CN102663348B - Marine ship detection method in optical remote sensing image - Google Patents
Marine ship detection method in optical remote sensing image Download PDFInfo
- Publication number
- CN102663348B CN102663348B CN 201210077407 CN201210077407A CN102663348B CN 102663348 B CN102663348 B CN 102663348B CN 201210077407 CN201210077407 CN 201210077407 CN 201210077407 A CN201210077407 A CN 201210077407A CN 102663348 B CN102663348 B CN 102663348B
- Authority
- CN
- China
- Prior art keywords
- image
- target
- ship
- ship target
- zone
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 230000003287 optical effect Effects 0.000 title abstract 2
- 238000000034 method Methods 0.000 claims abstract description 21
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims description 12
- 230000000007 visual effect Effects 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 6
- 239000004744 fabric Substances 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000012706 support-vector machine Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 9
- 238000002474 experimental method Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000004064 recycling Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides an optical remote sensing image marine ship detection method based on local contrast information and a space pyramid characteristic. A technical scheme is characterized by: firstly, sliding a window in a sea area based on local contrast so as to carry out suspected object detection of a ship and reducing a false alarm of ship detection; then, for a suspected object area obtained through segmentation, taking a neighborhood according to a certain size of the window, using a space pyramid matching model to extract space context information so as to carry out classification, deleting background interference, acquiring a ship detection result and reducing the false alarm of the ship detection. By using the method of the invention, white polarity performance and black polarity performance problems of the ship can be effectively inhibited. Simultaneously, for a similarity problem of the ship object and the other interference and a difference problem possessed by ship object, the local neighborhood context information is introduced to carry out characteristic description and identification of the ship. The object and the background interference is distinguished and a false alarm rate of the ship detection can be effectively inhibited.
Description
Technical field
The present invention relates to the intelligent Ship Target Detection technology in remote Sensing Image Analysis field, more particularly, relate to the marine Ship Detection of remote sensing image under a kind of complicated sea condition.
Background technology
In remote sensing image, the ShipTargets test problems under the complicated sea background is difficult point always.On the one hand, because the many factors such as image device, atmosphere, shooting angle, time, meteorology, and different wave of the sea situations is different to the reflection potential of illumination, so that there are great changes in the information such as the brightness of remote sensing image, contrast, Sea background has instability, there is fluctuating in mean flow rate, and its high-frequency information is altered a great deal by the impact of wave, flight path in amplitude.Detect specific to the naval vessel, owing to being subjected to the impact of illumination, naval vessel surface coating, the Ship Target expressing gradation is uncertain, and the visible light Ship Target may be higher or lower than Sea background brightness in brightness, is called the white polarity performance on naval vessel and deceives the polarity performance.At this moment, traditional detection method based on Threshold segmentation can't select a suitable threshold value with target and background separation, causes higher false alarm rate.
On the other hand, because visual light imaging easily is subjected to the impact of the factors such as weather, so often there are a large amount of situations such as cloud in the visible images, the naval vessel detects and often is subject to the interference such as cloud, wave, often false-alarm is more to cause present Ship Detection, although at present some marine Ship Detections after the rough detection of sea regional aim obtains suspected target, increase is carried out analysis confirmation to the target unique characteristics of Ship Target candidate region or naval vessel suspected target, remove the part false-alarm, obtain the naval vessel testing result.Ship Target unique characteristics commonly used in these methods has: gray scale, size, shape, textural characteristics, after extracting the description of target self various features, adopt sorter that doubtful Ship Target is carried out class validation identification.Yet, disturb for suspected targets such as cloud, wave, islands, sometimes the target signature and the actual Ship Target that extract at the suspected target piece are closely similar, and different resolution, the feature that Ship Target does not reflect in the phase images simultaneously are variant again, really be familiar with for the naval vessel and do not bring a lot of difficulties, so still there is more false-alarm in based target unique characteristics analytical approach at present, hindered the application of the marine Ship Detection of visible images.
Summary of the invention
The present invention provides a kind of remote sensing image based on local contrast information and space pyramid feature marine Ship Detection in order effectively to solve the marine ship detection problem of complicated sea remote sensing image.This method establishment the naval vessel the performance of white polarity and deceive polarity performance problem, the variability issues that has for Similarity Problem and the Ship Target itself of Ship Target and other interference simultaneously, this method is introduced the local neighborhood contextual information of target and is described and identification for the feature on naval vessel, distinguish target and background interference, the false alarm rate that the establishment naval vessel detects.
Basic ideas of the present invention are: at first, for the black and white polarity problems of Ship Target, detect at the suspected target that zone, sea moving window carries out the naval vessel based on local contrast, reduce the false dismissal that the naval vessel detects; Then, get its neighborhood to cutting apart the suspected target zone that obtains by certain window size, utilize space pyramid Matching Model to extract spatial context information and classify, the deletion background interference is obtained the naval vessel testing result, reduces the false-alarm that the naval vessel detects.
Technical scheme of the present invention is: a kind of remote sensing image ShipTargets detection method specifically comprises the steps:
Known: a width of cloth input picture I1 is remote sensing image.
The first step: extra large land Region Segmentation
Input picture I1 is carried out cutting apart of zone, sea and land area, obtain the zone, sea is carried out the marine site image I 2 of mark.
Second step: the rough detection of doubtful Ship Target
To marine site image I 2, adopt the Contrast Box algorithm of having revised decision rule to process, detect and obtain one group of rectangular area that comprises doubtful Ship Target, be referred to as doubtful Ship Target zone.
Especially, being defined as follows decision rule in Contrast Box algorithm judges:
|(μ
T-μ
B)|/δ
B＞K (1)
μ wherein
TThe gray average of target window in the expression Contrast Box algorithm, μ
BThe gray average of expression backdrop window, δ
BThe gray standard deviation of expression backdrop window, K is detection threshold.Satisfy criterion (1) and then think the suspected target point, it namely is doubtful Ship Target zone that the minimum area boundary rectangle is tried to achieve in the zone that a plurality of adjacent suspected target points are formed, and the zone of usually selecting to comprise adjacent suspected target point more than 2 is regional as doubtful Ship Target.
The 3rd step: the contextual feature of doubtful Ship Target is extracted
Each doubtful Ship Target zone made obtain with the following method suspected target neighborhood image piece: centered by the center in doubtful Ship Target zone, length and width is respectively 2 times of sizes of doubtful Ship Target zone length and width, obtain comprising the rectangular area of suspected target and neighborhood thereof, be referred to as suspected target neighborhood image piece.Utilize space pyramid Matching Model to extract the spatial context feature of doubtful Ship Target to suspected target neighborhood image piece.
The 4th step: doubtful Ship Target recognition and verification
To the spatial context feature of doubtful Ship Target, utilize based on SVM (the support vector machine) sorter of histogram intersection nuclear and classify, obtain the whether affirmation result on naval vessel of this doubtful Ship Target.
The invention has the beneficial effects as follows: on the one hand, redefined the judgment criterion of Contrast Box algorithm according to the Ship Imaging characteristic, thereby overcome the black and white polarity problems that the naval vessel exists in the marine site, reduced the target false dismissed rate.On the other hand, after getting access to doubtful Ship Target, be different from traditional method of only Ship Target extraction feature itself being analyzed, the present invention is directed to Ship Target and neighborhood thereof extraction spatial context feature analyzes, solved to a certain extent the not strong problem of the feature property distinguished between Ship Target and the false-alarm, the ShipTargets that can be directly used in remote sensing image detects.
Description of drawings
Fig. 1 is the marine Ship Detection process flow diagram in the remote sensing image provided by the present invention;
Fig. 2 is that second step carries out doubtful Ship Target rough detection schematic diagram in the emulation experiment;
Fig. 3 is the spatial context character description method schematic diagram of doubtful Ship Target of the 3rd step in the emulation experiment;
Fig. 4 is the process flow diagram that the 4th step classified based on the svm classifier device of histogram intersection nuclear to the spatial context characteristic use in the emulation experiment.
Embodiment
Below in conjunction with accompanying drawing remote sensing image Ship Target Detection method provided by the invention is elaborated.
Fig. 1 is the marine Ship Detection process flow diagram in the remote sensing image provided by the present invention.The first step of this process flow diagram is extra large land Region Segmentation, obtain the zone, sea by extra large land Region Segmentation, at first determine the binarization segmentation threshold value based on the gray difference on sea and land according to the OTSU method, obtain marine site and land initial segmentation, in the marine site, select again Seed Points to adopt region-growing method to obtain the zone, sea.The rough detection of the doubtful Ship Target of second step, to utilize the Contrast Box algorithm of having revised decision rule in the zone, sea, to obtain one group of doubtful Ship Target zone by the detection of pixel moving window, Contrast Box algorithm specific implementation sees article " CASASENT D.P.SU W.; TURAGA D.; et al, SAR ship detection using new conditional contrast box filter[C], SPIE; 1999,372l:274-284. " for details.The 3rd step contextual feature of doubtful Ship Target is extracted, and doubtful Ship Target zone is obtained suspected target neighborhood image piece, utilizes space pyramid Matching Model to extract the spatial context feature.Doubtful Ship Target recognition and verification of the 4th step is to use the svm classifier device of examining based on histogram intersection that the spatial context feature of doubtful Ship Target is classified to be confirmed the result.
Fig. 2 is that second step carries out doubtful Ship Target rough detection schematic diagram in the emulation experiment.Fig. 2 (a) is the marine site image of a width of cloth remote sensing image of input, clear in order to represent, at this marine site image middle finger naval vessel, island and cloud are shown, Fig. 2 (b) is depicted as 4 doubtful Ship Target zones that obtain by detection, and Fig. 2 (c) is depicted as the suspected target neighborhood image piece that respectively 4 doubtful Ship Target zones is got access to.The Contrast Box algorithm of having revised decision rule in utilization carries out in the rough detection of doubtful Ship Target, the size of target window T is maximum Ship Target size in the remote sensing images, be 4 times of target window T area for guaranteeing that backdrop window covers enough sufficient background characteristics data, getting backdrop window B.Detection threshold K control detects false alarm rate, usually gets 1.25.
Fig. 3 is the spatial context character description method schematic diagram of doubtful Ship Target of the 3rd step in the emulation experiment.At first suspected target neighborhood image piece is carried out the division of regular grid, recycling space pyramid Matching Model is extracted the spatial context feature of doubtful Ship Target.Specifically describe as follows:
Fig. 3 left-side images is a width of cloth suspected target neighborhood image piece that detects, its uniform grid that carries out rule is cut apart, shown in Fig. 3 intermediate image, be referred to as the image block node, extract again SIFT (the Scale Invariant Feature Transform) feature of each image block node, the SIFT specific implementation sees article " David G.Lowe; Distinctive image features from scale-invariant keypoints.InternationalJournal of Computer Vision, 200460 (2): 91-110 " for details.The image block node size is determined desirable 16 * 16 pixels of middle high-resolution, 8 * 8 pixels, desirable 4 * 4 pixels of low resolution according to the height of remote sensing images resolution is different.Simultaneously select at random a part of suspected target neighborhood image piece as training image, the SIFT feature that each image block nodes of these training images extracts is carried out the K mean cluster, K mean cluster number in the experiment, namely the visual vocabulary number is made as 100, obtains the visual vocabulary code book of image.Then the SIFT feature of the image block Node extraction of all suspected target neighborhood image pieces all quantized according to this visual vocabulary code book.Each image block node is with visual vocabulary of correspondence like this.Thereby the pixel space of having finished image represents the conversion of the visual vocabulary space representation of image, and the figure that is converted to is referred to as vocabulary figure, the vocabulary figure that Fig. 3 image right is converted to for the suspected target neighborhood image piece of inputting.
Recycle the spatial context feature that space pyramid Matching Model is obtained doubtful Ship Target for the vocabulary figure that obtains.Wherein pyramid Matching Model in space specifically sees document " Lazibnik etc.Beyond Bags of Features:Spatial Pyramid Matching for Recognizing Natural Scene Categories.Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.New York:2006:2169-2178 " for details.The spatial context of remembering is characterized as P
w
Fig. 4 is the process flow diagram that the 4th step classified based on the svm classifier device of histogram intersection nuclear to the spatial context characteristic use in the emulation experiment of the present invention.This sorting algorithm flow process is divided into two parts of training and testing, and the dotted line left side of Fig. 4 vertical direction is the training stage, and the right side is test phase.Sample Storehouse comprises this four classes sample of island, cloud, wave and naval vessel altogether.In the training stage, training sample database consists of by randomly drawing a part of suspected target neighborhood image piece in each class sample, the top, Fig. 4 dotted line left side is the training sample signal of input, then obtain the visual vocabulary code book according to the 3rd one step process, extract the spatial context feature P of all the suspected target neighborhood image pieces in the training sample database
w, the svm classifier device is trained by training sample based on the histogram intersection kernel function again, obtain the svm classifier model.The top, Fig. 4 dotted line right side is a width of cloth suspected target neighborhood image piece of input, at test phase, the visual vocabulary code book that suspected target neighborhood image piece was obtained according to the training stage, obtain its spatial context feature, recycling svm classifier model is classified to test sample book, obtains the whether affirmation result of Ship Target of suspected target.Wherein, the histogram intersection kernel function sees article " Barla A; Odone F; and Verri A.Histogram intersection kernel for imageclassification[C] .Proceedings of the International Conference on Image Processing, Barcelona, Catalonia; Spain; Sept.14-17,2003, Vol.2:513-516. " for details.
Claims (2)
1. remote sensing image ShipTargets detection method, a known width of cloth input picture I1 is remote sensing image, it is characterized in that, specifically comprises the steps:
The first step: extra large land Region Segmentation;
Input picture I1 is carried out cutting apart of zone, sea and land area, and mark is carried out in the zone, sea, obtain marine site image I 2;
Second step: the rough detection of doubtful Ship Target;
To marine site image I 2, adopt the Contrast Box algorithm of having revised decision rule to process, detect and obtain one group of rectangular area that comprises doubtful Ship Target, be referred to as doubtful Ship Target zone; Namely being defined as follows decision rule in Contrast Box algorithm judges:
|(μ
T-μ
B)|/δ
B>K
μ wherein
TThe gray average of target window in the expression Contrast Box algorithm, μ
BThe gray average of expression backdrop window, δ
BThe gray standard deviation of expression backdrop window, K is detection threshold, satisfies above-mentioned decision rule and just thinks the suspected target point; It namely is doubtful Ship Target zone that the minimum area boundary rectangle is tried to achieve in the zone that a plurality of adjacent suspected target points are formed;
The 3rd step: the contextual feature of doubtful Ship Target is extracted;
Each doubtful Ship Target zone made obtain with the following method suspected target neighborhood image piece: centered by the center in doubtful Ship Target zone, length and width is respectively 2 times of sizes of doubtful Ship Target zone length and width, obtain comprising the rectangular area of suspected target and neighborhood thereof, be referred to as suspected target neighborhood image piece; Utilize space pyramid Matching Model to extract the spatial context feature of doubtful Ship Target to suspected target neighborhood image piece, detailed process is:
The uniform grid that the suspected target neighborhood image is carried out rule is cut apart, and each uniform grid is called the image block node, extracts the SIFT feature of each image block node again; Simultaneously select at random a part of suspected target neighborhood image piece as training image, the SIFT feature that each image block nodes of these training images extracts is carried out the K mean cluster, obtain the visual vocabulary code book of image; Then the SIFT feature of the image block Node extraction of all suspected target neighborhood image pieces all quantized according to this visual vocabulary code book, make the corresponding visual vocabulary of each image block node, thereby the pixel space of finishing image represents the conversion of the visual vocabulary space representation of image, and the figure that is converted to is called vocabulary figure; Recycle the spatial context feature that space pyramid Matching Model is obtained doubtful Ship Target for the vocabulary figure that obtains;
The 4th step: doubtful Ship Target recognition and verification;
To the spatial context feature of doubtful Ship Target, utilizing the SVM(support vector machine based on histogram intersection nuclear) sorter classifies, and obtains the whether affirmation result on naval vessel of this doubtful Ship Target.
2. remote sensing image ShipTargets detection method according to claim 1 is characterized in that, selects to comprise the zone of adjacent suspected target point more than 2 as doubtful Ship Target zone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210077407 CN102663348B (en) | 2012-03-21 | 2012-03-21 | Marine ship detection method in optical remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210077407 CN102663348B (en) | 2012-03-21 | 2012-03-21 | Marine ship detection method in optical remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102663348A CN102663348A (en) | 2012-09-12 |
| CN102663348B true CN102663348B (en) | 2013-10-16 |
Family
ID=46772832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201210077407 Expired - Fee Related CN102663348B (en) | 2012-03-21 | 2012-03-21 | Marine ship detection method in optical remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102663348B (en) |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102867196B (en) * | 2012-09-13 | 2015-10-21 | 武汉大学 | Based on the complicated sea remote sensing image Ship Detection of Gist feature learning |
| CN103020592B (en) * | 2012-11-22 | 2015-09-23 | 北京大学 | A kind of remote sensing images naval vessel based on local context detects false-alarm elimination method |
| CN103150582B (en) * | 2013-02-04 | 2015-11-25 | 北京理工大学 | Modified result method is sentenced based on contextual remote sensing image cloud |
| CN103150567B (en) * | 2013-02-04 | 2016-01-20 | 北京理工大学 | The remote sensing image cloud method of discrimination of integration mode identification and Context Knowledge |
| CN103544502A (en) * | 2013-10-29 | 2014-01-29 | 上海市城市建设设计研究总院 | High-resolution remote-sensing image ship extraction method based on SVM |
| CN103617628A (en) * | 2013-12-16 | 2014-03-05 | 北京航空航天大学 | Naval ship detection method based on polarized azimuth secant function characteristic |
| CN103996017B (en) * | 2014-02-24 | 2018-08-31 | 航天恒星科技有限公司 | One kind being based on HuThe not Ship Detection of bending moment and support vector machines |
| CN104166853B (en) * | 2014-04-24 | 2019-06-25 | 中国人民解放军海军航空大学 | The method that rapidly extracting regularization naval vessel is sliced from high-resolution remote sensing image |
| CN103984945A (en) * | 2014-05-14 | 2014-08-13 | 武汉大学 | Optical remote sensing image ship detection method |
| CN104182731B (en) * | 2014-08-12 | 2017-05-17 | 中国科学技术大学 | A grid-shaped radar detection method based on cross detection |
| CN105095846B (en) * | 2014-09-28 | 2019-07-30 | 航天恒星科技有限公司 | Region growing seed point extracting method and system towards the segmentation of remote sensing images sea land |
| CN104992140B (en) * | 2015-05-27 | 2019-07-26 | 上海海事大学 | A kind of sea exception floating object detecting method based on remote sensing images |
| CN105139034B (en) * | 2015-08-27 | 2019-06-28 | 北京市遥感信息研究所 | A kind of ship detecting method that combination spectrum filters out |
| CN105303526B (en) * | 2015-09-17 | 2017-11-07 | 哈尔滨工业大学 | A kind of Ship Target Detection method based on coastline data and spectrum analysis |
| CN105976345A (en) * | 2016-04-27 | 2016-09-28 | 北京航空航天大学 | Visible light remote sensing image synthesis method |
| CN106056084B (en) * | 2016-06-01 | 2019-05-24 | 北方工业大学 | Remote sensing image port ship detection method based on multi-resolution hierarchical screening |
| CN106295498B (en) * | 2016-07-20 | 2019-04-16 | 湖南大学 | Remote sensing image target area detection device and method |
| CN106709426B (en) * | 2016-11-29 | 2020-04-14 | 上海航天测控通信研究所 | Ship target detection method based on infrared remote sensing image |
| CN106815553B (en) * | 2016-12-13 | 2019-07-23 | 华中科技大学 | A kind of infrared forward sight image Ship Detection based on edge matching |
| CN106845372B (en) * | 2016-12-31 | 2018-01-26 | 华中科技大学 | The ship target detection recognition method and system of a kind of space remote sensing optical imagery |
| CN107704865A (en) * | 2017-05-09 | 2018-02-16 | 北京航空航天大学 | Fleet Targets Detection based on the extraction of structure forest edge candidate region |
| CN108229342B (en) * | 2017-12-18 | 2021-10-26 | 西南技术物理研究所 | Automatic sea surface ship target detection method |
| CN108376247A (en) * | 2018-02-05 | 2018-08-07 | 北方工业大学 | Strategic coarse-fine combined sea-land separation method applied to optical remote sensing ship detection |
| CN108681691A (en) * | 2018-04-09 | 2018-10-19 | 上海大学 | A kind of marine ships and light boats rapid detection method based on unmanned water surface ship |
| CN110120073B (en) * | 2019-03-28 | 2020-12-22 | 上海大学 | Method for guiding recovery of unmanned ship based on lamp beacon visual signal |
| CN110443201B (en) * | 2019-08-06 | 2023-01-10 | 哈尔滨工业大学 | Target identification method based on multi-source image joint shape analysis and multi-attribute fusion |
| CN110910421B (en) * | 2019-11-11 | 2022-11-11 | 西北工业大学 | Weak and small moving object detection method based on block characterization and variable neighborhood clustering |
| CN111160354B (en) * | 2019-12-30 | 2022-06-17 | 哈尔滨工程大学 | Ship image segmentation method based on joint image information under sea and sky background |
| CN111582089B (en) * | 2020-04-27 | 2021-07-09 | 中国人民解放军军事科学院国防科技创新研究院 | Maritime target information fusion method based on satellite infrared and visible light images |
| CN111931688A (en) * | 2020-08-27 | 2020-11-13 | 珠海大横琴科技发展有限公司 | Ship recognition method and device, computer equipment and storage medium |
| CN112070083A (en) * | 2020-09-04 | 2020-12-11 | 北京灵汐科技有限公司 | Image content detection method and device, electronic equipment and storage medium |
| CN112101250B (en) * | 2020-09-18 | 2022-07-15 | 电子科技大学 | Method for detecting offshore ship target based on context semantic perception |
| CN113674308B (en) * | 2021-05-06 | 2024-02-13 | 西安电子科技大学 | SAR image ship target rapid detection method based on image enhancement and multiple detection |
| CN115331113A (en) * | 2022-10-12 | 2022-11-11 | 浙江华是科技股份有限公司 | Ship target detection model training method and system and computer storage medium |
| 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 |
| CN117761695B (en) * | 2024-02-22 | 2024-04-30 | 中国科学院空天信息创新研究院 | Multi-angle SAR three-dimensional imaging method based on self-adaptive partition SIFT |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024156A (en) * | 2010-11-16 | 2011-04-20 | 中国人民解放军国防科学技术大学 | Method for positioning lip region in color face image |
-
2012
- 2012-03-21 CN CN 201210077407 patent/CN102663348B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024156A (en) * | 2010-11-16 | 2011-04-20 | 中国人民解放军国防科学技术大学 | Method for positioning lip region in color face image |
Non-Patent Citations (8)
| Title |
|---|
| Learning Mid-Level Features For Recognition;Y-Lan Boureau等;《Proc 23rd International conference on computer vision and pattern recognition》;20100618;第2559-2566页 * |
| Y-Lan Boureau等.Learning Mid-Level Features For Recognition.《Proc 23rd International conference on computer vision and pattern recognition》.2010, |
| 利用贝叶斯网络融合空间上下文的高分辨遥感图像分类;程环环等;《计算机工程与科学》;20110131;第33卷(第1期);第70-76页 * |
| 可见光遥感图像中舰船目标检测方法;赵英海等;《光电工程》;20080831;第35卷(第8期);第102-106页 * |
| 周晖等.引入PLSA模型的光学遥感图像舰船检测.《遥感学报》.2010,第14卷(第4期), |
| 引入PLSA模型的光学遥感图像舰船检测;周晖等;《遥感学报》;20100430;第14卷(第4期);第672-680页 * |
| 程环环等.利用贝叶斯网络融合空间上下文的高分辨遥感图像分类.《计算机工程与科学》.2011,第33卷(第1期), |
| 赵英海等.可见光遥感图像中舰船目标检测方法.《光电工程》.2008,第35卷(第8期), |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102663348A (en) | 2012-09-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102663348B (en) | Marine ship detection method in optical remote sensing image | |
| CN108510467B (en) | SAR image target identification method based on depth deformable convolution neural network | |
| CN102509091B (en) | Airplane tail number recognition method | |
| Pan et al. | A robust system to detect and localize texts in natural scene images | |
| CN107610114B (en) | optical satellite remote sensing image cloud and snow fog detection method based on support vector machine | |
| CN107480620B (en) | Remote sensing image automatic target identification method based on heterogeneous feature fusion | |
| Kim et al. | Multi-task convolutional neural network system for license plate recognition | |
| CN104680127A (en) | Gesture identification method and gesture identification system | |
| Al-Hmouz et al. | License plate localization based on a probabilistic model | |
| CN102509112A (en) | Number plate identification method and identification system thereof | |
| CN110008900B (en) | Method for extracting candidate target from visible light remote sensing image from region to target | |
| Islam et al. | Text detection and recognition using enhanced MSER detection and a novel OCR technique | |
| CN109902618A (en) | A kind of sea ship recognition methods and device | |
| Fung et al. | A review of evaluation of optimal binarization technique for character segmentation in historical manuscripts | |
| CN107704865A (en) | Fleet Targets Detection based on the extraction of structure forest edge candidate region | |
| Kobchaisawat et al. | Thai text localization in natural scene images using convolutional neural network | |
| CN105787475A (en) | Traffic sign detection and identification method under complex environment | |
| Sheng et al. | Real-time anti-interference location of vehicle license plates using high-definition video | |
| Liu et al. | A robust algorithm for text detection in color images | |
| Huang et al. | Ship detection using texture statistics from optical satellite images | |
| Chowdhury et al. | Scene text detection using sparse stroke information and MLP | |
| Deb et al. | Statistical characteristics in HSI color model and position histogram based vehicle license plate detection | |
| Zheng et al. | Character segmentation for license plate recognition by K-means algorithm | |
| Qiu et al. | An optimized license plate recognition system for complex situations | |
| Zhang et al. | A novel detection and recognition system for Chinese traffic signs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131016 |