CN105426903A - Cloud determination method and system for remote sensing satellite images - Google Patents
Cloud determination method and system for remote sensing satellite images Download PDFInfo
- Publication number
- CN105426903A CN105426903A CN201510705957.0A CN201510705957A CN105426903A CN 105426903 A CN105426903 A CN 105426903A CN 201510705957 A CN201510705957 A CN 201510705957A CN 105426903 A CN105426903 A CN 105426903A
- Authority
- CN
- China
- Prior art keywords
- cloud
- remote sensing
- image data
- sensing satellite
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a cloud determination method and a system for remote sensing satellite images. The method includes: receiving remote sensing satellite image data; performing format-removing processing of the remote sensing satellite image data; according to a preset data sharding rule, partitioning the remote sensing satellite image data after format-removing processing, and obtaining a plurality of sub-image data; and with the combination of texture features of visible light images, conducting cloud detection of each sub-image data by employing a classifier of a support vector machine in order to recognize a cloud mask in the remote sensing satellite image data. According to the method and the system, compared with cloud mask recognition of multi-spectrum images in the remote sensing satellite images in the prior art, cloud mask recognition is realized by employing the classifier of the support vector machine with the combination of the remote sensing satellite images and the texture features of the visible light images, the accuracy of the scheme can be effectively guaranteed via a general and engineering design, the cloud mask and other target substances such as snow and sand are effectively distinguished and classified, and the recognition rate can be high.
Description
Technical field
The present invention relates to technical field of image processing, particularly a kind of cloud of remote sensing satellite image sentences method and system.
Background technology
In remote sensing satellite image data, owing to being subject to the restriction of the conditions such as weather, existed in a large number by the data of cloud cover, these data itself do not possess intelligence value, but occupy the storage resources of the data handling system on ground, bring interference to the interpretation work in later stage, information making etc.
Therefore, in order to solve this interference, need the cloud mask that identifies in the remote sensing satellite image data received, so can remove in subsequent operation these cloud masks corresponding by the invalid data of cloud cover.
Summary of the invention
In view of this, the cloud that the invention provides a kind of remote sensing satellite image sentences method and system, in order to solve in prior art the technical matters needing to identify the remote sensing satellite image data medium cloud mask received.
The cloud that the invention provides a kind of remote sensing satellite image sentences method, and described method comprises:
Receive remote sensing satellite image data;
Solution format analysis processing is carried out to described remote sensing satellite image data;
According to the data segmentation rules preset, the remote sensing satellite image data of separating after format analysis processing are carried out piecemeal, obtains multiple sub-image data;
In conjunction with the textural characteristics of visible images, the sorter of support vector machine is adopted to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data.
Said method, preferably, at the textural characteristics in conjunction with visible images, adopt the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, after identifying the cloud mask in described remote sensing satellite image data, described method also comprises:
Data corresponding for described cloud mask are carried out record.
Said method, preferably, the described textural characteristics in conjunction with visible images, adopts the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data, comprising:
According to the intensity slicing threshold value preset, Threshold segmentation process is fixed respectively to sub-image data described in every block, to judge the first cloud sector view data in sub-image data described in every block, the image area of described first cloud sector view data is less than the image area of the sub-image data belonging to it;
In conjunction with the textural characteristics of visible images, the sorter of at least one support vector machine is adopted to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block;
Based on the first cloud sector view data medium cloud mask described in every block, determine the cloud mask in described remote sensing satellite image data.
Said method, preferably, the described textural characteristics in conjunction with visible images, the sorter of at least one support vector machine is adopted to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block, comprising:
Calculate the cloud sector gray level co-occurrence matrixes of the first cloud sector view data described in every block;
Based on described cloud sector gray level co-occurrence matrixes, calculate the textural characteristics of the visible images of the first cloud sector view data described in every block, described textural characteristics comprises: angle second moment, contrast, average, entropy, variance, contrast divide square;
With described textural characteristics for characteristic of division, the sorter of at least one support vector machine is utilized to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
Said method, preferably, carries out solution format analysis processing to described remote sensing satellite image data, comprising:
Frame synchronous ssearching is carried out to described remote sensing satellite image data, to search frame synchronization head;
Based on described frame synchronization head, determine the target data frame in described remote sensing satellite image data;
To between the different spectral coverage in described remote sensing satellite image data according to the frame count alignment operation of described target data frame, to complete the solution format analysis processing of described remote sensing satellite image data.
The cloud that present invention also offers a kind of remote sensing satellite image sentences system, and described system comprises:
Image receiving unit, for receiving remote sensing satellite image data;
Separate format analysis processing unit, for carrying out solution format analysis processing to described remote sensing satellite image data;
The remote sensing satellite image data of separating after format analysis processing, for the data segmentation rules according to budget, are carried out piecemeal, are obtained multiple sub-image data by image block unit;
Image detecting element, for the textural characteristics in conjunction with visible images, adopts the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data.
Said system, preferably, also comprises:
Data corresponding for described cloud mask, for identify the cloud mask in described remote sensing satellite image data at described image detecting element after, are carried out record by data record unit.
Said system, preferably, described image detecting element comprises:
First classification subelement, for the intensity slicing threshold value that foundation is preset, Threshold segmentation process is fixed respectively to sub-image data described in every block, to judge the first cloud sector view data in sub-image data described in every block, the image area of described first cloud sector view data is less than the image area of the sub-image data belonging to it;
Second classification subelement, for adopting at least one support vector machine classifier to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block;
Cloud mask determination subelement, for based on the first cloud sector view data medium cloud mask described in every block, determines the cloud mask in described remote sensing satellite image data.
Said system, preferably, described second classification subelement, comprising:
Co-occurrence matrix computing module, for calculating the cloud sector gray level co-occurrence matrixes of the first cloud sector view data described in every block;
Textural characteristics computing module, for based on described cloud sector gray level co-occurrence matrixes, calculate the textural characteristics of the visible images of the first cloud sector view data described in every block, described textural characteristics comprises: angle second moment, contrast, average, entropy, variance, contrast divide square;
Image classification module, for with described textural characteristics for characteristic of division, the sorter of at least one support vector machine is utilized to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
Said system, preferably, described solution format analysis processing unit comprises:
Frame synchronous ssearching subelement, for carrying out frame synchronous ssearching to described remote sensing satellite image data, to search frame synchronization head;
Frame determination subelement, for based on described frame synchronization head, determines the target data frame in described remote sensing satellite image data;
Frame count alignment subelement, between the different spectral coverage in described remote sensing satellite image data according to the frame count alignment operation of described target data frame, to complete the solution format analysis processing of described remote sensing satellite image data.
From such scheme, the cloud of a kind of remote sensing satellite image provided by the invention sentences method and system, by to real-time reception to remote sensing satellite image carry out solution format analysis processing after, the sorter of support vector machine is utilized to carry out the textural characteristics of cloud detection in conjunction with visible images, to identify the cloud mask in visible images data, be different from prior art and the identification of cloud mask is carried out to the multispectral image comprising multiple spectral information in remote sensing satellite image, the present invention adopts the sorter of support vector machine to carry out the identification of cloud mask to remote sensing satellite image in conjunction with the textural characteristics of visible images, pass through universalization, the design of through engineering approaches, can the accuracy of effective assured plan, effectively cloud mask and other target substances are carried out difference classify as avenged sand etc., and then reach higher discrimination.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only embodiments of the invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to the accompanying drawing provided.
The cloud of a kind of remote sensing satellite image that Fig. 1 provides for the embodiment of the present invention one sentences the process flow diagram of method;
The cloud of a kind of remote sensing satellite image that Fig. 2 provides for the embodiment of the present invention two sentences the realization flow figure of method;
The cloud of a kind of remote sensing satellite image that Fig. 3 provides for the embodiment of the present invention three sentences the partial process view of method;
Fig. 4 a ~ Fig. 4 c and Fig. 4 d1 ~ Fig. 4 d6 is respectively the application example figure of the embodiment of the present invention;
Fig. 5 is another part process flow diagram of the embodiment of the present invention three;
The cloud of a kind of remote sensing satellite image that Fig. 6 provides for the embodiment of the present invention four sentences the partial process view of method;
The cloud of a kind of remote sensing satellite image data that Fig. 7 provides for the embodiment of the present invention five sentences the structural representation of system;
The cloud of a kind of remote sensing satellite image that Fig. 8 provides for the embodiment of the present invention six sentences the structural representation of system;
The cloud of a kind of remote sensing satellite image that Fig. 9 provides for the embodiment of the present invention seven sentences the part-structure schematic diagram of system;
Figure 10 is another part structural representation of the embodiment of the present invention seven;
The cloud of a kind of remote sensing satellite image that Figure 11 provides for the embodiment of the present invention eight sentences the part-structure schematic diagram of system.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
With reference to figure 1, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention one sentences the process flow diagram of method, and wherein, described method can comprise the following steps:
Step 101: receive remote sensing satellite image data.
Wherein, described remote sensing satellite image data are the high speed remote sensing satellite bit stream data after decompressing, and high speed can be adopted in the present embodiment to receive board to receive the remote sensing satellite bit stream data after decompression.
Concrete, first initialization is carried out, Gains resources parameter in initialization procedure, device parameter is set, starting outfit starts to carry out data receiver collection as received at a high speed board, and the view data gathered arranges with the form of loop blocks queue, be supplied to subsequent operation by the shared drive arranged.
It should be noted that, 2 shared drives can be set here, such as, there is the internal memory of 2 pieces of 8MB, adopt the mode of table tennis to obtain view data.
Step 102: solution format analysis processing is carried out to described remote sensing satellite image data.
Concrete, the present embodiment is when carrying out solution format analysis processing to described remote sensing satellite image data, the shared drive of a 256MB can be set, and adopt the remote sensing satellite image data that the mode of circulation obtains the intermediate data in solution format analysis processing process and separates after format analysis processing.
The remote sensing satellite image data of separating after format analysis processing are carried out piecemeal, are obtained multiple sub-image data by step 103: according to the data segmentation rules preset.
Wherein, in described data segmentation rules, the image resolution ratio based on described remote sensing satellite image data carries out piecemeal, and a point block size can be arranged according to image resolution ratio, is applicable in the image real time transfer of different resolution.Such as, be chosen as 8*, 16*16,32*32 or self-defined setting, the image resolution ratio of described remote sensing satellite image data is higher, can arrange piecemeal larger, and the interval of piecemeal can be set to the size of piecemeal, 1/2 or self-defined setting of point block size.
That is, described data segmentation rules is corresponding with the image resolution ratio of described remote sensing satellite image data, image resolution ratio according to described remote sensing satellite image data carries out piecemeal to described remote sensing satellite image data, so that be chunked into different threads, run by each thread parallel, to carry out subsequent treatment, to accelerate the treatment effeciency of view data to the remote sensing satellite image data of separating after format analysis processing simultaneously.Such as, according to the data line (multiple of 32 row, such as 320 row) cutting to 8 threads, carry out follow-up image procossing respectively in each thread, concrete, first sample, reduce resolution, sampling rate is 4*4, then carries out the piecemeal of data, be set to 32*32 pixel (can customize), point block gap can be set to 16.
Step 104: in conjunction with the textural characteristics of visible images, adopts the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data.
Wherein, the sorter of described support vector machine can carry out cloud detection in conjunction with the textural characteristics of visible images in described remote sensing satellite image data, distinguish each described sub-image data medium cloud mask and other materials as husky etc. in avenged, and then identify the cloud mask in described remote sensing satellite image data.
Concrete, the present embodiment is when carrying out detection and identifying to described remote sensing satellite image data, the shared drive of a 512MB can be set, and adopt the mode of circulation to obtain the intermediate data in detection identification processing procedure and detect data corresponding to the cloud mask after identifying.
From such scheme, the cloud of a kind of remote sensing satellite image that the embodiment of the present invention one provides sentences method, by to real-time reception to remote sensing satellite image carry out solution format analysis processing after, the sorter of support vector machine is utilized to carry out the textural characteristics of cloud detection in conjunction with visible images, to identify the cloud mask in visible images data, be different from prior art and the identification of cloud mask is carried out to the multispectral image comprising multiple spectral information in remote sensing satellite image, the present embodiment adopts the sorter of support vector machine to carry out the identification of cloud mask to remote sensing satellite image in conjunction with the textural characteristics of visible images, pass through universalization, the design of through engineering approaches, can the accuracy of effective assured plan, effectively cloud mask and other target substances are carried out difference classify as avenged sand etc., and then reach higher discrimination.
With reference to figure 2, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention two sentences the realization flow figure of method, and after described step 104, described method can also comprise the following steps:
Step 105: data corresponding for described cloud mask are carried out record.
Concrete, data corresponding for described cloud mask are carried out real time record according to optical camera data format specifications, the data of record comprise: view data, auxiliary data and index data etc., be supplied to target identification system, with further recognition and verification so that follow-up.
