CN108363958B - Oil tank detection method based on high-resolution optical remote sensing image - Google Patents
Oil tank detection method based on high-resolution optical remote sensing image Download PDFInfo
- Publication number
- CN108363958B CN108363958B CN201810054662.5A CN201810054662A CN108363958B CN 108363958 B CN108363958 B CN 108363958B CN 201810054662 A CN201810054662 A CN 201810054662A CN 108363958 B CN108363958 B CN 108363958B
- Authority
- CN
- China
- Prior art keywords
- oil tank
- area
- image
- remote sensing
- detected
- 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 31
- 230000003287 optical effect Effects 0.000 title claims abstract description 11
- 238000012706 support-vector machine Methods 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000004891 communication Methods 0.000 claims abstract description 8
- 238000013507 mapping Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 4
- 230000008901 benefit Effects 0.000 abstract description 3
- 230000011218 segmentation Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003703 image analysis method Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000000007 visual effect 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
- G06V20/176—Urban or other man-made structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06V10/20—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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Geophysics And Detection Of Objects (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an oil tank detection technology based on a high-resolution optical remote sensing image, and belongs to the field of remote sensing image target detection. Firstly, selecting a to-be-detected area of an oil tank, binarizing an image to be detected through a support vector machine model, and obtaining a suspected area of the oil tank; and then further judging whether the suspected area of the oil tank is the oil tank, if the length-width ratio of the area is close to 1 and other white pixel communication areas exist in a certain range around the area, judging the area of the oil tank, and if not, judging the area of the oil tank is a non-oil tank area. And finally mapping to the geographical position of the oil tank by the image position. The method disclosed by the invention has the advantages that manual intervention is not needed, the oil tank can be automatically detected, the oil tank can be accurately extracted, the detection precision is greatly improved, and the application requirements are met; the method is efficient, rapid and high in accuracy, and mass remote sensing image data can be detected in a short time.
Description
Technical Field
The invention belongs to the field of remote sensing image target detection, and particularly relates to an oil tank detection technology based on a high-resolution optical remote sensing image.
Background
At present, with the development of remote sensing technology, the resolution of optical remote sensing images is higher and higher, and a more efficient processing method is needed in the face of massive high-resolution remote sensing image data. The oil tank is used as an important strategic material storage means, and the oil tank detection based on the remote sensing image has important significance.
The existing general ship detection method based on the high-resolution optical remote sensing image mostly adopts two modes: 1) based on a visual interpretation method, the position of the oil tank is manually selected in the remote sensing image, so that the accurate position of the oil tank can be obtained; 2) based on a threshold segmentation method, each pixel point of an image is generally distinguished according to a certain gray threshold, and then the position of the oil tank is obtained through some simple image analysis.
Although the first method has high detection precision, the detection efficiency is low, the method depends on manpower, is difficult to apply on a large scale, is difficult to process massive remote sensing data, and is difficult to popularize and apply on a large scale because the input manpower cost is far higher than the benefit of oil tank detection.
The second method can realize automatic detection and identification of the oil tank, but the detection effect is poor, and satisfactory precision is difficult to achieve; moreover, the angle, illumination and radiation of different remote sensing images during imaging all change, so that the applicable segmentation threshold values are different, and thus, it is difficult to select a uniform segmentation threshold value for oil tank detection. If different thresholds are chosen for each sheet, it is still difficult to perform automated tank detection and identification on a large scale.
Disclosure of Invention
In order to solve the problems, the invention provides an oil tank detection method based on a high-resolution optical remote sensing image, and the purpose of automatic detection of an oil tank is achieved through a mode recognition and image analysis technology.
The invention discloses an oil tank detection method based on high-resolution optical remote sensing images, which comprises the following steps:
selecting an area to be detected of an oil tank;
and after a river course shoreline and a port area are detected in the remote sensing image, an image of 1000 meters near the river course and the port area is intercepted and used as an interest area to be detected of the oil tank.
Carrying out binarization on an image to be detected and obtaining a suspected area of the oil tank;
and training a support vector machine model which divides the oil tank area and the non-oil tank area, putting the image to be detected into the trained vector machine model pixel by pixel to judge whether the image is the oil tank area or the non-oil tank area, and giving the gray value of the oil tank area to white and the gray value of the non-oil tank area to black so as to obtain a binary image only containing the oil tank area and the non-oil tank area. And analyzing the communicated region of the binary image, wherein each independent white pixel communicated region is a suspected oil tank region.
Step three, further judging whether the suspected area of the oil tank is the oil tank according to the following judgment criterion;
the judgment criterion is as follows: and for the suspected area of the oil tank, if the aspect ratio of the area is close to 1 and other white pixel communication areas exist in a certain range around the area, judging the area as the area of the oil tank, otherwise, judging the area as the area of the non-oil tank.
Mapping the image position to a geographic position;
after the pixel coordinates of the oil tanks in the image are obtained, the geographic position of each oil tank is positioned according to the mapping relation between the pixel coordinates and the geographic coordinates in the remote sensing image, and the automatic detection of the oil tanks is completed.
