CN109086701B - Automatic identification method for luminous fishing boat for luminous remote sensing data - Google Patents
Automatic identification method for luminous fishing boat for luminous remote sensing data Download PDFInfo
- Publication number
- CN109086701B CN109086701B CN201810808844.7A CN201810808844A CN109086701B CN 109086701 B CN109086701 B CN 109086701B CN 201810808844 A CN201810808844 A CN 201810808844A CN 109086701 B CN109086701 B CN 109086701B
- Authority
- CN
- China
- Prior art keywords
- image
- value
- remote sensing
- maximum value
- fishing boat
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000001914 filtration Methods 0.000 claims abstract description 17
- 238000005260 corrosion Methods 0.000 claims abstract description 11
- 230000007797 corrosion Effects 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 7
- 238000012937 correction Methods 0.000 claims description 4
- 238000003711 image thresholding Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 230000007935 neutral effect Effects 0.000 abstract 1
- 238000001514 detection method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 241000238366 Cephalopoda Species 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation 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
- 230000000717 retained effect Effects 0.000 description 1
- 230000011218 segmentation Effects 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to an automatic identification method of a lamplight fishing boat for NPP (neutral Point protocol) luminous remote sensing data, which comprises the following steps of: screening luminous remote sensing data; preprocessing the screened noctilucent remote sensing data, and cutting out a subregion image of a sea area near the lamplight fishing boat; calculating a threshold value of the subarea image by adopting a central maximum value method, segmenting the image according to the calculated threshold value, reserving points larger than the threshold value, and determining that the calculated target is a suspected target image; carrying out connected region marking on the suspected target image, counting the maximum value of each region and the number of pixels of each region, counting target regions smaller than the corrosion operator, and recording the maximum value points of the target regions; carrying out corrosion operation on the image; carrying out maximum value filtering on the image; and subtracting the suspected target image from the maximum value filtered image, and selecting a zero value point as a target point for output. The invention can improve the automation level of service monitoring of the light fishing boat.
Description
Technical Field
The invention relates to the technical field of fishing boat service monitoring, in particular to a light fishing boat automatic identification method for noctilucent remote sensing data.
Background
The operating fishery of the pelagic fishery comprises sea areas such as the North Pacific, the south east Pacific, the south West Atlantic and the like, mainly catches the fishes and the squids at the middle and upper layers, and generally catches the fishes and the squids by adopting a light fishing boat for trapping at night. How to quickly and automatically obtain the information such as the number, nationality, light types and the like of fishing boats in a certain fishing area has application requirements on management and evaluation of the number of fishing boats in each fishing area, fishing effort force, calculation of fishing cost and income and the like by departments such as fishery management and the like.
In the traditional light fishing boat detection of the noctilucent remote sensing data, after noise is removed by adopting methods such as median filtering and the like, a threshold value is manually selected to identify the light fishing boat according to the brightness difference between the image light fishing boat and the background. The method has the problems that the luminous remote sensing data volume is large, one scene of image is about 1G, the remote sensing images in different time phases are greatly interfered by noise such as cloud and fog, the efficiency is low if all the remote sensing images adopt a manual and manual threshold selection rule, and the reliability of threshold selection is influenced by the subjective experience of different mappers. The existing light fishing boat detection method based on the noctilucent remote sensing data cannot meet the application requirements of the ocean fishery on automatic monitoring of the light fishing boat.
Disclosure of Invention
The invention aims to solve the technical problem of providing an automatic identification method of a light fishing boat for noctilucent remote sensing data, and improving the automation level of service monitoring of the light fishing boat.
The technical scheme adopted by the invention for solving the technical problems is as follows: the automatic identification method of the lamplight fishing boat for the noctilucent remote sensing data comprises the following steps:
(1) screening luminous remote sensing data;
(2) preprocessing the screened noctilucent remote sensing data, and cutting out a subregion image of a sea area near the lamplight fishing boat;
(3) calculating a threshold value of the subarea image by adopting a central maximum value method, segmenting the image according to the calculated threshold value, reserving points larger than the threshold value, and determining that the calculated target is a suspected target image;
(4) carrying out connected region marking on the suspected target image, counting the maximum value of each region and the number of pixels of each region, counting target regions smaller than the corrosion operator, and recording the maximum value points of the target regions;
(5) carrying out corrosion operation on the image;
(6) carrying out maximum value filtering on the image;
(7) and subtracting the suspected target image from the maximum value filtered image, and selecting a zero value point as a target point for output.
The conditions for screening the noctilucent remote sensing data in the step (1) are as follows: the screened area range is required to contain the movable sea area of the lamplight fishing boat, the time is required to have the working time of the lamplight fishing boat, and the cloud cover in the movable sea area of the lamplight fishing boat is required to be less than 10%.
The preprocessing in the step (2) comprises remote sensing image projection correction and noise removal; wherein, the remote sensing image projection correction adopts Lambert projection; the noise removal is realized by adopting median filtering and wiener filtering.
The step (3) includes the substeps of:
(31) solving the gradients of the horizontal direction and the vertical direction of the sub-area image;
(32) calculating the gradient mean value in the transverse direction and the longitudinal direction, and then counting the position information of the gradient image points larger than the mean value;
(33) calculating DN values of the preprocessed image points corresponding to the points with the gradient larger than the gradient mean value, searching the maximum value and the minimum value in the DN values, and taking the average value of the maximum value and the minimum value;
(34) comparing the threshold values in the two directions, and taking the minimum value; if the minimum value is greater than the DN value corresponding to 98% of the positions, the minimum value is reserved; and if the minimum value is less than the DN value corresponding to 98%, taking the DN value corresponding to 99% as a threshold value. Advantageous effects
Due to the adoption of the technical scheme, compared with the prior art, the invention has the following advantages and positive effects: the invention adopts a self-adaptive extreme value threshold value selection method and central extreme value filtering, can automatically identify the luminous remote sensing image lamplight fishing boat, and obviously improves the identification precision. The method is high in applicability, not only suitable for the noctilucent remote sensing image, but also suitable for other types of images, and has practical value. According to the method, the gradient of two directions of an image and the characteristics of a target image are combined, firstly, a suspected target area is calculated in an original image, and then, central maximum value filtering is carried out in the suspected target area. This reduces false detections caused by detection of fishing vessel points by other noise to some extent.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is an NPP primitive diagram;
FIG. 3 is a diagram of a preprocessed sub-graph;
FIG. 4 is a suspected target map;
fig. 5 is a fishing boat information recognition diagram.
Detailed Description
The invention will be further illustrated with reference to the following specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.
The embodiment of the invention relates to an automatic identification method of a light fishing boat for noctilucent remote sensing data, which comprises the following steps as shown in figure 1: screening luminous remote sensing data; preprocessing the screened noctilucent remote sensing data, and cutting out a subregion image of a sea area near the lamplight fishing boat; calculating a threshold value of the subarea image by adopting a central maximum value method, segmenting the image according to the calculated threshold value, reserving points larger than the threshold value, and determining that the calculated target is a suspected target image; carrying out connected region marking on the suspected target image, counting the maximum value of each region and the number of pixels of each region, counting target regions smaller than the corrosion operator, and recording the maximum value points of the target regions; carrying out corrosion operation on the image; carrying out maximum value filtering on the image; and subtracting the suspected target image from the maximum value filtered image, and selecting a zero value point as a target point for output.
Therefore, the method preprocesses the noctilucent remote sensing image, then obtains the gradient mean value based on the gradient image, calculates the maximum value and the minimum value of the gradient image which are larger than the gradient mean value, takes the DN value of more than 99 percent of the positions as the threshold value of the light fishing boat, compares the threshold values in the two directions, and reserves the smaller value as the final threshold value. The method combines the gradients in two directions of the image and the characteristics of the target image, firstly calculates the area of a suspected target in the original image, and then carries out central maximum value filtering in the area of the suspected target, thereby reducing false detection caused by the detection of other noises on the fishing boat point to a certain extent.
The invention is further illustrated by the following specific example.
The method comprises the steps of firstly, screening noctilucent remote sensing data, wherein the area range is required to contain the movable sea area of the lamplight fishing boat, the time is required to be the time when the lamplight fishing boat is operating, and the cloud cover in the movable sea area of the lamplight fishing boat is required to be less than 10%. The screened noctilucent remote sensing data of the embodiment are as follows: image phase 2015 3, 1, 21 of UTC time: 31: 12.8, image size 4064X 3072, study area located in the Indian ocean area at 56 ° E-60 ° E, 12 ° N-16 ° N.
The original NPP data is data that is not projected, and the image needs to be projected after the original image (see fig. 2) is obtained. Firstly, an original image is projected into an equal longitude and latitude image by using a throwing point pursuit method. Equal longitude and latitude projection is not satisfactory in order to make better use of the detected data of the fishing vessel point. Because the research area is a middle-low latitude area, the equal longitude and latitude data is subjected to projection conversion, and the projection of the final data is Lambert projection. And cutting out the research area according to the longitude and latitude of the research area. Many noises exist in the sub-regions, and the original image is subjected to background suppression by using median filtering and wiener filtering. The image after pre-processing is shown in figure 3.
Calculating gradients in the horizontal direction and the vertical direction according to a formula image _ gradients (i) ═ image (i +1) -image (i), wherein image _ gradient (i) is gradient data, and image (i) represents a DN value at the ith pixel in the image. Counting the position information of points larger than the gradient mean value and passing
And calculating the mean value of the maximum value and the minimum value of the gradient image which is larger than the gradient mean value, wherein max is the maximum value of the corresponding value of the point which is larger than the gradient mean value in the gradient image in the projection image, and min is the minimum value of the corresponding value of the point which is larger than the gradient mean value in the gradient image in the projection image.
And comparing the threshold solved in the horizontal direction and the vertical direction, and selecting the smaller value as the threshold. Considering that the fishing boat point is similar to the noise point in the image, a limit of 98% is added. And comparing the solved threshold with DN values corresponding to 98%, if the threshold is larger, taking the threshold as a threshold of the operation in the step, otherwise, selecting a threshold corresponding to 99%. Based on the obtained threshold segmentation image, points larger than the threshold are retained (the target is represented by 1, and the background is represented by 0), and the obtained target is a pseudo-target image, which is an image a, that is, the final fishing vessel point is definitely present in the pseudo-target image, as shown in fig. 4.
And a communicated region marking step of marking the communicated region of the suspected image, and counting the maximum value of each marked region and the number of pixels of each region. And counting a target area smaller than the corrosion operator, and recording the maximum value point of the target area as an image b. (prevent the corrosion algorithm from corroding off targets smaller than the corrosion operator).
And carrying out maximum value filtering on the image to obtain an image c. And (c-a) subtracting the image c with the maximum value filtering from the suspected target image a, and selecting zero points as target points to be output, wherein the points are detected fishing boat points and are images d. The final detected image is e: e ═ d + b. As shown in fig. 5.
As can be easily found, the invention adopts a self-adaptive extreme value threshold value selection method and central maximum value filtering, can automatically identify the luminous remote sensing image light fishing boat, and obviously improves the identification precision.
Claims (3)
1. A light fishing boat automatic identification method for luminous remote sensing data is characterized by comprising the following steps:
(1) screening luminous remote sensing data;
(2) preprocessing the screened noctilucent remote sensing data, and cutting out a subregion image of a sea area near the lamplight fishing boat;
(3) calculating a threshold value of the subarea image by adopting a central maximum value method, segmenting the image according to the calculated threshold value, reserving points larger than the threshold value, and determining that the calculated target is a suspected target image; the sub-region image thresholding by adopting the central maximum value method comprises the following sub-steps:
(31) solving the gradients of the horizontal direction and the vertical direction of the sub-area image;
(32) calculating the gradient mean value in the transverse direction and the longitudinal direction, and then counting the position information of the gradient image points larger than the mean value;
(33) calculating DN values of the preprocessed image points corresponding to the points with the gradient larger than the gradient mean value, searching the maximum value and the minimum value in the DN values, and taking the average value of the maximum value and the minimum value;
(34) comparing the average values of the two directions, and taking the minimum value; if the minimum value is greater than the DN value corresponding to 98% of the positions, the minimum value is reserved; if the minimum value is less than 98% of the corresponding DN value, taking the DN value corresponding to 99% as a threshold value;
(4) carrying out connected region marking on the suspected target image, counting the maximum value of each region and the number of pixels of each region, counting target regions smaller than the corrosion operator, and recording the maximum value points of the target regions;
(5) carrying out corrosion operation on the image;
(6) carrying out maximum value filtering on the image;
(7) subtracting the suspected target image from the maximum value filtered image, selecting a zero value point as a target point, and taking the target point and the maximum value point of the target area as final output.
2. The automatic identification method for the luminous fishing boat for the remote sensing data according to claim 1, wherein the conditions for screening the luminous remote sensing data in the step (1) are as follows: the screened area range is required to contain the movable sea area of the lamplight fishing boat, the time is required to have the working time of the lamplight fishing boat, and the cloud cover in the movable sea area of the lamplight fishing boat is required to be less than 10%.
3. A light fishing boat automatic identification method for luminous remote sensing data according to claim 1, characterized in that the preprocessing in the step (2) comprises remote sensing image projection correction and noise removal; wherein, the remote sensing image projection correction adopts Lambert projection; the noise removal is realized by adopting median filtering and wiener filtering.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810808844.7A CN109086701B (en) | 2018-07-19 | 2018-07-19 | Automatic identification method for luminous fishing boat for luminous remote sensing data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810808844.7A CN109086701B (en) | 2018-07-19 | 2018-07-19 | Automatic identification method for luminous fishing boat for luminous remote sensing data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109086701A CN109086701A (en) | 2018-12-25 |
| CN109086701B true CN109086701B (en) | 2021-08-20 |
Family
ID=64838452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810808844.7A Expired - Fee Related CN109086701B (en) | 2018-07-19 | 2018-07-19 | Automatic identification method for luminous fishing boat for luminous remote sensing data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109086701B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993087A (en) * | 2019-03-22 | 2019-07-09 | 中国海洋大学 | A method of saury fishing boat is identified using remote sensing nighttime light data |
| CN112836560A (en) * | 2020-06-01 | 2021-05-25 | 新亚优华(青岛)海洋水产科技有限公司 | Method for identifying lamplight fishing boat in Korean fishery agreement area in yellow sea based on different thresholds |
| CN112785513B (en) * | 2020-08-25 | 2023-04-18 | 青岛经济技术开发区海尔热水器有限公司 | Self-adaptive median filtering method for filtering impulse noise |
| CN115661434B (en) * | 2022-10-17 | 2023-05-02 | 中国人民解放军61540部队 | Night marine ship automatic identification method, system, electronic equipment and medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4727614B2 (en) * | 2007-03-30 | 2011-07-20 | シャープ株式会社 | Image processing apparatus, control program, computer-readable recording medium, electronic apparatus, and control method for image processing apparatus |
| CN104318544B (en) * | 2014-09-25 | 2017-08-11 | 中国水产科学研究院东海水产研究所 | Operation fishing boat quantity survey (surveying) method is caught based on night lights satellite remote sensing date photo-induction |
| CN106709523B (en) * | 2016-12-29 | 2020-01-10 | 北京理工雷科电子信息技术有限公司 | Optical remote sensing image ship identification method based on S-HOG characteristics |
-
2018
- 2018-07-19 CN CN201810808844.7A patent/CN109086701B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN109086701A (en) | 2018-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109086701B (en) | Automatic identification method for luminous fishing boat for luminous remote sensing data | |
| CN110414411B (en) | Sea surface ship candidate area detection method based on visual saliency | |
| Chen et al. | Ship detection from coastal surveillance videos via an ensemble Canny-Gaussian-morphology framework | |
| JP6319785B2 (en) | Abnormal tide level fluctuation detection device, abnormal tide level fluctuation detection method, and abnormal tide level fluctuation detection program | |
| CN104217215B (en) | A kind of classifying identification method of water surface foggy image and picture rich in detail | |
| US9031285B2 (en) | Detection of floating objects in maritime video using a mobile camera | |
| CN110544211B (en) | Method, system, terminal and storage medium for detecting lens attached object | |
| US20070110309A1 (en) | Shadow detection in images | |
| CN103488992B (en) | A kind of oil spilling detection method towards complicated SAR image scene | |
| CN112149476B (en) | Target detection method, device, equipment and storage medium | |
| CN106530313A (en) | Sea-sky line real-time detection method based on region segmentation | |
| CN112417955B (en) | Method and device for processing tour inspection video stream | |
| US20170178341A1 (en) | Single Parameter Segmentation of Images | |
| CN112102288A (en) | Water body identification and water body change detection method, device, equipment and medium | |
| CN113255537A (en) | Image enhancement denoising method for identifying sailing ship | |
| CN110516565A (en) | A kind of sea ship recognition methods based on Zernike and SVM | |
| CN112307943B (en) | Water area man-boat target detection method, system, terminal and medium | |
| CN116797941A (en) | Marine oil spill risk source rapid intelligent identification and classification method for high-resolution remote sensing image | |
| CN104820818A (en) | Fast recognition method for moving object | |
| CN114219996A (en) | SAR image berthing ship target detection method | |
| CN117095015A (en) | Image segmentation method, device, computer equipment and readable storage medium | |
| CN114283322A (en) | SAR image ship detection method | |
| Chapple | Unsupervised detection of mine-like objects in seabed imagery from autonomous underwater vehicles | |
| CN115619801A (en) | Monitoring video image occlusion detection method based on deep learning | |
| Nandini et al. | A survey of the existing shadow detection techniques |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210820 |