AU2021102441A4 - Method for automatically identifying abnormal category pattern spots and system thereof - Google Patents
Method for automatically identifying abnormal category pattern spots and system thereof Download PDFInfo
- Publication number
- AU2021102441A4 AU2021102441A4 AU2021102441A AU2021102441A AU2021102441A4 AU 2021102441 A4 AU2021102441 A4 AU 2021102441A4 AU 2021102441 A AU2021102441 A AU 2021102441A AU 2021102441 A AU2021102441 A AU 2021102441A AU 2021102441 A4 AU2021102441 A4 AU 2021102441A4
- Authority
- AU
- Australia
- Prior art keywords
- pattern
- pattern spot
- category
- index
- image
- 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.)
- Ceased
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000003595 spectral effect Effects 0.000 claims abstract description 77
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 28
- 238000005520 cutting process Methods 0.000 claims abstract description 17
- 230000009191 jumping Effects 0.000 claims abstract description 6
- 238000007689 inspection Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000002420 orchard Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present disclosure discloses a method for automatically identifying
abnormal category pattern spots and a system thereof. The method
comprises: acquiring a multispectral remote sensing image to be identified;
cutting the multispectral remote sensing image to obtain the image of the
pattern spot whose category is error-prone or confusable; calculating the
comprehensive spectral index of each pattern spot image; according to the
size of the comprehensive spectral index of each pattern spot image, sorting
each pattern spot image according to the element category; determining
whether there are abnormal category pattern spots in a set number of pattern
spot images before and after the pattern spot image sequence; if so, adding
a set number of pattern spot images before and/or after the pattern spot image
sequence, and identifying the abnormal category of the added pattern spot
images; and if there are abnormal category pattern spots in the added pattern
spot images, jumping to the step of "adding a set number of pattern spot
images before and/or after the pattern spot image sequence for identification"
until there are no abnormal category pattern spots in the added pattern spot
image. According to the present disclosure, the abnormal category pattern
spots can be automatically identified.
-1/2
101
acquiring a multispectral remote sensing image to be identified
cutting the multispectral remote sensing image based on the geometric 102
range of the pattern spot to obtain the image of the pattern spot whose
category is error-prone or confusable
calculating the comprehensive spectral index of each pattern spot image, 103
wherein the comprehensive spectral index consists of a weighted
combination of a vegetation index, a water body index and a differential
building index for distinguishing the category of pattern spot elements
104
according to the size of the comprehensive spectral index of each pattern
spot image, sorting each pattern spot image according to the element
category
determining whether there are abnormal category pattern spots in a set 105
number of pattern spot images before and after the pattern spot image
sequence
there are abnormal
category pattern spots
adding a set number of pattern spot images before and/or after the pattern
spot image sequence, and identifying the abnormal category of the added 106
pattern spot images, and if there are abnormal category pattern spots in the
added pattern spot images, jumping to the step of "adding a set number of
pattern spot images before and/or after the pattern spot image sequence for
identification" until there are no abnormal category pattern spots in the
added pattern spot image
I107
outputting an abnormal pattern spot image
FIG.1
Description
-1/2
101 acquiring a multispectral remote sensing image to be identified
cutting the multispectral remote sensing image based on the geometric 102 range of the pattern spot to obtain the image of the pattern spot whose category is error-prone or confusable
calculating the comprehensive spectral index of each pattern spot image, 103 wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements
104 according to the size of the comprehensive spectral index of each pattern spot image, sorting each pattern spot image according to the element category
determining whether there are abnormal category pattern spots in a set 105 number of pattern spot images before and after the pattern spot image sequence
there are abnormal category pattern spots
adding a set number of pattern spot images before and/or after the pattern spot image sequence, and identifying the abnormal category of the added 106 pattern spot images, and if there are abnormal category pattern spots in the added pattern spot images, jumping to the step of "adding a set number of pattern spot images before and/or after the pattern spot image sequence for identification" until there are no abnormal category pattern spots in the added pattern spot image
I107
outputting an abnormal pattern spot image
FIG.1
Patents Act 1990
Invention title:
Applicant:
Associated provisional applications:
The following statement is a full description of the invention, including the best method of performing it known to me:
Technical Field
[0001] The present disclosure relates to that technical field of image processing, in particular to a method for automatically identifying abnormal category pattern spots and a system thereof.
