CN103020590A - Vehicle identification system and method based on three-dimensional model and image matching - Google Patents
Vehicle identification system and method based on three-dimensional model and image matching Download PDFInfo
- Publication number
- CN103020590A CN103020590A CN2012104721999A CN201210472199A CN103020590A CN 103020590 A CN103020590 A CN 103020590A CN 2012104721999 A CN2012104721999 A CN 2012104721999A CN 201210472199 A CN201210472199 A CN 201210472199A CN 103020590 A CN103020590 A CN 103020590A
- Authority
- CN
- China
- Prior art keywords
- image
- vehicle
- dimensional model
- checked
- online
- 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
Abstract
The invention provides a vehicle identification system and a vehicle identification method based on three-dimensional model and image matching. The method comprises the following steps of performing initial detection location to an appointed object in a real-time video by the online algorithm, and performing more accurate analysis and detection to the detection result by the offline algorithm. According to the invention, the three-dimensional model of the vehicle is converted to be a two-dimensional image with object structure information at a specific visual angle and then is subjected to matching detection with the image to be detected provided by the online algorithm; meanwhile, for the successful matching result, the structure information can be provided; by adoption of the vehicle identification method, the high accuracy of the online vehicle detection result; in addition, the disadvantages that the offline vehicle detection method is complex and can not be used in a real-time system are overcome; and the purpose of obtaining real-time accurate detection results can be achieved by combining the online and offline detection methods to form complementary of algorithms.
Description
Technical field
The present invention relates to the technical field of vehicle identification system, particularly a kind of vehicle identification system and method thereof based on three-dimensional model and images match.
Background technology
Along with the foreign country that holds in New York at the beginning of 2008 11 months concerns in the council, president IBM Peng Mingsheng has delivered the key-note speech of " earth of wisdom: leader's agenda of future generation ", and " the wisdom earth " and " wisdom city " just becomes the focus of attracting attention in the current world gradually." wisdom city " is that the developing intelligent digital city management with insertion type, interactive function in digital city is used; its core content is as the basis take the data activation; set up dynamic data center, city; application by the wisdom technology; make the city productive life with dynamic, real-time, mode is implemented management more accurately; resource utilization and labour productivity can improve greatly; energy resource consumption rate and disposal of pollutants rate can obviously descend; resources and environment will obtain more effective protection, and the relation of man and nature will be more harmonious.
Under this background, vehicle and road synthetic are got up to consider, use systematically solving road traffic problems of various new and high technologies, produced thus new research and application---intelligent transportation system ITS (Intelligent Transportation System).ITS carries out effective integration with advanced person's technology for information acquisition, data communication technology, automatic control technology, the information processing technology and data activating technology etc., and combine with traffic programme, traffic engineering and management, apply to whole traffic control system and set up a kind of on a large scale, the comprehensive comprehensive intelligent control and management system that transports in real time, accurately and efficiently that plays a role.
Simultaneously, although online vehicle checking method is very ripe, but can not guarantee the high correctness of testing result; On the other hand, although the off-line vehicle detection can provide the testing result of high correctness, because its method is very complicated, can't in real-time system, use.If yet can combine both the complementation of formation algorithm, can reach and obtain in real time the accurately purpose of testing result.The present invention produces under this idea.
Summary of the invention
The technical problem to be solved in the present invention is: a kind of vehicle identification system based on three-dimensional model and images match and method thereof are provided, guarantee online vehicle detection result's high correctness; On the other hand, overcome the off-line vehicle checking method complicated, the shortcoming that can't use in real-time system combines both the complementation of formation algorithm, then can reach to obtain in real time the accurately purpose of testing result.
