CN102572220A - Bionic compound eye moving object detection method adopting new 3-2-2 spatial information conversion model - Google Patents
Bionic compound eye moving object detection method adopting new 3-2-2 spatial information conversion model Download PDFInfo
- Publication number
- CN102572220A CN102572220A CN201210048378XA CN201210048378A CN102572220A CN 102572220 A CN102572220 A CN 102572220A CN 201210048378X A CN201210048378X A CN 201210048378XA CN 201210048378 A CN201210048378 A CN 201210048378A CN 102572220 A CN102572220 A CN 102572220A
- Authority
- CN
- China
- Prior art keywords
- moving target
- bionic
- camera
- information
- bionic compound
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a three-level physical realization means for dividing biological vision into a three-level structure and conducting bionic object detection, and discloses a bionic compound eye moving object detection method adopting a new 3-2-2 spatial information conversion model. A low-resolution imaging sensor technology based on a bionic compound eye structure is adopted to preliminarily detect moving objects to obtain three-dimensional outline information, so that redundant information is reduced by orders of magnitude; a high-resolution imaging sensor technology based on a bionic compound eye function is adopted to stare at an object area to accurately identify the moving objects, and identified images are stabilized to obtain two-dimensional high-definition data of the moving objects, so that image stabilization tracking control over the moving objects and identification control over the moving objects from three-dimensional outlines to two-dimension details are realized under the condition that the information quantity is not increased, three-dimensional data are lowered to two-dimensional data, and the three-dimensional precise data quantity is reduced, thereby realizing the bionic means such as remote sensing observation, and the like, which has the advantages of realizing the maintenance of high frequency detail information and lossless compression of effective target information. The three-level physical realization means can be widely used in the acquisition and detection processes of various moving objects.
Description
Technical field
The present invention relates to a kind of moving target detecting method; Particularly opened up remote sensing information temporal resolution new application, and be useful in the wild environment bionic compound eyes moving object detection of the 3-3-2 spatial information conversion new model of catching, discern moving target about a kind of.
Background technology
Existing method of carrying out moving object detection through single high-resolution lens images acquired; Receive the constraint of imaging sensor and plane image expression; The first 3 dimension information with real space of general needs change 2 dimension information into and detect moving target; Carrying out 3 dimension information of image processing reproduction of moving target then on computers, is the information translation pattern of 3-2-3 with this procedure definition usually.3-2-3 information translation pattern not only receives hardware constraints; And amount of calculation is big; Real-time is also poor, in the transfer process of 3 dimensional objects to 2 dimension image, can lose a large amount of detailed information, makes the accuracy rate that detects lower greatly; And in the process of 3 d reconstruction, can increase many redundant datas again at 2 dimension images; Not only do not had practical significance but also caused the sharp increase of data volume, and particularly realized becoming the data transaction of dimension twice of the 3-2-3 information translation, will inevitably produce model error through transformation model.In order not make model error influence high frequency minutia information, can only use more redundant information, all remain model error, the lossless compression method of this reserved high-frequency details almost gets into the technological blind alley that can't have an effect.More serious is; If between object of observation and the observation carrier relative motion is arranged; Data redundancy when moving target is caught and discerned so will be bigger on the basis of the above, and Real time identification and monitoring be difficulty more just, and this also is the bottleneck place that the high time resolution technology of being difficult to of remote sensing information realizes.
Summary of the invention
To the problems referred to above; The purpose of this invention is to provide that a kind of real-time is good, detection efficiency is high, detection speed is fast, the little moving target detecting method of required detection data volume; The method is the bionic compound eyes moving object detection that a kind of brand-new 3-3-2 spatial information is changed new model, has filled up the technological gap that remote sensing field temporal resolution is used.
For realizing above-mentioned purpose; The present invention takes following technical scheme: a kind of bionic compound eyes moving object detection of 3-3-2 spatial information conversion new model; May further comprise the steps: 1. be provided with and include the bionic compound eyes synchronous imaging device that a plurality of low resolution cameras and high-resolution camera are combined into sphere; Said bionic compound eyes synchronous imaging device also includes control system; Said control system includes DSP master control core cell, fpga logic control unit and graphics processing unit, and said graphics processing unit comprises profile detection algorithm, coupling algorithm for reconstructing and surely looks like algorithm; 2. said bionic compound eyes synchronous imaging device is placed on and observes on the carrier, said observation carrier carries said bionic compound eyes synchronous imaging device the moving target in the wild environment is carried out preliminary position probing; 3. said DSP master control core cell carries out label respectively to said high-resolution and low resolution camera, writes down all adjacent two low initial positions of differentiating camera in the said low resolution camera simultaneously; When 4. beginning the capture movement target; DSP master control core cell transmits a signal to the fpga logic control unit; The fpga logic control unit transmit control signal control all low resolution camera the moving target in the wild environment is carried out preliminary position probing; And the data of gathering are sent to graphics processing unit, graphics processing unit obtains the three-D profile point cloud of moving target through coupling; The camera site of the Position Control high-resolution camera that 5. feeds back according to the low data of differentiating camera collection, and adopt high-resolution camera that the moving target position is accurately extracted and surely look like algorithm process, moving target clearly obtained.
Said step detailed process 5. is: the initial position that said DSP master control core cell will photograph two adjacent low resolution cameras of moving target primary data sends to the fpga logic control unit; Fpga logic control unit control high-resolution camera rotates to said initial position; The fpga logic control unit transmits control signal and controls the staring imaging that the real-time taking moving target of high-resolution camera is accomplished moving target simultaneously, and the data of gathering are sent to the clear two dimensional image that obtains moving target after graphics processing unit is handled.
The present invention is owing to take above technical scheme; It has the following advantages: when 1, adopting bionic compound eyes moving target detecting method of the present invention that target in the wild environment is discerned; The camera that at first adopts a plurality of low resolution carries out Preliminary detection to the three-D profile of moving target; The mode of utilizing the high-resolution camera to stare through two dimension after the preliminary lock motion target is accurately discerned moving target; The data acquisition modes of distinguishing through this variation not only can effectively reduce data volume, improves detection efficiency, but also can solve the contradictory problems of visual field, resolution and real-time.2, the present invention is based on bionic insect compound eye detection mechanism moving target is carried out detection method; Not only can become order of magnitude ground to reduce the redundancy of three-dimensional data; Make the exponential quantity reduction of amount of information, the minimizing of spatial information dimension; And can be at short notice capture movement target rapidly, the information processing bottleneck problem of effectively having avoided the 3-2-3 translative mode to bring.The present invention can be widely used in various moving targets catch with testing process in.
Description of drawings
Fig. 1 is the principle exploded view that insect compound eye of the present invention detects moving target;
Fig. 2 is the physics realization classification figure of bionic compound eyes measuring ability of the present invention;
Fig. 3 is a bionic compound eyes synchronous imaging apparatus structure sketch map of the present invention;
Fig. 4 is a 3-3-2 data capture method sketch map of the present invention.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is carried out detailed description.
As shown in Figure 1; The compound eye of occurring in nature insect is a variable resolution sphere imaging system; The mechanism of capture movement target in the insect lowered in field environment: at first each ommatidium of insect strengthens through the lateral inhibition edge elementary motion detection is carried out in the position of moving target in the wild environment; Capture movement Three-dimension Target profile is stablized the effect that steady picture is played in the visual field through the visual servo control system adjustment compound eye attitude of self then fast, and it is zone (the large scene data acquisition that the lock motion target belongs to this moment that moving target is locked to Snazzi band (ommatidium close quarters); Rough detection); Last in the Snazzi band insect turn to the Snazzi band that moving target is carried out staring imaging (little contextual data collection, the smart extraction) through self its high-resolution ommatidium of apparent motion control system control, thereby follow the tracks of the capture movement target.
Like Fig. 2, shown in 3, the mechanism that the present invention is based on insect capture movement target proposes bionic compound eyes and detects principle, respectively from high to low successively through function bionics, control is bionical and bionic structure detects moving target.At first one bionic compound eyes synchronous imaging device is set through bionic structure imitation insect compound eye imaging arrangement; Bionic compound eyes synchronous imaging device can by a center be poroid folder establish device and around this folder establish device circumferentially extended six supports form; Folder establishes that vertical fixing has high-resolution camera in the hole of device; There is a low resolution camera end of six supports vertical fixing respectively; And the shooting angle of each low resolution camera is adjustable; The shooting angle of seven cameras is all regulated through a neural control system of function bionics imitation insect compound eye; The function of the quick capture movement target through bionic insect is carried out the Preliminary detection and the high-resolution accurate identification of low resolution respectively to the moving target in the wild environment; Wherein control system includes a DSP master control core cell, a fpga logic control unit and a graphics processing unit; DSP master control core cell configures various parameters and transmits a signal to the fpga logic control unit, and the rotation of fpga logic control unit output control signal control bionic compound eyes synchronous imaging device is also controlled each camera taking moving target simultaneously, accomplishes collection, transmission and the storage of data; And the data of storage are sent to graphics processing unit and handle, graphics processing unit comprises that the profile to moving target detects, the moving target three-dimensional point cloud is rebuild and the realization of steady picture algorithm.
In the foregoing description, the concrete structure of bionic compound eyes synchronous imaging device and the number of camera can be provided with according to the needs of actual acquisition.
The present invention's specific embodiment as shown in Figure 4; Adopt bionic compound eyes synchronous imaging device promptly to adopt the camera of six low resolution that the moving target in the wild environment is carried out Preliminary detection; Gather the three-D profile data of moving target, utilize high-resolution camera to stare target area pursuit movement target according to collection result then, obtain the two-dimentional high definition data of moving target; The present invention adopts 3-3-2 information translation pattern to the motion target detection method, may further comprise the steps:
1) with bionic compound eyes synchronous imaging device be placed on one observe carrier the bottom; And bionic compound eyes synchronous imaging device is can observe carrier relatively to rotate; The observation carrier carries bionic compound eyes synchronous imaging device Preliminary detection is carried out in the position of the moving target in the wild environment, observes carrier and can adopt space shuttle.
2) DSP master control core cell carries out label respectively to all high-resolution and low resolution camera; Write down plane and the angle of horizontal plane, the i.e. initial position of definite all two adjacent low resolution cameras at the support place of every adjacent two low resolution camera in six low resolution camera simultaneously.
As shown in Figure 4; DSP master control core cell can be 0 with the labelled notation of high-resolution camera; Circumferentially six adjacent low resolution camera are labeled as 1,2,3,4,5 and 6 successively successively, and every adjacent two of record places the low plane and the angle of horizontal plane of differentiating the support place of cameras, and this is the initial position of low resolution camera; The initial position of all two adjacent low resolution cameras has six angles, and angle can be got acute angle.
When 3) beginning the capture movement target; DSP master control core cell transmits a signal to the fpga logic control unit; The fpga logic control unit transmit control signal control six circumferential low resolution camera the moving target in the lowered in field environment of going is carried out preliminary position probing; And the data of gathering are sent to graphics processing unit, and graphics processing unit obtains the three-D profile point cloud of moving target through coupling, and detailed process is:
The camera that the fpga logic control unit is controlled six low resolution simultaneously detects the moving target on the road in the open air of going with multi-angle; And the image of six low resolution camera collections of extract real-time; If wherein some cameras of circumferential six cameras for example moving target occurs in the field range of camera 1; Check in the camera 2 or 6 that circumferentially is adjacent whether have moving target this moment; If with do not find moving target in its circumferentially adjacent camera 2 or 6; Then continue to search; Up to its circumferentially adjacent camera 2 or 6 in moving target (target that for example occurs moving object in the camera 6 simultaneously) also appears simultaneously, the image of the moving target that gather camera 1 and the camera 6 that is adjacent respectively this moment is stored and is sent to graphics processing unit and carries out three-dimensional reconstruction, obtains the sparse three-dimensional point cloud profile of moving target.
The camera site of the Position Control high-resolution camera that 4) feeds back according to the low data of differentiating camera collection, and adopt high-resolution camera that the moving target position is accurately extracted and surely look like algorithm process, moving target clearly obtained.
DSP master control core cell is that the angle of initial position is that α sends to the fpga logic control unit with the angle of camera 1 and camera plane, 6 place and horizontal plane; Fpga logic control unit control high-resolution camera rotation alpha makes the high-resolution camera rotate to camera 1 and plane, camera 6 place; The lock motion target area; The real-time taking moving target of the high-resolution camera of control that transmits control signal is simultaneously accomplished the staring imaging of moving target, and the data of gathering are sent to graphics processing unit through surely obtain the clear two dimensional image of moving target as algorithm process.
Above-mentioned each embodiment only is used to explain the present invention; Wherein the structure of each parts, connected mode and implementation method etc. all can change to some extent; Every equivalents of on the basis of technical scheme of the present invention, carrying out and improvement all should not got rid of outside protection scope of the present invention.
Claims (6)
1. the bionic compound eyes moving object detection of 3-3-2 spatial information conversion new model comprises with the lower part:
1) employing is carried out the three-D profile information that preliminary position probing obtains moving target based on the bionical low resolution imaging sensor technology of compound eye structural to moving target in the wild environment; Realized that the simple three-D profile under the complex three-dimensional observing environment characterizes, made amount of information become the order of magnitude to reduce;
2) adopt and to stare the target area based on the bionical high-resolution imaging sensor technology of compound eye structural moving target is accurately discerned; And the image of identification is carried out surely as processing; Obtain the two-dimentional high definition data of moving target; Not only realized under the condition that does not increase amount of information steady picture tracking Control, and realized the identification control of the three-D profile of moving target, made the three-dimensional data amount be reduced to two dimension to two-dimentional details to moving target; Three-dimensional fine data amount reduces, and has realized reserved high-frequency detailed information such as remote sensing observations, has realized that the lossless compress of effective target provides the bionics means.
2. the bionic compound eyes moving object detection of 3-3-2 spatial information conversion new model as claimed in claim 1, it is characterized in that: the realization of bionics mechanism can be divided into according to biological vision: 1) the elementary detection of variable resolution sphere of visual field low resolution moving target fast detecting greatly; 2) apparent motion of visual servo control; 3) the tertiary structure method that detects in the Snazzi band of moving target identification, and three level physicals of corresponding with it bionic structure-big view field imaging transducer, function bionics-image stabilization, principle and function bionics-moving object detection are realized the division means.
3. according to claim 1 or claim 2 the bionic compound eyes moving object detection of 3-3-2 spatial information conversion new model, it is characterized in that: concrete implementation procedure is:
1. the bionic compound eyes synchronous imaging device that includes a plurality of low resolution cameras and a high-resolution camera combination is set; Said bionic compound eyes synchronous imaging device also includes control system; Said control system includes DSP master control core cell, fpga logic control unit and graphics processing unit, and said graphics processing unit comprises profile detection algorithm, coupling algorithm for reconstructing and surely looks like algorithm;
2. said bionic compound eyes synchronous imaging device is placed on and observes on the carrier, said observation carrier carries said bionic compound eyes synchronous imaging device the moving target in the wild environment is carried out preliminary position probing;
3. said DSP master control core cell carries out label respectively to said high-resolution and low resolution camera, writes down all adjacent two low initial positions of differentiating camera in the said low resolution camera simultaneously;
When 4. beginning the capture movement target; DSP master control core cell transmits a signal to the fpga logic control unit; The fpga logic control unit transmit control signal control all low resolution camera the moving target in the wild environment is carried out preliminary position probing; And the data of gathering are sent to graphics processing unit, graphics processing unit obtains the three-D profile point cloud of moving target through coupling;
The camera site of the Position Control high-resolution camera that 5. feeds back according to the low data of differentiating camera collection, and adopt high-resolution camera that the moving target position is accurately extracted and surely look like algorithm process, moving target clearly obtained.
4. the bionic compound eyes moving object detection of 3-3-2 spatial information conversion new model as claimed in claim 3; It is characterized in that: said step detailed process 5. is: the initial position that said DSP master control core cell will photograph two adjacent low resolution cameras of moving target primary data sends to the fpga logic control unit; Fpga logic control unit control high-resolution camera rotates to said initial position; The fpga logic control unit transmits control signal and controls the staring imaging that the real-time taking moving target of high-resolution camera is accomplished moving target simultaneously, and the data of gathering are sent to the clear two dimensional image that obtains moving target after graphics processing unit is handled.
5. change the bionic compound eyes moving object detection of new model like each described 3-3-2 spatial information of claim 1～4; It is characterized in that: it at first converts 3 dimension point clouds into by 3 dimension moving target objects and carries out a whole set of moving target that 2 dimensions stare at last and catch the bionics method with imaging representation; Catch the exponential rising of the amount of information that stores 3 d reconstruction into respect to conventional 3-2-3 pattern by 3 dimensional objects to 2 dimension; 3-3-2 information translation pattern has realized the minimizing of spatial information dimension and the exponential reduction of amount of information; Fundamentally improved the efficient that detects, using for the temporal resolution of remotely-sensed data provides a kind of new model.
6. change the bionic compound eyes moving object detection of new model like the described 3-3-2 spatial information of each claim of claim 1～5; It is characterized in that:, can realize that big visual field, multiresolution, fast-moving target detect 1) based on the bionical high low resolution combined imaging sensor technology of compound eye structural; 2) profile based on compound eye mechanism characterizes detection method; 3) manage and the FEEDBACK CONTROL means based on the steady camera of bionic compound eyes profile servo tracking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210048378XA CN102572220A (en) | 2012-02-28 | 2012-02-28 | Bionic compound eye moving object detection method adopting new 3-2-2 spatial information conversion model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210048378XA CN102572220A (en) | 2012-02-28 | 2012-02-28 | Bionic compound eye moving object detection method adopting new 3-2-2 spatial information conversion model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102572220A true CN102572220A (en) | 2012-07-11 |
Family
ID=46416611
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210048378XA Pending CN102572220A (en) | 2012-02-28 | 2012-02-28 | Bionic compound eye moving object detection method adopting new 3-2-2 spatial information conversion model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102572220A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103108164A (en) * | 2013-02-01 | 2013-05-15 | 南京迈得特光学有限公司 | Compound eye type panorama continuous tracking and monitoring system |
| CN104202504A (en) * | 2014-08-19 | 2014-12-10 | 昆明理工大学 | Processing method of real-time electronic image stabilization circuit system based on FPGA (Field Programmable Gate Array) |
| CN105227929A (en) * | 2015-10-16 | 2016-01-06 | 中国民航科学技术研究院 | A kind of safety monitoring device for airport |
| CN105352482A (en) * | 2015-11-02 | 2016-02-24 | 北京大学 | Bionic compound eye microlens technology-based 3-3-2 dimension object detection method and system |
| CN107105147A (en) * | 2017-06-05 | 2017-08-29 | 北京理工大学 | A kind of bionical super-resolution imaging sensor and imaging method |
| CN110456377A (en) * | 2019-08-15 | 2019-11-15 | 中国人民解放军63921部队 | It is a kind of that foreign matter detecting method and system are attacked based on the satellite of three-dimensional laser radar |
| WO2021027323A1 (en) * | 2019-08-14 | 2021-02-18 | 北京理工大学 | Hybrid image stabilization method and device based on bionic eye platform |
| CN112669396A (en) * | 2020-12-18 | 2021-04-16 | 深圳智慧林网络科技有限公司 | Image lossless compression method and device |
| CN112998693A (en) * | 2021-02-01 | 2021-06-22 | 上海联影医疗科技股份有限公司 | Head movement measuring method, device and equipment |
| CN113612929A (en) * | 2021-10-11 | 2021-11-05 | 北京创米智汇物联科技有限公司 | Compound-eye imaging device, control method thereof, electronic apparatus, and storage medium |
| CN113892130A (en) * | 2019-06-07 | 2022-01-04 | 马自达汽车株式会社 | Moving body external environment recognition device |
| CN116681732A (en) * | 2023-08-03 | 2023-09-01 | 南昌工程学院 | Target motion recognition method and system based on compound eye morphological vision |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1932841A (en) * | 2005-10-28 | 2007-03-21 | 南京航空航天大学 | Petoscope based on bionic oculus and method thereof |
| US20110075018A1 (en) * | 2009-09-29 | 2011-03-31 | Fujifilm Corporation | Compound-eye image pickup apparatus |
-
2012
- 2012-02-28 CN CN201210048378XA patent/CN102572220A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1932841A (en) * | 2005-10-28 | 2007-03-21 | 南京航空航天大学 | Petoscope based on bionic oculus and method thereof |
| US20110075018A1 (en) * | 2009-09-29 | 2011-03-31 | Fujifilm Corporation | Compound-eye image pickup apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 高鹏骐: "无人机仿生复眼运动目标检测机理与方法研究", 《万方数据知识服务平台》, 31 May 2010 (2010-05-31) * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103108164A (en) * | 2013-02-01 | 2013-05-15 | 南京迈得特光学有限公司 | Compound eye type panorama continuous tracking and monitoring system |
| CN104202504A (en) * | 2014-08-19 | 2014-12-10 | 昆明理工大学 | Processing method of real-time electronic image stabilization circuit system based on FPGA (Field Programmable Gate Array) |
| CN104202504B (en) * | 2014-08-19 | 2017-06-06 | 昆明理工大学 | A kind of processing method of the real time electronic steady image circuit system based on FPGA |
| CN105227929A (en) * | 2015-10-16 | 2016-01-06 | 中国民航科学技术研究院 | A kind of safety monitoring device for airport |
| CN105352482A (en) * | 2015-11-02 | 2016-02-24 | 北京大学 | Bionic compound eye microlens technology-based 3-3-2 dimension object detection method and system |
| CN105352482B (en) * | 2015-11-02 | 2017-12-26 | 北京大学 | 332 dimension object detection methods and system based on bionic compound eyes micro lens technology |
| CN107105147A (en) * | 2017-06-05 | 2017-08-29 | 北京理工大学 | A kind of bionical super-resolution imaging sensor and imaging method |
| CN107105147B (en) * | 2017-06-05 | 2019-06-21 | 北京理工大学 | A kind of bionical super-resolution imaging sensor and imaging method |
| CN113892130A (en) * | 2019-06-07 | 2022-01-04 | 马自达汽车株式会社 | Moving body external environment recognition device |
| WO2021027323A1 (en) * | 2019-08-14 | 2021-02-18 | 北京理工大学 | Hybrid image stabilization method and device based on bionic eye platform |
| CN110456377A (en) * | 2019-08-15 | 2019-11-15 | 中国人民解放军63921部队 | It is a kind of that foreign matter detecting method and system are attacked based on the satellite of three-dimensional laser radar |
| CN110456377B (en) * | 2019-08-15 | 2021-07-30 | 中国人民解放军63921部队 | Satellite foreign matter attack detection method and system based on three-dimensional laser radar |
| CN112669396A (en) * | 2020-12-18 | 2021-04-16 | 深圳智慧林网络科技有限公司 | Image lossless compression method and device |
| CN112669396B (en) * | 2020-12-18 | 2023-09-12 | 深圳智慧林网络科技有限公司 | Lossless image compression method and device |
| CN112998693A (en) * | 2021-02-01 | 2021-06-22 | 上海联影医疗科技股份有限公司 | Head movement measuring method, device and equipment |
| CN112998693B (en) * | 2021-02-01 | 2023-06-20 | 上海联影医疗科技股份有限公司 | Head movement measuring method, device and equipment |
| CN113612929A (en) * | 2021-10-11 | 2021-11-05 | 北京创米智汇物联科技有限公司 | Compound-eye imaging device, control method thereof, electronic apparatus, and storage medium |
| CN116681732A (en) * | 2023-08-03 | 2023-09-01 | 南昌工程学院 | Target motion recognition method and system based on compound eye morphological vision |
| CN116681732B (en) * | 2023-08-03 | 2023-10-20 | 南昌工程学院 | Target motion recognition method and system based on compound eye morphological vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102572220A (en) | Bionic compound eye moving object detection method adopting new 3-2-2 spatial information conversion model | |
| Li et al. | Deep learning-based object detection techniques for remote sensing images: A survey | |
| WO2020119140A1 (en) | Method, apparatus and smart device for extracting keyframe in simultaneous localization and mapping | |
| CN109685768B (en) | Pulmonary nodule automatic detection method and system based on pulmonary CT sequence | |
| CN110706248A (en) | Visual perception mapping algorithm based on SLAM and mobile robot | |
| CN107481315A (en) | A kind of monocular vision three-dimensional environment method for reconstructing based on Harris SIFT BRIEF algorithms | |
| CN106529538A (en) | Method and device for positioning aircraft | |
| CN110637461B (en) | Compact optical flow handling in computer vision systems | |
| CN103426165A (en) | Precise registration method of ground laser-point clouds and unmanned aerial vehicle image reconstruction point clouds | |
| CN110084243A (en) | It is a kind of based on the archives of two dimensional code and monocular camera identification and localization method | |
| CN103099623A (en) | Extraction method of kinesiology parameters | |
| CN105760831A (en) | Pedestrian tracking method based on low-altitude aerial photographing infrared video | |
| CN110852182A (en) | Depth video human body behavior recognition method based on three-dimensional space time sequence modeling | |
| CN109543576A (en) | Train driver detection method based on bone detection and three-dimensional reconstruction | |
| CN105516661B (en) | Principal and subordinate's target monitoring method that fisheye camera is combined with ptz camera | |
| JP2019185787A (en) | Remote determination of containers in geographical region | |
| CN107038758A (en) | A kind of augmented reality three-dimensional registration method based on ORB operators | |
| CN111105486A (en) | Multi-view-angle dynamic three-dimensional reconstruction method for mice | |
| DE102021130626A1 (en) | TRIGGER-RESPONSIVE CLIP EXTRACTION BASED ON REMOTE ANALYSIS | |
| US20220215620A1 (en) | Geospatial modeling system providing 3d geospatial model update based upon predictively registered image and related methods | |
| Drobnitzky et al. | Survey and systematization of 3D object detection models and methods | |
| CN103175527A (en) | Large-visual-field and low-power-consumption earth sensor system applied to microsatellite | |
| CN105352482A (en) | Bionic compound eye microlens technology-based 3-3-2 dimension object detection method and system | |
| Marek | Image-based parking space occupancy classification: Dataset and baseline | |
| CN109492513B (en) | Face space duplication eliminating method for light field monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120711 |