Wherein, the present embodiment carries out in recording process in the data corresponding to described cloud mask, can arrange the internal memory of a 512MB, and adopts the mode of circulation to place data corresponding to cloud mask.
With reference to figure 3, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention three sentences the realization flow figure of real time steps 104 in method, and wherein, real time steps 104 can be realized by following steps:
Step 141: according to the intensity slicing threshold value preset, Threshold segmentation process is fixed respectively to sub-image data described in every block, to judge the first cloud sector view data in sub-image data described in every block.
Wherein, the image area of described first cloud sector view data is less than the image area of the sub-image data belonging to it.
That is, the minimum average (lowering 20%) of the cloud that the present embodiment is added up a large amount of cloud piecemeal districts (200) of the visible data according to described remote sensing satellite is set to the threshold value (100) of first time intensity slicing, as known in the histogram structure in Fig. 4 a, cloud layer region and most landforms (" other " class) have less coincidence in gray scale, adopt described intensity slicing threshold value to be fixed Threshold segmentation, most cloudless region (in real image, situation that is husky and snow occupies the minority) can be removed.Concrete, when interpretation, according to the number percent of the number of pixels of block statistics more than threshold value, such as, 85% (can arrange) can be greater than and then judge the data of this blocks of data as doubtful cloud sector, otherwise be the data in non-cloud sector.
Step 142: in conjunction with the textural characteristics of visible images, the sorter of at least one support vector machine is adopted to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
Wherein, the present embodiment is after distinguishing the doubtful cloud sector in described remote sensing satellite image data and non-cloud sector, again the cloud sector in doubtful cloud sector and described first cloud sector view data and non-cloud sector are distinguished, namely identify cloud mask image and target substance image as husky or avenge image.
Concrete, after cloud sector once and non-cloud region class time, the textural characteristics in conjunction with visible images carries out the classification in cloud sector and non-cloud sector.
Step 143: based on the first cloud sector view data medium cloud mask described in every block, determine the cloud mask in described remote sensing satellite image data.
That is, go out the cloud mask in the first cloud sector view data described in every block in Classification and Identification, and then piece together out the cloud mask in whole described remote sensing satellite image data, to complete the present embodiment object.
Concrete, with reference to figure 5, be the realization flow figure of step 142 described in the embodiment of the present invention three, wherein, described step 142 can be realized by following steps:
Step 501: the cloud sector gray level co-occurrence matrixes calculating the first cloud sector view data described in every block.
Step 502: based on described cloud sector gray level co-occurrence matrixes, calculates the textural characteristics of the visible images of the first cloud sector view data described in every block.
Wherein, described textural characteristics comprises: angle second moment, contrast, average, entropy, variance, contrast divide square.
Step 503: with described textural characteristics for characteristic of division, the sorter of at least one support vector machine is utilized to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
Wherein, in the present embodiment when classifying, multidimensional feature space is difficult to graphically represent, the two-dimensional feature space distribution plan of angle second moment and contrast composition is given in Fig. 4 b, cloud and snow can be distinguished preferably in this two-dimensional space, therefore, the classification of support vector machine (SVM) is adopted to classify when this subseries.
Concrete, can adopt in the present embodiment the sorter of three support vector machine further to cloud mask image and target substance image as husky, to avenge or other are classified, be respectively: the support vector machine classifier 1 (SVM1) that cloud and snow are formed, cloud and the husky support vector machine classifier 2 (SVM2) formed, the support vector machine classifier 3 (SVM3) that cloud and other classes are formed.In the present embodiment, select the every class of sample 220 (Fig. 4 c) of cloud, snow, sand and other classes, the kernel function of the sorter of support vector machine adopts radial basis RBF kernel function:
K(x,x
i)＝exp(-γ||x-x
i||
2)
Wherein, kernel functional parameter σ=0.5.
The feature of classification selects the textural characteristics based on co-occurrence matrix, the textural characteristics one of gray level co-occurrence matrixes has 12, detect by sample training the verification and measurement ratio obtained again to judge, select angle second moment, contrast, average, entropy, variance, contrast to divide these 6 textural characteristics of square as characteristic of division, can the highest verification and measurement ratio be obtained.
Wherein, gray level co-occurrence matrixes P may be defined as:
Molecule on the right of above formula equal sign be there is certain spatial relationship, number that pixel that gray-scale value is i and j is right, denominator is the summation number (# represents quantity) that pixel is right.The P obtained like this is normalized.P is the matrix of Ng × Ng, if (x
1, y
1) and (x
2, y
2) spacing is d, the angle of both and abscissa line is θ, then can obtain the gray level co-occurrence matrixes P (i, j, d, θ) of various spacing and angle.
(1) angle second moment (AngularSecondMoment):
(2) contrast (Contrast):
(3) average (Mean):
(4) entropy (Entropy):
(5) variance (Variance):
(6) contrast divides square (InverseDifferenceMoment):
Wherein,
When calculating gray level co-occurrence matrixes, color range is compressed to 16 looks, and the size of such gray level co-occurrence matrixes is 16 × 16, and compression color range directly can reduce gray level co-occurrence matrixes, thus reduces the calculated amount of textural characteristics.
Following table is the training result table of typical sample, is 100.00%, for the Yun Xuesha with grey similarity, can reaches the classification accuracy rate of 97%, particularly cloud and sand to the training precision of cloud and other classes, and their grey level histogram distribution is extremely similar.Further, the averaging time that training sample detects again is 0.016s.
Training result is as shown in table 1 below
Table 1 training result
Fig. 4 d1 ~ Fig. 4 d6 is some visible light image cloud recognition result figure, and wherein, Fig. 4 d1 is a width cloud atlas, and Fig. 4 d4 is the recognition result figure of Fig. 4 d1; The recognition result figure of Fig. 4 d2 to be snow figure, Fig. 4 d5 be Fig. 4 d2; In Fig. 4 d3, major part is sand ground, has a small amount of cloud, and Fig. 4 d6 is the recognition result figure of Fig. 4 d3, and picture size is 512 × 512.In recognition result figure, white portion represents the region being identified as cloud, and black represents non-cloud region.
After identifying cloud mask, namely cloud sentence reason export cloud sentence result data (data that cloud mask is corresponding) to shared drive, cloud is sentenced result data and is recorded in dish battle array by data record, simultaneously, solution formatted data is carried out record according to 0 grade of call format of standard by data record, and generates 0 grade of index file (comprising line number, this information such as latitude of passing through) and be recorded to dish battle array.
From such scheme, technical scheme in the embodiment of the present invention, with data acquisition, separate format analysis processing, data record integrated treatment, can require to carry out the settings such as sampling rate, piecemeal, color range, parallel processing threads for different satellite data bit rate, there is certain universalization, engineering ability, there is stronger adaptability.
With reference to figure 6, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention four sentences the realization flow figure of step 102 described in method, and wherein, described step 102 can be realized by following steps:
Step 121: carry out frame synchronous ssearching to described remote sensing satellite image data, to search frame synchronization head.
Step 122: based on described frame synchronization head, determines the target data frame in described remote sensing satellite image data.
Step 123: between the different spectral coverage in described remote sensing satellite image data according to the frame count alignment operation of described target data frame, to complete the solution format analysis processing of described remote sensing satellite image data.
The embodiment of the present invention is when carrying out solution format analysis processing, frame synchronous ssearching is carried out to optical camera data, search frame synchronous head determines effective Frame, the error code fault-tolerant strategy of frame synchronization head can be taked when search, concrete, whether is first frame synchronization head in the position judgment of fixing frame length, error code can be set to 1-3 bit (can arrange); Frame count alignment is pressed between the different CCD of optical camera data or spectral coverage, if have 1-5 second (can arrange) data according to frame count to not going up, then carry out the filling of this sheet or these spectral coverage data, if more than 5 seconds (can arrange), be set to segmentation (i.e. the satellite image of different time shooting).
With reference to figure 7, the cloud of a kind of remote sensing satellite image data provided for the embodiment of the present invention five sentences the structural representation of system, and wherein, described system can comprise following structural unit:
Image receiving unit 701, for receiving remote sensing satellite image data.
Wherein, can by arranging main control unit in the present embodiment, as shown in Figure 7, described main control unit mainly starts in the mode of thread the several unit carrying out data processing, and open up public shared drive, the work such as the stopping of unit and the recovery of shared drive is carried out when data processing terminates.
And described remote sensing satellite image data are the high speed remote sensing satellite bit stream data after decompressing, can adopt in the present embodiment and receive at a high speed board to receive the remote sensing satellite bit stream data after decompression.
Concrete, first the present embodiment carries out initialization, Gains resources parameter in initialization procedure, device parameter is set, starting outfit starts to carry out data receiver collection as received at a high speed board, and the view data gathered arranges with the form of loop blocks queue, be supplied to subsequent operation by the shared drive arranged.
It should be noted that, 2 shared drives can be set here, such as, there is the internal memory of 2 pieces of 8MB, adopt the mode of table tennis to obtain view data.
Separate format analysis processing unit 702, for carrying out solution format analysis processing to described remote sensing satellite image data.
Concrete, the present embodiment is when carrying out solution format analysis processing to described remote sensing satellite image data, the shared drive of a 256MB can be set, and adopt the remote sensing satellite image data that the mode of circulation obtains the intermediate data in solution format analysis processing process and separates after format analysis processing.
The remote sensing satellite image data of separating after format analysis processing, for the data segmentation rules according to budget, are carried out piecemeal, are obtained multiple sub-image data by image block unit 703.
Wherein, in described data segmentation rules, the image resolution ratio based on described remote sensing satellite image data carries out piecemeal, and a point block size can be arranged according to image resolution ratio, is applicable in the image real time transfer of different resolution.Such as, be chosen as 8*, 16*16,32*32 or self-defined setting, the image resolution ratio of described remote sensing satellite image data is higher, can arrange piecemeal larger, and the interval of piecemeal can be set to the size of piecemeal, 1/2 or self-defined setting of point block size.
That is, described data segmentation rules is corresponding with the image resolution ratio of described remote sensing satellite image data, image resolution ratio according to described remote sensing satellite image data carries out piecemeal to described remote sensing satellite image data, so that be chunked into different threads, run by each thread parallel, to carry out subsequent treatment, to accelerate the treatment effeciency of view data to the remote sensing satellite image data of separating after format analysis processing simultaneously.Such as, according to the data line (multiple of 32 row, such as 320 row) cutting to 8 threads, carry out follow-up image procossing respectively in each thread, concrete, first sample, reduce resolution, sampling rate is 4*4, then carries out the piecemeal of data, be set to 32*32 pixel (can customize), point block gap can be set to 16.
Image detecting element 704, for the textural characteristics in conjunction with visible images, adopts the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data.
Wherein, the sorter of described support vector machine can carry out cloud detection in conjunction with the textural characteristics of visible images in described remote sensing satellite image data, distinguish each described sub-image data medium cloud mask and other materials as husky etc. in avenged, and then identify the cloud mask in described remote sensing satellite image data.
Concrete, the present embodiment is when carrying out detection and identifying to described remote sensing satellite image data, the shared drive of a 512MB can be set, and adopt the mode of circulation to obtain the intermediate data in detection identification processing procedure and detect data corresponding to the cloud mask after identifying.
That is, it is 8MB × 2 that the present embodiment can arrange the shared drive size of carrying out between the image receiving unit 701 of data receiver and decompressed data by described main control unit, has the internal memory of 2 pieces of 8MB, takes the mode of rattling to obtain data; Memory size between described solution format analysis processing unit 702 and described image receiving unit 701 is 256MB, takes the mode circulated to obtain data; Carrying out the memory size that cloud sentences between the described image detecting element 704 of reason and described solution format analysis processing unit 702 is 512MB, takes the mode circulated to obtain data; The memory size that cloud sentences result data and data corresponding to described cloud mask is set to 512MB, takes the mode circulated to obtain data.In the acquisition position of data in internal memory, be also provided with the pointer pointing to latest data in this internal memory and the data total amount of having loaded respectively.
From such scheme, the cloud of a kind of remote sensing satellite image that the embodiment of the present invention five provides sentences system, by to real-time reception to remote sensing satellite image carry out solution format analysis processing after, the sorter of support vector machine is utilized to carry out the textural characteristics of cloud detection in conjunction with visible images, to identify the cloud mask in visible images data, be different from prior art and the identification of cloud mask is carried out to the multispectral image comprising multiple spectral information in remote sensing satellite image, the present embodiment adopts the sorter of support vector machine to carry out the identification of cloud mask to remote sensing satellite image in conjunction with the textural characteristics of visible images, pass through universalization, the design of through engineering approaches, can the accuracy of effective assured plan, effectively cloud mask and other target substances are carried out difference classify as avenged sand etc., and then reach higher discrimination.
With reference to figure 8, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention six sentences the structural representation of system, and wherein, described system can also comprise following structural unit:
Data corresponding for described cloud mask, after identifying the cloud mask in described remote sensing satellite image data at described image detecting element 704, are carried out record by data record unit 705.