Background Art
[0002] With the development of society, in the field of monitoring geographical conditions, higher requirements are put forward for the accuracy and fineness of surface coverage data. However, due to the factors that surface conditions in China are complex, the quality of remote sensing images is different, and interpretation is greatly influenced by the subjectivity of people, the problem of pattern spot classification errors often occurs in surface coverage data, which seriously affects the quality of monitoring results of geographical conditions. Every year, production units and quality inspection units need to spend a lot of manpower, material resources and financial resources to check errors of large pattern spots and foreign objects with the same spectrum in surface coverage data.
Summary
[0003] The purpose of the present disclosure is to provide a method for automatically identifying abnormal category pattern spots and a system thereof.
[0004] To achieve the above purpose, the present disclosure provides the following scheme.
[0005] The present disclosure relates to a method for automatically identifying abnormal category pattern spots, comprising: acquiring a multispectral remote sensing image to be identified; cutting the multispectral remote sensing image based on the geometric range of the pattern spot to obtain the image of the pattern spot whose category is error-prone or confusable; calculating the comprehensive spectral index of each pattern spot image, wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements; according to the size of the comprehensive spectral index of each pattern spot image, sorting each pattern spot image according to the element category; determining whether there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot image sequence; if there are abnormal category pattern spots, adding a set number of pattern spot images before and/or after the pattern spot image sequence, and identifying the abnormal category of the added pattern spot images, and if there are abnormal category pattern spots in the added pattern spot images, jumping to the step of "adding a set number of pattern spot images before and/or after the pattern spot image sequence for identification" until there are no abnormal category pattern spots in the added pattern spot image; outputting an abnormal pattern spot image.
[0006] Preferably, prior to cutting the multispectral remote sensing image based on the geometric range of the pattern spot, the method further comprises: according to the historical data, determining the error-prone or confusable element category and the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index used to distinguish the error-prone or confusable element category.
[0007] Preferably, according to the size of the comprehensive spectral index of each pattern spot image, sorting each pattern spot image according to the element category specifically comprises: determining the intermediate value or average value of the comprehensive spectral index of each pattern spot image; according to the size of the intermediate value or average value of the comprehensive spectral index of the pattern spot image, sorting the pattern spot image according to the element category.
[0008] Preferably, the method for detecting abnormal category pattern spots comprises: judging whether the difference of the comprehensive spectral indices of adjacent pattern spot images is greater than a set threshold; if so, it means that there are abnormal category pattern spots, the abnormal category pattern spots are all the pattern spots between the adjacent pattern spot images and the first end of the pattern spot sequence, and the first end is the end of the pattern spot sequence closer to the adjacent pattern spot images.
[0009] Preferably, prior to calculating the comprehensive spectral index of each pattern spot image, the method further comprises: normalizing the vegetation index and the water body index; stretching the differential building index and the normalized difference vegetation index and water body index.
[0010] Preferably, the method further comprises: outputting the image of the abnormal category pattern spot identified by a computer to a human-computer interaction terminal for manual secondary identification.
[0011] Preferably, the cutting process of the multispectral remote sensing image is carried out in a computer memory.
[0012] The present disclosure further provides a system for automatically identifying abnormal category pattern spots, comprising: a multispectral remote sensing image-to-be-identified acquiring module, which is configured to acquire a multispectral remote sensing image to be identified; a pattern spot image cutting module, which is configured to cut the multispectral remote sensing image based on the geometric range of the pattern spot to obtain the image of the pattern spot whose category is error prone or confusable; a comprehensive spectral index calculating module, which is configured to calculate the comprehensive spectral index of each pattern spot image, wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements; a sorting module, which is configured to according to the size of the comprehensive spectral index of each pattern spot image, sort each pattern spot image according to the element category; an abnormal category determining module, which is configured to determine whether there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot image sequence, and determine whether there are abnormal category pattern spots in the added pattern spot images.
[0013] Preferably, the system further comprises: an initial parameter determining module, which is configured to, according to the historical data, determine the error-prone or confusable element category and the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index used to distinguish the error-prone or confusable element category.
[0014] Preferably, the sorting module specifically comprises: a numerical value determining unit, which is configured to determine the intermediate value or average value of the comprehensive spectral index of each pattern spot image; a sorting unit, which is configured to, according to the size of the intermediate value or average value of the comprehensive spectral index of the pattern spot image, sort the pattern spot image according to the element category.
[0015] According to the specific embodiment provided by the present disclosure, the present disclosure discloses the following technical effects: the method for automatically identifying abnormal category pattern spots and the system thereof provided by the present disclosure first extract the pattern spots which are error-prone or confusable in the multispectral remote sensing image to be identified, then calculate the spectral index of each pattern spot which can distinguish the pattern spots which are error-prone or confusable, sort each pattern spot according to the size of the spectral index, and finally, determine the abnormal category pattern spots by performing abnormal category detection on a set number of pattern spots at both ends of the sequence. According to the present disclosure, the abnormal category pattern spots is automatically identified, and the detection efficiency is improved.
Brief Description of the Drawings
[0016] In order to explain the embodiments of the present disclosure or the technical scheme in the prior art more clearly, the drawings needed in the embodiments will be briefly introduced hereinafter. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those skilled in the art, other drawings can be obtained according to these drawings without paying creative labour.
[0017] FIG. 1 is a schematic flow chart of a method for automatically identifying abnormal category pattern spots according to embodiment 1 of the present disclosure;
[0018] FIG. 2 is a structural schematic diagram of a system for automatically identifying abnormal category pattern spots according to embodiment 2 of the present disclosure.
Detailed Description of Preferred Embodiments
[0019] The technical scheme in the embodiments of the present disclosure will be described clearly and completely hereinafter with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only some embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without paying creative labour belong to the scope of protection of the present disclosure.
[0020] In order to make the above objects, features and advantages of the present disclosure more obvious and understandable, the present disclosure will be further explained in detail hereinafter with reference to the drawings and specific embodiments.
Embodiment 1
[0021] FIG. 1 is a schematic flow chart of a method for automatically identifying abnormal category pattern spots according to embodiment 1 of the present disclosure. Referring to FIG. 1, the method for automatically identifying abnormal category pattern spots according to the embodiment comprises the following steps:
[0022] Step 101: acquiring a multispectral remote sensing image to be identified;
[0023] Step 102: cutting the multispectral remote sensing image based on the geometric range of the pattern spot to obtain the image of the pattern spot whose category is error-prone or confusable;
[0024] Step 103: calculating the comprehensive spectral index of each pattern spot image, wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements;
[0025] Step 104: according to the size of the comprehensive spectral index of each pattern spot image, sorting each pattern spot image according to the element category;
[0026] Step 105: determining whether there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot image sequence;
[0027] Step 106: if there are abnormal category pattern spots, adding a set number of pattern spot images before and/or after the pattern spot image sequence, and identifying the abnormal category of the added pattern spot images, and if there are abnormal category pattern spots in the added pattern spot images, jumping to the step of "adding a set number of pattern spot images before and/or after the pattern spot image sequence for identification" until there are no abnormal category pattern spots in the added pattern spot image;
[0028] Step 107: outputting an abnormal pattern spot image.
