image and the corrected image;

By calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index, the comprehensive score of the detector is obtained, and the detector with the highest comprehensive score of the detector is selected as the reference detector for relative calibration .

Further, the average brightness change index of the image calculated based on the original remote sensing image and the corrected image, specifically includes:

Select any detector as a temporary reference detector, and construct a first relative calibration look-up table;

Correcting the original remote sensing image based on the first relative calibration lookup table to obtain a first corrected original image;

Calculate the absolute difference value of the average code value of the first corrected original image and the original remote sensing image;

Select the remaining detectors in the channel as the temporary reference detectors in turn, and repeat the above steps to construct the average code value absolute difference data set;

The average brightness variation index of the image is calculated based on the average code value absolute difference data set.

Further, the image information entropy index calculated based on the original remote sensing image and the corrected image, specifically includes:

Select any detector as a temporary reference detector, and construct a second relative calibration look-up table;

Correcting the original remote sensing image based on the second relative calibration lookup table to obtain a second corrected original image;

calculating the corrected image information entropy of the second corrected original image;

Select the remaining detectors in the channel as temporary reference detectors in turn, and repeat the above steps to construct an information entropy data set;

The image information entropy index is calculated based on the information entropy data set.

Further, the image radiation resolution index calculated based on the original remote sensing image and the corrected image, specifically includes:

Select any detector as a temporary reference detector, and construct a third relative calibration lookup table;

Correcting the original remote sensing image based on the third relative calibration lookup table to obtain a third corrected original image;

calculating the average effective code value of the third corrected original image;

Select the remaining detectors in the channel as temporary reference detectors in turn, and repeat the above steps to construct an average effective code value data set;

The image radiation resolution index is calculated based on the average effective code value data set.

Further, the comprehensive score of the probe obtained by calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index, specifically includes:

A weighted algorithm is used to calculate the comprehensive score of the probe.

Further, the weighting algorithm includes average weighting or non-average weighting.

Further, the relative calibration is realized by a cumulative probability method.

In a second aspect, an embodiment of the present invention also provides a system for selecting reference detectors of a line-scanning remote sensor, including:

an acquisition module, used for acquiring an original remote sensing image, and correcting the original remote sensing image through relative calibration to obtain a corrected image;

a first processing module, configured to calculate the average brightness change index of the image based on the original remote sensing image and the corrected image;

A second processing module, configured to calculate and obtain an image information entropy index based on the original remote sensing image and the corrected image;

a third processing module, configured to calculate and obtain an image radiation resolution index based on the original remote sensing image and the corrected image;

The comprehensive module is used to obtain the comprehensive score of the probe by calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index, and select the probe with the highest comprehensive score of the probe as the relative fixed score. Target datum probe.

In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements any of the above-mentioned programs when executing the program. The steps of the method for selecting the reference detector of the linear array scanning remote sensor.

In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the line scan remote sensor benchmark as described in any of the above The steps of the selection method of the probe.

The method and system for selecting a reference detector of a linear array scanning remote sensor provided by the embodiment of the present invention, by aiming at the problem of how to select a reference detector in the relative calibration process of a linear scanning remote sensor, the output code value distribution based on the detector is given. According to the scoring method of the characteristic, the detector with the highest score is selected as the reference detector, and the most suitable reference detector is selected from the perspective of the statistical characteristics of the earth observation data.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a schematic flowchart of a method for selecting a reference detector element of a linear scanning remote sensor provided by an embodiment of the present invention;

FIG. 2 is a schematic diagram of the frequency distribution of each probe code value and the cumulative probability of the code value in channel 3 provided by an embodiment of the present invention;

3 is a schematic diagram of a relative calibration look-up table of channel 3 provided by an embodiment of the present invention;

4 is a schematic diagram of the image correction average code value and the absolute difference value of the original image and the distribution of the average brightness index of the image provided by an embodiment of the present invention;

5 is a schematic diagram of the distribution of the image information entropy along with the probe number and the distribution of the image information entropy index provided by the embodiment of the present invention;

6 is a schematic diagram of the distribution of the average effective code value of the image with the detector number and the distribution of the image radiation resolution index provided by the embodiment of the present invention;