Concrete, data corresponding for described cloud mask are carried out real time record according to optical camera data format specifications, the data of record comprise: view data, auxiliary data and index data etc., be supplied to target identification system, with further recognition and verification so that follow-up.
Wherein, the present embodiment can be carried out in recording process in the data corresponding to described cloud mask by described main control unit, the internal memory of a 512MB is set, and adopt the mode of circulation to place data corresponding to cloud mask, and the memory size arranging the solution formatted data between described data record unit 705 and described image detecting element 704 is set to 512MB.
With reference to figure 9, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention seven sentences the structural representation of image detecting element 704 described in system, and wherein, described image detecting element 704 can comprise following structure:
First classification subelement 741, for according to the intensity slicing threshold value preset, is fixed Threshold segmentation process to sub-image data described in every block, respectively to judge the first cloud sector view data in sub-image data described in every block.
Wherein, the image area of described first cloud sector view data is less than the image area of the sub-image data belonging to it.
That is, the minimum average (lowering 20%) of the cloud that the present embodiment is added up a large amount of cloud piecemeal districts (200) of the visible data according to described remote sensing satellite is set to the threshold value (100) of first time intensity slicing, as known in the histogram structure in Fig. 4 a, cloud layer region and most landforms (" other " class) have less coincidence in gray scale, adopt described intensity slicing threshold value to be fixed Threshold segmentation, most cloudless region (in real image, situation that is husky and snow occupies the minority) can be removed.Concrete, when interpretation, according to the number percent of the number of pixels of block statistics more than threshold value, such as, 85% (can arrange) can be greater than and then judge the data of this blocks of data as doubtful cloud sector, otherwise be the data in non-cloud sector.
Second classification subelement 742, for adopting at least one support vector machine classifier to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
Wherein, the present embodiment is after distinguishing the doubtful cloud sector in described remote sensing satellite image data and non-cloud sector, again the cloud sector in doubtful cloud sector and described first cloud sector view data and non-cloud sector are distinguished, namely identify cloud mask image and target substance image as husky or avenge image.
Concrete, after cloud sector once and non-cloud region class time, the textural characteristics in conjunction with visible images carries out the classification in cloud sector and non-cloud sector.
Cloud mask determination subelement 743, for based on the first cloud sector view data medium cloud mask described in every block, determines the cloud mask in described remote sensing satellite image data.
That is, go out the cloud mask in the first cloud sector view data described in every block in Classification and Identification, and then piece together out the cloud mask in whole described remote sensing satellite image data, to complete the present embodiment object.
Concrete, with reference to Figure 10, be the structural representation of the second classification subelement 742 described in the embodiment of the present invention, wherein, described second classification subelement 742 can comprise with lower module:
Co-occurrence matrix computing module 1001, for calculating the cloud sector gray level co-occurrence matrixes of the first cloud sector view data described in every block.
Textural characteristics computing module 1002, for based on described cloud sector gray level co-occurrence matrixes, calculates the textural characteristics of the visible images of the first cloud sector view data described in every block.
Wherein, described textural characteristics comprises: angle second moment, contrast, average, entropy, variance, contrast divide square.
Image classification module 1003, for with described textural characteristics for characteristic of division, the sorter of at least one support vector machine is utilized to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
Wherein, in the present embodiment when classifying, multidimensional feature space is difficult to graphically represent, the two-dimensional feature space distribution plan of angle second moment and contrast composition is given in Fig. 4 b, cloud and snow can be distinguished preferably in this two-dimensional space, therefore, the classification of support vector machine (SVM) is adopted to classify when this subseries.
Concrete, can adopt in the present embodiment the sorter of three support vector machine further to cloud mask image and target substance image as husky, to avenge or other are classified, be respectively: the support vector machine classifier 1 (SVM1) that cloud and snow are formed, cloud and the husky support vector machine classifier 2 (SVM2) formed, the support vector machine classifier 3 (SVM3) that cloud and other classes are formed.In the present embodiment, select the every class of sample 220 (Fig. 4 c) of cloud, snow, sand and other classes, the kernel function of the sorter of support vector machine adopts radial basis RBF kernel function:
K(x,x
i)＝exp(-γ||x-x
i||
2)
Wherein, kernel functional parameter σ=0.5.
The feature of classification selects the textural characteristics based on co-occurrence matrix, the textural characteristics one of gray level co-occurrence matrixes has 12, detect by sample training the verification and measurement ratio obtained again to judge, select angle second moment, contrast, average, entropy, variance, contrast to divide these 6 textural characteristics of square as characteristic of division, can the highest verification and measurement ratio be obtained.
Wherein, gray level co-occurrence matrixes P may be defined as:
Molecule on the right of above formula equal sign be there is certain spatial relationship, number that pixel that gray-scale value is i and j is right, denominator is the summation number (# represents quantity) that pixel is right.The P obtained like this is normalized.P is the matrix of Ng × Ng, if (x
1, y
1) and (x
2, y
2) spacing is d, the angle of both and abscissa line is θ, then can obtain the gray level co-occurrence matrixes P (i, j, d, θ) of various spacing and angle.
(1) angle second moment (AngularSecondMoment):
(2) contrast (Contrast):
(3) average (Mean):
(4) entropy (Entropy):
(5) variance (Variance):
(6) contrast divides square (InverseDifferenceMoment):
Wherein,
When calculating gray level co-occurrence matrixes, color range is compressed to 16 looks, and the size of such gray level co-occurrence matrixes is 16 × 16, and compression color range directly can reduce gray level co-occurrence matrixes, thus reduces the calculated amount of textural characteristics.
Following table is the training result table of typical sample, is 100.00%, for the Yun Xuesha with grey similarity, can reaches the classification accuracy rate of 97%, particularly cloud and sand to the training precision of cloud and other classes, and their grey level histogram distribution is extremely similar.Further, the averaging time that training sample detects again is 0.016s.Training result as shown in Table 1.
Fig. 4 d1 ~ Fig. 4 d6 is some visible light image cloud recognition result figure, and wherein, Fig. 4 d1 is a width cloud atlas, and Fig. 4 d4 is the recognition result figure of Fig. 4 d1; The recognition result figure of Fig. 4 d2 to be snow figure, Fig. 4 d5 be Fig. 4 d2; In Fig. 4 d3, major part is sand ground, has a small amount of cloud, and Fig. 4 d6 is the recognition result figure of Fig. 4 d3, and picture size is 512 × 512.In recognition result figure, white portion represents the region being identified as cloud, and black represents non-cloud region.
After identifying cloud mask, namely cloud sentence reason export cloud sentence result data (data that cloud mask is corresponding) to shared drive, cloud is sentenced result data and is recorded in dish battle array by data record, simultaneously, solution formatted data is carried out record according to 0 grade of call format of standard by data record, and generates 0 grade of index file (comprising line number, this information such as latitude of passing through) and be recorded to dish battle array.
From such scheme, technical scheme in the embodiment of the present invention, with data acquisition, separate format analysis processing, data record integrated treatment, can require to carry out the settings such as sampling rate, piecemeal, color range, parallel processing threads for different satellite data bit rate, there is certain universalization, engineering ability, there is stronger adaptability.
With reference to Figure 11, the cloud of a kind of remote sensing satellite image provided for the embodiment of the present invention eight sentences the structural representation separating format analysis processing unit 702 described in system, and wherein, described solution format analysis processing unit 702 can comprise following structure:
Frame synchronous ssearching subelement 721, for carrying out frame synchronous ssearching to described remote sensing satellite image data, to search frame synchronization head.
Frame determination subelement 722, for based on described frame synchronization head, determines the target data frame in described remote sensing satellite image data.
Frame count alignment subelement 723, between the different spectral coverage in described remote sensing satellite image data according to the frame count alignment operation of described target data frame, to complete the solution format analysis processing of described remote sensing satellite image data.
The embodiment of the present invention is when carrying out solution format analysis processing, frame synchronous ssearching is carried out to optical camera data, search frame synchronous head determines effective Frame, the error code fault-tolerant strategy of frame synchronization head can be taked when search, concrete, whether is first frame synchronization head in the position judgment of fixing frame length, error code can be set to 1-3 bit (can arrange); Frame count alignment is pressed between the different CCD of optical camera data or spectral coverage, if have 1-5 second (can arrange) data according to frame count to not going up, then carry out the filling of this sheet or these spectral coverage data, if more than 5 seconds (can arrange), be set to segmentation (i.e. the satellite image of different time shooting).
For aforesaid each embodiment of the method, in order to simple description, therefore it is all expressed as a series of combination of actions, but those skilled in the art should know, the application is not by the restriction of described sequence of movement, because according to the application, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art also should know, the embodiment described in instructions all belongs to preferred embodiment, and involved action and module might not be that the application is necessary.
In this instructions, each embodiment adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar portion mutually see.For device disclosed in embodiment, because it corresponds to the method disclosed in Example, so description is fairly simple, relevant part illustrates see method part.
It should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.
For convenience of description, various unit is divided into describe respectively with function when describing above device.Certainly, the function of each unit can be realized in same or multiple software and/or hardware when implementing the application.
As seen through the above description of the embodiments, those skilled in the art can be well understood to the mode that the application can add required general hardware platform by software and realizes.Based on such understanding, the technical scheme of the application can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product can be stored in storage medium, as ROM/RAM, magnetic disc, CD etc., comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform the method described in some part of each embodiment of the application or embodiment.
To the above-mentioned explanation of the disclosed embodiments, professional and technical personnel in the field are realized or uses the application.To be apparent for those skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein when not departing from the spirit or scope of the application, can realize in other embodiments.Therefore, the application can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
Claims (10)
1. the cloud of remote sensing satellite image sentences a method, it is characterized in that, described method comprises:
Receive remote sensing satellite image data;
Solution format analysis processing is carried out to described remote sensing satellite image data;
According to the data segmentation rules preset, the remote sensing satellite image data of separating after format analysis processing are carried out piecemeal, obtains multiple sub-image data;
In conjunction with the textural characteristics of visible images, the sorter of support vector machine is adopted to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data.
2. method according to claim 1, it is characterized in that, at the textural characteristics in conjunction with visible images, the sorter of support vector machine is adopted to carry out cloud detection respectively to sub-image data described in every block, after identifying the cloud mask in described remote sensing satellite image data, described method also comprises:
Data corresponding for described cloud mask are carried out record.
3. method according to claim 1 and 2, it is characterized in that, the described textural characteristics in conjunction with visible images, adopt the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data, comprising:
According to the intensity slicing threshold value preset, Threshold segmentation process is fixed respectively to sub-image data described in every block, to judge the first cloud sector view data in sub-image data described in every block, the image area of described first cloud sector view data is less than the image area of the sub-image data belonging to it;
In conjunction with the textural characteristics of visible images, the sorter of at least one support vector machine is adopted to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block;
Based on the first cloud sector view data medium cloud mask described in every block, determine the cloud mask in described remote sensing satellite image data.
4. method according to claim 3, it is characterized in that, the described textural characteristics in conjunction with visible images, the sorter of at least one support vector machine is adopted to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block, comprising:
Calculate the cloud sector gray level co-occurrence matrixes of the first cloud sector view data described in every block;
Based on described cloud sector gray level co-occurrence matrixes, calculate the textural characteristics of the visible images of the first cloud sector view data described in every block, described textural characteristics comprises: angle second moment, contrast, average, entropy, variance, contrast divide square;
With described textural characteristics for characteristic of division, the sorter of at least one support vector machine is utilized to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
5. method according to claim 1, is characterized in that, carries out solution format analysis processing, comprising described remote sensing satellite image data:
Frame synchronous ssearching is carried out to described remote sensing satellite image data, to search frame synchronization head;
Based on described frame synchronization head, determine the target data frame in described remote sensing satellite image data;
To between the different spectral coverage in described remote sensing satellite image data according to the frame count alignment operation of described target data frame, to complete the solution format analysis processing of described remote sensing satellite image data.
6. the cloud of remote sensing satellite image sentences a system, it is characterized in that, described system comprises:
Image receiving unit, for receiving remote sensing satellite image data;
Separate format analysis processing unit, for carrying out solution format analysis processing to described remote sensing satellite image data;
The remote sensing satellite image data of separating after format analysis processing, for the data segmentation rules according to budget, are carried out piecemeal, are obtained multiple sub-image data by image block unit;
Image detecting element, for the textural characteristics in conjunction with visible images, adopts the sorter of support vector machine to carry out cloud detection respectively to sub-image data described in every block, to identify the cloud mask in described remote sensing satellite image data.
7. system according to claim 6, is characterized in that, also comprises:
Data corresponding for described cloud mask, for identify the cloud mask in described remote sensing satellite image data at described image detecting element after, are carried out record by data record unit.