[0029] In this embodiment, the geometric range of each pattern spot is sequentially acquired from the surface coverage vector data, and an image cutting function is defined. The cutting range is the geometric range of the pattern spot. The cut image is a multispectral remote sensing image to be identified. The number of bands of the multispectral remote sensing image to be identified should be no less than 4, including at least red, green, blue, near infrared and other bands. The image cutting function defines the image data processing mode facing the spectral index of the pattern spot, wherein preferably, the processing process is carried out in the computer memory, and the cutting result is not output to the computer hard disk. The comprehensive spectral index of each pattern spot image is calculated. The comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements. Of course, in other embodiments, other spectral category data that can distinguish the element category of the pattern spots can also be used. After calculating the comprehensive spectral index of each pattern spot image, the intermediate value of the comprehensive spectral index of the pattern spot image can be selected to represent that the pattern spots are sorted according to the size. Of course, in other embodiments, the average value of the comprehensive spectral index of the pattern spot image can also be selected to represent that the pattern spots are sorted. During sorting, it is necessary to sort according to the element category. That is to say, not all the pattern spots of the element categories are mixed for sorting, but pattern spots belonging to the same element category are sorted separately. For example, all the pattern spots with the element category attribute as paddy fields are sorted as a group, and all the pattern spots with the element category attribute as lakes are sorted as a group. After obtaining the pattern spot sequence, first, it is judged whether there are abnormal category pattern spots in a set number (for example, two) of pattern spot images before and after the pattern spot sequence. If there are no abnormal category pattern spots, it is considered that the pattern spots in the whole pattern spot sequence are normal and there are no abnormal category pattern spots. If there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot sequence, for example, there are abnormal category pattern spots in a set number of pattern spot images at the front end (back end) of the pattern spot sequence, a set number (for example, 2-3) of pattern spots are added at the front end (back end) of the pattern spot sequence to perform abnormal category detection until the set number of added pattern spots are all normal pattern spots. At this time, the last added pattern spots are considered as normal pattern spots, and the pattern spots identified before this are all abnormal category pattern spots.
[0030] The intermediate value of the comprehensive spectral index of a pattern spot image is determined as follows: when the number of pixels in a pattern spot is odd, the middlemost value in the pixel comprehensive spectral index sequence is the intermediate value of the comprehensive spectral index of the pattern spot; when the number of pixels in the pattern spot is even, two middlemost values in the pixel comprehensive spectral index sequence are taken out. The average value of the two values is calculated, which is the intermediate value of the comprehensive spectral index of the pattern spot.
[0031] In this embodiment, after the abnormal pattern spots are determined, the abnormal category pattern spots can be directly output, or the abnormal category pattern spots can be output to the human-computer interaction terminal for manual secondary identification.
[0032] In this embodiment, prior to step 102, the method may further comprise according to the historical data, determining the error-prone or confusable element category and the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index used to distinguish the error-prone or confusable element category. For example, the quality problems of large pattern spots and foreign objects with the same spectrum found in the monitoring data of national basic geographical conditions over the years are statistically analyzed, focusing on the error-prone and confusable element category, such as paddy fields and dry lands, orchards and woodlands, etc. The spectral information of non visible light bands and visible light bands of their images is studied and analyzed, so as to obtain the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index of pattern spots which can distinguish the error-prone and confusable element category.
[0033] As an embodiment, the method for detecting abnormal category pattern spots can specifically comprise: judging whether the difference of the comprehensive spectral indices of adjacent pattern spot images is greater than a set threshold; if so, it means that there are abnormal category pattern spots, the abnormal category pattern spots are all the pattern spots between the adjacent pattern spot images and the first end of the pattern spot sequence, and the first end is the end of the pattern spot sequence closer to the adjacent pattern spot images.
[0034] For example, the specific operation method of judging whether there are abnormal category pattern spots in the first five pattern spots in the pattern spot sequence is as follows: judging whether the difference of the comprehensive spectral index (such as the intermediate value of the comprehensive spectral index) between two adjacent pattern spots in the five pattern spots is greater than the set threshold which can generally be set as 0.04. If the difference of the comprehensive spectral index (such as the intermediate value of the comprehensive spectral index) between the fourth and fifth pattern spots is greater than the set threshold, the first four pattern spots are considered as abnormal category pattern spots.
[0035] In this embodiment, when calculating the comprehensive spectral index of each pattern spot, it is necessary to preprocess each index. The preprocessing process can comprise the normalizing processing and the stretching processing, and the specific process is as follows.
[0036] First, the normalized difference vegetation index (NDVI) of the pattern spot is calculated by the following formula to highlight the vegetation information in the image.
NDVI = B,,, - B,d
[0037] Bi+Bred where Bair is the near infrared band of a remote
sensing image, and Bred is the red band of a remote sensing image.
[0038] Then, the normalized water body index NDWI of the pattern spot is calculated by the following formula to highlight the water information in the image.
NDVI = Bn,, - Bgee
[0039] B g,,,+Been where Bmir is the near infrared band of a remote
sensing image, and Bgree" is the green band of a remote sensing image.
[0040] Then, the differential building index DSBI is calculated by the following formula.
[0041] DSBI=k*(Bh,- B,,)+(1- k)*(Bhl,,11 - ) where k is the calculating
coefficient which can generally be set to 0.5, Bblue is the blue band of a remote
sensing image, Bed is the red band of a remote sensing image, and Bgeen is
the green band of a remote sensing image.
[0042] Finally, the normalized difference vegetation index, the normalized water body index and the differential building index are stretched, the range is stretched to [0,1], and the normalized difference vegetation index and the normalized water body index are stretched by the following formula.
Pixel', Pixel+, ±1
[0043] V 2 where Pixel is the pixel value on the original
normalized water body or vegetation index, and veis the pixel value after stretching.
[0044] The differential building index is stretched by the following formula.
Pixel' - Pixelv - Pixelv m
[0045] Pie-ta - P -"el where ie 1 is the pixel value on the
original differential building index,P is the pixel value after stretching, Pixel- 1 i" is the minimum pixel value of the whole original differential building
index, and -"Pixelvax is the maximum pixel value of the whole original differential building index.
[0046] The vegetation index NDVI, the water body index NDWI and the building index NSBI after stretching are comprehensively weighted by the following formula, and the comprehensive spectral index NCI of the pattern spot is calculated.
rNCI=k* NDVI+k2* NDWI+k3 * NSBI
[0047] jki+k 2 +k 3 =1 where ki is the coefficient of
NDVI, k2 is the coefficient of NDWI, and k3 is the coefficient of NSBI. The
values of k k k 3 can be determined according to the statistical analysis results of spectral information of various category elements. When it is easier for the NDVI value to distinguish a certain element category from other element categories, when calculating the NCI of such category elements, the
value of ki is set to 1, the values of k2 and k3 are set to 0. When it is easier for the NDWI value to distinguish a certain element category from other element categories, when calculating the NCI of such category elements, the
value of k2 is set to 1, and the values of ki and k3 are set to 0. When it is easier for the NSBI value to distinguish a certain element category from other element categories, when calculating the NCI of such category elements, the
value of k3 is set to 1, and the values of ki and k2 are set to 0.
Embodiment 2
[0048] FIG. 2 is a structural schematic diagram of a system for automatically identifying abnormal category pattern spots according to embodiment 2 of the present disclosure. Referring to FIG. 2, the system for automatically identifying abnormal category pattern spots according to the embodiment comprises:
[0049] a multispectral remote sensing image-to-be-identified acquiring module 201, which is configured to acquire a multispectral remote sensing image to be identified;
[0050] a pattern spot image cutting module 202, which is configured to cut the multispectral remote sensing image based on the geometric range of the pattern spot to obtain the image of the pattern spot whose category is error prone or confusable;
[0051] a comprehensive spectral index calculating module 203, which is configured to calculate the comprehensive spectral index of each pattern spot image, wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements;
[0052] a sorting module 204, which is configured to according to the size of the comprehensive spectral index of each pattern spot image, sort each pattern spot image according to the element category;
[0053] an abnormal category determining module 205, which is configured to determine whether there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot image sequence, and determine whether there are abnormal category pattern spots in the added pattern spot images.
[0054] As an embodiment, the system further comprises: an initial parameter determining module, which is configured to, according to the historical data, determine the error-prone or confusable element category and the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index used to distinguish the error-prone or confusable element category.
[0055] As an embodiment, the sorting module 204 specifically comprises: a numerical value determining unit, which is configured to determine the intermediate value or average value of the comprehensive spectral index of each pattern spot image; a sorting unit, which is configured to, according to the size of the intermediate value or average value of the comprehensive spectral index of the pattern spot image, sort the pattern spot image according to the element category.
[0056] As an embodiment, the abnormal category determining module 205 specifically comprises: a judging unit, which is configured to judge whether the difference of the comprehensive spectral indices of adjacent pattern spot images is greater than a set threshold; an abnormal category pattern spot determining unit, which is configured to determine all the pattern spots between the adjacent pattern spot images and the first end of the pattern spot sequence as the abnormal category pattern spots, and the first end as the end of the pattern spot sequence closer to the adjacent pattern spot images when the difference of the comprehensive spectral indices of adjacent pattern spot images is greater than a set threshold.
[0057] As an embodiment, the system further comprises: a normalizing module, which is configured to normalize the vegetation index and the water body index; a stretching module, which is configured to stretch the differential building index and the normalized difference vegetation index and water body index.
[0058] The method for automatically identifying abnormal category pattern spots and the system thereof according to the present disclosure automatically extract errors of large pattern spots and foreign objects with the same spectrum, improve the efficiency and reliability of quality detection of surface coverage data, and effectively improve the quality of geographical conditions.
[0059] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. It is sufficient to refer to the same and similar parts among each embodiment. For the system disclosed in the embodiment, because it corresponds to the method disclosed in the embodiment, it is described relatively simply, and the relevant points can be found in the description of the method.
[0060] In the present disclosure, a specific example is applied to illustrate the principle and implementation of the present disclosure, and the explanation of the above embodiments is only used to help understand the method and its core idea of the present disclosure. At the same time, according to the idea of the present disclosure, there will be some changes in the specific implementation and application scope for those skilled in the art. To sum up, the contents of this specification should not be construed as limiting the present disclosure.
Claims (5)
- Claims 1. A method for automatically identifying abnormal category pattern spots, comprising: acquiring a multispectral remote sensing image to be identified; cutting the multispectral remote sensing image based on the geometric range of the pattern spot to obtain the image of the pattern spot whose category is error-prone or confusable; calculating the comprehensive spectral index of each pattern spot image, wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements; according to the size of the comprehensive spectral index of each pattern spot image, sorting each pattern spot image according to the element category; determining whether there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot image sequence; if there are abnormal category pattern spots, adding a set number of pattern spot images before and/or after the pattern spot image sequence, and identifying the abnormal category of the added pattern spot images, and if there are abnormal category pattern spots in the added pattern spot images, jumping to the step of "adding a set number of pattern spot images before and/or after the pattern spot image sequence for identification" until there are no abnormal category pattern spots in the added pattern spot image; outputting an abnormal pattern spot image.
- 2. The method for automatically identifying abnormal category pattern spots according to claim 1, wherein prior to cutting the multispectral remote sensing image based on the geometric range of the pattern spot, the method further comprises: according to the historical data, determining the error-prone or confusable element category and the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index used to distinguish the error-prone or confusable element category; wherein according to the size of the comprehensive spectral index of each pattern spot image, sorting each pattern spot image according to the element category specifically comprises: determining the intermediate value or average value of the comprehensive spectral index of each pattern spot image; according to the size of the intermediate value or average value of the comprehensive spectral index of the pattern spot image, sorting the pattern spot image according to the element category; wherein the method for detecting abnormal category pattern spots comprises: judging whether the difference of the comprehensive spectral indices of adjacent pattern spot images is greater than a set threshold; if so, it means that there are abnormal category pattern spots, the abnormal category pattern spots are all the pattern spots between the adjacent pattern spot images and the first end of the pattern spot sequence, and the first end is the end of the pattern spot sequence closer to the adjacent pattern spot images; wherein prior to calculating the comprehensive spectral index of each pattern spot image, the method further comprises: normalizing the vegetation index and the water body index; stretching the differential building index and the normalized difference vegetation index and water body index; wherein the method further comprises: outputting the image of the abnormal category pattern spot identified by a computer to a human-computer interaction terminal for manual secondary identification; wherein the cutting process of the multispectral remote sensing image is carried out in a computer memory.