7 is a schematic diagram of the distribution of the detector comprehensive score with the detector number provided by the embodiment of the present invention;

8 is a schematic structural diagram of a system for selecting reference detectors of a linear scanning remote sensor provided by an embodiment of the present invention;

FIG. 9 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The terms "first", "second" and "third" in the embodiments of the present application are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implying the number of indicated technical features. Thus, features defined as "first", "second" and "third" may expressly or implicitly include at least one of such features. In the description of this application, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a system, product or device comprising a series of components or units is not limited to the listed components or units, but may optionally also include components or units not listed, or Other parts or units inherent in the equipment. In the description of the present application, "a plurality of" means at least two, such as two, three, etc., unless otherwise expressly and specifically defined.

In view of the problems existing in the prior art, the embodiment of the present invention adopts a comprehensive scoring method to integrate image brightness change, image information entropy and image radiation resolution into a single index for evaluating the potential ability of each detector as a reference detector.

1 is a schematic flowchart of a method for selecting a reference detector of a linear scanning remote sensor provided by an embodiment of the present invention, as shown in FIG. 1 , including:

S1, obtaining an original remote sensing image, and correcting the original remote sensing image through relative calibration to obtain a corrected image;

Specifically, taking Fengyun-3 as an example, FY-3/MERSI uses multi-element linear array scanning to obtain Earth observation images, in which channels 1-4 are channels with a resolution of 250 meters, and 40 detectors are scanned in parallel; channels 6-20 For the 1000-meter resolution channel, 10 detectors are scanned in parallel. Due to the natural radiation response difference of the linear detector, the original image has streaks. Therefore, the relative calibration of the image needs to be completed. The calibration look-up table is used to correct the output code values of other detectors to complete the relative calibration.

S2, calculating the average brightness change index of the image based on the original remote sensing image and the corrected image;

The average brightness of the original remote sensing image can be expressed by the average value of all observed output code values. Considering that the image brightness should change little before and after relative calibration, an average brightness change index is constructed to evaluate the difference before and after correction. Taking a detector in the channel as a temporary reference detector, a relative calibration look-up table is constructed using the cumulative probability method. Use the relative calibration lookup table to correct the original remote sensing data of other detectors. Calculate the average code value of the corrected image and the average code value of the original image, and then calculate the absolute difference between the two. Taking each detector in the channel as a temporary reference detector in turn, the above steps are repeated to obtain a data set of absolute difference values of average code values corresponding to each detector. Taking the minimum value of the absolute difference value of the average code value as 100 points and the maximum value as 0 point, the linear score of the absolute difference value of the average code value of each detector is given as the average brightness change index of the image. The average brightness change index of the image, which describes the difference between the average brightness of the corrected image and the average brightness of the original image when the detector is used as a temporary reference detector.

S3, based on the original remote sensing image and the corrected image, the image information entropy index is obtained by calculating;

Image information entropy is used to describe the amount of information contained in the image, which can be calculated by the probability corresponding to each output code value. Based on the relative scaled images of different detectors, there are differences in their image information entropy. Considering that the relatively scaled image should provide as much information as possible, an information entropy index is constructed to evaluate the difference in information entropy. Taking a detector in the channel as a temporary reference detector, a relative calibration look-up table is constructed using the cumulative probability method. Use the relative calibration lookup table to correct the original remote sensing data of other detectors. Calculate the information entropy of the corrected image. Taking each detector in the channel as a temporary reference detector in turn, the above steps are repeated to obtain an image information entropy data set corresponding to each detector. Taking the maximum value of the image information entropy as 100 points and the minimum value as 0 points, the linear score of the image information entropy of each probe is given as the image information entropy index. The image information entropy index describes the amount of image information after correcting the original image when the detector is used as a temporary reference detector. The higher the image information entropy index score, the greater the amount of information contained.