The invention innovatively provides a coarse-to-fine multilevel cascade oil tank detection method, which comprises the following steps of firstly, carrying out pixel-level oil tank binary classification by using a support vector machine model in the second step, and extracting a suspected oil tank area; and thirdly, accurately extracting the oil tank by using an image morphology method on the basis of initial extraction, and having high efficiency, rapidness and high accuracy. Compared with the prior art, the method of the invention has the following advantages and beneficial effects:
1) the oil tank can be automatically detected without manual intervention;
2) by using a support vector machine and other pattern recognition methods and communication area analysis and other image analysis methods, the detection precision is greatly improved, and the application requirements are met;
3) the method is efficient and rapid, and can detect mass remote sensing image data in a short time.
Drawings
Fig. 1 is a flow chart of oil tank detection based on high-resolution optical remote sensing images.
Detailed Description
The technical solution of the present invention is further explained with reference to the accompanying drawings and the detailed description.
The oil tank detection method based on the high-resolution optical remote sensing image of the invention is shown in fig. 1, and the steps of the oil tank detection of the invention are described in detail by taking a remote sensing image as an example.
Step one, selecting an interest area to be detected of the oil tank from the remote sensing image.
A large number of oil tanks are gathered near a river channel and a port, and for detecting the oil tanks in remote sensing images, the river channel and the port area need to be detected firstly, and after the detection of the river channel and the port is finished, all the oil tanks can be contained by extending about 1000 meters in the vertical direction of the river channel, so that the remote sensing images which are 1000 meters near the river channel and the port area are intercepted as interest areas for detecting the oil tanks.
In this embodiment, the region to be detected of the oil tank in the remote sensing image is intercepted according to the information of the shore line and the port, as shown in (1) in fig. 1.
And secondly, carrying out image binary segmentation on the image to be detected by using a Support Vector Machine (SVM), and carrying out connected region analysis on the binary image to obtain a suspected region of the oil tank.
In the step, a trained SVM model is used for carrying out binary segmentation on the image of the area to be detected to obtain a suspected oil tank area and a non-oil tank area, and the specific process is as follows:
201) firstly, training an SVM model capable of separating an oil tank region from a non-oil tank region: selecting representative remote sensing images, respectively selecting an oil tank part and other parts from the representative remote sensing images, recording the gray values of the oil tank part and the other parts, and then training by utilizing the gray value samples of the two parts to obtain the SVM model. The output result of the SVM model is a tank area or a non-tank area.
202) And after the trained SVM model is obtained, putting the image to be detected into the trained SVM model pixel by pixel to judge whether the image is an oil tank area or a non-oil tank area, and assigning the gray value of the oil tank area as white and the gray value of the non-oil tank area as black, so that a binary image only containing the oil tank area and the non-oil tank area can be obtained.
203) And (3) analyzing the connected regions of the binary images to obtain regions with the same gray values gathered together, namely obtaining all regions judged as connected oil tank regions, wherein each independent connected region may be an oil tank as shown in (2) in fig. 1.
And step three, establishing a judgment criterion, and judging whether the communicated area is an oil tank.
Since the oil tanks are generally circular and regularly gathered together, the length-width ratio of the communication areas formed by the oil tanks should be close to 1, and the independent parts without other communication areas around are unlikely to be the oil tanks, so that the white pixel communication areas with the length-width ratio close to 1 and other white pixel communication areas around are judged to be the oil tanks in the invention.
And (3) analyzing the communicated areas of the binary images to obtain mutually independent communicated areas, and judging whether the communicated areas are actually oil tanks according to the aspect ratio of the communicated areas and whether the communicated areas are close to each other to obtain oil tank detection results, as shown in fig. 1.
And step four, the image coordinates of the oil tank correspond to the geographical coordinates of the oil tank.
And (4) storing the mapping relation between the pixel coordinates and the geographic coordinates in the remote sensing image, and mapping the pixel coordinates to the geographic coordinates from the oil tank detection result obtained in the step (3), as shown in (4) in fig. 1, thereby completing the automatic detection of the oil tank.
Claims (2)
1. An oil tank detection method based on high-resolution optical remote sensing images is characterized by comprising the following steps:
step 1, selecting an area to be detected of an oil tank;
after a river course shoreline and a port area are detected in the remote sensing image, an image of 1000 meters near the river course and the port area is intercepted and used as an interest area to be detected of the oil tank;
step 2, carrying out binarization on an image to be detected and obtaining a suspected area of the oil tank;
training a support vector machine model which divides an oil tank area and a non-oil tank area, putting an image to be detected into the trained vector machine model pixel by pixel to judge whether the image is the oil tank area or the non-oil tank area, and giving the gray value of the oil tank area to white and the gray value of the non-oil tank area to black to obtain a binary image only containing the oil tank area and the non-oil tank area; analyzing communicated areas of the binary image, wherein each independent white pixel communicated area is an oil tank suspected area;
step 3, further judging whether the suspected area of the oil tank is the oil tank according to the following judgment criteria;
the judgment criterion is as follows: for the suspected area of the oil tank, if the length-width ratio of the area is close to 1 and other white pixel communication areas exist in a certain range around the area, the area is judged to be the area of the oil tank, otherwise, the area is judged to be the area of a non-oil tank;
step 4, mapping the image position to a geographic position;
after the pixel coordinates of the oil tanks in the image are obtained, the geographic position of each oil tank is positioned according to the mapping relation between the pixel coordinates and the geographic coordinates in the remote sensing image, and the automatic detection of the oil tanks is completed.
2. The method according to claim 1, wherein in the step 2, when the support vector machine model is trained, the oil tank part and the other parts are respectively checked out from the remote sensing image, the gray values of the two parts are recorded, then the support vector machine model is trained by using the gray value samples of the two parts, and the output result of the model is the oil tank area or the non-oil tank area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810054662.5A CN108363958B (en) | 2018-01-19 | 2018-01-19 | Oil tank detection method based on high-resolution optical remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810054662.5A CN108363958B (en) | 2018-01-19 | 2018-01-19 | Oil tank detection method based on high-resolution optical remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108363958A CN108363958A (en) | 2018-08-03 |
| CN108363958B true CN108363958B (en) | 2021-10-01 |
Family
ID=63006421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810054662.5A Active CN108363958B (en) | 2018-01-19 | 2018-01-19 | Oil tank detection method based on high-resolution optical remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108363958B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110210453B (en) * | 2019-06-14 | 2021-06-29 | 中国资源卫星应用中心 | Remote sensing image feature-based oil tank storage capacity determination method and system |
| CN113808134B (en) * | 2021-11-19 | 2022-05-20 | 中科星睿科技(北京)有限公司 | Oil tank layout information generation method, oil tank layout information generation device, electronic apparatus, and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217196A (en) * | 2014-08-26 | 2014-12-17 | 武汉大学 | A method for detecting automatically a circular oil tank with a remote sensing image |
| CN105488481A (en) * | 2015-12-04 | 2016-04-13 | 清华大学 | Detection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9195876B2 (en) * | 2013-06-05 | 2015-11-24 | Digitalglobe Inc. | Oil tank farm storage monitoring |
| US9922265B2 (en) * | 2014-07-25 | 2018-03-20 | Digitalglobe, Inc. | Global-scale object detection using satellite imagery |
-
2018
- 2018-01-19 CN CN201810054662.5A patent/CN108363958B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217196A (en) * | 2014-08-26 | 2014-12-17 | 武汉大学 | A method for detecting automatically a circular oil tank with a remote sensing image |
| CN105488481A (en) * | 2015-12-04 | 2016-04-13 | 清华大学 | Detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108363958A (en) | 2018-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109145830B (en) | Intelligent water gauge identification method | |
| CN101576956B (en) | On-line character detection method based on machine vision and system thereof | |
| CN112149667B (en) | Automatic reading method of pointer instrument based on deep learning | |
| CN110334760B (en) | Optical component damage detection method and system based on RESUnet | |
| CN108921099A (en) | Moving ship object detection method in a kind of navigation channel based on deep learning | |
| CN111126393A (en) | Vehicle appearance refitting judgment method and device, computer equipment and storage medium | |
| CN108363958B (en) | Oil tank detection method based on high-resolution optical remote sensing image | |
| CN110866915A (en) | Circular inkstone quality detection method based on metric learning | |
| CN115184380A (en) | Printed circuit board welding spot abnormity detection method based on machine vision | |
| CN112734729A (en) | Water gauge water level line image detection method and device suitable for night light supplement condition and storage medium | |
| CN109670498A (en) | A kind of license plate locating method | |
| Tian et al. | Corrosion identification of fittings based on computer vision | |
| CN110084587B (en) | Automatic dinner plate settlement method based on edge context | |
| CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
| CN105740827B (en) | A kind of stop line detection and distance measuring method based on Fast Labeling connection | |
| CN105740831B (en) | A kind of stopping line detecting method applied to intelligent driving | |
| CN116681677A (en) | Lithium battery defect detection method, device and system | |
| CN116381053A (en) | Ultrasonic detection method and system for welding metal materials | |
| CN115496716A (en) | Single and double micronucleus cell image detection method based on deep learning and related equipment | |
| CN111583341B (en) | Cloud deck camera shift detection method | |
| CN105930813B (en) | A method of detection composes a piece of writing this under any natural scene | |
| CN112581424B (en) | Classification extraction method for surface and subsurface defects of optical element | |
| CN115205193A (en) | Steel plate surface defect detection method based on microdefect YOLO network | |
| CN114882375A (en) | Intelligent identification method and device for tailing pond | |
| CN112990064A (en) | Dial pointer identification method based on color segmentation and probability model |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211207 Address after: Floor 9, block a, Jiahao International Center, 116 Zizhuyuan Road, Haidian District, Beijing 100097 Patentee after: CHINA HIGHWAY ENGINEERING CONSULTING Corp. Patentee after: Zhongzi Data Co., Ltd Address before: 100089 courtyard 17, Changyun palace, West Third Ring Road, Haidian District, Beijing Patentee before: CHINA HIGHWAY ENGINEERING CONSULTING Corp. |