The technical solution used in the present invention: a kind of vehicle identification system based on three-dimensional model and images match comprises: online processing host, processed offline main frame, these hardware devices of camera and operate in image capture module, online processing module on the online processing host, operate in three software modules of processed offline module on the processed offline main frame; Image capture module is by the realtime image data in camera collection vehicle monitoring zone, for online processing module analysis; Online processing module is carried out fast detecting to the image that image capture module collects, and locates the vehicle object that wherein comprises, and testing result is offered the processed offline module; The processed offline module is done the testing result of online processing module and more accurate analyzed and detect, at last with result feedback to the user;
Wherein, described online processing host is one of core component of this system, and it links to each other with camera simultaneously and the processed offline main frame links to each other; In addition, the image capture module of native system, online processing module also are to run on the online processing host; Described processed offline main frame is one of core component of this system, and it links to each other with online processing host simultaneously; In addition, the processed offline module of native system also is to run on the processed offline main frame; Described camera is the harvester of native system view data, and shooting area is vehicle monitoring zone, its view data that collects take frame as unit passes to online processing host, the Data Source of processing online as online processing host;
Further, the regional zone that can cover for described image capture module of described vehicle monitoring.
Further, described off-line target detection adopts the vehicle identification based on three-dimensional model and images match, namely utilize the three-dimensional model of vehicle as sorter, when carrying out target detection, location parameter and image parameter to be checked according to current camera, choose suitable visual angle three-dimensional model is converted into two dimensional image with object structures information, and the matching degree between the resulting image of the pre-service of Two-dimensional image and image to be checked, thereby realize the off-line vehicle checking method.
Further, this system mainly comprises quick location and the accurate measuring ability of vehicle, and described system reaches the separately shortcoming that remedies two kinds of detection methods by online and off-line checking method are merged, the purpose of the in real time accurate testing result of acquisition.
The present invention provides a kind of vehicle identification method based on three-dimensional model and images match in addition, it is characterized in that: vehicle identification comprises three steps: the coupling of the pre-service of image to be checked, three-dimensional model and image to be checked, demarcation Output rusults;
The pre-treatment step of described image to be checked is carried out noise reduction and gray processing pretreatment operation to providing view data;
The coupling step of described three-dimensional model and image to be checked is according to location parameter and the image parameter to be checked of current camera, choose suitable visual angle three-dimensional model is converted into two dimensional image with object structures information and the matching degree between the resulting image of the pre-treatment step of Two-dimensional image and image to be checked;
Described demarcation Output rusults step is according to the testing result of the coupling step of three-dimensional model and image to be checked, the structural information in the two dimensional image is used for the architectural feature of demarcating image to be checked, export at last detected object with and structural information.
Further, the location parameter of described current camera is the information such as the height of image capture device that image to be checked is provided, angle.
Further, the information such as assumed position of the described image parameter to be checked size that is image, resolution, vehicle.
The present invention is with the advantage that traditional vehicle detecting system is compared:
1, the present invention can take full advantage of system resource, improves simultaneously the accuracy that real-time vehicle detects.Because when on road specific vehicle being detected, vehicle to be checked not necessarily can all appear in the visual field of camera all the time, has in this case the resource of many time systems to be in idle state; And when vehicle to be checked occurs, require again system in real time it to be captured, therefore need to use online object detection method, its inevitable requirement that needs to sacrifice accuracy and guarantee real-time.And method of the present invention can catch in real time to the target that occurs, utilize simultaneously the idle time to use off-line algorithm that target is analyzed more accurately and detected, can reach the purpose that takes full advantage of system resource and improve the real-time vehicle detection accuracy.
2, the present invention can reduce the sample data amount of sorter.The target vehicle sample that does not need a large amount of difformities and state based on the vehicle target recognition methods of three-dimensional model and images match, vehicle vector three-dimensional model of its needs, and generate two dimensional image by projected position and the angle information that calculates, this can reach simplifies sorter, the purpose that reduces to take up room.