S4, calculating and obtaining an image radiation resolution index based on the original remote sensing image and the corrected image;

Radiometric resolution is the ability to distinguish small changes in energy in an image. The number of gray levels output by the remote sensor for a certain range of incident energy. The more gray levels used, the finer the image describes the energy change. Taking a detector in the channel as a temporary reference detector, a relative calibration look-up table is constructed using the cumulative probability method. Use the relative calibration lookup table to correct the original remote sensing data of other detectors. After the image is corrected, the minimum code value and the maximum code value corresponding to the medium energy are taken. Count the frequency of each code value between the minimum code value and the maximum code value of each detector in the corrected image, count the number of code values whose code value frequency is not zero, and record it as the amount of valid code values. The average value of valid code values of all probes is calculated as the average value of valid code values of the temporary reference probe. Each detector in the channel is taken as a temporary reference detector in turn, and the above steps are repeated to obtain an average valid code value data set corresponding to each detector. Taking the maximum value of the average effective code value as 100 points and the minimum value as 0 points, the linear score of the average effective code value of each detector is given as the image radiation resolution index. The radiometric resolution index describes the radiometric resolution capability of the image after correcting the original remote sensing image when the detector is used as a temporary reference detector. The larger the index, the better the radiometric resolution.

S5, by calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index to obtain a probe comprehensive score, and selecting the probe with the highest comprehensive score of the probe as a benchmark for relative scaling Probe.

Precise and stable relative calibration is an important component of the remote sensing data ground preprocessing system, which is of great significance to subsequent data applications. The selection of the reference detector is an important part of the relative calibration, and the comprehensive influence of the reference detector used on the brightness of the corrected image, the amount of image information, and the image radiation resolution should be considered. The comprehensive score of the detector is based on the average brightness change index of the image, the image information entropy index and the radiation resolution index, and the value is calculated by the method of weighted average. The weight of the three indices can be the same, or the weight of each index can be adjusted according to the requirements of subsequent applications. The detector with the highest comprehensive score of the detector is used as the reference detector for relative calibration for relative calibration.

In the embodiment of the present invention, a scoring method based on the distribution characteristics of the output code value of the detector is provided for the problem of how to select the reference detector during the relative calibration process of the linear array scanning remote sensor, and the detector with the highest score is selected as the reference detector, From the point of view of the statistical characteristics of the earth observation data, the most suitable reference detector is selected.

Based on the above embodiment, step S2 in the method specifically includes:

Select any detector as a temporary reference detector, and construct a first relative calibration look-up table;

Correcting the original remote sensing image based on the first relative calibration lookup table to obtain a first corrected original image;

Calculate the absolute difference value of the average code value of the first corrected original image and the original remote sensing image;

Select the remaining detectors in the channel as the temporary reference detectors in turn, and repeat the above steps to construct the average code value absolute difference data set;

The average brightness variation index of the image is calculated based on the average code value absolute difference data set.

Specifically, in the first step, a relative calibration look-up table is constructed with one detector as a temporary reference detector. A detector i of the selected channel is used as a temporary reference detector. First, the frequency hist _i (DN ^* ) of each original code value DN ^* output by this detector in the earth observation image is counted, and then the code value of the earth observation image is calculated. The cumulative probability distribution function is as follows:

In the formula, Pi is the cumulative probability distribution function of the code value, _maxDN is the maximum design output code value of the channel, and the quantization level of the MERSI channel is 12bit, then the maxDN is 4095.

The undetermined detector j is processed in the same way to obtain the code value cumulative probability distribution function P _j . After relative scaling, the code value distribution output by detector j should be the same as that of detector i, then the output code value of detector j is relative to the scaled output code value. The corresponding relationship with the original code value can be obtained by using the following formula:

为P_i的反函数。In the formula,

is the inverse function of _Pi .

Due to the discrete value, we can construct a static code value look-up table for relative scaling of detector j to temporary reference detector i:

In the formula, LUT _j is the look-up table (Look-Up-Table, LUT) of the detector j.

In the second step, the original image is corrected using the relative scaling lookup table. The code value of each pixel in the original image, according to the detector number and the size of the code value, is converted into the corrected code value using a look-up table to complete the relative calibration of the remote sensing image, as follows:

In the formula, N is the number of all detectors in the channel. For MERSI's 250-meter resolution channels (channels 1-4, 24, and 25), N is 40; for 1000-meter resolution channels (channels 5-19), N is 10.

