CN105678812A - Method for tracking ice floe locus - Google Patents
Method for tracking ice floe locus Download PDFInfo
- Publication number
- CN105678812A CN105678812A CN201610121999.4A CN201610121999A CN105678812A CN 105678812 A CN105678812 A CN 105678812A CN 201610121999 A CN201610121999 A CN 201610121999A CN 105678812 A CN105678812 A CN 105678812A
- Authority
- CN
- China
- Prior art keywords
- ice raft
- target
- field picture
- area
- target ice
- 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.)
- Pending
Links
Classifications
-
- 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
- G06T2207/10016—Video; Image sequence
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method for tracking an ice floe locus. The method comprises: a high-frequency camera is used for shooting an ice floe area to obtain a plurality of images; target ice floes are extracted and tracked; in two continuous frames of images, target ice floes of the two continuous frames of images are matched by using a particle tracking speed measuring matching algorithm so as to obtain corresponding target ice floes; areas of the corresponding target ice floes in the two continuous frames of images are calculated respectively; if an area of a target ice floe in a next frame of image is smaller than the area of the target ice floe in the current frame of image, the target ice floe of the current frame is broken and whether other ice floes are around the target ice floe is determined; the area of the current frame of target ice floe is calculated and is compared with a total area of ice floes in a connection area of a next frame after expansion; if the total area of ice floes in the connection area after expansion is larger than the area of the current frame of target ice floe, the expansion factor is reduced until the total area of the ice floes in the connection area after expansion is equal to or smaller than the area of target ice floe of the current frame; and then the locus of the target ice floe is drawn.
Description
Technical field
The present invention relates to a kind of method for following the tracks of ice raft track, utilized the method for image procossing to carry out broken ice raft tracking especially for a kind of, thus obtain the method for following the tracks of ice raft track.
Background technology
When following the tracks of ice raft motion track, it is necessary to according to the two continuous frames image of collected by camera, follow the tracks of the motion track of ice raft. But during for ice raft fragmentation, a target fissions into multiple target, target coupling mistake can be caused, thus target can trail-and-error. Domestic still not for the treatment process of this phenomenon at present.
In the process that big ice raft drifts about in following current, fixture on sea, lake or in river can be caused certain damage, but its shape of big ice raft is not fixed, can be broken gradually along with the process of drift, in order to reinforce the fixture on sea, lake or in river, it is desired to pay close attention to the track of ice raft in drift process.
Summary of the invention
The goal of the invention of the application is to provide a kind of method for following the tracks of ice raft track, and the method in ice raft moving process, can judge that whether ice raft is broken, and when fragmentation, determine the fritter ice raft after fragmentation, and follow the tracks of, finally obtain good ice raft motion track.
In order to complete the goal of the invention of the application, the application by the following technical solutions:
The method for following the tracks of ice raft track of the present invention, it comprises: following step:
(1) take ice raft region continuously with high frequency camera, obtain some images;
(2) process above-mentioned some images by background method of finite difference, extract the target ice raft to be followed the tracks of;
(3) in the image of n and n+1 two frame, utilize particle to follow the trail of the matching algorithm that tests the speed and the target ice raft in n and n+1 two two field picture is mated, obtain corresponding target ice raft;
(4) calculate target ice raft in n two field picture area and in n+1 two field picture the area of target ice raft, if the area of the target ice raft in n+1 two field picture is less than the area of target ice raft in n two field picture, then target ice raft is broken, enters step (5);If the area of the target ice raft in n+1 two field picture equals the area of target ice raft in n two field picture, then target ice raft does not have fragmentation, returns step (2), and at this moment, n=n+1, proceeds mating of n two field picture and the target ice raft in n+1 two field picture;
(5) judge in n+1 two field picture around target ice raft with or without other ice rafts, by the area expanded around target ice raft, other adjoining ice rafts and target ice raft are linked together, area and target ice raft area ratio around expansion target ice raft are the coefficient of expansion, calculate the total area of target ice raft and other ice rafts, if the above-mentioned total area in n+1 two field picture is equal to or less than the target ice raft area in n two field picture, then the above-mentioned total area thought in n+1 two field picture does not comprise in n two field picture the ice raft outside target ice raft; If the above-mentioned total area in n+1 two field picture is greater than the target ice raft area in n two field picture, then reduce the coefficient of expansion, other ice rafts closed on obtained like this become few, and the above-mentioned total area diminishes, until the target ice raft area that in n+1 two field picture, the above-mentioned total area is equal to or less than in n two field picture is for extremely;
(6) the target ice raft in step (5) and other ice rafts are regarded as target ice raft, repeating step (3) again, until the area of target ice raft is less than the threshold value of setting;
(7) the target ice raft obtained according to step (4) to (6), draws the track of target ice raft.
The method for following the tracks of ice raft track of the present invention, wherein: the span of the described coefficient of expansion is between 1-2;
The method for following the tracks of ice raft track of the present invention, wherein: in step (3), utilizing particle to follow the trail of the matching algorithm that tests the speed is taking the target ice raft center point coordinate in n two field picture as the center of circle, n+1 image is found the ice raft nearest with the target ice raft central point in n two field picture, if it is nearest with target ice raft in n two field picture, the ice raft then thought in n+1 two field picture mates with the target ice raft in n two field picture, otherwise for not mate;
The method for following the tracks of ice raft track of the present invention, wherein: described background method of finite difference has the image of obvious aberration to make background by the color with target ice raft, extracts the method for target ice raft;
The method for following the tracks of ice raft track of the present invention, wherein: the threshold value of the setting in described step (6) is 10 × 10 pixels.
The invention solves drift about at sea, problem that ice raft on lake or in river does not have track data, can judge that whether ice raft is broken in this scenario, if fragmentation, determine broken fritter ice raft, and follow the tracks of broken fritter ice raft, for the hydraulic structure damage intensity research on sea, lake or in river is provided foundation by ice raft.
Accompanying drawing explanation
Fig. 1 is in n frame, wants tracking target ice raft schematic diagram with the shooting of high frequency camera;
Fig. 2 is in n frame, with the schematic diagram of Fig. 1 of background method of finite difference process;
Fig. 3 is in n+1 frame, wants tracking target ice raft schematic diagram with the shooting of high frequency camera;
Fig. 4 is in n+1 frame, with the schematic diagram of Fig. 3 of background method of finite difference process;
Fig. 5 is the schematic diagram of the shape after target ice raft 1 and other ice rafts 2 expand;
Fig. 6 is the ice raft track schematic diagram of the present invention, in order to conveniently, this figure only depicts part ice raft track.
In Fig. 1 to Fig. 6, label 1 is target ice raft; Label 2 is other ice rafts, and label 11 is the shape after target ice raft 1 expands; Label 22 is the shape after other ice rafts 2 expand.
Embodiment
As shown in Fig. 1 to figure, the present invention only for n frame and n+1 two field picture, is described in detail, wherein n be more than or equal to 1 natural number.
The method for following the tracks of ice raft track of the present invention, it comprises: following step:
(1) as shown in figures 1 and 3, take continuously with high frequency camera and want tracking target ice raft 1, obtain some images, such as: obtain 2 shown in Fig. 1 and Fig. 3 image;
(2) process above-mentioned 2 images by background method of finite difference, extract the target ice raft 1 to be followed the tracks of, as shown in Figure 2 and Figure 4, extract n frame and n+1 frame target ice raft 1;
(3) in the image of n and n+1 two frame shown in Fig. 2 and Fig. 4, follow the trail of, according to particle, the matching algorithm that tests the speed to be mated by the target ice raft 1 in n and n+1 two two field picture, it is taking target ice raft 1 center point coordinate in n two field picture as the center of circle, n+1 image is found the ice raft nearest with target ice raft 1 central point in n two field picture, if it is nearest with target ice raft 1 in n two field picture, the ice raft then thought in n+1 two field picture mates with the target ice raft in n two field picture, otherwise for not mate;
(4) calculate target ice raft 1 in n two field picture area and in n+1 two field picture the area of target ice raft 1, if the area of the target ice raft in n+1 two field picture is less than the area of target ice raft in n two field picture, as shown in Figure 3, then target ice raft is broken, enters step (5); If the area of the target ice raft in n+1 two field picture equals the area of target ice raft in n two field picture, then target ice raft does not have fragmentation, returns step (2), and at this moment, n=n+1, proceeds mating of n two field picture and the target ice raft in n+1 two field picture;
(5) as shown in Figure 5, by the area expanded around target ice raft, other adjoining ice rafts and target ice raft are linked together, area and target ice raft area ratio around expansion target ice raft are the coefficient of expansion, calculate the total area of target ice raft and other ice rafts, solid line 1 in figure, 2 is target ice raft 1 and other ice rafts 2 of the broken rear formation of n+1 frame target ice raft 1, dotted line 11, 22 is the shape after target ice raft 1 and other ice rafts 2 expand, 11 can be seen, 22 form connection region, calculate the total area being connected in region ice raft after expanding, if the above-mentioned total area in n+1 two field picture is equal to or less than target ice raft 1 area in n two field picture, then think not comprise the ice raft outside target ice raft 1 in n two field picture, if the target ice raft total area and target ice raft 1 area being greater than in n two field picture in n+1 two field picture, then reduce the coefficient of expansion, other ice rafts closed on obtained like this become few, the total area diminishes, until the target ice raft total area and target ice raft 1 area being equal to or less than in n two field picture are for extremely in n+1 two field picture, namely other ice rafts 2 are determined, also likely because target ice raft 1 naturally melts or melts after fragmentation or melt, so the total area of target ice raft 1 can diminish, and can not find other ice rafts 2,
(6) the target ice raft 1 in step (5) and other ice rafts being regarded as target ice raft again, repeating step (3), until the area of target ice raft 1 is less than the threshold value of setting, the threshold value of setting is 10 × 10 pixels;
(7) the target ice raft obtained according to step (4) to (6), as shown in Figure 6, draws the track of target ice raft 1.
Above-mentioned enforcement mode illustrates principle and effect thereof of the present invention, but not for limiting the present invention. Any those skilled in the art all can under the spirit not running counter to the present invention and category, it is possible to above-mentioned enforcement mode is carried out any modification and change.Therefore, the scope of the present invention, should as listed by claim book.
Claims (6)
1. for following the tracks of a method for ice raft track, it comprises: following step:
(1) clap ice raft region continuously with high frequency camera, obtain some images;
(2) process above-mentioned some images by background method of finite difference, extract the target ice raft to be followed the tracks of;
(3) in the image of n and n+1 two frame, utilize particle to follow the trail of the matching algorithm that tests the speed and the target ice raft in n and n+1 two two field picture is mated, obtain corresponding target ice raft; It is characterized in that:
(4) calculate target ice raft in n two field picture area and in n+1 two field picture the area of target ice raft, if the area of target ice raft is less than the area of target ice raft in n two field picture in n+1 two field picture, then target ice raft is broken, enters step (5); If the area of the target ice raft in n+1 two field picture equals the area of target ice raft in n two field picture, then target ice raft does not have fragmentation, returns step (2), at this moment, n will be newly defined as n+1, proceed mating of n two field picture and the target ice raft in n+1 two field picture;
(5) judge in n+1 two field picture around target ice raft with or without other ice rafts, by the area expanded around target ice raft, other adjoining ice rafts and target ice raft are linked together, area and target ice raft area ratio around expansion target ice raft are the coefficient of expansion, calculate the total area of target ice raft and other ice rafts, if the above-mentioned total area in n+1 two field picture is equal to or less than the target ice raft area in n two field picture, then the above-mentioned total area thought in n+1 two field picture does not comprise in n two field picture the ice raft outside target ice raft; If the above-mentioned total area in n+1 two field picture is greater than the target ice raft area in n two field picture, then reduce the coefficient of expansion, other ice rafts closed on obtained like this become few, and the above-mentioned total area diminishes, until the target ice raft area that in n+1 two field picture, the above-mentioned total area is equal to or less than in n two field picture;
(6) the target ice raft in step (5) and other ice rafts are regarded as target ice raft, repeating step (3) again, until the area of target ice raft is less than the threshold value of setting;
(7) the target ice raft obtained according to step (4) to (6), draws the track of target ice raft.
2. as claimed in claim 1 for following the tracks of the method for ice raft track, it is characterised in that: the span of the described coefficient of expansion is between 1-2.
3. as claimed in claim 1 or 2 for following the tracks of the method for ice raft track, it is characterized in that: in step (3), utilizing particle to follow the trail of the matching algorithm that tests the speed is taking the target ice raft center point coordinate in n two field picture as the center of circle, n+1 image is found the ice raft nearest with the target ice raft central point in n two field picture, if it is nearest with target ice raft in n two field picture, the ice raft then thought in n+1 two field picture mates with the target ice raft in n two field picture, otherwise for not mate.
4. the method for following the tracks of ice raft track as described in claim 1 or 2 or 3, it is characterised in that: described n be more than or equal to 1 natural number.
5. as claimed in claim 4 for following the tracks of the method for ice raft track, it is characterised in that: described background method of finite difference has the image of obvious aberration to make background by the color with target ice raft, extracts the method for target ice raft.
6. as claimed in claim 5 for following the tracks of the method for ice raft track, it is characterised in that: the threshold value of the setting in described step (6) is 10 × 10 pixels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610121999.4A CN105678812A (en) | 2016-03-04 | 2016-03-04 | Method for tracking ice floe locus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610121999.4A CN105678812A (en) | 2016-03-04 | 2016-03-04 | Method for tracking ice floe locus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105678812A true CN105678812A (en) | 2016-06-15 |
Family
ID=56307908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610121999.4A Pending CN105678812A (en) | 2016-03-04 | 2016-03-04 | Method for tracking ice floe locus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105678812A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018152106A (en) * | 2018-05-02 | 2018-09-27 | 株式会社日立国際電気 | System for detecting invasion on water and method for the same |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542551A (en) * | 2010-12-13 | 2012-07-04 | 北京师范大学 | Automatic change detection technology for floating ice at edges of polar ice sheets |
| CN102567726A (en) * | 2010-12-13 | 2012-07-11 | 北京师范大学 | Technology for automatically extracting floating ice in polar ice sheet edge regions |
| CN102708369A (en) * | 2012-05-14 | 2012-10-03 | 大连理工大学 | Sea ice parameter extraction method on basis of satellite image |
-
2016
- 2016-03-04 CN CN201610121999.4A patent/CN105678812A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542551A (en) * | 2010-12-13 | 2012-07-04 | 北京师范大学 | Automatic change detection technology for floating ice at edges of polar ice sheets |
| CN102567726A (en) * | 2010-12-13 | 2012-07-11 | 北京师范大学 | Technology for automatically extracting floating ice in polar ice sheet edge regions |
| CN102708369A (en) * | 2012-05-14 | 2012-10-03 | 大连理工大学 | Sea ice parameter extraction method on basis of satellite image |
Non-Patent Citations (3)
| Title |
|---|
| ZHEN LIU 等: "Automated change detection of multi-level icebergs near Mertz Glacier region using feature vector matching", 《PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, COMPUTER VISION, AND PATTERN RECOGNITION》 * |
| 韩雷 等: "基于图像处理技术的浮冰破碎跟踪试验研究", 《中国水利》 * |
| 韩雷 等: "基于多彩色CCD图像处理的浮冰测量研究", 《黑龙江大学工程学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018152106A (en) * | 2018-05-02 | 2018-09-27 | 株式会社日立国際電気 | System for detecting invasion on water and method for the same |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105809712A (en) | Effective estimation method for large displacement optical flows | |
| JP6000699B2 (en) | Cell division process tracking device and cell division process tracking program | |
| CN105872345A (en) | Full-frame electronic image stabilization method based on feature matching | |
| CN104268899A (en) | Moving object detection method based on frame difference and background difference | |
| WO2007050707A3 (en) | Video foreground segmentation method | |
| CN101894380B (en) | Method for tracing target object in panoramic video automatically | |
| CN103440484A (en) | Flame detection method adaptive to large outdoor space | |
| CN106846288B (en) | A kind of more algorithm fusion methods of bimodal infrared image difference characteristic Index | |
| CN103745449A (en) | Rapid and automatic mosaic technology of aerial video in search and tracking system | |
| Chen et al. | Temporally consistent multi-class video-object segmentation with the video graph-shifts algorithm | |
| CN102831617A (en) | Method and system for detecting and tracking moving object | |
| Zhou et al. | A UAV patrol system using panoramic stitching and object detection | |
| CN101739692A (en) | Fast correlation tracking method for real-time video target | |
| CN105678812A (en) | Method for tracking ice floe locus | |
| CN114266299A (en) | Method and system for detecting defects of steel structure of railway bridge based on unmanned aerial vehicle operation | |
| CN103500451B (en) | A kind of independent floating ice extracting method for satellite data | |
| Wu et al. | A real-time robust approach for tracking uavs in infrared videos | |
| Huang et al. | Learning to detect fine-grained change under variant imaging conditions | |
| CN103530635A (en) | Coastline extracting method based on satellite microwave remote sensing image | |
| CN112598739B (en) | Mobile robot infrared target tracking method, system and storage medium based on space-time characteristic aggregation network | |
| CN101272450B (en) | Global motion estimation exterior point removing and kinematic parameter thinning method in Sprite code | |
| CN104376538A (en) | Image sparse denoising method | |
| CN102254186B (en) | Method for detecting infrared target by using local difference of structure similarity | |
| Nguyen et al. | A multi-modal particle filter based motorcycle tracking system | |
| Gao et al. | Monocular 3D Vehicle Detection on Roadside based on Foreground Depth Fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160615 |