3, the structural information that the present invention can export target.Because the vehicle structure information that the three-dimensional model of vehicle comprises has been delivered in the two dimensional image that generates equally, so after itself and images match to be checked, this information also can be delivered on the vehicle of identifying, thereby can when detecting whole vehicle, obtain the structural informations such as this vehicle glazing, car door, car light.
Description of drawings
Fig. 1 is a kind of structural representation of the vehicle identification system based on three-dimensional model and images match;
Fig. 2 is a kind of workflow diagram of the vehicle identification system based on three-dimensional model and images match;
Fig. 3 is image capture module Filter link schematic diagram;
Fig. 4 is online module of target detection workflow diagram;
Fig. 5 is off-line module of target detection workflow diagram.
Embodiment
Based on main design idea of the present invention, system of the present invention comprises: online processing host, processed offline main frame, the first-class hardware device of shooting and operate in image capture module, online processing module on the online processing host, operate in three software modules of processed offline module on the processed offline main frame.
The quick location of vehicle and accurate testing process are as follows: image capture module is by the realtime image data in camera collection vehicle monitoring zone, for online processing module analysis; Online processing module is carried out fast detecting to the image that image capture module collects, and locates the vehicle object that wherein comprises, and testing result is offered the processed offline module; The processed offline module is done the testing result of online processing module and more accurate analyzed and detect, at last with result feedback to the user.
Wherein, described online processing host is one of core component of this system, and it links to each other with camera simultaneously and the processed offline main frame links to each other; In addition, the image capture module of native system, online processing module also are to run on the online processing host.
Described processed offline main frame is one of core component of this system, and it links to each other with online processing host simultaneously; In addition, the processed offline module of native system also is to run on the processed offline main frame.
Described camera is the harvester of native system view data, and shooting area is vehicle monitoring zone, its view data that collects take frame as unit passes to online processing host, the Data Source of processing online as online processing host.
Described image capture module is finished the collection of view data.The relevant interface structure Filter link that can adopt the DirectShow technology to provide is realized image data acquiring.
Described online processing module namely utilizes online algorithm of target detection that the image that camera transmits is carried out target detection.If detect vehicle, then determine its size and coordinate.
Described processed offline module namely take result that online processing module was provided as input, carries out analyzing and detecting more accurately to image by the off-line object detection method.
Further, described off-line object detection method adopts the vehicle identification method based on three-dimensional model and images match, namely utilizes the three-dimensional model of vehicle as sorter.When carrying out target detection, location parameter and image parameter to be checked according to current camera, choose suitable visual angle three-dimensional model is converted into two dimensional image with object structures information, and the matching degree between the resulting image of the pre-treatment step of Two-dimensional image and image to be checked, thereby realize the off-line vehicle checking method.
Further, described vehicle size, namely the pixel of the detected rectangular area that comprises vehicle is wide and pixel is high.The coordinate of vehicle, namely with the coordinate of certain any the pixel coordinate in the rectangular area as whole vehicle, for example the central point with rectangular area, vehicle place represents vehicle coordinate.
Based on the vehicle identification method of three-dimensional model and images match, its characteristics are:
System of the present invention comprises: vehicle identification comprises three steps: the coupling of the pre-service of image to be checked, three-dimensional model and image to be checked, demarcation Output rusults.
The identification process of vehicle is as follows: providing view data is carried out noise reduction, the pretreatment operation such as gray processing; According to location parameter and the image parameter to be checked of current camera, choose suitable visual angle three-dimensional model is converted into two dimensional image with object structures information and the matching degree between Two-dimensional image and the resulting image of pre-service; According to testing result, the structural information in the two dimensional image is used for the architectural feature demarcate image to be checked, export at last detected object with and structural information.
The present invention is described in detail below in conjunction with drawings and the specific embodiments.
As shown in Figure 1, the native system hardware device comprises online processing host, processed offline main frame and camera.Camera collection vehicle monitoring area image also sends online processing host to; Online processing host is done preliminary processing to image, detects in real time vehicle wherein, and sends the result to the off-line main frame; The off-line main frame is by the three-dimensional model of vehicle, is translated into the two dimensional image with object structures information at specific visual angle, and the result that online detection main frame is provided does more accurately and detects, and net result is fed back to the user.