计算式如下：The third step is to calculate the absolute difference between the average code values of the images before and after the correction. Calculate the average code value of the relative scaled image

The calculation formula is as follows:

In the formula, the numerator is the sum of the code values of all pixels, and the denominator Number _AllPixals is the number of all pixels.

计算式如下：Calculate the average code value of the original image

The calculation formula is as follows:

Furthermore, the absolute difference value ΔDN _mean of the average code value of the relative scaled image and the average code value of the original image can be calculated by the following formula.

In the fourth step, each detector in the channel is used as the temporary reference detector in turn, and the first to the third step are repeated to construct the average code value absolute difference data set ΔDNSet _mean , as follows:

ΔDNSet _mean = {ΔDN _mean (i), i=1,2,...,N}

The fifth step is to calculate the overall brightness change index. The minimum value of the absolute difference of the average code value is 100 points, and the maximum value is 0 points. When each detector is used as a temporary reference detector, the linear score of the average brightness change before and after the relative calibration of the image is calculated as the average brightness change index of the image. , denoted as I _bright , the calculation formula is as follows:

When the absolute difference of the average code value is the minimum value, the value of I _bright is 100, and the average brightness change of the image before and after the relative calibration is the smallest; when the absolute difference of the average code value is the maximum value, the value of I _bright is 0, and in other cases , the value of I _bright is distributed between 0 and 100.

Based on any of the above embodiments, step S3 in the method specifically includes:

Select any detector as a temporary reference detector, and construct a second relative calibration look-up table;

Correcting the original remote sensing image based on the second relative calibration lookup table to obtain a second corrected original image;

calculating the corrected image information entropy of the second corrected original image;

Select the remaining detectors in the channel as temporary reference detectors in turn, and repeat the above steps to construct an information entropy data set;

The image information entropy index is calculated based on the information entropy data set.

Specifically, in the first step, a relative calibration look-up table is constructed by using a cumulative probability method with a detector in the channel as a temporary reference detector. A detector i of the selected channel is used as a temporary reference detector. First, the frequency hist _i (DN ^* ) of each original code value DN ^* input by this detector in the earth observation image is counted, and then the code value of the earth observation image is calculated. The cumulative probability distribution function is as follows:

In the formula, Pi is the cumulative probability distribution function of the code value, and _maxDN is the maximum design output code value of the channel. If the channel quantization level is 12bit, the maxDN is 4095.

The undetermined detector j is processed in the same way to obtain the code value cumulative probability distribution function P _j . After relative scaling, the code value distribution output by the detector j should be the same as that of the detector i, then the output code value of the detector j is relative to the scaled output code value. The corresponding relationship with the original code value can be obtained by using the following formula:

为P_i的反函数。In the formula,

is the inverse function of _Pi .

Due to the discrete value, we can construct a static code value look-up table for relative scaling of detector j to temporary reference detector i:

In the formula, LUT _j is the look-up table (Look-Up-Table, LUT) of the detector j.

The second step is to correct the original remote sensing data of other detectors by using the relative calibration look-up table. The code value of each pixel in the original image, according to the detector number and the size of the code value, is converted into the corrected code value using a look-up table to complete the relative calibration of the remote sensing image, as follows:

where N is the number of all detectors in the channel.

The third step is to calculate the information entropy of the corrected image. After calculating the relative scaling, the frequency hist(DN) corresponding to each code value of the image is calculated, and then the probability p(DN) corresponding to each code value is calculated, as follows:

In the formula, Number _AllPixals is the number of all pixels.

The image information entropy H after relative scaling can be calculated by the following formula:

In the fourth step, each detector in the channel is used as the temporary reference detector in turn, and the first to the third step are repeated to obtain the image information entropy data set HSet corresponding to each detector.

HSet _effective-DN = {H(i), i=1,2,...,N}

The fifth step is to calculate the image information entropy index. Taking the maximum value of the image information entropy as 100 points and the minimum value as 0 points, when calculating each detector as a temporary reference detector, the linear score of the image information entropy is taken as the image information entropy index, denoted as I _entropy , and the calculation formula is as follows: shown:

When the image information entropy is the maximum value, the value of I _entropy is 100, and the relative scaled image information is the largest when the probe i is the temporary reference detector; when the image information entropy is the minimum value, the value of I _entropy is 0, otherwise, the value of I _entropy is distributed between 0 and 100.