- 3. A system for automatically identifying abnormal category pattern spots, comprising: a multispectral remote sensing image-to-be-identified acquiring module, which is configured to acquire a multispectral remote sensing image to be identified; a pattern spot image cutting module, which is configured to cut the multispectral remote sensing image based on the geometric range of the pattern spot to obtain the image of the pattern spot whose category is error prone or confusable; a comprehensive spectral index calculating module, which is configured to calculate the comprehensive spectral index of each pattern spot image, wherein the comprehensive spectral index consists of a weighted combination of a vegetation index, a water body index and a differential building index for distinguishing the category of pattern spot elements; a sorting module, which is configured to according to the size of the comprehensive spectral index of each pattern spot image, sort each pattern spot image according to the element category; an abnormal category determining module, which is configured to determine whether there are abnormal category pattern spots in a set number of pattern spot images before and after the pattern spot image sequence, and determine whether there are abnormal category pattern spots in the added pattern spot images.
- 4. The system for automatically recognizing abnormal category pattern spots according to claim 3, wherein the system further comprises: an initial parameter determining module, which is configured to, according to the historical data, determine the error-prone or confusable element category and the weight coefficient of a vegetation index, a water body index and a differential building index in the comprehensive spectral index used to distinguish the error-prone or confusable element category.
- 5. The system for automatically identifying abnormal category pattern spots according to claim 3 or claim 4, wherein the sorting module specifically comprises: a numerical value determining unit, which is configured to determine the intermediate value or average value of the comprehensive spectral index of each pattern spot image; a sorting unit, which is configured to, according to the size of the intermediate value or average value of the comprehensive spectral index of the pattern spot image, sort the pattern spot image according to the element category.SICHUAN SURVEYING AND MAPPING PRODUCT QUALITY SUPERVISION AND INSPECTION STATION OF THE MINISTRY OF NATURAL RESOURCES (SICHUAN SURVEYING AND MAPPING PRODUCT QUALITY SUPERVISION AND INSPECTION STATION) By its Patent Attorneys ARMOUR IPP2428AU00
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010534638.9A CN111709927B (en) | 2020-06-12 | 2020-06-12 | Automatic identification method and system for type abnormal pattern spots |
CN202010534638.9 | 2020-06-12 |
Publications (1)
Publication Number | Publication Date |
---|---|
AU2021102441A4 true AU2021102441A4 (en) | 2021-06-24 |
Family
ID=72540346
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU2021102441A Ceased AU2021102441A4 (en) | 2020-06-12 | 2021-05-10 | Method for automatically identifying abnormal category pattern spots and system thereof |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN111709927B (en) |
AU (1) | AU2021102441A4 (en) |
WO (1) | WO2021248599A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113484245B (en) * | 2021-07-05 | 2022-11-22 | 重庆市规划和自然资源调查监测院 | Remote sensing rapid monitoring method and system for paddy field planting pattern in hilly and mountainous areas and computer readable storage medium |
CN114491187B (en) * | 2022-01-20 | 2022-11-18 | 重庆市规划和自然资源调查监测院 | Intelligent natural resource monitoring integrated system |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120271609A1 (en) * | 2011-04-20 | 2012-10-25 | Westerngeco L.L.C. | Methods and computing systems for hydrocarbon exploration |
CN103593852A (en) * | 2013-11-29 | 2014-02-19 | 中国科学院光电研究院 | Hyperspectral image abnormality detection method based on homogeneous patches |
CN103593853B (en) * | 2013-11-29 | 2016-05-11 | 武汉大学 | The multiple dimensioned object-oriented classification method of remote sensing image of expressing based on joint sparse |
CN104408463B (en) * | 2014-10-15 | 2020-05-12 | 中国土地勘测规划院 | High-resolution construction land pattern spot identification method |
EP3173975A1 (en) * | 2015-11-30 | 2017-05-31 | Delphi Technologies, Inc. | Method for identification of candidate points as possible characteristic points of a calibration pattern within an image of the calibration pattern |
KR101756380B1 (en) * | 2017-03-06 | 2017-07-10 | 한국지질자원연구원 | Detection methods of porphyry copper deposits using malachite hyperspectral imagery |
CN109726679B (en) * | 2018-12-28 | 2020-11-27 | 常州市星图测绘科技有限公司 | Remote sensing classification error spatial distribution mapping method |
CN110135310A (en) * | 2019-04-30 | 2019-08-16 | 云南财经大学 | A kind of crops remote sensing recognition method based on single argument feature selection approach |
CN110363798B (en) * | 2019-07-24 | 2022-02-18 | 宁波市测绘和遥感技术研究院 | Method for generating remote sensing image interpretation sample set |
CN111178169B (en) * | 2019-12-12 | 2023-03-10 | 广州地理研究所 | Urban surface covering fine classification method and device based on remote sensing image |
CN111144249B (en) * | 2019-12-16 | 2022-05-10 | 广州地理研究所 | Ground surface coverage type determination method based on automatic optimization MESMA |
-
2020
- 2020-06-12 CN CN202010534638.9A patent/CN111709927B/en active Active
- 2020-07-06 WO PCT/CN2020/100451 patent/WO2021248599A1/en active Application Filing
-
2021
- 2021-05-10 AU AU2021102441A patent/AU2021102441A4/en not_active Ceased
Also Published As
Publication number | Publication date |
---|---|
WO2021248599A1 (en) | 2021-12-16 |
CN111709927B (en) | 2022-05-27 |
CN111709927A (en) | 2020-09-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
AU2021102441A4 (en) | Method for automatically identifying abnormal category pattern spots and system thereof | |
Hayes et al. | Comparison of change-detection techniques for monitoring tropical forest clearing and vegetation regrowth in a time series | |
Cihlar et al. | Classification by progressive generalization: A new automated methodology for remote sensing multichannel data | |
CN111626269B (en) | Practical large-space-range landslide extraction method | |
CN110675588B (en) | Forest fire detection device and method | |
CN107239795A (en) | SAR image change detecting system and method based on sparse self-encoding encoder and convolutional neural networks | |
CN103942777A (en) | Mobile phone glass cover plate defect detecting method based on principal component analysis | |
CN110376202B (en) | Tea tree anthracnose lesion identification method based on imaging hyperspectral technology | |
CN111739067A (en) | Remote sensing image change detection method and device | |
Homayouni et al. | Hyperspectral image analysis for material mapping using spectral matching | |
CN113657294B (en) | Crop disease and insect pest detection method and system based on computer vision | |
CN117636185B (en) | Pine wood nematode disease detecting system based on image processing | |
CN110555395A (en) | Classified evaluation method for nitrogen content grade of rape canopy | |
Elberink et al. | Detection of collapsed buildings by classifying segmented airborne laser scanner data | |
CN113011354A (en) | Unmanned aerial vehicle hyperspectral image pine wood nematode disease identification method based on deep learning | |
CN111007013A (en) | Crop rotation fallow remote sensing monitoring method and device for northeast cold region | |
CN110570462A (en) | flood inundation range automatic extraction method based on polarized radar remote sensing image | |
CN109115719A (en) | A kind of Citrus Huanglongbing pathogen Band fusion rapid detection method based on high light spectrum image-forming technology | |
CN111060455B (en) | Northeast cold-cool area oriented remote sensing image crop marking method and device | |
Teomete et al. | Digital image processing for pavement distress analyses | |
CN113962929A (en) | Photovoltaic cell assembly defect detection method and system and photovoltaic cell assembly production line | |
CN106169086B (en) | High-resolution optical image under navigation data auxiliary damages method for extracting roads | |
CN110929739B (en) | Automatic impervious surface range remote sensing iterative extraction method | |
CN116310826B (en) | High-resolution remote sensing image forest land secondary classification method based on graphic neural network | |
CN111882573A (en) | Cultivated land plot extraction method and system based on high-resolution image data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FGI | Letters patent sealed or granted (innovation patent) | ||
MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |