CN111597909B - Fire detection judging method based on visual saliency - Google Patents
Fire detection judging method based on visual saliency Download PDFInfo
- Publication number
- CN111597909B CN111597909B CN202010319755.3A CN202010319755A CN111597909B CN 111597909 B CN111597909 B CN 111597909B CN 202010319755 A CN202010319755 A CN 202010319755A CN 111597909 B CN111597909 B CN 111597909B
- Authority
- CN
- China
- Prior art keywords
- image
- suspicious region
- region
- equal
- visual saliency
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000000007 visual effect Effects 0.000 title claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 4
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000000903 blocking effect Effects 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 230000035945 sensitivity Effects 0.000 abstract description 3
- 238000012544 monitoring process Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002265 prevention 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
-
- G06T3/02—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06V10/56—Extraction of image or video features relating to colour
-
- 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
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Abstract
A fire detection judging method based on visual saliency relates to the technical field of fire detection judgment, solves the problem of low accuracy, and comprises the following steps: s1, acquiring a high-resolution remote sensing image of a region to be detected; s2, processing the high-resolution remote sensing image to obtain a color enhanced image; s3, obtaining a suspicious region I according to priori knowledge that R is more than or equal to G and more than or equal to B and a visual saliency detection algorithm; s1 and S2 are executed again, after S2 is completed, a suspicious region II is obtained according to priori knowledge that R is more than or equal to G and more than or equal to B and a visual saliency detection algorithm, and S5 is carried out; s5, comparing the suspicious region I with the suspicious region II, and judging fire if the suspicious region II overlaps with the suspicious region I and the suspicious region II is larger than the suspicious region I; otherwise, the suspicious region I is removed, and the suspicious region II is returned to S4 as the suspicious region I. The invention has high sensitivity and accuracy for fire detection and judgment.
Description
Technical Field
The invention relates to the technical field of fire detection and judgment, in particular to a fire detection and judgment method based on visual saliency.
Background
Fire prevention is a hotspot problem beneficial to national folks, and at present, fire monitoring and judging technologies mainly comprise manual monitoring, observation tower monitoring, manned/unmanned aerial vehicle monitoring, satellite monitoring and the like, and detection, identification and judging aspects mainly depend on naked eye observation of experienced operators or semi-automatic auxiliary interpretation of computers. The large-area fire disaster can only be detected by naked eyes, and the fire disaster is difficult to detect for a small-range forest fire. Meanwhile, when the fire disaster is identified manually, the problems of fatigue, negligence and low observation efficiency are easy to occur to the staff, so that the timely discovery of the fire disaster is influenced. To overcome these problems, a fire detection judgment method based on visual saliency is gradually generated, but the problem of low detection accuracy still exists.
Disclosure of Invention
The invention provides a fire detection judging method based on visual saliency, which aims to solve the problem of low detection accuracy of the existing fire detection judging method.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a fire detection judging method based on visual saliency comprises the following steps:
s1, acquiring a high-resolution remote sensing image of a region to be detected;
s2, processing the high-resolution remote sensing image to obtain a color enhanced image;
s3, extracting the obtained color enhanced image by using a visual saliency detection algorithm according to priori knowledge that R is more than or equal to G and more than or equal to B in a real fire RGB image, wherein the region that R is more than or equal to G and more than or equal to B is a suspicious region I, and carrying out S4; if there is no region where R.gtoreq.G > B, then S1 is returned.
S4, executing S1 and S2 again, extracting the color enhanced image by using a visual saliency detection algorithm according to priori knowledge that R is more than or equal to G and more than or equal to B in the real fire RGB image after S2 is completed, obtaining a suspicious region II, and carrying out S5;
s5, comparing the suspicious region I with the suspicious region II, and judging fire if the suspicious region II overlaps with the suspicious region I and the suspicious region II is larger than the suspicious region I; otherwise, the suspicious region I is removed, and the suspicious region II is returned to S4 as the suspicious region I.
The beneficial effects of the invention are as follows:
the fire detection judging method based on visual saliency replaces manual observation, the fire detection judging sensitivity and accuracy are high through the method, the detection accuracy is improved through the S2, and the accuracy of fire detection judgment is improved through the comparison judgment of suspicious areas in the S3-S5.
Drawings
Fig. 1 is a flow chart of a fire detection judging method based on visual saliency.
Detailed Description
The invention will be described in further detail with reference to the drawings and examples.
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
A fire detection judging method based on visual saliency, as shown in figure 1, comprises the following steps:
s1, acquiring a high-resolution remote sensing image of a region to be detected, and performing S2. The high-resolution remote sensing image is a full-color image and a multispectral image.
S2, processing the high-resolution remote sensing image to obtain a color enhanced image. The specific process is as follows:
s2.1, correcting the full-color image and the multispectral image by adopting a collineation equation model to obtain a corrected full-color image and a corrected multispectral image.
S2.2, registering the corrected multispectral image by taking the corrected panchromatic image as a reference to obtain a registered image.
Graying the corrected multispectral image, and blocking the corrected full-color image and the grayed multispectral image; calculating information entropy of each image block, extracting SURF feature points of the image blocks meeting the information entropy threshold, performing feature point matching, and screening feature points after the matching is completed; and calculating an affine transformation matrix by using all the reserved matching feature points, and carrying out affine transformation on the multispectral image according to the affine transformation matrix, so as to finish registration and obtain a registered image.
S2.3, fusing the registered images by using an HIS method to obtain a fused image;
and converting the multispectral image from RGB to HIS by adopting an HIS method, carrying out histogram specification dividing on the registered panchromatic image according to the I component, replacing the original I component with the specified dividing result, and carrying out HIS conversion on the RGB space to obtain the final fusion image.
S2.4, performing color enhancement on the fusion image by adopting the McCann method to obtain a color enhanced image.
S3, extracting the obtained color enhanced image by using a visual saliency detection algorithm according to priori knowledge that R is more than or equal to G and more than or equal to B in a real fire RGB image, wherein the region that R is more than or equal to G and more than or equal to B is a suspicious region I, namely obtaining a suspicious region I, and carrying out S4; if there is no region where R.gtoreq.G > B, then S1 is returned.
S4, executing S1 and S2 again, extracting the color enhanced image by using a visual saliency detection algorithm according to priori knowledge that R is more than or equal to G and more than or equal to B in the real fire RGB image after S2 is completed, obtaining a suspicious region II, and carrying out S5;
s2 mentioned in S4 is completed without S3, but directly with S4 extraction of suspicious region two.
S5, comparing the suspicious region I with the suspicious region II, if the suspicious region II overlaps with the suspicious region I and the suspicious region II is larger than the suspicious region I, judging that the suspicious region I is fire, otherwise, removing the suspicious property of the suspicious region I, and returning the suspicious region II as the suspicious region I to S4.
And (2) executing the step (S1) twice before and after, wherein the overlapping rate of the high-resolution remote sensing image of the region to be detected is more than 40%.
The fire detection judging method based on visual saliency replaces manual observation. The fire disaster judgment by the detection and judgment method has high sensitivity and high accuracy. The invention ensures high geometric correction precision through a collinear equation model, fuses a large-scale search image through an HIS method, has high registration efficiency and high registration precision, improves detection precision through color enhancement, shortens calculation time through the McCann method when being applied to fire detection, and improves fire detection judgment accuracy through comparison judgment of suspicious areas.
Claims (5)
1. The fire detection judging method based on visual saliency is characterized by comprising the following steps:
s1, acquiring a high-resolution remote sensing image of a region to be detected;
s2, processing the high-resolution remote sensing image to obtain a color enhanced image;
the step S2 comprises the following steps:
s2.1, correcting the full-color image and the multispectral image by adopting a collineation equation model to obtain a corrected full-color image and a corrected multispectral image;
s2.2, registering the corrected multispectral image by taking the corrected panchromatic image as a reference to obtain a registered image;
s2.3, fusing the registered images by using an HIS method to obtain a fused image;
s2.4, performing color enhancement on the fusion image by adopting a McCann method to obtain a color enhanced image;
s3, extracting the color enhanced image obtained in the S2 by utilizing a visual saliency detection algorithm according to priori knowledge that R is more than or equal to G and more than or equal to B in a real fire RGB image, wherein the region with R is more than or equal to G and more than or equal to B is a suspicious region I, and carrying out S4; if no region with R more than or equal to G > B exists, returning to S1;
s4, executing S1 and S2 again, extracting the color enhanced image by using a visual saliency detection algorithm according to priori knowledge that R is more than or equal to G and more than or equal to B in the real fire RGB image after S2 is completed, obtaining a suspicious region II, and carrying out S5;
s5, comparing the suspicious region I with the suspicious region II, and judging fire if the suspicious region II overlaps with the suspicious region I and the suspicious region II is larger than the suspicious region I; otherwise, the suspicious region I is removed, and the suspicious region II is returned to S4 as the suspicious region I.
2. The fire detection and judgment method based on visual saliency according to claim 1, wherein the high-resolution remote sensing image is a full-color image or a multispectral image.
3. The fire detection and judgment method based on visual saliency as claimed in claim 1, wherein the specific process of S2.2 is as follows: graying the corrected multispectral image, and blocking the corrected full-color image and the grayed multispectral image; calculating information entropy of each image block, extracting SURF feature points of the image blocks meeting the information entropy threshold, performing feature point matching, and screening feature points after the matching is completed; and calculating an affine transformation matrix by using all the reserved matching feature points, and carrying out affine transformation on the multispectral image according to the affine transformation matrix, so as to finish registration and obtain a registered image.
4. The fire detection and judgment method based on visual saliency as claimed in claim 1, wherein the specific process of S2.3 is to use an HIS method to convert the multispectral image from RGB to HIS space, to perform histogram specification on the registered panchromatic image according to the I component, to replace the original I component with the specified result, and to perform HIS conversion to RGB space, thereby obtaining the final fusion image.
5. The fire detection and judgment method based on visual saliency according to claim 1, wherein the overlapping rate of the high-resolution remote sensing image is greater than 40% when S1 is executed twice before and after.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010319755.3A CN111597909B (en) | 2020-04-22 | 2020-04-22 | Fire detection judging method based on visual saliency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010319755.3A CN111597909B (en) | 2020-04-22 | 2020-04-22 | Fire detection judging method based on visual saliency |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111597909A CN111597909A (en) | 2020-08-28 |
| CN111597909B true CN111597909B (en) | 2023-05-02 |
Family
ID=72185180
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010319755.3A Active CN111597909B (en) | 2020-04-22 | 2020-04-22 | Fire detection judging method based on visual saliency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111597909B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853058A (en) * | 2019-11-12 | 2020-02-28 | 电子科技大学 | High-resolution remote sensing image road extraction method based on visual saliency detection |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7944547B2 (en) * | 2006-05-20 | 2011-05-17 | Zheng Wang | Method and system of generating 3D images with airborne oblique/vertical imagery, GPS/IMU data, and LIDAR elevation data |
| US9020256B2 (en) * | 2011-04-22 | 2015-04-28 | Exelis Inc. | System and method for combining color information with spatial information in multispectral images |
| AU2011253982B9 (en) * | 2011-12-12 | 2015-07-16 | Canon Kabushiki Kaisha | Method, system and apparatus for determining a subject and a distractor in an image |
| CN108765860A (en) * | 2018-05-02 | 2018-11-06 | 北京准视科技有限公司 | A kind of image-type fire detecting and alarm method |
| CN109118461A (en) * | 2018-07-06 | 2019-01-01 | 航天星图科技（北京）有限公司 | A kind of HIS fusion method based on Distributed Architecture |
| CN109766763A (en) * | 2018-12-15 | 2019-05-17 | 中国科学院长春光学精密机械与物理研究所 | A kind of forest fire detection method and system |
| CN110930439B (en) * | 2019-12-04 | 2022-11-29 | 长光卫星技术股份有限公司 | High-grade product automatic production system suitable for high-resolution remote sensing image |
| CN111046827A (en) * | 2019-12-20 | 2020-04-21 | 哈尔滨理工大学 | Video smoke detection method based on convolutional neural network |
-
2020
- 2020-04-22 CN CN202010319755.3A patent/CN111597909B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853058A (en) * | 2019-11-12 | 2020-02-28 | 电子科技大学 | High-resolution remote sensing image road extraction method based on visual saliency detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111597909A (en) | 2020-08-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109447068A (en) | A method of it separating seal from image and calibrates seal | |
| CN103824091B (en) | A kind of licence plate recognition method for intelligent transportation system | |
| CN110033431B (en) | Non-contact detection device and detection method for detecting corrosion area on surface of steel bridge | |
| CN111079545A (en) | Three-dimensional target detection method and system based on image restoration | |
| CN104506857A (en) | Camera position deviation detection method and device | |
| CN106056594A (en) | Double-spectrum-based visible light image extraction system and method | |
| WO2017221643A1 (en) | Image processing device, image processing system, image processing method, and program | |
| CN112419298B (en) | Bolt node plate rust detection method, device, equipment and storage medium | |
| CN104966054B (en) | Detection method of small target in unmanned plane visible images | |
| WO2017221644A1 (en) | Image processing device, image processing system, image processing method, and program | |
| CN111563896A (en) | Image processing method for catenary anomaly detection | |
| CN107507140B (en) | Method for suppressing vehicle shadow interference in open scene of highway based on feature fusion | |
| CN107862319A (en) | A kind of heterologous high score optical image matching error elimination method based on neighborhood ballot | |
| CN114943841A (en) | Method and device for assisting operation safety control based on image recognition | |
| CN104966283A (en) | Imaging layered registering method | |
| CN108803651B (en) | Unmanned aerial vehicle road inspection method based on image characteristics | |
| CN114359149A (en) | Dam bank dangerous case video detection method and system based on real-time image edge enhancement | |
| CN111597909B (en) | Fire detection judging method based on visual saliency | |
| CN112528994B (en) | Free angle license plate detection method, license plate recognition method and recognition system | |
| CN105205829A (en) | Transformer substation infrared image segmentation method based on improved two-dimensional Otsu algorithm | |
| CN108074264A (en) | A kind of classification multi-vision visual localization method, system and device | |
| CN107886530A (en) | A kind of improved image registration algorithm based on SIFT feature | |
| CN117197746A (en) | Safety monitoring system and method based on deep learning | |
| CN112200769B (en) | Fixed point monitoring new and old time phase image change detection method for illegal building detection | |
| CN110852203B (en) | Multi-factor suspicious person identification method based on video feature learning |
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 |