Based on any of the above embodiments, step S4 in the method specifically includes:

Select any detector as a temporary reference detector, and construct a third relative calibration lookup table;

Correcting the original remote sensing image based on the third relative calibration lookup table to obtain a third corrected original image;

calculating the average effective code value of the third corrected original image;

Select the remaining detectors in the channel as temporary reference detectors in turn, and repeat the above steps to construct an average effective code value data set;

The image radiation resolution index is calculated based on the average effective code value data set.

Specifically, in the first step, a relative calibration look-up table is constructed by using a cumulative probability method with a detector in the channel as a temporary reference detector. A detector i of the selected channel is used as a temporary reference detector. First, the frequency hist _i (DN ^* ) of each original code value DN ^* input by this detector in the earth observation image is counted, and then the code value of the earth observation image is calculated. The cumulative probability distribution function is as follows

In the formula, Pi is the cumulative probability distribution function of the code value, and _maxDN is the maximum design output code value of the channel. If the channel quantization level is 12bit, the maxDN is 4095.

The undetermined detector j is processed in the same way to obtain the code value cumulative probability distribution function P _j . After relative scaling, the code value distribution output by the detector j should be the same as that of the detector i, then the output code value of the detector j is relative to the scaled output code value. The corresponding relationship with the original code value can be obtained by using the following formula:

为P_i的反函数。In the formula,

is the inverse function of _Pi .

Due to the discrete value, we can construct a static code value look-up table for relative scaling of detector j to temporary reference detector i:

In the formula, LUT _j is the look-up table (Look-Up-Table, LUT) of the detector j.

The second step is to correct the original remote sensing data of other detectors by using the relative calibration look-up table. The code value of each pixel in the original image, according to the detector number and the size of the code value, is converted into the corrected code value using a look-up table to complete the relative calibration of the remote sensing image, as follows:

where N is the number of all detectors in the channel.

The third step is to calculate the average effective code value. The minimum code value and the maximum code value corresponding to the medium incident energy after image correction. After the image is corrected, the frequency of each code value between the minimum code value and the maximum code value of each detector is counted, and the number of code values whose code value frequency is not zero is counted as the amount of valid code values. The average value of valid code values of all probes is calculated as the average value of valid code values of the temporary reference probe.

After the image is corrected, the cumulative probability distribution P(DN) of all image code values is counted, and the image cumulative probability between 1% and 99% is set as the medium energy distribution range of the image. The code value corresponding to 1% is marked as the minimum valid code value DN _min , and the code value corresponding to 99% is marked as the maximum valid code value DN _max , which is calculated as follows:

DN _min = arg (P(DN) = 1%), DN _max = arg (P(DN) = 99%)

The probability corresponding to the minimum valid code value and the maximum valid code value may also use other probability values other than 1% and 99%, and may be appropriately adjusted according to the requirements of subsequent specific application scenarios of relative scaling.

The frequency of occurrence of each detector code value hist _j (DN),j∈[0,N] is counted separately. After the lookup table is corrected, the individual code values of some detectors will not appear in the corrected image. Therefore, the corresponding hist _j (DN) = 0. The effective output code value flag _j (DN) of each detector is marked, as shown in the following formula:

Between the minimum valid code value and the maximum valid code value, count the number of valid output code values Num _DN of each detector, and the calculation formula is as follows:

Calculate the average Num _effective-DN of the number of effective output code values of all detectors in the channel. The calculation formula is as follows:

The average number of valid code values describes the average number of valid code values required for the corrected image to describe the medium incident energy when the detector is used as a temporary reference detector, that is, the radiation resolution capability of the corrected image.

In the fourth step, each detector in the channel is used as the temporary reference detector in turn, and the first to the third step are repeated to construct the average effective code value data set NumSet _effective-DN corresponding to each detector.

NumSet _effective-DN = {Num _effective-DN (i), i=1,2,...,N}

The fifth step is to calculate the radiation resolution index. Taking the maximum value of the average effective code value as 100 points and the minimum value as 0 points, calculate the linear score of the average effective code value of each detector as the image radiation resolution index, denoted as I _{energy-resolution} , the calculation formula is as follows Show.

When the average valid code value is the maximum value, the value of I _{energy-resolution} is 100; when the average valid code value is the minimum value, the value of I _{energy-resolution} is 0. In other cases, the value distribution of I _{energy-resolution} between 0 and 100.

Based on any of the above embodiments, step S5 in the method specifically includes:

A weighted algorithm is used to calculate the comprehensive score of the probe.

Wherein, the weighting algorithm includes average weighting or non-average weighting.

Specifically, the comprehensive score of the detector is based on the average brightness change index of the image, the image information entropy index and the radiation resolution index, and the value is calculated by the method of weighted average. The weight of the three indices can be the same, or the weight of each index can be adjusted according to the requirements of subsequent applications. The detector with the highest comprehensive score of the detector is used as the reference detector for relative calibration.

The probe comprehensive score can be calculated using the following formula:

I _estimate (i)=ω _bright I _bright (i)+ω _entropy I _entropy (i)+ω _{energy-resolution} I _{energy-resolution} (i)

In the formula, i is the detector number, I _estimate is the comprehensive score of the detector, I _bright is the average brightness change index of the image, I _entropy is the image information entropy index, and I _{energy-resolution} is the image radiation resolution index; ω _bright , ω _entropy and ω _{energy-resolution} are the weights of three exponents, respectively, which requires ω _bright + ω _entropy + ω _{energy-resolution} = 1, and generally takes ω _bright = ω _entropy = ω _{energy-resolution} = 1/3.

Take the detector with the highest I _estimate as the reference detector, i.e. i _standard =argmax(I _estimate (i)), and i _standard is the reference detector number.

The following is the implementation process of the FY-3D medium resolution spectral imager (MERSI) in the embodiment of the present invention. Taking channel 3 as an example, the Earth observation data of MERSI from 2018-4-10 to 2018-4-18 were counted.

The frequency map (hist) of each probe code value is counted, as shown in the left figure in FIG. 2 , and the code value cumulative probability distribution function P(DN) is shown in the right figure in FIG. 2 . From the data distribution in the figure, it can be known that there is a difference in radiation response among the detectors in channel 3.

Taking the 26th detector as the temporary reference detector, according to the method described in the foregoing embodiment, the relative calibration look-up table is constructed as shown in FIG. 3 .

Likewise, other detectors in the channel can be used as temporary reference detectors to construct a relative scaling lookup table.

First, according to the method described in the foregoing embodiment, the absolute difference of the average code value of each probe is calculated, as shown in the left figure in FIG. 4 , and the calculated overall brightness variation index is shown in the right figure in FIG. 4 . Using different detectors as temporary reference pixels, the corrected image differs from the original average brightness. The left panel in Figure 4 shows that detector 10 has the smallest difference, and detector 14 has the largest difference. After linear scoring, the overall brightness change index of detector 10 is 100, while the overall brightness change index of detector 14 is 0, and the index of other detectors is between 0 and 100.

Next, according to the method described in the foregoing embodiment, the image information entropy of each detector is calculated, as shown in the left figure in FIG. 5 , and the calculated image information entropy index is shown in the right figure in FIG. 5 . Using different detectors as temporary reference pixels, the corrected image information entropy. The left image in Fig. 5 shows that the image information entropy of detector 20 is the largest, and the information entropy of the image of detector 14 is the smallest. After linear scoring, the image information entropy index of probe No. 20 is 100, while the image information entropy index of probe No. 14 is 0.

Thirdly, according to the method described in the foregoing embodiment, the average effective code value of each detector is calculated, as shown in the left figure in FIG. 6 , and the calculated image radiation resolution index is shown in the right figure in FIG. 6 . Using different detectors as temporary reference pixels, the average effective code value after correction. The left figure in Fig. 6 shows that the average effective code value of the detector No. 20 is the largest, and the average effective code value of the detector No. 14 is the smallest. After linear scoring, the image radiometric resolution index of detector 20 is 100, while the image radiometric resolution index of detector 14 is 0.

Further, based on the average brightness change index of the image, the image information entropy index and the radiation resolution index, the method of equal weighted weighted average is used to calculate the comprehensive score of the probe, as shown in Figure 7. The highest score is No. 19 probe, while the benchmark probe currently used in business is No. 26 probe. Its comprehensive score is 76.9 points, ranking only 15th. Therefore, it is more appropriate to use probe No. 19 as the reference probe in business.

The following describes a system for selecting a reference detector of a line scan remote sensor provided by an embodiment of the present invention, the system for selecting a reference detector for a line scan remote sensor described below and the selection of a reference detector for a line scan remote sensor described above The methods can refer to each other correspondingly.

FIG. 8 is a schematic structural diagram of a system for selecting reference detectors of a line scanning remote sensor provided by an embodiment of the present invention. As shown in FIG. 8 , it includes: an acquisition module 81 , a first processing module 82 , a second processing module 83 , The third processing module 84 and the synthesis module 85; wherein:

The acquisition module 81 is used to acquire the original remote sensing image, and the original remote sensing image is corrected through relative calibration to obtain a corrected image; the first processing module 82 is used to calculate and obtain an image based on the original remote sensing image and the corrected image. The average brightness change index; the second processing module 83 is used to calculate the image information entropy index based on the original remote sensing image and the corrected image; the third processing module 84 is used to calculate the image information entropy index based on the original remote sensing image and the corrected image The image radiation resolution index is obtained by calculating; the comprehensive module 85 is configured to obtain the comprehensive score of the probe by calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index, and select the comprehensive score of the probe. The detector with the highest score serves as the reference detector for relative scaling.

In the embodiment of the present invention, a scoring method based on the distribution characteristics of the output code value of the detector is provided for the problem of how to select the reference detector during the relative calibration process of the linear array scanning remote sensor, and the detector with the highest score is selected as the reference detector, From the point of view of the statistical characteristics of the earth observation data, the most suitable reference detector is selected.

FIG. 9 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG. 9 , the electronic device may include: a processor (processor) 910, a communication interface (Communications Interface) 920, a memory (memory) 930, and a communication bus 940, The processor 910 , the communication interface 920 , and the memory 930 communicate with each other through the communication bus 940 . The processor 910 can invoke the logic instructions in the memory 930 to execute a method for selecting a reference detector of a line scan remote sensor, the method comprising: acquiring an original remote sensing image, revising the original remote sensing image through relative calibration, and obtaining the revision. image; based on the original remote sensing image and the revised image, the average brightness change index of the image is calculated; based on the original remote sensing image and the revised image, the image information entropy index is calculated; based on the original remote sensing image and the After the correction, the image is calculated to obtain the image radiation resolution index; by calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index, the comprehensive score of the detector is obtained, and the comprehensive score of the detector is selected. The highest detector serves as the reference detector for relative scaling.

In addition, the above-mentioned logic instructions in the memory 930 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

On the other hand, an embodiment of the present invention also provides a computer program product, the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, when the program instructions When executed by a computer, the computer can execute the method for selecting a reference detector of a line scan remote sensor provided by the above method embodiments, the method includes: acquiring an original remote sensing image, and correcting the original remote sensing image through relative calibration, obtain the corrected image; calculate the average brightness change index based on the original remote sensing image and the corrected image; calculate the image information entropy index based on the original remote sensing image and the corrected image; calculate the image information entropy index based on the original remote sensing image and the corrected image to obtain the image radiation resolution index; by calculating the average brightness change index of the image, the image information entropy index and the image radiation resolution index, the comprehensive score of the detector is obtained, and the detector is selected. The detector with the highest comprehensive score is used as the reference detector for relative calibration.

In yet another aspect, embodiments of the present invention further provide a non-transitory computer-readable storage medium on which a computer program is stored, and the computer program is implemented by a processor to execute the line scan remote sensor provided by the above embodiments when the computer program is executed. A method for selecting a reference detector, the method comprising: acquiring an original remote sensing image, correcting the original remote sensing image through relative calibration to obtain a corrected image; calculating an average image based on the original remote sensing image and the corrected image brightness change index; image information entropy index is calculated based on the original remote sensing image and the corrected image; image radiation resolution index is calculated based on the original remote sensing image and the corrected image; The change index, the image information entropy index, and the image radiation resolution index are calculated to obtain a detector comprehensive score, and the detector with the highest detector comprehensive score is selected as a reference detector for relative calibration.

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on this understanding, the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A method for selecting a reference probe element of a linear array scanning remote sensor is characterized by comprising the following steps:

acquiring an original remote sensing image, and correcting the original remote sensing image through relative calibration to obtain a corrected image;

calculating to obtain an image average brightness change index based on the original remote sensing image and the corrected image;

calculating to obtain an image information entropy index based on the original remote sensing image and the corrected image;

calculating to obtain an image radiation resolution index based on the original remote sensing image and the corrected image;

and calculating the image average brightness change index, the image information entropy index and the image radiation resolution index to obtain a comprehensive probing element score, and selecting the probing element with the highest comprehensive probing element score as a relative calibration reference probing element.

2. The method for selecting the linear array scanning remote sensor reference probe element according to claim 1, wherein the calculating based on the original remote sensing image and the corrected image to obtain an image average brightness change index specifically comprises:

selecting any probe element as a temporary reference probe element, and constructing a first relative calibration lookup table;

correcting the original remote sensing image based on the first relative calibration lookup table to obtain a first corrected original image;

calculating the absolute difference value of the average code values of the first correction original image and the original remote sensing image;

sequentially selecting the rest probe elements in the channel as temporary reference probe elements, repeating the steps, and constructing an average code value absolute difference data set;

and calculating the image average brightness change index based on the average code value absolute difference data set.

3. The method for selecting the linear array scanning remote sensor reference probe element according to claim 1, wherein the step of calculating an image information entropy index based on the original remote sensing image and the corrected image specifically comprises:

selecting any probe element as a temporary reference probe element, and constructing a second relative calibration lookup table;

correcting the original remote sensing image based on the second relative calibration lookup table to obtain a second corrected original image;

calculating a correction image information entropy of the second correction original image;

sequentially selecting the rest probe elements in the channel as temporary reference probe elements, repeating the steps, and constructing an information entropy data set;

and calculating the image information entropy index based on the information entropy data set.

4. The method for selecting the remote sensor reference probe element of the linear array scanning according to claim 1, wherein the step of calculating the image radiation resolution index based on the original remote sensing image and the corrected image specifically comprises the steps of:

selecting any probe element as a temporary reference probe element, and constructing a third relative calibration lookup table;

correcting the original remote sensing image based on the third relative calibration lookup table to obtain a third corrected original image;

calculating the average effective code value amount of the third correction original image;

sequentially selecting the rest probe elements in the channel as temporary reference probe elements, repeating the steps, and constructing an average effective code value data set;

and calculating the image radiation resolution index based on the average effective code value data set.

5. The method for selecting the remote sensor reference probe element for linear array scanning according to claim 1, wherein the step of calculating the image average brightness change index, the image information entropy index and the image radiation resolution index to obtain the probe element comprehensive score specifically comprises the steps of:

and calculating by adopting a weighting algorithm to obtain the comprehensive score of the probe element.

6. The method of claim 5, wherein the weighting algorithm comprises an average weighting or a non-average weighting.

7. A method for selecting a remote sensor reference probe as claimed in any one of claims 1 to 6, wherein the relative scaling is achieved by cumulative probability.

8. A system for selecting a reference probe element of a linear array scanning remote sensor is characterized by comprising:

the acquisition module is used for acquiring an original remote sensing image, and correcting the original remote sensing image through relative calibration to obtain a corrected image;

the first processing module is used for calculating to obtain an image average brightness change index based on the original remote sensing image and the corrected image;

the second processing module is used for calculating to obtain an image information entropy index based on the original remote sensing image and the corrected image;

the third processing module is used for calculating to obtain an image radiation resolution index based on the original remote sensing image and the corrected image;

and the comprehensive module is used for calculating the image average brightness change index, the image information entropy index and the image radiation resolution index to obtain a comprehensive probe element score, and selecting the probe element with the highest comprehensive probe element score as a relative calibration reference probe element.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the method of selecting a remote sensor reference probe as claimed in any one of claims 1 to 7.

10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of a method for reference probe selection for a line scanning remote sensor according to any of claims 1 to 7.

Images