As shown in Figure 2, be a kind of workflow diagram of the vehicle identification system based on three-dimensional model and images match.
Step 2-1 extracts view data, and this step is finished by image capture module, adopts the DirectShow technology to realize.
Step 2-2, Preliminary detection is namely utilized online algorithm of target detection, and the view data that camera transmits is carried out analyzing and processing, whether has vehicle in the fast detecting present image.
Step 2-3 judges whether to detect vehicle; If do not detect, leap to step 2-5.
Step 2-4 sends to the off-line main frame, namely take the subimage of detected vehicle as parameter, shares the mode of data by network or database, and these data are sent to the processed offline main frame.
Step 2-5 checks end mark, if true, then program withdraws from; Otherwise, jump to step 2-1 and continue operation.
Step 2-6 obtains the data of online main frame, namely according to the employed shared data mode of two sub-systems, reads the subimage of the vehicle that online main frame sends over from buffer zone or database.
Step 2-7 accurately detects, and namely utilizes the vehicle identification algorithm based on three-dimensional model and images match, and view data is carried out analyzing and processing, detects in the present image whether have corresponding vehicle.
Step 2-8 judges whether to detect vehicle; If do not detect, leap to step 2-10.
Step 2-9 feeds back to the user, and the classification and the structural information that are about to detected vehicle feed back to the user.
Step 2-10 checks end mark, if true, then program withdraws from; Otherwise, jump to step 2-6 and continue operation.
As shown in Figure 3, be image capture module Filter link schematic diagram.
Described Filter link is the interface that utilizes DirectShow to provide, and defines the Filter link that three Filter consist of a connection.View data flows in the Filter link, intercepts and preserve the collection that these data are finished view data.Wherein, three Filter are respectively: Device Source Filter(device data Filter), be made of camera; Transform Filter is a Sample Grabber Filter(sample collection Filter), utilize its GetCurrentBuffer () to obtain view data, whenever call once this function, just can obtain a current two field picture; Last Render Filter is that a Null RenderFilter(sky is played up Filter), it just simply abandons view data, is left intact.These three Filter need to add in the management through figures device.Then use the intelligent link technology among the DirectShow that three Filter are connected into a unimpeded data channel.
As shown in Figure 4, be online module of target detection workflow diagram.
Step 4-1 detects vehicle, utilizes characteristic in the vehicle classification device and the zone to be detected in the image to carry out characteristic matching;
Step 4-2 if there is vehicle, then jumps to step 5-3; Otherwise online target detection process finishes;
Step 4-3 according to testing result, calculates size and the coordinate of vehicle.
As shown in Figure 5, be off-line module of target detection workflow diagram.
Step 5-1 generates two dimensional image, according to the parameter of importing into and image to be detected, three-dimensional model is converted into two dimensional image with object structures information at specific visual angle.
Step 5-2 detects vehicle, and the two dimensional image that utilizes 5-1 to generate carries out characteristic matching as the zone to be detected in sorter and the image to be checked;
Step 5-3 if there is vehicle, then jumps to step 5-4; Otherwise off-line target detection process finishes;
Step 5-4 obtains the structural information of vehicle, according to the object structures information in the two dimensional image, the structure of image to be checked is demarcated;
Step 5-5, Output rusults is exported to the user with testing result and structural information.
Claims (7)
1. vehicle identification system based on three-dimensional model and images match is characterized in that: comprising: online processing host, processed offline main frame, these hardware devices of camera and operate in image capture module, online processing module on the online processing host, operate in three software modules of processed offline module on the processed offline main frame; Image capture module is by the realtime image data in camera collection vehicle monitoring zone, for online processing module analysis; Online processing module is carried out fast detecting to the image that image capture module collects, and locates the vehicle object that wherein comprises, and testing result is offered the processed offline module; The processed offline module is done the testing result of online processing module and more accurate analyzed and detect, at last with result feedback to the user;
Wherein, described online processing host is one of core component of this system, and it links to each other with camera simultaneously and the processed offline main frame links to each other; In addition, the image capture module of native system, online processing module also are to run on the online processing host; Described processed offline main frame is one of core component of this system, and it links to each other with online processing host simultaneously; In addition, the processed offline module of native system also is to run on the processed offline main frame; Described camera is the harvester of native system view data, and shooting area is vehicle monitoring zone, its view data that collects take frame as unit passes to online processing host, the Data Source of processing online as online processing host.
2. the vehicle identification system based on three-dimensional model and images match according to claim 1 is characterized in that: the zone that described vehicle monitoring zone can cover for described image capture module.
3. the vehicle identification system based on three-dimensional model and images match according to claim 1, it is characterized in that: described off-line target detection adopts the vehicle identification based on three-dimensional model and images match, namely utilize the three-dimensional model of vehicle as sorter, when carrying out target detection, location parameter and image parameter to be checked according to current camera, choose suitable visual angle three-dimensional model is converted into two dimensional image with object structures information, and the matching degree between the resulting image of the pre-service of Two-dimensional image and image to be checked, thereby realize the off-line vehicle checking method.
4. the vehicle identification system based on three-dimensional model and images match according to claim 1, it is characterized in that: this system mainly comprises quick location and the accurate measuring ability of vehicle, described system is by merging online and off-line checking method, reach the separately shortcoming that remedies two kinds of detection methods, obtain in real time the accurately purpose of testing result.
5. vehicle identification method based on three-dimensional model and images match, it is characterized in that: vehicle identification comprises three steps: the coupling of the pre-service of image to be checked, three-dimensional model and image to be checked, demarcate Output rusults;
The pre-treatment step of described image to be checked is carried out noise reduction and gray processing pretreatment operation to providing view data;
The coupling step of described three-dimensional model and image to be checked is according to location parameter and the image parameter to be checked of current camera, choose suitable visual angle three-dimensional model is converted into two dimensional image with object structures information and the matching degree between the resulting image of the pre-treatment step of Two-dimensional image and image to be checked;
Described demarcation Output rusults step is according to the testing result of the coupling step of three-dimensional model and image to be checked, the structural information in the two dimensional image is used for the architectural feature of demarcating image to be checked, export at last detected object with and structural information.
6. the vehicle identification method based on three-dimensional model and images match according to claim 5 is characterized in that: the location parameter of described current camera is the information such as the height of image capture device that image to be checked is provided, angle.
7. the vehicle identification method based on three-dimensional model and images match according to claim 5 is characterized in that: the information such as assumed position of the size that described image parameter to be checked is image, resolution, vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210472199.9A CN103020590B (en) | 2012-11-20 | 2012-11-20 | A kind of vehicle identification system based on three-dimensional model and images match and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210472199.9A CN103020590B (en) | 2012-11-20 | 2012-11-20 | A kind of vehicle identification system based on three-dimensional model and images match and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103020590A true CN103020590A (en) | 2013-04-03 |
| CN103020590B CN103020590B (en) | 2015-12-09 |
Family
ID=47969181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210472199.9A Expired - Fee Related CN103020590B (en) | 2012-11-20 | 2012-11-20 | A kind of vehicle identification system based on three-dimensional model and images match and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103020590B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809437A (en) * | 2015-04-28 | 2015-07-29 | 无锡赛睿科技有限公司 | Real-time video based vehicle detecting and tracking method |
| CN106710221A (en) * | 2017-03-15 | 2017-05-24 | 深圳市鹏之艺建筑设计有限公司 | Smart city intelligent transportation system and evaluation method thereof |
| CN109871761A (en) * | 2019-01-16 | 2019-06-11 | 中德（珠海）人工智能研究院有限公司 | A kind of identifying system of threedimensional model |
| CN110070025A (en) * | 2019-04-17 | 2019-07-30 | 上海交通大学 | Objective detection system and method based on monocular image |
| CN111368651A (en) * | 2020-02-18 | 2020-07-03 | 杭州海康威视系统技术有限公司 | Vehicle identification method and device and electronic equipment |
| CN112488178A (en) * | 2020-11-26 | 2021-03-12 | 推想医疗科技股份有限公司 | Network model training method and device, image processing method and device, and equipment |
| CN113283268A (en) * | 2020-02-19 | 2021-08-20 | 百度在线网络技术（北京）有限公司 | Image processing method and device |
| CN113865665A (en) * | 2021-09-23 | 2021-12-31 | 华帝股份有限公司 | Method for detecting water level and device applying same |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2251104A1 (en) * | 1997-11-13 | 1999-05-13 | Michael Greenspan | Method of recognising objects within two-dimensional and three-dimensional images |
| CN1506903A (en) * | 2002-12-06 | 2004-06-23 | 中国科学院自动化研究所 | Automatic fingerprint distinguishing system and method based on template learning |
| CN101261722A (en) * | 2008-01-17 | 2008-09-10 | 北京航空航天大学 | Electronic police background intelligent management and automatic implementation system |
| CN101286240A (en) * | 2008-04-22 | 2008-10-15 | 北京航空航天大学 | Quickly configurable and real time rendering safety protection three-dimensional electronic map |
| CN101951502A (en) * | 2010-10-19 | 2011-01-19 | 北京硅盾安全技术有限公司 | Three-dimensional intelligent video monitoring method |
| US8090160B2 (en) * | 2007-10-12 | 2012-01-03 | The University Of Houston System | Automated method for human face modeling and relighting with application to face recognition |
-
2012
- 2012-11-20 CN CN201210472199.9A patent/CN103020590B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2251104A1 (en) * | 1997-11-13 | 1999-05-13 | Michael Greenspan | Method of recognising objects within two-dimensional and three-dimensional images |
| CN1506903A (en) * | 2002-12-06 | 2004-06-23 | 中国科学院自动化研究所 | Automatic fingerprint distinguishing system and method based on template learning |
| US8090160B2 (en) * | 2007-10-12 | 2012-01-03 | The University Of Houston System | Automated method for human face modeling and relighting with application to face recognition |
| CN101261722A (en) * | 2008-01-17 | 2008-09-10 | 北京航空航天大学 | Electronic police background intelligent management and automatic implementation system |
| CN101286240A (en) * | 2008-04-22 | 2008-10-15 | 北京航空航天大学 | Quickly configurable and real time rendering safety protection three-dimensional electronic map |
| CN101951502A (en) * | 2010-10-19 | 2011-01-19 | 北京硅盾安全技术有限公司 | Three-dimensional intelligent video monitoring method |
Non-Patent Citations (1)
| Title |
|---|
| 熊斌等: "《基于三维模型的前视红外目标匹配识别方法》", 《激光与红外》, vol. 41, no. 3, 31 March 2011 (2011-03-31), pages 344 - 350 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809437B (en) * | 2015-04-28 | 2018-04-13 | 无锡赛睿科技有限公司 | A kind of moving vehicles detection and tracking method based on real-time video |
| CN104809437A (en) * | 2015-04-28 | 2015-07-29 | 无锡赛睿科技有限公司 | Real-time video based vehicle detecting and tracking method |
| CN106710221A (en) * | 2017-03-15 | 2017-05-24 | 深圳市鹏之艺建筑设计有限公司 | Smart city intelligent transportation system and evaluation method thereof |
| CN109871761A (en) * | 2019-01-16 | 2019-06-11 | 中德（珠海）人工智能研究院有限公司 | A kind of identifying system of threedimensional model |
| CN110070025B (en) * | 2019-04-17 | 2023-03-31 | 上海交通大学 | Monocular image-based three-dimensional target detection system and method |
| CN110070025A (en) * | 2019-04-17 | 2019-07-30 | 上海交通大学 | Objective detection system and method based on monocular image |
| CN111368651A (en) * | 2020-02-18 | 2020-07-03 | 杭州海康威视系统技术有限公司 | Vehicle identification method and device and electronic equipment |
| CN111368651B (en) * | 2020-02-18 | 2024-03-08 | 杭州海康威视系统技术有限公司 | Vehicle identification method and device and electronic equipment |
| CN113283268A (en) * | 2020-02-19 | 2021-08-20 | 百度在线网络技术（北京）有限公司 | Image processing method and device |
| CN112488178A (en) * | 2020-11-26 | 2021-03-12 | 推想医疗科技股份有限公司 | Network model training method and device, image processing method and device, and equipment |
| CN112488178B (en) * | 2020-11-26 | 2022-07-26 | 推想医疗科技股份有限公司 | Network model training method and device, image processing method and device, and equipment |
| CN113865665B (en) * | 2021-09-23 | 2023-11-21 | 华帝股份有限公司 | Method for detecting water level and device applying same |
| CN113865665A (en) * | 2021-09-23 | 2021-12-31 | 华帝股份有限公司 | Method for detecting water level and device applying same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103020590B (en) | 2015-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103020590B (en) | A kind of vehicle identification system based on three-dimensional model and images match and method thereof | |
| CN102915432B (en) | A kind of vehicle-mounted microcomputer image/video data extraction method and device | |
| CN112990310A (en) | Artificial intelligence system and method for serving electric power robot | |
| CN102915638A (en) | Surveillance video-based intelligent parking lot management system | |
| CN104751110A (en) | Bio-assay detection method and device | |
| CN110119726A (en) | A kind of vehicle brand multi-angle recognition methods based on YOLOv3 model | |
| CN105022999A (en) | Man code company real-time acquisition system | |
| CN104616002A (en) | Facial recognition equipment used for judging age groups | |
| CN111160202A (en) | AR equipment-based identity verification method, AR equipment-based identity verification device, AR equipment-based identity verification equipment and storage medium | |
| KR102309111B1 (en) | Ststem and method for detecting abnomalous behavior based deep learning | |
| CN108064389B (en) | A kind of target identification method, device and intelligent terminal | |
| CN110276265A (en) | Pedestrian monitoring method and device based on intelligent three-dimensional solid monitoring device | |
| CN113284144B (en) | Tunnel detection method and device based on unmanned aerial vehicle | |
| CN112200077A (en) | Artificial intelligent image processing method and system based on intelligent traffic | |
| CN111476192A (en) | Intercepted image synthesis method based on intelligent traffic and big data cloud server | |
| CN102789719A (en) | Diver sight line recognition instrument in road driving skill test process and realizing method thereof | |
| CN111797787B (en) | Waste image detection and classification system based on Internet of things technology | |
| CN115567563B (en) | Comprehensive transportation hub monitoring and early warning system based on end edge cloud and control method thereof | |
| CN115862128A (en) | Human body skeleton-based customer abnormal behavior identification method | |
| WO2020027778A1 (en) | Neural network identification of objects in 360-degree images | |
| CN115908886A (en) | Image classification method, image processing apparatus, and storage device | |
| CN111310602A (en) | System and method for analyzing attention of exhibit based on emotion recognition | |
| CN113496200A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN206946507U (en) | A kind of machine vision passenger flow statistic device based on thermal imaging | |
| CN209216110U (en) | Support to realize the device of the top-type passenger flow monitor processing based on stereovision technique |
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 |
Granted publication date: 20151209 Termination date: 20191120 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |