CN110929572A - Forest fire identification method and system - Google Patents
Forest fire identification method and system Download PDFInfo
- Publication number
- CN110929572A CN110929572A CN201910994446.3A CN201910994446A CN110929572A CN 110929572 A CN110929572 A CN 110929572A CN 201910994446 A CN201910994446 A CN 201910994446A CN 110929572 A CN110929572 A CN 110929572A
- Authority
- CN
- China
- Prior art keywords
- classification
- area
- forest fire
- interpolation
- time phase
- 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 27
- 238000004364 calculation method Methods 0.000 claims abstract description 25
- 230000011218 segmentation Effects 0.000 claims abstract description 19
- 239000013598 vector Substances 0.000 claims description 18
- 238000002955 isolation Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 230000003595 spectral effect Effects 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 230000000694 effects Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
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
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4007—Interpolation-based scaling, e.g. bilinear interpolation
Abstract
The invention discloses a forest fire identification method and a forest fire identification system, wherein two time phase images of an observation area in two time slots can be obtained through a camera arranged in a forest area, a satellite monitoring mode and the like, bicubic spline interpolation calculation is carried out on the two time phase images respectively to obtain two interpolation images, the two interpolation images are classified based on a mean shift segmentation algorithm to obtain a plurality of classification sets, the classification sets are segmented respectively to obtain changed area point sets, all the area point sets are combined to obtain two time phase surface primitives, finally the similarity calculation is carried out on the two time phase surface primitives to separate out suspected fire type area characteristics, and the suspected fire type area characteristics are compared with a standard fire type library to determine fire area locations, characteristic types and the like; the invention reduces the comparison area by dividing the image area, reduces the processing amount of comparison data, and has the technical effects of improving the identification speed and improving the overall response efficiency.
Description
Technical Field
The invention belongs to the technical field of forest fire monitoring, and particularly relates to a forest fire identification method and system.
Background
Forest is a green treasury of human beings, and forest fire is one of the important reasons for the global forest resource degradation, is a natural disaster which has great destructiveness and is difficult to rescue suddenly, and can cause great harm and loss.
The deep forest fire prevention work is the important part of the forestry work, and scientific and accurate real-time monitoring not only is an inevitable requirement on forest fire prevention, but also is an important means for effectively controlling fire spread and reducing economic loss, and is also an important reference for fire suppression command decision.
Disclosure of Invention
The invention aims to provide a forest fire identification method and a forest fire identification system, which are used for carrying out segmentation identification on the suspected forest fire area characteristics of a forest by combining double-three interpolation calculation and a mean shift segmentation algorithm, comparing the suspected forest fire area characteristics with a standard fire variety library to determine the area location, the characteristic type and the like of a fire, and improving the identification speed of the forest fire.
In order to solve the technical problems, the invention adopts the following technical scheme:
a forest fire identification method is provided, which comprises the following steps: acquiring two time phase images of an observation area in two time slots; respectively carrying out bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images; processing the two interpolation images based on a mean shift segmentation algorithm to obtain two time phase surface elements; carrying out similarity calculation on the two time-phase surface elements to separate out the characteristics of the suspected fire area; forest fire information is identified based on the characteristics of the suspected fire regions.
A forest fire identification system is proposed, comprising: the image acquisition module is used for acquiring two time phase images of the observation area in two time slots; the object construction module is used for respectively carrying out bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images; processing the two interpolation images based on a mean shift segmentation algorithm to obtain two time phase surface elements; the object similarity calculation module is used for calculating the similarity of the two time phase surface elements and separating the characteristics of the suspected fire area; and the identification module is used for identifying forest fire information based on the characteristics of the suspected fire area.
Compared with the prior art, the invention has the advantages and positive effects that: the forest fire identification method and system provided by the invention can acquire two time phase images of an observation area in two time slots by means of a camera arranged in a forest area, satellite monitoring and the like, perform bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images respectively, classify the two interpolation images respectively based on a mean shift segmentation algorithm to obtain a plurality of classification sets, divide the classification sets respectively to obtain changed area point sets, combine all the area point sets to obtain two time phase surface primitives, perform similarity calculation on the two time phase surface primitives finally, separate out suspected fire type area characteristics, and compare the suspected fire type area characteristics with a standard fire type base to determine fire area locations, characteristic types and the like; the invention reduces the comparison area by dividing the image area, reduces the processing amount of comparison data, and has the technical effects of improving the identification speed and improving the overall response efficiency.
Other features and advantages of the present invention will become more apparent from the detailed description of the embodiments of the present invention when taken in conjunction with the accompanying drawings.
Drawings
FIG. 1 is a flow chart of a method of forest fire identification in accordance with the present invention;
fig. 2 is an architecture diagram of a forest fire recognition system according to the present invention;
FIG. 3 is a schematic diagram of a mean shift segmentation process in the forest fire identification method according to the present invention;
fig. 4 is a schematic classification diagram in step S2 of the forest fire identification method according to the present invention.
Detailed Description
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
The forest fire identification method provided by the invention is realized based on a forest fire identification system shown in fig. 2, wherein the forest fire identification system comprises an image acquisition module 1, an object construction module 2, an object similarity calculation module 3 and an identification module 4, wherein the image acquisition module 1 is such as a camera arranged in a forest area, satellite monitoring and the like, and the object construction module 2, the object similarity calculation module 3 and the identification module 4 realize identification of forest fire, specifically, as shown in fig. 1, the forest fire identification method comprises the following steps:
step S1: two time phase images of the observation area in two time slots are obtained.
For the target observation region, the first phase image f (x, y) and the second phase image g (u, v) are acquired at time slots t1 and t2, respectively.
Step S2: and respectively carrying out bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images.
Performing bicubic spline interpolation calculation on the first time phase image f (x, y) to obtain
Wherein x 'is x-1+ i, y' is y-1+ i, pjRepresents the intermediate value of the j-th line interpolation in the x direction, [ x ]]、[y]Respectively representing the rounding-down of x and y;
a is a forest region image extraction coefficient, and values are taken according to the conditions of different forest regions, and generally 1 is taken.
Performing bicubic spline interpolation calculation on the second time phase image g (u, v) to obtain
Wherein u '-u-1 + i, v' -v-1 + i, pjRepresents the intermediate value of the j-th line interpolation in the u direction, [ u ]]、[v]Respectively representing the rounding of u and v;
a is a forest region image extraction coefficient, and values are taken according to the conditions of different forest regions, and generally 1 is taken.
Step S3: and processing the two interpolation images based on a mean shift segmentation algorithm to obtain two time phase surface elements.
Respectively carrying out mean shift segmentation on the two interpolation image data obtained in the step S2 for classification; after classification, the classification result is divided according to the classification of each interpolation image data to obtain a changed region point set, and finally all the region point sets are combined to obtain a time phase surface element.
Specifically, the classification process includes: 1) as shown in fig. 3, a central point O is randomly selected from the interpolated image data, step 2) a set M is calculated by using the data points within a set bandwidth h from the central point O, and the vectors from the central point O to each point in the set M are added to obtain an offset vector:
3) Judging whether the offset vector meets the set threshold requirement, if not, moving the central point O along the direction of the offset vector p, as shown in fig. 3, wherein the moving distance is the modulus of the offset vector, and determining a new central point O1Then by a distance O1Setting data points in the range of the bandwidth h as a set M1Calculating from O1Starting with set M1Adding the vectors of each point in the vector table to obtain a new offset vector; 4) repeating the steps 1) to 3) until the offset vector meets the set threshold range, and recording the central point O at the momentiAnd a classification set, wherein the classification set is a data point which is accessed with a frequency higher than a set frequency in the processes from step 1) to step 4)A simple embodiment is shown in FIG. 4 as the range of black and bold lines, M1、.....、MiThe intersection of all sets.
And 5) repeating the operation of randomly selecting a new central point, and obtaining a plurality of new classification sets again according to the methods from the step 1) to the step 4) until all data points in the interpolation image are classified.
In the embodiment of the invention, the classification sets with the data point number higher than the set value are divided to be used as all region point sets which are possible to generate fire, namely changed region point sets, and finally the changed region point sets are combined to obtain a time phase plane element.
Step S4: and (4) carrying out similarity calculation on the two time-phase surface elements, and classifying the characteristics of the suspected fire area.
Comparing the two time phase surface primitives D (x, y) and DN (u, v) obtained in step S3, and obtaining an isolation value between the two by making a difference:
wherein, N is an image feature discrimination coefficient, β and gamma are values of different key features of the image respectively, and the values of N, β and gamma are determined according to the spectral feature, edge, gradient feature, texture feature and/or elevation feature of the target object.
The images can be effectively separated by calculating the isolation value, and the characteristics of the suspected fire area are separated.
Step S5: forest fire information is identified based on the characteristics of the suspected fire regions.
And acquiring data information according to forest fire characteristics in the fire type characteristic library, comparing the characteristics of the suspected fire type area separated in the step S4 with the fire type characteristic library, and determining the area location, the characteristic type and the like of the fire.
Based on the forest fire identification method, the image acquisition module 1 of the forest fire identification system is used for acquiring two time phase images of an observation area in two time slots; the object construction module 2 is used for respectively carrying out bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images; processing the two interpolation images based on a mean shift segmentation algorithm to obtain two time phase surface elements; the object similarity calculation module 3 is used for performing similarity calculation on the two time phase surface primitives and separating out the characteristics of the suspected fire area; the identification module 4 is used for identifying forest fire information based on the characteristics of the suspected fire area.
Specifically, the object construction module 2 includes an interpolation image classification unit 21, a segmentation unit 22 and a merging unit 23; the interpolation image classification unit 21 is configured to classify interpolation image data based on a mean shift segmentation algorithm; the dividing unit 22 is configured to divide the classification result to obtain a changed region point set; the merging unit 23 is configured to merge all the region point sets to obtain a phase plane primitive.
The interpolated image classification unit 21 is specifically configured to perform the following steps: step 1) randomly selecting a central point in the interpolation image data; step 2) calculating an offset vector by taking data points within a set bandwidth range from a central point as a set; step 3), the central point moves along the offset vector; step 4) repeating the steps 1) to 3) with the new central point until the offset vector meets a set threshold value, and recording the central point and the classification set at the moment; wherein, the data points in the classification set are the data points with the visited frequency higher than the set frequency in the processes from step 1) to step 4); and 5) repeating the steps 1) to 4) until all data points in the interpolation image are classified to obtain a plurality of classification sets.
The dividing unit 22 is specifically configured to divide the classification set with the data point number higher than the set value from the classification sets obtained in step 5) into region point sets with changed data points.
The object similarity calculation module 3 is specifically configured to: the isolation values of the two time-plane primitives are calculated as:
d (x, y) and DN (u, v) are two time phase surface elements respectively, N is an image feature discrimination coefficient, β and gamma are values of different key features of an image respectively, and the values of N, β and gamma are determined according to spectral features, edges, gradient features, texture features and/or elevation features of a target object.
The identification process of the specific forest fire identification system has been described in detail in the above forest fire identification method, and is not described herein again.
It should be noted that the above description is not intended to limit the present invention, and the present invention is not limited to the above examples, and those skilled in the art should also make changes, modifications, additions or substitutions within the spirit and scope of the present invention.
Claims (10)
1. A forest fire identification method, comprising:
acquiring two time phase images of an observation area in two time slots;
respectively carrying out bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images;
processing the two interpolation images based on a mean shift segmentation algorithm to obtain two time phase surface elements;
carrying out similarity calculation on the two time-phase surface elements to separate out the characteristics of the suspected fire area;
forest fire information is identified based on the characteristics of the suspected fire regions.
2. The forest fire identification method according to claim 1, wherein the processing of the interpolated image based on a mean shift segmentation algorithm to obtain a time phase surface primitive comprises:
classifying the interpolation image data based on a mean shift segmentation algorithm;
dividing the classification result to obtain a changed region point set;
and combining all the regional point sets to obtain a phase surface element.
3. The forest fire identification method according to claim 2, wherein the classification of the interpolated image data based on a mean shift segmentation algorithm specifically comprises:
step 1) randomly selecting a central point in the interpolation image data;
step 2) calculating an offset vector by taking data points within a set bandwidth range from the central point as a set;
step 3) the central point moves along the offset vector;
step 4) repeating the steps 1) to 3) with the new central point until the offset vector meets a set threshold value, and recording the central point and the classification set at the moment; wherein, the data points in the classification set are the data points with the visited frequency higher than the set frequency in the processes from step 1) to step 4);
and 5) repeating the steps 1) to 4) until all data points in the interpolation image are classified to obtain a plurality of classification sets.
4. The forest fire recognition method according to claim 3, wherein the step of dividing the classification result to obtain the changed region point set specifically comprises the steps of:
and (5) dividing the classification set with the data point number higher than the set value from the classification sets obtained in the step 5) to be used as the changed region point set.
5. The forest fire identification method according to claim 1, wherein the similarity calculation is performed on the two time-phase surface primitives, and the characteristics of the suspected fire area are separated, specifically comprising:
the isolation values of the two time-plane primitives are calculated as:
separating out a suspected fire characteristic area based on the isolation value;
d (x, y) and DN (u, v) are two time phase surface elements respectively, N is an image feature discrimination coefficient, β and gamma are values of different key features of an image respectively, and the values of N, β and gamma are determined according to spectral features, edges, gradient features, texture features and/or elevation features of a target object.
6. Forest fire identification system, its characterized in that includes:
the image acquisition module is used for acquiring two time phase images of the observation area in two time slots;
the object construction module is used for respectively carrying out bicubic spline interpolation calculation on the two time phase images to obtain two interpolation images; processing the two interpolation images based on a mean shift segmentation algorithm to obtain two time phase surface elements;
the object similarity calculation module is used for calculating the similarity of the two time phase surface elements and separating the characteristics of the suspected fire area;
and the identification module is used for identifying forest fire information based on the characteristics of the suspected fire area.
7. The forest fire identification system of claim 6 wherein the object construction module comprises:
the interpolation image classification unit is used for classifying interpolation image data based on a mean shift segmentation algorithm;
a dividing unit, which is used for dividing the classification result to obtain a changed region point set;
and the merging unit is used for merging all the region point sets to obtain a phase surface primitive.
8. The forest fire recognition system of claim 7, wherein the interpolated image classification unit is specifically configured to:
step 1) randomly selecting a central point in the interpolation image data;
step 2) calculating an offset vector by taking data points within a set bandwidth range from the central point as a set;
step 3) the central point moves along the offset vector;
step 4) repeating the steps 1) to 3) with the new central point until the offset vector meets a set threshold value, and recording the central point and the classification set at the moment; wherein, the data points in the classification set are the data points with the visited frequency higher than the set frequency in the processes from step 1) to step 4);
and 5) repeating the steps 1) to 4) until all data points in the interpolation image are classified to obtain a plurality of classification sets.
9. The forest fire recognition system of claim 8, wherein the segmentation unit is specifically configured to: and (5) dividing the classification set with the data point number higher than the set value from the classification sets obtained in the step 5) to be used as the changed region point set.
10. The forest fire system of claim 6, wherein the object similarity calculation module is specifically configured to:
the isolation values of the two time-plane primitives are calculated as:
separating out a suspected fire characteristic area based on the isolation value;
d (x, y) and DN (u, v) are two time phase surface elements respectively, N is an image feature discrimination coefficient, β and gamma are values of different key features of an image respectively, and the values of N, β and gamma are determined according to spectral features, edges, gradient features, texture features and/or elevation features of a target object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910994446.3A CN110929572B (en) | 2019-10-18 | Forest fire identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910994446.3A CN110929572B (en) | 2019-10-18 | Forest fire identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110929572A true CN110929572A (en) | 2020-03-27 |
| CN110929572B CN110929572B (en) | 2023-11-10 |
Family
ID=
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111931648A (en) * | 2020-08-10 | 2020-11-13 | 成都思晗科技股份有限公司 | Hiwari 8 waveband data-based mountain fire real-time monitoring method |
| CN111931645A (en) * | 2020-08-10 | 2020-11-13 | 成都思晗科技股份有限公司 | Real-time mountain fire risk monitoring method based on remote sensing data |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10283482A (en) * | 1997-04-03 | 1998-10-23 | Toshiba Corp | Device for detecting moving target and method therefor |
| US6701005B1 (en) * | 2000-04-29 | 2004-03-02 | Cognex Corporation | Method and apparatus for three-dimensional object segmentation |
| CN101127078A (en) * | 2007-09-13 | 2008-02-20 | 北京航空航天大学 | Unmanned machine vision image matching method based on ant colony intelligence |
| CN101251893A (en) * | 2008-03-28 | 2008-08-27 | 西安电子科技大学 | Self-adapting multi-dimension veins image segmenting method based on wavelet and average value wander |
| CN102314688A (en) * | 2011-09-09 | 2012-01-11 | 华中科技大学 | Image segmentation method and image set segmentation method |
| KR101237089B1 (en) * | 2011-10-12 | 2013-02-26 | 계명대학교 산학협력단 | Forest smoke detection method using random forest classifier method |
| CN106997477A (en) * | 2017-03-30 | 2017-08-01 | 联想(北京)有限公司 | A kind of image processing method and device |
| CN107194944A (en) * | 2017-06-21 | 2017-09-22 | 南京林业大学 | Forest fire image dividing method and device |
| CN107301664A (en) * | 2017-05-25 | 2017-10-27 | 天津大学 | Improvement sectional perspective matching process based on similarity measure function |
| US20180101934A1 (en) * | 2016-10-06 | 2018-04-12 | Macau University Of Science And Technology | Device and a method for creating an image |
| CN109359554A (en) * | 2018-09-21 | 2019-02-19 | 南京信息工程大学 | A kind of forest fire recognition methods based on more composograph processing techniques |
| CN109657616A (en) * | 2018-12-19 | 2019-04-19 | 四川立维空间信息技术有限公司 | A kind of remote sensing image land cover pattern automatic classification method |
| CN110047241A (en) * | 2019-04-27 | 2019-07-23 | 刘秀萍 | A kind of forest fire unmanned plane cruise monitoring system |
| CN110334776A (en) * | 2019-07-15 | 2019-10-15 | 哈尔滨理工大学 | A kind of image classification recognition methods based on region bicubic interpolation technology |
| CN110334581A (en) * | 2019-05-09 | 2019-10-15 | 宁波市测绘设计研究院 | A kind of multi-source Remote Sensing Images change detecting method |
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10283482A (en) * | 1997-04-03 | 1998-10-23 | Toshiba Corp | Device for detecting moving target and method therefor |
| US6701005B1 (en) * | 2000-04-29 | 2004-03-02 | Cognex Corporation | Method and apparatus for three-dimensional object segmentation |
| CN101127078A (en) * | 2007-09-13 | 2008-02-20 | 北京航空航天大学 | Unmanned machine vision image matching method based on ant colony intelligence |
| CN101251893A (en) * | 2008-03-28 | 2008-08-27 | 西安电子科技大学 | Self-adapting multi-dimension veins image segmenting method based on wavelet and average value wander |
| CN102314688A (en) * | 2011-09-09 | 2012-01-11 | 华中科技大学 | Image segmentation method and image set segmentation method |
| KR101237089B1 (en) * | 2011-10-12 | 2013-02-26 | 계명대학교 산학협력단 | Forest smoke detection method using random forest classifier method |
| US20180101934A1 (en) * | 2016-10-06 | 2018-04-12 | Macau University Of Science And Technology | Device and a method for creating an image |
| CN106997477A (en) * | 2017-03-30 | 2017-08-01 | 联想(北京)有限公司 | A kind of image processing method and device |
| CN107301664A (en) * | 2017-05-25 | 2017-10-27 | 天津大学 | Improvement sectional perspective matching process based on similarity measure function |
| CN107194944A (en) * | 2017-06-21 | 2017-09-22 | 南京林业大学 | Forest fire image dividing method and device |
| CN109359554A (en) * | 2018-09-21 | 2019-02-19 | 南京信息工程大学 | A kind of forest fire recognition methods based on more composograph processing techniques |
| CN109657616A (en) * | 2018-12-19 | 2019-04-19 | 四川立维空间信息技术有限公司 | A kind of remote sensing image land cover pattern automatic classification method |
| CN110047241A (en) * | 2019-04-27 | 2019-07-23 | 刘秀萍 | A kind of forest fire unmanned plane cruise monitoring system |
| CN110334581A (en) * | 2019-05-09 | 2019-10-15 | 宁波市测绘设计研究院 | A kind of multi-source Remote Sensing Images change detecting method |
| CN110334776A (en) * | 2019-07-15 | 2019-10-15 | 哈尔滨理工大学 | A kind of image classification recognition methods based on region bicubic interpolation technology |
Non-Patent Citations (2)
| Title |
|---|
| 张祎勃: "基于空间相关性的SAR图像变化检测研究" * |
| 沈佳洁: "基于均值漂移算法的高分辨率遥感图像分割方法研究" * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111931648A (en) * | 2020-08-10 | 2020-11-13 | 成都思晗科技股份有限公司 | Hiwari 8 waveband data-based mountain fire real-time monitoring method |
| CN111931645A (en) * | 2020-08-10 | 2020-11-13 | 成都思晗科技股份有限公司 | Real-time mountain fire risk monitoring method based on remote sensing data |
| CN111931645B (en) * | 2020-08-10 | 2023-05-23 | 成都思晗科技股份有限公司 | Real-time mountain fire risk monitoring method based on remote sensing data |
| CN111931648B (en) * | 2020-08-10 | 2023-08-01 | 成都思晗科技股份有限公司 | Mountain fire real-time monitoring method based on Himaware 8-band data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639692B (en) | Shadow detection method based on attention mechanism | |
| US11120556B2 (en) | Iterative method for salient foreground detection and multi-object segmentation | |
| Maurya et al. | Road extraction using k-means clustering and morphological operations | |
| JP2002288658A (en) | Object extracting device and method on the basis of matching of regional feature value of segmented image regions | |
| EP3073443B1 (en) | 3d saliency map | |
| CN109858438B (en) | Lane line detection method based on model fitting | |
| CN110866926B (en) | Infrared remote sensing image rapid and fine sea-land segmentation method | |
| US11887346B2 (en) | Systems and methods for image feature extraction | |
| CN104657980A (en) | Improved multi-channel image partitioning algorithm based on Meanshift | |
| CN113763427B (en) | Multi-target tracking method based on coarse-to-fine shielding processing | |
| CN111915583A (en) | Vehicle and pedestrian detection method based on vehicle-mounted thermal infrared imager in complex scene | |
| CN111274964B (en) | Detection method for analyzing water surface pollutants based on visual saliency of unmanned aerial vehicle | |
| CN110852327A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN113111878B (en) | Infrared weak and small target detection method under complex background | |
| US20040131236A1 (en) | Method and apparatus for processing an image | |
| CN107977608B (en) | Method for extracting road area of highway video image | |
| CN112101113B (en) | Lightweight unmanned aerial vehicle image small target detection method | |
| CN107170004B (en) | Image matching method for matching matrix in unmanned vehicle monocular vision positioning | |
| CN104881670A (en) | Rapid target extraction method used for SAR azimuth estimation | |
| CN110929572A (en) | Forest fire identification method and system | |
| CN110929572B (en) | Forest fire identification method and system | |
| CN106951831B (en) | Pedestrian detection tracking method based on depth camera | |
| JP2000306104A (en) | Method and device for picture area division | |
| CN114092486A (en) | Automatic segmentation method and device for image texture background | |
| CN113763474A (en) | Scene geometric constraint-based indoor monocular depth estimation 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 |