8. the system according to claim 6 or 7, is characterized in that, described image detecting element comprises:
First classification subelement, for the intensity slicing threshold value that foundation is preset, Threshold segmentation process is fixed respectively to sub-image data described in every block, to judge the first cloud sector view data in sub-image data described in every block, the image area of described first cloud sector view data is less than the image area of the sub-image data belonging to it;
Second classification subelement, for adopting at least one support vector machine classifier to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block;
Cloud mask determination subelement, for based on the first cloud sector view data medium cloud mask described in every block, determines the cloud mask in described remote sensing satellite image data.
9. system according to claim 8, is characterized in that, described second classification subelement, comprising:
Co-occurrence matrix computing module, for calculating the cloud sector gray level co-occurrence matrixes of the first cloud sector view data described in every block;
Textural characteristics computing module, for based on described cloud sector gray level co-occurrence matrixes, calculate the textural characteristics of the visible images of the first cloud sector view data described in every block, described textural characteristics comprises: angle second moment, contrast, average, entropy, variance, contrast divide square;
Image classification module, for with described textural characteristics for characteristic of division, the sorter of at least one support vector machine is utilized to classify to the cloud mask image in the first cloud sector view data described in every block and target substance image, to identify the cloud mask in the first cloud sector view data described in every block.
10. system according to claim 6, is characterized in that, described solution format analysis processing unit comprises:
Frame synchronous ssearching subelement, for carrying out frame synchronous ssearching to described remote sensing satellite image data, to search frame synchronization head;
Frame determination subelement, for based on described frame synchronization head, determines the target data frame in described remote sensing satellite image data;
Frame count alignment subelement, between the different spectral coverage in described remote sensing satellite image data according to the frame count alignment operation of described target data frame, to complete the solution format analysis processing of described remote sensing satellite image data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510705957.0A CN105426903A (en) | 2015-10-27 | 2015-10-27 | Cloud determination method and system for remote sensing satellite images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510705957.0A CN105426903A (en) | 2015-10-27 | 2015-10-27 | Cloud determination method and system for remote sensing satellite images |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105426903A true CN105426903A (en) | 2016-03-23 |
Family
ID=55505101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510705957.0A Pending CN105426903A (en) | 2015-10-27 | 2015-10-27 | Cloud determination method and system for remote sensing satellite images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105426903A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106846285A (en) * | 2016-12-30 | 2017-06-13 | 苏州中科天启遥感科技有限公司 | High-performance remote sensing image synthetic method and device |
| CN106920231A (en) * | 2017-02-17 | 2017-07-04 | 北京理工大学 | A kind of remote sensing image clouds appraisal procedure based on full-colour image statistical nature |
| CN107610114A (en) * | 2017-09-15 | 2018-01-19 | 武汉大学 | Optical satellite remote sensing image cloud snow mist detection method based on SVMs |
| CN107895152A (en) * | 2017-11-24 | 2018-04-10 | 西安空间无线电技术研究所 | A kind of in-orbit clearance quick determination method |
| CN108280810A (en) * | 2018-01-09 | 2018-07-13 | 北方工业大学 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
| CN110175638A (en) * | 2019-05-13 | 2019-08-27 | 北京中科锐景科技有限公司 | A kind of fugitive dust source monitoring method |
| CN110855638A (en) * | 2019-10-28 | 2020-02-28 | 中国资源卫星应用中心 | Remote sensing satellite data decompression processing system and method based on cloud computing |
| CN110889840A (en) * | 2019-11-28 | 2020-03-17 | 航天恒星科技有限公司 | Effectiveness detection method of high-resolution 6 # remote sensing satellite data for ground object target |
| CN111158020A (en) * | 2020-01-06 | 2020-05-15 | 中国科学院微小卫星创新研究院 | Satellite-borne real-time cloud judgment system and method for satellite |
| CN111508239A (en) * | 2020-04-16 | 2020-08-07 | 成都旸谷信息技术有限公司 | Intelligent vehicle flow identification method and system based on mask matrix |
| CN112084844A (en) * | 2020-07-28 | 2020-12-15 | 北京空间飞行器总体设计部 | Task re-planning method based on satellite-borne real-time cloud judgment |
| CN113377506A (en) * | 2021-04-30 | 2021-09-10 | 航天东方红卫星有限公司 | Optical remote sensing satellite data transmission task planning method applying posterior meteorological information |
| CN113516633A (en) * | 2021-05-28 | 2021-10-19 | 西安空间无线电技术研究所 | Cloud detection identification filling method for satellite image |
| CN115100541A (en) * | 2022-07-21 | 2022-09-23 | 文娟 | Satellite remote sensing data processing method and system and cloud platform |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6990410B2 (en) * | 2003-11-24 | 2006-01-24 | The Boeing Company | Cloud cover assessment: VNIR-SWIR |
| CN101419282A (en) * | 2008-12-05 | 2009-04-29 | 航天恒星科技有限公司 | Integration high speed remote sensing data receiving and processing equipment |
| CN102608579A (en) * | 2012-02-29 | 2012-07-25 | 西安空间无线电技术研究所 | SAR (synthetic aperture radar) data receiving and processing system and method |
| CN102829869A (en) * | 2012-08-16 | 2012-12-19 | 中国科学院西安光学精密机械研究所 | Ground test system aiming at large-aperture static interference imaging spectrometer |
| CN103500449A (en) * | 2013-09-26 | 2014-01-08 | 西安空间无线电技术研究所 | Satellite visible light remote sensing image cloud detection method |
| CN103927759A (en) * | 2014-04-30 | 2014-07-16 | 武汉大学 | Automatic cloud detection method of aerial images |
| CN104502999A (en) * | 2014-12-10 | 2015-04-08 | 中国科学院遥感与数字地球研究所 | Day and night cloud detection method and device based on static satellite multichannel data |
| US20150222861A1 (en) * | 2014-02-05 | 2015-08-06 | Panasonic Intellectual Property Management Co., Ltd. | Monitoring apparatus, monitoring system, and monitoring method |
| CN104867139A (en) * | 2015-05-12 | 2015-08-26 | 中国科学院遥感与数字地球研究所 | Remote sensing image cloud and shadow detection method based on radiation field |
-
2015
- 2015-10-27 CN CN201510705957.0A patent/CN105426903A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6990410B2 (en) * | 2003-11-24 | 2006-01-24 | The Boeing Company | Cloud cover assessment: VNIR-SWIR |
| CN101419282A (en) * | 2008-12-05 | 2009-04-29 | 航天恒星科技有限公司 | Integration high speed remote sensing data receiving and processing equipment |
| CN102608579A (en) * | 2012-02-29 | 2012-07-25 | 西安空间无线电技术研究所 | SAR (synthetic aperture radar) data receiving and processing system and method |
| CN102829869A (en) * | 2012-08-16 | 2012-12-19 | 中国科学院西安光学精密机械研究所 | Ground test system aiming at large-aperture static interference imaging spectrometer |
| CN103500449A (en) * | 2013-09-26 | 2014-01-08 | 西安空间无线电技术研究所 | Satellite visible light remote sensing image cloud detection method |
| US20150222861A1 (en) * | 2014-02-05 | 2015-08-06 | Panasonic Intellectual Property Management Co., Ltd. | Monitoring apparatus, monitoring system, and monitoring method |
| CN103927759A (en) * | 2014-04-30 | 2014-07-16 | 武汉大学 | Automatic cloud detection method of aerial images |
| CN104502999A (en) * | 2014-12-10 | 2015-04-08 | 中国科学院遥感与数字地球研究所 | Day and night cloud detection method and device based on static satellite multichannel data |
| CN104867139A (en) * | 2015-05-12 | 2015-08-26 | 中国科学院遥感与数字地球研究所 | Remote sensing image cloud and shadow detection method based on radiation field |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106846285A (en) * | 2016-12-30 | 2017-06-13 | 苏州中科天启遥感科技有限公司 | High-performance remote sensing image synthetic method and device |
| CN106920231A (en) * | 2017-02-17 | 2017-07-04 | 北京理工大学 | A kind of remote sensing image clouds appraisal procedure based on full-colour image statistical nature |
| CN107610114A (en) * | 2017-09-15 | 2018-01-19 | 武汉大学 | Optical satellite remote sensing image cloud snow mist detection method based on SVMs |
| CN107610114B (en) * | 2017-09-15 | 2019-12-10 | 武汉大学 | optical satellite remote sensing image cloud and snow fog detection method based on support vector machine |
| CN107895152A (en) * | 2017-11-24 | 2018-04-10 | 西安空间无线电技术研究所 | A kind of in-orbit clearance quick determination method |
| CN107895152B (en) * | 2017-11-24 | 2020-02-14 | 西安空间无线电技术研究所 | On-orbit cloud seam rapid detection method |
| CN108280810A (en) * | 2018-01-09 | 2018-07-13 | 北方工业大学 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
| CN110175638B (en) * | 2019-05-13 | 2021-04-30 | 北京中科锐景科技有限公司 | Raise dust source monitoring method |
| CN110175638A (en) * | 2019-05-13 | 2019-08-27 | 北京中科锐景科技有限公司 | A kind of fugitive dust source monitoring method |
| CN110855638A (en) * | 2019-10-28 | 2020-02-28 | 中国资源卫星应用中心 | Remote sensing satellite data decompression processing system and method based on cloud computing |
| CN110889840A (en) * | 2019-11-28 | 2020-03-17 | 航天恒星科技有限公司 | Effectiveness detection method of high-resolution 6 # remote sensing satellite data for ground object target |
| CN111158020A (en) * | 2020-01-06 | 2020-05-15 | 中国科学院微小卫星创新研究院 | Satellite-borne real-time cloud judgment system and method for satellite |
| CN111508239A (en) * | 2020-04-16 | 2020-08-07 | 成都旸谷信息技术有限公司 | Intelligent vehicle flow identification method and system based on mask matrix |
| CN112084844A (en) * | 2020-07-28 | 2020-12-15 | 北京空间飞行器总体设计部 | Task re-planning method based on satellite-borne real-time cloud judgment |
| CN112084844B (en) * | 2020-07-28 | 2024-03-19 | 北京空间飞行器总体设计部 | Task re-planning method based on satellite-borne real-time cloud judgment |
| CN113377506A (en) * | 2021-04-30 | 2021-09-10 | 航天东方红卫星有限公司 | Optical remote sensing satellite data transmission task planning method applying posterior meteorological information |
| CN113377506B (en) * | 2021-04-30 | 2024-03-26 | 航天东方红卫星有限公司 | Optical remote sensing satellite data transmission task planning method applying posterior meteorological information |
| CN113516633A (en) * | 2021-05-28 | 2021-10-19 | 西安空间无线电技术研究所 | Cloud detection identification filling method for satellite image |
| CN115100541A (en) * | 2022-07-21 | 2022-09-23 | 文娟 | Satellite remote sensing data processing method and system and cloud platform |
| CN115100541B (en) * | 2022-07-21 | 2023-06-06 | 米脂县宇宝北斗农业发展有限公司 | Satellite remote sensing data processing method, system and cloud platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105426903A (en) | Cloud determination method and system for remote sensing satellite images | |
| Hu et al. | Fast forest fire smoke detection using MVMNet | |
| CN104166841B (en) | The quick detection recognition methods of pedestrian or vehicle is specified in a kind of video surveillance network | |
| WO2017190574A1 (en) | Fast pedestrian detection method based on aggregation channel features | |
| CN107273832B (en) | License plate recognition method and system based on integral channel characteristics and convolutional neural network | |
| CN104866616B (en) | Monitor video Target Searching Method | |
| Derpanis et al. | Classification of traffic video based on a spatiotemporal orientation analysis | |
| CN104978567B (en) | Vehicle checking method based on scene classification | |
| CN105809205B (en) | A kind of classification method and its system of high spectrum image | |
| Faria et al. | Fusion of time series representations for plant recognition in phenology studies | |
| CN109919223B (en) | Target detection method and device based on deep neural network | |
| CN109902576B (en) | Training method and application of head and shoulder image classifier | |
| CN113658192B (en) | Multi-target pedestrian track acquisition method, system, device and medium | |
| CN106156777A (en) | Textual image detection method and device | |
| CN106295498A (en) | Remote sensing image target area detection apparatus and method | |
| CN108764302A (en) | A kind of bill images sorting technique based on color characteristic and bag of words feature | |
| CN111259756A (en) | Pedestrian re-identification method based on local high-frequency features and mixed metric learning | |
| CN108073940B (en) | Method for detecting 3D target example object in unstructured environment | |
| CN111724566A (en) | Pedestrian falling detection method and device based on intelligent lamp pole video monitoring system | |
| CN113688709A (en) | Intelligent detection method, system, terminal and medium for wearing safety helmet | |
| CN114049572A (en) | Detection method for identifying small target | |
| Mammeri et al. | North-American speed limit sign detection and recognition for smart cars | |
| He et al. | Robust illumination invariant texture classification using gradient local binary patterns | |
| CN110008899B (en) | Method for extracting and classifying candidate targets of visible light remote sensing image | |
| Hua et al. | Detection of collapsed buildings with the aerial images captured from UAV |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160323 |