CN118015475A - Method for detecting ground conflict area in Lin Quanque right - Google Patents
Method for detecting ground conflict area in Lin Quanque right Download PDFInfo
- Publication number
- CN118015475A CN118015475A CN202410417206.8A CN202410417206A CN118015475A CN 118015475 A CN118015475 A CN 118015475A CN 202410417206 A CN202410417206 A CN 202410417206A CN 118015475 A CN118015475 A CN 118015475A
- Authority
- CN
- China
- Prior art keywords
- square
- density
- blocks
- detecting
- marking
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000004891 communication Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 5
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 abstract description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012850 discrimination method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/457—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by analysing connectivity, e.g. edge linking, connected component analysis or slices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Linguistics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for detecting a ground conflict area in Lin Quanque rights, which belongs to the technical field of image data processing and comprises the steps of obtaining a remote sensing image of a forest rights evidence issuing scope, and dividing the remote sensing image into m blocks; extracting high-level semantic features of each square; calculating the square density of each block in turn; marking the core square and communicating the density; and detecting the ground type conflict area based on density communication. The method can effectively abstract and extract high-level semantic information contained in the image to calculate density, define the density based on the similarity of the square blocks and the neighbors thereof, accurately capture the local information of the data, hierarchically merge the square blocks based on the density, and enable the clustering of irregularly shaped remote sensing images to be more accurate. When the method is used for detecting the abnormality, the average similarity between the blocks in the cluster and the similarity difference between all the blocks in other clusters can be fully considered, and the problem of ground type conflict in Lin Quanque weight is better solved.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to a method for detecting a ground conflict area in Lin Quanque rights.
Background
The forest rights refer to ownership rights and use rights of forests, woods and woodlands, and the forest rights are required to be in the range of issuing a certificate, but other lands such as cultivated lands, grasslands and the like exist in the range of issuing the certificate due to inconsistent identification of the lands by different management departments; or constructing roads, buildings and the like in the range of the later forest rights and certificates, and the forest land is changed, so that land collision occurs. The land-based conflict area is a non-woodland with a very small duty cycle. At present, the identification mode of the land conflict area is mainly used for identifying the land conflict area in the forest land by manpower according to remote sensing images, but the manual identification speed is low and the time cost is high.
Noun interpretation: instance Discrimination, abbreviated as InstDisc, is an example discrimination method, and is an unsupervised contrast learning algorithm. The example discrimination refers to that each training data is regarded as a category different from other data, so that the model has the capability of identifying visual features of images, and meanwhile, the feature vectors output by the model are ensured to be closer to the images similar in vision and further away from the images dissimilar in vision.
Disclosure of Invention
The invention aims to provide a method for detecting the ground type conflict area in Lin Quanque-oriented rights, which can quickly identify the ground type conflict area and solve the problems.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: a method for detecting a ground type conflict area in Lin Quanque-oriented rights comprises the following steps of;
s1, acquiring a remote sensing image of a forest rights and license issuing scope, dividing the remote sensing image into m blocks, and sequentially marking the blocks as x 1~xm;
s2, extracting high-level semantic features of each square, wherein the high-level semantic features corresponding to x 1~xm are v 1~vm in sequence;
S3, sequentially calculating the density rho 1~ρm of m square blocks, wherein the calculation method of the density rho i of the ith square block x i comprises S31-S32;
S31, finding the neighbors of x i, n i, and marking as ，1≤i≤m；
S32, calculating the density rho i of x i;
，
Wherein v i,j is the high-level semantic feature of the square corresponding to x i,j, j is more than or equal to 1 and less than or equal to n i, To calculate the Euclidean distance;
S4, sequentially processing x 1~xm, marking a core block and communicating the density, wherein the processing method of x i comprises S41-S42;
S41, if Marking x i as a core square, otherwise, not marking, wherein ρ i,j is the density of the square corresponding to x i,j;
S42, if for two adjacent blocks x i and x k, at least 1 is a core block and Marking x i、xk as density communication, wherein ρ k is the density of x k, and α is a density communication parameter;
S5, detecting a ground-based conflict area based on density communication, wherein the method comprises the steps of S51-S57;
s51, constructing a block set ；
S52, searching the square with the maximum density in X 1~xm from X, and if the square is X i, generating a cluster C 1={xi;
S53, adding neighbors in density communication with X i to C 1, forming a set A by core blocks in C 1, traversing all blocks in A, adding blocks in density communication with the blocks in A to C 1, subtracting blocks in C 1 from X, and updating X;
S54, generating a cluster C 2 from the updated X according to the steps S52 and S53;
S55, repeating the step S54 until X is empty, generating C clusters, and sequentially forming C 1~Cc clusters;
S56, selecting a cluster C δ with the largest square, and calculating the average similarity avg (C δ) of the square, wherein delta is more than or equal to 1 and less than or equal to C;
S57, for the rest clusters C t, if the following formula is met, marking the square in C t as abnormal, wherein t is more than or equal to 1 and less than or equal to C, and t is not equal to delta;
，
Where v δ is the high-level semantic feature of the highest density square in C δ, x t is the square in C t, and v t is the high-level semantic feature of x t.
As preferable: in S1, the remote sensing image comprises N 1×N2 pixels, each square comprises M multiplied by M pixels, and M < < N 1、M<<N2.
As preferable: s2, taking each square as a category, taking the ith square x i as the ith category, and extracting features by using Instance Discrimination algorithm to obtain high-level semantic features。
As preferable: density connectivity parameter α=0.5.
As preferable: in S56, avg (C δ) is obtained according to the following formula;
，
wherein, |C δ | is the number of elements contained in set A, v p、vq is the corresponding high-level semantic features of the p-th and q-th blocks in C δ, respectively.
Compared with the prior art, the invention has the advantages that: firstly dividing a remote sensing image into a plurality of blocks, extracting the characteristics of the blocks by applying a deep learning method, finally defining a new density communication concept according to the densities of the blocks and the neighbors thereof, and detecting a forest land conflict area based on the density communication, thereby having the following remarkable advantages:
(1) The remote sensing image is modularized, and the blocks are taken as the dividing basic elements instead of the pixel points, so that the dividing speed is greatly increased.
(2) Features of the square blocks are extracted by using an unsupervised contrast learning method, high-level semantic information contained in the extracted image can be effectively abstracted, and accurate estimation of density of the square blocks is facilitated.
(3) The density of the square is defined based on the similarity between the square and its neighbor, so that the local information of the data can be accurately captured.
(4) Based on the density, the strategy of hierarchically combining the squares enables the clustering of irregularly shaped remote sensing images to be more accurate.
(5) When the method detects abnormality, the average similarity between the blocks in the cluster and the similarity difference between all the blocks in other clusters are fully considered, so that the problem of ground type conflict in Lin Quanque weight can be better solved.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of a neighbor block;
FIG. 3 is a remote sensing image of a forest rights issuing scope;
FIG. 4 shows the result of abnormality detection performed in the method of FIG. 3 according to the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
Example 1: referring to fig. 1 and 2, a method for detecting a ground type conflict area in Lin Quanque-oriented rights includes the following steps;
s1, acquiring a remote sensing image of a forest rights and license issuing scope, dividing the remote sensing image into m blocks, and sequentially marking the blocks as x 1~xm;
s2, extracting high-level semantic features of each square, wherein the high-level semantic features corresponding to x 1~xm are v 1~vm in sequence;
S3, sequentially calculating the density rho 1~ρm of m square blocks, wherein the calculation method of the density rho i of the ith square block x i comprises S31-S32;
S31, finding the neighbors of x i, n i, and marking as ，1≤i≤m；
S32, calculating the density rho i of x i;
，
Wherein v i,j is the high-level semantic feature of the square corresponding to x i,j, j is more than or equal to 1 and less than or equal to n i, To calculate the Euclidean distance;
S4, sequentially processing x 1~xm, marking a core block and communicating the density, wherein the processing method of x i comprises S41-S42;
S41, if Marking x i as a core square, otherwise, not marking, wherein ρ i,j is the density of the square corresponding to x i,j;
S42, if for two adjacent blocks x i and x k, at least 1 is a core block and Marking x i、xk as density communication, wherein ρ k is the density of x k, and α is a density communication parameter;
S5, detecting a ground-based conflict area based on density communication, wherein the method comprises the steps of S51-S57;
s51, constructing a block set ；
S52, searching the square with the maximum density in X 1~xm from X, and if the square is X i, generating a cluster C 1={xi;
S53, adding neighbors in density communication with X i to C 1, forming a set A by core blocks in C 1, traversing all blocks in A, adding blocks in density communication with the blocks in A to C 1, subtracting blocks in C 1 from X, and updating X;
S54, generating a cluster C 2 from the updated X according to the steps S52 and S53;
S55, repeating the step S54 until X is empty, generating C clusters, and sequentially forming C 1~Cc clusters;
S56, selecting a cluster C δ with the largest square, and calculating the average similarity avg (C δ) of the square, wherein delta is more than or equal to 1 and less than or equal to C;
S57, for the rest clusters C t, if the following formula is met, marking the square in C t as abnormal, wherein t is more than or equal to 1 and less than or equal to C, and t is not equal to delta;
，
Where v δ is the high-level semantic feature of the highest density square in C δ, x t is the square in C t, and v t is the high-level semantic feature of x t.
In S1, the remote sensing image comprises N 1×N2 pixels, each square comprises M multiplied by M pixels, and M < < N 1、M<<N2.
S2, taking each square as a category, taking the ith square x i as the ith category, and extracting features by using Instance Discrimination algorithm to obtain high-level semantic features。
Density connectivity parameter α=0.5.
In S56, avg (C δ) is obtained according to the following formula;
，
wherein, |C δ | is the number of elements contained in set A, v p、vq is the corresponding high-level semantic features of the p-th and q-th blocks in C δ, respectively.
In addition, regarding the calculation of the density of each block in S3, the neighboring blocks are referred to in S31, and in fig. 2, 8 neighbors of x i are sequentially labeled as x i,1~xi,8, and 3 neighbors of x m of the lower right corner are sequentially labeled as x m,1~xm,3.
Example 2: referring to fig. 3 and 4, in order to illustrate the effect of the present invention, the remote sensing image shown in fig. 3 is selected in this embodiment, fig. 3 is a forest land including a pond, an elliptical partial area in the figure is a pond, and in the detection of a land-like conflict area by the method of the present invention, abnormal squares are marked to obtain fig. 4, which exactly corresponds to the position of the pond. The set of blocks marked as anomalies is the local class conflict area. As can be seen by comparing FIG. 3 with FIG. 4, the present invention can accurately detect abnormal squares contained in the remote sensing image of the woodland and identify the conflict area of the land.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (5)
1. A method for detecting a ground type conflict area in Lin Quanque weight is characterized by comprising the following steps: comprises the following steps of;
s1, acquiring a remote sensing image of a forest rights and license issuing scope, dividing the remote sensing image into m blocks, and sequentially marking the blocks as x 1~xm;
s2, extracting high-level semantic features of each square, wherein the high-level semantic features corresponding to x 1~xm are v 1~vm in sequence;
S3, sequentially calculating the density rho 1~ρm of m square blocks, wherein the calculation method of the density rho i of the ith square block x i comprises S31-S32;
S31, finding the neighbors of x i, n i, and marking as ，1≤i≤m；
S32, calculating the density rho i of x i;
，
Wherein v i,j is the high-level semantic feature of the square corresponding to x i,j, j is more than or equal to 1 and less than or equal to n i, To calculate the Euclidean distance;
S4, sequentially processing x 1~xm, marking a core block and communicating the density, wherein the processing method of x i comprises S41-S42;
S41, if Marking x i as a core square, otherwise, not marking, wherein ρ i,j is the density of the square corresponding to x i,j;
S42, if for two adjacent blocks x i and x k, at least 1 is a core block and Marking x i、xk as density communication, wherein ρ k is the density of x k, and α is a density communication parameter;
S5, detecting a ground-based conflict area based on density communication, wherein the method comprises the steps of S51-S57;
s51, constructing a block set ；
S52, searching the square with the maximum density in X 1~xm from X, and if the square is X i, generating a cluster C 1={xi;
S53, adding neighbors in density communication with X i to C 1, forming a set A by core blocks in C 1, traversing all blocks in A, adding blocks in density communication with the blocks in A to C 1, subtracting blocks in C 1 from X, and updating X;
S54, generating a cluster C 2 from the updated X according to the steps S52 and S53;
S55, repeating the step S54 until X is empty, generating C clusters, and sequentially forming C 1~Cc clusters;
S56, selecting a cluster C δ with the largest square, and calculating the average similarity avg (C δ) of the square, wherein delta is more than or equal to 1 and less than or equal to C;
S57, for the rest clusters C t, if the following formula is met, marking the square in C t as abnormal, wherein t is more than or equal to 1 and less than or equal to C, and t is not equal to delta;
，
Where v δ is the high-level semantic feature of the highest density square in C δ, x t is the square in C t, and v t is the high-level semantic feature of x t.
2. The method for detecting a ground-based conflict area in Lin Quanque-oriented rights according to claim 1, wherein: in S1, the remote sensing image comprises N 1×N2 pixels, each square comprises M multiplied by M pixels, and M < < N 1、M<<N2.
3. The method for detecting a ground-based conflict area in Lin Quanque-oriented rights according to claim 1, wherein: s2, taking each square as a category, taking the ith square x i as the ith category, and extracting features by using Instance Discrimination algorithm to obtain high-level semantic features。
4. The method for detecting a ground-based conflict area in Lin Quanque-oriented rights according to claim 1, wherein: density connectivity parameter α=0.5.
5. The method for detecting a ground-based conflict area in Lin Quanque-oriented rights according to claim 1, wherein: in S56, avg (C δ) is obtained according to the following formula;
，
wherein, |C δ | is the number of elements contained in set A, v p、vq is the corresponding high-level semantic features of the p-th and q-th blocks in C δ, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410417206.8A CN118015475B (en) | 2024-04-09 | 2024-04-09 | Method for detecting ground conflict area in Lin Quanque right |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410417206.8A CN118015475B (en) | 2024-04-09 | 2024-04-09 | Method for detecting ground conflict area in Lin Quanque right |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118015475A true CN118015475A (en) | 2024-05-10 |
| CN118015475B CN118015475B (en) | 2024-06-21 |
Family
ID=90950384
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410417206.8A Active CN118015475B (en) | 2024-04-09 | 2024-04-09 | Method for detecting ground conflict area in Lin Quanque right |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118015475B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5875471U (en) * | 1981-11-16 | 1983-05-21 | 三菱電機株式会社 | Density balance device for cooling gas in cooling equipment for rotating electric machines |
| CN110659774A (en) * | 2019-09-23 | 2020-01-07 | 北京交通大学 | Big data method driven parking demand prediction method |
| US20210073692A1 (en) * | 2016-06-12 | 2021-03-11 | Green Grid Inc. | Method and system for utility infrastructure condition monitoring, detection and response |
| WO2023076958A1 (en) * | 2021-10-26 | 2023-05-04 | Csp Technologies, Inc. | Antimicrobial activity of chlorite-entrained polymers |
| CN117197467A (en) * | 2023-09-19 | 2023-12-08 | 郑州吉奥地理信息技术有限公司 | Multi-source remote sensing image ground semantic segmentation application and evaluation method |
-
2024
- 2024-04-09 CN CN202410417206.8A patent/CN118015475B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5875471U (en) * | 1981-11-16 | 1983-05-21 | 三菱電機株式会社 | Density balance device for cooling gas in cooling equipment for rotating electric machines |
| US20210073692A1 (en) * | 2016-06-12 | 2021-03-11 | Green Grid Inc. | Method and system for utility infrastructure condition monitoring, detection and response |
| CN110659774A (en) * | 2019-09-23 | 2020-01-07 | 北京交通大学 | Big data method driven parking demand prediction method |
| WO2023076958A1 (en) * | 2021-10-26 | 2023-05-04 | Csp Technologies, Inc. | Antimicrobial activity of chlorite-entrained polymers |
| CN117197467A (en) * | 2023-09-19 | 2023-12-08 | 郑州吉奥地理信息技术有限公司 | Multi-source remote sensing image ground semantic segmentation application and evaluation method |
Non-Patent Citations (5)
| Title |
|---|
| HSUEH-YI LAI: "A Holistic Team Cognition With Better Information Transparency: A Mixing Pattern of Team Interaction in Landing Systems", 《IEEE》, 15 June 2021 (2021-06-15) * |
| 吐尔逊・艾山;赛迪古丽・哈西木;吐热尼古丽・阿木提;: "新疆开都河流域下游绿洲景观格局分析", 新疆师范大学学报(自然科学版), no. 01, 30 March 2017 (2017-03-30) * |
| 易凤佳;李仁东;常变蓉;: "基于面向对象的长株潭地区遥感影像分类方法", 华中师范大学学报(自然科学版), no. 06, 15 December 2014 (2014-12-15) * |
| 李伟: "林权确权登记整村集中办理模式的应用探索", 《城市勘测》, 8 August 2023 (2023-08-08) * |
| 潘峰;唐晓岚;黄滢;顾哲衍;杨传清;: "太湖西山岛景观格局演变及风景开发趋向研究", 西北林学院学报, no. 05, 15 September 2017 (2017-09-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118015475B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107330396B (en) | Pedestrian re-identification method based on multi-attribute and multi-strategy fusion learning | |
| Zhai et al. | Detecting vanishing points using global image context in a non-manhattan world | |
| CN109670528B (en) | Data expansion method facing pedestrian re-identification task and based on paired sample random occlusion strategy | |
| CN109685135B (en) | Few-sample image classification method based on improved metric learning | |
| CN107230267B (en) | Intelligence In Baogang Kindergarten based on face recognition algorithms is registered method | |
| CN101980250B (en) | Method for identifying target based on dimension reduction local feature descriptor and hidden conditional random field | |
| CN111222434A (en) | Method for obtaining evidence of synthesized face image based on local binary pattern and deep learning | |
| CN109063649B (en) | Pedestrian re-identification method based on twin pedestrian alignment residual error network | |
| CN109871875B (en) | Building change detection method based on deep learning | |
| CN112668557B (en) | Method for defending image noise attack in pedestrian re-identification system | |
| CN108596010B (en) | Implementation method of pedestrian re-identification system | |
| CN111310662B (en) | Flame detection and identification method and system based on integrated deep network | |
| CN111027421A (en) | Graph-based direct-push type semi-supervised pedestrian re-identification method | |
| CN108345850A (en) | The scene text detection method of the territorial classification of stroke feature transformation and deep learning based on super-pixel | |
| CN105809678B (en) | A kind of line segment feature global registration method between two views under short base line condition | |
| CN103164693B (en) | A kind of monitor video pedestrian detection matching process | |
| CN113128600A (en) | Structured depth incomplete multi-view clustering method | |
| CN110472495B (en) | Deep learning face recognition method based on graphic reasoning global features | |
| Davy et al. | Reducing anomaly detection in images to detection in noise | |
| CN105740915A (en) | Cooperation segmentation method fusing perception information | |
| CN106570499A (en) | Object tracking method based on probability graph model | |
| CN111009005A (en) | Scene classification point cloud rough registration method combining geometric information and photometric information | |
| CN114861761B (en) | Loop detection method based on twin network characteristics and geometric verification | |
| Zhang et al. | Multimodal remote sensing image matching combining learning features and delaunay triangulation | |
| CN112926557B (en) | Method for training multi-mode face recognition model and multi-mode face recognition method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |