CN110675449A - Binocular camera-based offshore flow detection method - Google Patents
Binocular camera-based offshore flow detection method Download PDFInfo
- Publication number
- CN110675449A CN110675449A CN201910821151.6A CN201910821151A CN110675449A CN 110675449 A CN110675449 A CN 110675449A CN 201910821151 A CN201910821151 A CN 201910821151A CN 110675449 A CN110675449 A CN 110675449A
- Authority
- CN
- China
- Prior art keywords
- offshore flow
- image
- camera
- offshore
- flow
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- G06T7/85—Stereo camera calibration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a binocular camera-based offshore flow detection method, which belongs to the field of computer vision and the technical field of safety guarantee, and comprises the following steps: firstly, training a convolutional neural network algorithm for identifying the offshore flow; secondly, identifying whether an offshore flow exists in the sea wave image acquired by the binocular camera in the actual detection process by using a trained convolutional neural network; finally, locating the identified offshore flow; the method can acquire the three-dimensional information of the offshore flow based on binocular stereo vision; the method has the advantages that the accurate offshore flow position information can be timely acquired, the method is simple, the operability is high, relevant workers can mark the position of the beach by the three-dimensional information based on the offshore flow, visitors are reminded, and drowning events caused by the offshore flow are reduced.
Description
Technical Field
The invention belongs to the field of computer vision and the technical field of safety guarantee, and particularly relates to a binocular camera-based offshore flow detection method.
Background
The flow velocity of the offshore flow is mostly 0.3-1 meter per second, the fastest flow velocity can reach 3 meters per second, the length of the offshore flow can reach 30-100 meters or even longer, the flow direction is almost vertical to the shore line, and strong swimmers can be quickly pulled into deep water to cause drowning. The offshore flow becomes another marine disaster which causes harm to people traveling on the coast after storm surge and sea wave. About 90% of seaside drowning is due to offshore flow. The off-shore flow brings a great deal of problems to attraction maintenance, beach management and accident dispute treatment of coastal tourism, and seriously influences the healthy development of coastal tourism economy. At present, the technical evaluation, safety management and the like of the offshore flow disasters in China just start, and related investigation evaluation, risk evaluation, fine forecast, safety management, public science popularization warning and the like are extremely lacking; blind areas and error areas also exist in public recognition of the offshore flow, a large number of drowning events occur in a plurality of hot tourist areas due to recognition errors and lack of alertness, the rescue workload is increased, and the difficulty in safety management of coastal tourism is increased. There is therefore a great need for an efficient and simple method of off-shore flow detection. At present, the traditional detection method for the offshore flow is to place a buoy or a current meter near the shore.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides the binocular camera-based offshore flow detection method which is reasonable in design, overcomes the defects of the prior art and has a good effect.
In order to achieve the purpose, the invention adopts the following technical scheme:
a binocular camera-based offshore flow detection method comprises the following steps:
step 1: collecting an image set for training a convolutional neural network, and training the convolutional neural network;
step 2: identifying the offshore flow of the collected image by using an image identification algorithm of the trained convolutional neural network; judging whether an offshore flow exists;
if: if the judgment result is that the offshore flow exists, finding the characteristic points of the offshore flow, and extracting the characteristic points of the offshore flow; the characteristic points comprise the middle point and the left and right edge points of the offshore flow;
or if the judgment result is that the offshore flow does not exist, acquiring the image again for processing;
and step 3: positioning the identified feature points of the offshore flow by using a binocular positioning algorithm, and identifying the position of the offshore flow by using the position of the feature points of the offshore flow;
acquiring three-dimensional coordinates of the offshore flow based on a binocular stereoscopic vision technology;
and 4, step 4: based on the unknown position information of the feature points of the offshore flow, dangerous areas are divided and fed back to relevant workers, and the workers remind tourists in a marking mode.
Preferably, in step 2, the binocular camera comprises a left camera and a right camera, and since the data collected by the left camera and the right camera are substantially identical, the single-side camera image is used for the offshore flow recognition.
Preferably, the specific steps of training the convolutional neural network in step 1 are as follows:
step 1.1: preprocessing image data; the method specifically comprises the following steps:
step 1.1.1: processing the training data acquired in the step 1; converting the acquired training data into a data format which can be identified by TensorFlow;
step 1.1.2: adding label according to the image, putting the image and the label into an array, and converting the array into a format which can be identified by Tensorflow;
step 1.1.3: carrying out standardization processing including cutting and supplementing on the image;
step 1.2: building a convolutional neural network model based on a Tensorflow framework;
adopting a classical convolution neural network LeNet-5 model; the model is divided into 7 layers: a convolution layer-a pooling layer-a full connection output layer; the method comprises the following steps that a convolution layer extracts primary offshore flow characteristics, a pooling layer extracts main characteristics of an offshore flow, and a full-connection layer collects all partial characteristics;
step 1.3: and training the neural network for identifying the offshore flow by using the constructed convolutional neural network model.
Preferably, in step 3, the method specifically comprises the following steps:
step 3.1: calibrating the left camera and the right camera;
calibrating the camera by adopting a Zhangyingyou calibration method to obtain internal and external parameters of the left camera and the right camera;
step 3.2: correcting the left camera image and the right camera image;
carrying out distortion correction and three-dimensional correction on the image by using internal and external parameters obtained by the calibrated left and right cameras;
step 3.3: carrying out stereo matching on the images;
carrying out stereo matching on the images by using an SGBM algorithm to finally obtain a disparity map;
step 3.4: obtaining three-dimensional coordinate information of the offshore flow;
obtaining a depth image by using the disparity map and the internal parameters of the left camera and the right camera, and obtaining three-dimensional coordinates of the offshore flow feature points according to the camera model and the parameters obtained by the calibration camera;
step 3.5: and taking the position information of the midpoint of the offshore flow as the position of the offshore flow, and dividing dangerous areas at the left and right edges of the obtained offshore flow, namely expanding the areas of the offshore flow according to the property of the offshore flow and the actual condition to obtain the dangerous areas.
Preferably, in step 3.3, the SGBM is a semi-global block matching algorithm, and the SGBM algorithm specifically includes the following steps:
s1: preprocessing an image;
processing the image by using a horizontal Sobel operator, mapping pixel points to obtain a new image, and preprocessing the obtained gradient information of the original image;
s2: calculating cost;
the cost calculation is divided into two steps: firstly, gradient information of an image obtained through preprocessing is subjected to gradient cost calculation obtained through a sampling-based method; calculating the SAD cost obtained by the original image through a sampling-based method;
s3: dynamic planning;
energy accumulation is carried out in each direction according to the idea of dynamic programming, and then the matching costs in each direction are added to obtain the total matching cost;
s4: post-treatment;
the post-processing part needs to perform uniqueness detection, sub-pixel interpolation and left-right consistency detection.
The invention has the following beneficial technical effects:
according to the method, images are collected through a binocular camera, the ashore flow is identified by using an image identification algorithm based on a convolutional neural network, the ashore flow is positioned by using a binocular positioning principle, and a new method is provided for the detection of the ashore flow; the method for detecting the offshore flow based on the binocular camera can timely acquire more accurate position information of the offshore flow, is simple, low in labor cost and strong in operability, and relevant workers can mark the position of the beach by using the position information of the acquired offshore flow, so that tourists are reminded, the occurrence of drowning events caused by the offshore flow is reduced, and a feasible method is provided for coastal travel management.
Drawings
FIG. 1 is a flow chart of an implementation of a binocular camera based off-shore flow detection method;
FIG. 2 is a flow chart of a binocular positioning algorithm;
FIG. 3 is a flow chart of an algorithm for recognizing images;
fig. 4 is an offshore flow model schematic.
Detailed Description
The invention is described in further detail below with reference to the following figures and detailed description:
as shown in fig. 1, a binocular camera-based offshore flow detection method includes the following steps:
step 1: and collecting an image set for training the convolutional neural network, and training the convolutional neural network.
The training data provides a basis for subsequent image recognition, and can be near-shore sea waves acquired by using a binocular camera, and the image can have an off-shore current or an off-shore current near-shore sea wave image; or offshore ocean wave image data which contains an offshore flow and does not contain the offshore flow and exists on the network; or the off-shore flow ideal model and the off-shore wave image data generated by a computer.
The specific steps for training the convolutional neural network in step 1 are as follows:
step 1.1: preprocessing image data; the method specifically comprises the following steps:
step 1.1.1: processing the training data acquired in the step 1, and converting the acquired training data into a data format which can be identified by TensorFlow;
step 1.1.2: adding label according to the image, putting the image and the label into an array, and converting the array into a format which can be identified by Tensorflow;
step 1.1.3: carrying out standardization processing including cutting and supplementing on the image;
step 1.2: building a convolutional neural network model based on a Tensorflow framework;
adopting a classical convolution neural network LeNet-5 model; the model is divided into 7 layers: a convolution layer-a pooling layer-a full connection output layer; the method comprises the following steps that a convolution layer extracts primary offshore flow characteristics, a pooling layer extracts main characteristics of an offshore flow, and a full-connection layer collects all partial characteristics;
step 1.3: and training the neural network for identifying the offshore flow by using the constructed convolutional neural network model.
Step 2: after training is finished, experimental operation is carried out, actual sea waves are collected to be called as actual data, and the actual data are images collected by the sea waves when the offshore flow observation is actually carried out. Identifying the offshore flow of the acquired image by using an image identification algorithm of the trained CNN convolutional neural network (the flow of the image identification algorithm is shown in FIG. 2); judging whether an offshore flow exists or not;
if: if the judgment result is that the offshore flow exists, finding out the characteristic point of the offshore flow; the characteristic points comprise the middle point and the left and right edge points of the offshore flow;
or if the judgment result is that the offshore flow does not exist, acquiring the image again for processing;
the binocular camera comprises a left camera and a right camera, and the data collected by the left camera and the right camera are basically consistent, so that the single-side camera image is adopted for identifying the offshore flow.
And step 3: positioning the identified feature points of the offshore flow by using a binocular positioning algorithm (the flow is shown in fig. 3), and identifying the position of the offshore flow by using the position of the feature points of the offshore flow;
acquiring three-dimensional coordinates of the offshore flow based on a binocular stereoscopic vision technology;
the method specifically comprises the following steps:
step 3.1: calibrating the left camera and the right camera;
calibrating the camera by adopting a Zhangyingyou calibration method to obtain internal and external parameters of the left camera and the right camera; the Zhangyingyou calibration method is a camera calibration method based on a moving plane template, and the method is based on a method between a traditional camera calibration method and a camera self-calibration method. Overcomes the disadvantages of the two and combines the advantages of the two. The method specifically comprises the steps of calculating a homography matrix, calculating an internal parameter matrix, calculating an external parameter matrix and calculating a distortion parameter.
Step 3.2: correcting the left camera image and the right camera image;
carrying out distortion correction and three-dimensional correction on the image by using internal and external parameters obtained by the calibrated left and right cameras;
step 3.3: carrying out stereo matching on the images;
and carrying out stereo matching on the image by using an SGBM algorithm to obtain a disparity map, and finally obtaining actual three-dimensional coordinate information of the midpoint of the image. The SGBM is a semi-global block matching algorithm and has the characteristics of good parallax effect and high speed; the steps of the SGBM algorithm are as follows:
s1: preprocessing an image;
processing the image by using a horizontal Sobel operator, mapping pixel points to obtain a new image, and preprocessing the obtained gradient information of the original image;
s2: calculating cost;
the cost calculation is divided into two steps: firstly, gradient information of an image obtained through preprocessing is subjected to gradient cost calculation obtained through a sampling-based method; calculating the SAD cost obtained by the original image through a sampling-based method;
s3: dynamic planning;
energy accumulation is carried out in each direction according to the idea of dynamic programming, and then the matching costs in each direction are added to obtain the total matching cost;
s4: post-treatment;
the post-processing part needs to perform uniqueness detection, sub-pixel interpolation and left-right consistency detection.
Step 3.4: obtaining three-dimensional coordinates of the offshore flow;
obtaining a depth image by using the disparity map and the internal parameters of the left camera and the right camera, and obtaining three-dimensional coordinates of the offshore flow feature points according to the camera model and the parameters obtained by the calibration camera;
step 3.5: regarding the position information of the midpoint of the offshore flow as the position of the offshore flow, dividing the dangerous area at the left and right edges of the obtained offshore flow, namely properly expanding the area of the offshore flow according to the property of the offshore flow to obtain the dangerous area, such as: increasing by 5m in the left direction of the left edge point and 5m to the right at the right edge point.
And 4, step 4: based on the unknown position information of the feature points of the offshore flow, dangerous areas are divided and fed back to relevant workers, and the workers remind tourists in a marking mode.
It is to be understood that the above description is not intended to limit the present invention, and the present invention is not limited to the above examples, and those skilled in the art may make modifications, alterations, additions or substitutions within the spirit and scope of the present invention.
Claims (5)
1. A binocular camera-based offshore flow detection method is characterized by comprising the following steps: the method comprises the following steps:
step 1: collecting an image set for training a convolutional neural network, and training the convolutional neural network;
step 2: identifying the offshore flow of the collected image by using an image identification algorithm of a convolutional neural network; judging whether an offshore flow exists;
if: if the judgment result is that the offshore flow exists, finding the characteristic points of the offshore flow, and extracting the characteristic points of the offshore flow; the characteristic points comprise the middle point and the left and right edge points of the offshore flow;
or if the judgment result is that the offshore flow does not exist, acquiring the image again for processing;
and step 3: positioning the identified feature points of the offshore flow by using a binocular positioning algorithm, and identifying the position of the offshore flow by using the position of the feature points of the offshore flow;
acquiring three-dimensional coordinates of the offshore flow based on a binocular stereoscopic vision technology;
and 4, step 4: based on the unknown position information of the feature points of the offshore flow, dangerous areas are divided and fed back to relevant workers, and the workers remind tourists in a marking mode.
2. The binocular camera-based offshore flow detection method of claim 1, wherein: in step 2, the binocular camera comprises a left camera and a right camera, and since data collected by the left camera and the right camera are basically consistent, a single-side camera image is adopted for identifying the offshore flow.
3. The binocular camera-based offshore flow detection method of claim 1, wherein: the specific steps for training the convolutional neural network in step 1 are as follows:
step 1.1: preprocessing image data; the method specifically comprises the following steps:
step 1.1.1: processing the training data acquired in the step 1, and processing the training data by using images acquired by a group of single-side cameras; converting the acquired training data into a data format which can be identified by TensorFlow;
step 1.1.2: adding label according to the image, putting the image and the label into an array, and converting the array into a format which can be identified by Tensorflow;
step 1.1.3: carrying out standardization processing including cutting and supplementing on the image;
step 1.2: building a convolutional neural network model based on a Tensorflow framework;
adopting a classical convolution neural network LeNet-5 model; the model is divided into 7 layers: a convolution layer-a pooling layer-a full connection output layer; the method comprises the following steps that a convolution layer extracts primary offshore flow characteristics, a pooling layer extracts main characteristics of an offshore flow, and a full-connection layer collects all partial characteristics;
step 1.3: and training the neural network for identifying the offshore flow by using the constructed convolutional neural network model.
4. The binocular camera-based offshore flow detection method of claim 1, wherein: in step 3, the method specifically comprises the following steps:
step 3.1: calibrating the left camera and the right camera;
calibrating the camera by adopting a Zhangyingyou calibration method to obtain internal and external parameters of the left camera and the right camera;
step 3.2: correcting the left camera image and the right camera image;
carrying out distortion correction and three-dimensional correction on the image by using internal and external parameters obtained by the calibrated left and right cameras;
step 3.3: carrying out stereo matching on the images;
carrying out stereo matching on the images by using an SGBM algorithm to finally obtain a disparity map;
step 3.4: obtaining three-dimensional coordinate information of the offshore flow;
obtaining a depth image by using the disparity map and the internal parameters of the left camera and the right camera, and obtaining three-dimensional coordinates of the offshore flow feature points according to the camera model and the parameters obtained by the calibration camera;
step 3.5: and taking the position information of the midpoint of the offshore flow as the position of the offshore flow, and dividing dangerous areas at the left and right edges of the obtained offshore flow, namely expanding the areas of the offshore flow according to the property of the offshore flow and the actual condition to obtain the dangerous areas.
5. The binocular camera-based offshore flow detection method of claim 4, wherein: in step 3.3, the SGBM is a semi-global block matching algorithm, and the SGBM algorithm specifically includes the following steps:
s1: preprocessing an image;
processing the image by using a horizontal Sobel operator, mapping pixel points to obtain a new image, and preprocessing the obtained gradient information of the original image;
s2: calculating cost;
the cost calculation is divided into two steps: firstly, gradient information of an image obtained through preprocessing is subjected to gradient cost calculation obtained through a sampling-based method; calculating the SAD cost obtained by the original image through a sampling-based method;
s3: dynamic planning;
energy accumulation is carried out in each direction according to the idea of dynamic programming, and then the matching costs in each direction are added to obtain the total matching cost;
s4: post-treatment;
the post-processing part needs to perform uniqueness detection, sub-pixel interpolation and left-right consistency detection.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910821151.6A CN110675449B (en) | 2019-09-02 | 2019-09-02 | Binocular camera-based offshore flow detection method |
| PCT/CN2019/115513 WO2021042490A1 (en) | 2019-09-02 | 2019-11-05 | Offshore current detection method based on binocular camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910821151.6A CN110675449B (en) | 2019-09-02 | 2019-09-02 | Binocular camera-based offshore flow detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110675449A true CN110675449A (en) | 2020-01-10 |
| CN110675449B CN110675449B (en) | 2020-12-08 |
Family
ID=69076671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910821151.6A Active CN110675449B (en) | 2019-09-02 | 2019-09-02 | Binocular camera-based offshore flow detection method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110675449B (en) |
| WO (1) | WO2021042490A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110763426A (en) * | 2019-09-29 | 2020-02-07 | 哈尔滨工程大学 | Method and device for simulating offshore flow in pool |
| CN112950610A (en) * | 2021-03-18 | 2021-06-11 | 河海大学 | Method and system for monitoring and early warning of fission flow |
| CN113936248A (en) * | 2021-10-12 | 2022-01-14 | 河海大学 | Beach personnel risk early warning method based on image recognition |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115663665B (en) * | 2022-12-08 | 2023-04-18 | 国网山西省电力公司超高压变电分公司 | Binocular vision-based protection screen cabinet air-open state checking device and method |
| CN117131799B (en) * | 2023-08-17 | 2024-02-23 | 浙江大学 | Bottom bed shear stress calculation method based on image |
| CN117395377B (en) * | 2023-12-06 | 2024-03-22 | 上海海事大学 | Multi-view fusion-based coastal bridge sea side safety monitoring method, system and medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050074161A1 (en) * | 2001-01-29 | 2005-04-07 | Nicola Ancona | System and method for the measurement of the relative position of an object with respect to a point of reference |
| CN103308000A (en) * | 2013-06-19 | 2013-09-18 | 武汉理工大学 | Method for measuring curve object on basis of binocular vision |
| CN106982359A (en) * | 2017-04-26 | 2017-07-25 | 深圳先进技术研究院 | A kind of binocular video monitoring method, system and computer-readable recording medium |
| CN107092893A (en) * | 2017-04-28 | 2017-08-25 | 无锡北斗星通信息科技有限公司 | A kind of recognition methods based on image procossing |
| CN108154134A (en) * | 2018-01-11 | 2018-06-12 | 天格科技（杭州）有限公司 | Internet live streaming pornographic image detection method based on depth convolutional neural networks |
| WO2018155777A1 (en) * | 2017-02-22 | 2018-08-30 | 한국과학기술원 | Apparatus and method for estimating distance on basis of thermal image, and neural network learning method therefor |
| CN108665484A (en) * | 2018-05-22 | 2018-10-16 | 国网山东省电力公司电力科学研究院 | A kind of dangerous source discrimination and system based on deep learning |
| CN109048926A (en) * | 2018-10-24 | 2018-12-21 | 河北工业大学 | A kind of intelligent robot obstacle avoidance system and method based on stereoscopic vision |
| CN109903507A (en) * | 2019-03-04 | 2019-06-18 | 上海海事大学 | A kind of fire disaster intelligent monitor system and method based on deep learning |
| CN110060299A (en) * | 2019-04-18 | 2019-07-26 | 中国测绘科学研究院 | Danger source identifies and positions method in passway for transmitting electricity based on binocular vision technology |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050271266A1 (en) * | 2001-06-01 | 2005-12-08 | Gregory Perrier | Automated rip current detection system |
| US9165453B2 (en) * | 2012-01-12 | 2015-10-20 | Earl Senchuk | Rip current sensor and warning system with anchor |
| KR101191944B1 (en) * | 2012-03-27 | 2012-10-17 | 대한민국(국토해양부 국립해양조사원장) | Method for issuing notice of warning for rip currents |
| CN104933718B (en) * | 2015-06-23 | 2019-02-15 | 广东省智能制造研究所 | A kind of physical coordinates localization method based on binocular vision |
| CN105389468B (en) * | 2015-11-06 | 2017-05-10 | 中国海洋大学 | Rip current forecasting method |
| JP2017133901A (en) * | 2016-01-27 | 2017-08-03 | ソニー株式会社 | Monitoring device and monitoring method, and program |
-
2019
- 2019-09-02 CN CN201910821151.6A patent/CN110675449B/en active Active
- 2019-11-05 WO PCT/CN2019/115513 patent/WO2021042490A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050074161A1 (en) * | 2001-01-29 | 2005-04-07 | Nicola Ancona | System and method for the measurement of the relative position of an object with respect to a point of reference |
| CN103308000A (en) * | 2013-06-19 | 2013-09-18 | 武汉理工大学 | Method for measuring curve object on basis of binocular vision |
| WO2018155777A1 (en) * | 2017-02-22 | 2018-08-30 | 한국과학기술원 | Apparatus and method for estimating distance on basis of thermal image, and neural network learning method therefor |
| CN106982359A (en) * | 2017-04-26 | 2017-07-25 | 深圳先进技术研究院 | A kind of binocular video monitoring method, system and computer-readable recording medium |
| CN107092893A (en) * | 2017-04-28 | 2017-08-25 | 无锡北斗星通信息科技有限公司 | A kind of recognition methods based on image procossing |
| CN108154134A (en) * | 2018-01-11 | 2018-06-12 | 天格科技（杭州）有限公司 | Internet live streaming pornographic image detection method based on depth convolutional neural networks |
| CN108665484A (en) * | 2018-05-22 | 2018-10-16 | 国网山东省电力公司电力科学研究院 | A kind of dangerous source discrimination and system based on deep learning |
| CN109048926A (en) * | 2018-10-24 | 2018-12-21 | 河北工业大学 | A kind of intelligent robot obstacle avoidance system and method based on stereoscopic vision |
| CN109903507A (en) * | 2019-03-04 | 2019-06-18 | 上海海事大学 | A kind of fire disaster intelligent monitor system and method based on deep learning |
| CN110060299A (en) * | 2019-04-18 | 2019-07-26 | 中国测绘科学研究院 | Danger source identifies and positions method in passway for transmitting electricity based on binocular vision technology |
Non-Patent Citations (2)
| Title |
|---|
| G.COSTANTINI 等: "A Binocular Sensor Interface for Moving Objects Detection", 《2007 2ND INTERNATIONAL WORKSHOP ON ADVANCES IN SENSORS AND INTERFACE》 * |
| 李彪: ""基于双目立体视觉三维重建技术研究"", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110763426A (en) * | 2019-09-29 | 2020-02-07 | 哈尔滨工程大学 | Method and device for simulating offshore flow in pool |
| CN110763426B (en) * | 2019-09-29 | 2021-09-10 | 哈尔滨工程大学 | Method and device for simulating offshore flow in pool |
| CN112950610A (en) * | 2021-03-18 | 2021-06-11 | 河海大学 | Method and system for monitoring and early warning of fission flow |
| CN113936248A (en) * | 2021-10-12 | 2022-01-14 | 河海大学 | Beach personnel risk early warning method based on image recognition |
| CN113936248B (en) * | 2021-10-12 | 2023-10-03 | 河海大学 | Beach personnel risk early warning method based on image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110675449B (en) | 2020-12-08 |
| WO2021042490A1 (en) | 2021-03-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110675449B (en) | Binocular camera-based offshore flow detection method | |
| CN104091369B (en) | Unmanned aerial vehicle remote-sensing image building three-dimensional damage detection method | |
| CN102494733B (en) | Water level monitoring system based on image processing and method | |
| CN109100741A (en) | A kind of object detection method based on 3D laser radar and image data | |
| CN106877237A (en) | A kind of method of insulator missing in detection transmission line of electricity based on Aerial Images | |
| CN109029381A (en) | A kind of detection method of tunnel slot, system and terminal device | |
| CN107862735B (en) | RGBD three-dimensional scene reconstruction method based on structural information | |
| CN104483326A (en) | High-voltage wire insulator defect detection method and high-voltage wire insulator defect detection system based on deep belief network | |
| CN109472802B (en) | Surface mesh model construction method based on edge feature self-constraint | |
| CN110298227B (en) | Vehicle detection method in unmanned aerial vehicle aerial image based on deep learning | |
| CN102032875A (en) | Image-processing-based cable sheath thickness measuring method | |
| CN103927758B (en) | Saliency detection method based on contrast ratio and minimum convex hull of angular point | |
| CN110009610A (en) | A kind of reservoir dam slope protection surface damage visible detection method and bionic device | |
| CN107358632A (en) | Underwater Camera scaling method applied to underwater binocular stereo vision | |
| CN104599281B (en) | A kind of based on the conforming panorama sketch in horizontal linear orientation and remote sensing figure method for registering | |
| CN103871072A (en) | Method for automatic extraction of orthoimage embedding line based on projection digital elevation models | |
| CN107631782A (en) | A kind of level testing methods based on Harris Corner Detections | |
| CN115060343B (en) | Point cloud-based river water level detection system and detection method | |
| CN103679740B (en) | ROI (Region of Interest) extraction method of ground target of unmanned aerial vehicle | |
| CN113902792A (en) | Building height detection method and system based on improved RetinaNet network and electronic equipment | |
| CN109544609A (en) | A kind of sidescan-sonar image matching process based on SIFT algorithm | |
| CN209708175U (en) | A kind of reservoir dam slope protection surface damage vision-based detection bionic device | |
| CN112101211A (en) | Personnel and suspension arm position calculation method based on target detection and binocular ranging | |
| CN115790539A (en) | Underwater photogrammetry method for cooperative target | |
| CN112964193A (en) | Novel bridge deformation monitoring method and system |
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: 20221130 Address after: 266555 Building 50, No. 1208, Qichangcheng Road, Huangdao District, Qingdao, Shandong Patentee after: Qingdao Jianguo Zhongji Surveying and Mapping Technology Information Co.,Ltd. Address before: 579 qianwangang Road, Huangdao District, Qingdao City, Shandong Province Patentee before: SHANDONG University OF SCIENCE AND TECHNOLOGY |