CN101685014B - Device and method for sensing object position - Google Patents
Device and method for sensing object position Download PDFInfo
- Publication number
- CN101685014B CN101685014B CN2008101618082A CN200810161808A CN101685014B CN 101685014 B CN101685014 B CN 101685014B CN 2008101618082 A CN2008101618082 A CN 2008101618082A CN 200810161808 A CN200810161808 A CN 200810161808A CN 101685014 B CN101685014 B CN 101685014B
- Authority
- CN
- China
- Prior art keywords
- image
- dimensional signal
- signal information
- process unit
- sensing object
- 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.)
- Active
Links
Images
Abstract
The invention discloses a device and a method for rapidly sensing an object position, in particular the device and the method for sensing the object position directly applied to various image systems. The invention mainly aims at carrying out image pick-up on an area in a direction to be sensed through the image system according to a user, and an image processing unit judges the position and relevant information of the object through selection of images of scene areas and an object recognition method. The method for processing image data comprises a step of converting two-dimension image datainto one-dimension signal information of a distance axle, and object recognition comprises steps of judging the position of the object by means of utilizing numerical value differential estimation tothe one-dimension signal information, and judging the approach state of the object to the difference of the one-dimension signal information of two adjacent images which are continuously input. The algorithm has small computation and favorable transportability.
Description
Technical field
The invention relates to a kind of sensing object position device and method, particularly a kind of device and method that utilizes image system to do the dynamic object position detecting.
Background technology
Along with development of science and technology, the modern is used in many daily lifes object spacing detection techniques cover to improve the convenience and the security of life.For instance, because the modern is more and more exquisite for the security of automobile, many in recent years depots propose technology such as new vehicle safety framework and crash protection device successively.Wherein be again one of technology that attracts most attention with the vehicle collision safety pre-warning system, therefore in recent years, many Vehicle security systems are developed in succession, such as: lane bias alarm, reversing aid system and vehicle blind spot caution system.Wherein, especially remarkable with vehicle blind spot caution system for the effect of security protection, paid attention to by market, wherein, above-mentioned so-called blind spot be meant the vehicle side rear because of the visual limit of rearview mirror angle cause human eye can't look and the zone, it is that mainly the video camera by two sides about being placed in and has judged whether vehicle to approach and proposes caution with the image in acquisition blind spot district, vehicle side rear, and the measurement means of therefore approaching vehicle distances are to be that this institute of system must application person; With regard to the distance measuring means, being widely used in most the arrangement for detecting that object approaches in the market is range radar, range radar mainly can be divided into ultrasonic/sonic wave range finding, infrared distance measuring etc., it mainly is that electromagnetism or acoustic wave energy are emitted among the space with oriented approach, by receiving the energy that exists object to reflect in the space, can calculate the direction of this object, height and speed.And can survey the shape of object, be the accurate shape that the radar of target can be surveyed ground with ground.Yet the visual field of Vehicle radar is narrower usually, along with the difference that quantity is installed will cause different dead angles.Come for example with common radar for backing car, the radar for backing car of the dead angle of the radar for backing car of four eyed than two is also little many, but installation cost also exceeds one times relatively, and the distance range that radar for backing car can be detected is limited, and the dynamic object that is difficult to be applied to is in a big way detected.Another kind of dynamic object position detecting apparatus is an installed video, can effectively overcome the problem at dead angle, allows driving person under steam, sees the image at each dead angle by display device.Yet in the dynamic vehicle identification in the various installed videos visual field, mostly utilize image gradient estimation and edge detection to find out vehicle location at present, with blob analysis definition vehicle possible position, its identification accuracy remains to be improved again; Or utilizing figure comparison mode to find the search of carrying out city, different images district, its comparison identification capability is set up the property of usefulness excessively of example edition according to it and is decided, and search time is longer; In addition or calculate various image feature values in different index modes and carry out the foundation of artificial intelligence pattern, carry out distinguishing of determinand again, its technology is difficult for carrying out the transplanting of embedded system.
Yet in the installed video system, the required reaction time of System Discrimination is to be close to instant message feedback, and the identification of system calculation rule do not influence by the external environment light source should, all can possess reliability to a certain degree.Yet with above-mentioned three kinds of methods, must expend huge image processing resource, and for satisfying the problem of different images situation at day night, need to use two kinds of different algorithms with in response to it usually.Therefore for addressing the above problem, the present invention proposes a kind of arrangement for detecting and method of object space, be converted into a kind of one-dimensional signal information of distance axis by the road image that bidimensional image that image system captured such as installed video system are captured, by comparison one of adjacent two images micro-signal information difference and behind the one-dimensional signal differential value of the single time point distance axis behind the differential numerical estimation, and then can not be subjected to have judged whether that precisely object approaches under the weather restriction.The present invention not only can overlap and be used for existing image system, and higher precision more is provided, and lower manufacturing cost.
Summary of the invention
Fundamental purpose of the present invention is a kind of sensing object position device and method to be provided, to need not to use radar, and can overcome existing greatly based on the object detecting method calculated amount of image, and real-time is poor, the shortcoming high to System Hardware Requirement.
Sensing object position method of the present invention is a kind ofly to utilize the complexity evaluation method to convert the one-dimensional signal information of distance axis to institute's picked image, divide estimation to obtain the one-dimensional signal differential value of single time point distance axis this one-dimensional signal information via numerical value dimension again, and one-dimensional signal information gap value and this one-dimensional signal differential value of the more at least two adjacent times of passing through, judge whether the state that object exists and object approaches.
Sensing object position device of the present invention mainly is made up of an at least one image acquisition unit and an image process unit.Image acquisition unit is responsible for the desire detection area of at least one direction is carried out image capture.Image process unit then is the one-dimensional signal information that institute's pick-up image is converted to distance axis by image complexity estimation mode, again one-dimensional signal information is carried out the numerical differentiation estimation to obtain the one-dimensional signal differential value of single time point distance axis, can judge the position of this object when this one-dimensional signal differential value more than or equal to a preset threshold value.And, can judge the state that approaches of this object more than or equal to a preset threshold value when this difference by calculating the one-dimensional signal information gap value of the adjacent time point of two-phase at least.In addition, image process unit then can further be connected to a caution unit, and this caution unit can send caution by screen display device, light-emittingdiode, hummer or loudspeaker to the user according to the judgement of object space.The present invention can directly overlap and be used for existing image platform, and more convenience of user and compatibility are provided.
Beneficial effect of the present invention is, by image information being changed into one-dimensional signal information, with one-dimensional signal information is carried out simple differential or difference computing and carries out the static and dynamically identification of object, calculated amount is little, calculation speed is fast, be convenient to detect in real time and low to hardware requirement, and algorithm also can realize by simple software, therefore, the present invention system that can almost allow all have image capture has the function of sensing object position under the situation that is not equipped with radar.Not only can allow all kinds of electronic systems significantly increase under the cost not needing, obtain the extension on the greater functionality, more can be because of not being subjected to the weather image in detecting, to possess more advantages of excellent stability and accuracy.
Beneath by the appended graphic explanation in detail of specific embodiment cooperation, when the effect that is easier to understand purpose of the present invention, technology contents, characteristics and is reached.
Description of drawings
Fig. 1 is a device architecture synoptic diagram of the present invention;
Fig. 2 approaches the detecting schematic flow sheet for object of the present invention;
Fig. 3 is a reconnaissance range synoptic diagram of the present invention;
Fig. 4 A-4D is a pick-up image synoptic diagram of the present invention;
Fig. 5 is that the object of example approaches the information processing synoptic diagram with the vehicle for the present invention;
Fig. 6 is the another embodiment of the present invention synoptic diagram.
Description of reference numerals:
The 2-image acquisition unit; The 4-image process unit; 6-warns the unit.
Embodiment
The present invention has disclosed a kind of sensing object position device and method, particularly a kind of desire detecting orientation is continued to carry out image capture, in complexity estimation mode institute's pick-up image is converted to the one-dimensional signal information of distance axis again, and one-dimensional signal information is carried out numerical differentiation estimate to obtain the one-dimensional signal differential value of single time point distance axis, calculate the difference of the one-dimensional signal information of the adjacent time point of two-phase at least at last, and whether judge the position of object according to this one-dimensional signal differential value more than or equal to a predetermined threshold value; Cooperate the state that approaches of whether judging object at least according to the one-dimensional signal information gap value of this two adjacent time point again more than or equal to a predetermined threshold value; Also can reach the purpose that the detecting dynamic object approaches state by this; Wherein above-mentioned two threshold values can be identical.
Fig. 1 is a system architecture synoptic diagram of the present invention, and sensing object position device of the present invention mainly is made up of an at least one image acquisition unit 2 and an image process unit 4, also can comprise a caution unit 6.Fig. 2 then is a detecting schematic flow sheet of the present invention, Fig. 5 is the object approximation signal processing synoptic diagram of example with the vehicle for the present invention, please refer to Fig. 2 and cooperation with reference to Fig. 5, at first when step S01, image acquisition unit 2 can carry out image capture at least one zone of desire detecting direction according to user's definition, and employed image acquisition unit 2 can be the photographic goods of common CCD or CMOS etc.2 picked image of image acquisition unit can enter in the image process unit 4, and according to the imagery zone of user definition, mode with the estimation of image complexity, for example the area image entropy is estimated mode, the image amount of information of obtaining the different distance position is shown in step S02, bidimensional image is converted to the one-dimensional signal information (I) of distance axis, in the embodiment of Fig. 5, this one-dimensional signal information is the track image amount of information among the figure, shown in solid line part among Fig. 5, below all with one-dimensional signal information explanation, shown in following equation 1.
In equation 1, I (x) is illustrated in the one-dimensional signal information that axial distance is the x position, in this formula, is the information entropy form, in concrete enforcement, can also have other formula formal definition that is used to characterize complexity now.X can or be converted into distance according to the pixel of the image that is captured.G
i(x) for the x place number appears for i gray-scale value axial distance in area image; T
Pixels(x) be axially apart from being x place valid pixel sum total in the area image; p
i(x) be gray-scale value i and probability occurs.Because on the characteristic of image amount of information, color and luster more less or dull more, its information magnitude is lower usually; Therefore if when also having other objects to occur on the orientation of detecting, then may provide higher image amount of information because of imagery zones such as object color, shadow of object or light modulation exist.And the use image amount of information is as the principle of object detecting, except allowing image acquisition unit 2 and image process unit 4 can not be subjected to weather to influence its function, mainly be significantly to reduce the required calculation function of image process unit 4, higher instant feedback rate is provided, reduce the system hardware demand.
After obtaining these one-dimensional signal information, image process unit 4 is shown in step S03, again one-dimensional signal information is carried out the numerical differentiation estimation to obtain the one-dimensional signal differential value of single time point distance axis, in this example, the track image information differential that this one-dimensional signal differential value is among Fig. 5 is handled, as dotted line among Fig. 5 partly shown in, below all illustrate with the one-dimensional signal differential value.Please refer to Fig. 5 left side, its Y-axis dotted line partly represent when have in institute's picked image automobile storage the time, the variation of this one-dimensional signal information and this one-dimensional signal differential value, wherein can see in the vehicle location place, this one-dimensional signal differential value is apparently higher than a threshold value, can learn by this automobile storage the position, the difference of the one-dimensional signal information of image process unit 4 by calculating the adjacent time point of two-phase is at least calculated the possibility that whether has object to approach according to this difference and this one-dimensional signal differential value more afterwards; Please refer to Fig. 5 right-hand part, its Y-axis dotted line is partly represented, when vehicle approaches, the variation of the one-dimensional signal information gap value of adjacent time point, wherein can see in approaching the position, this one-dimensional signal information difference is apparently higher than a threshold value, can learn the state that approaches of vehicle by this; So when this one-dimensional signal differential value and this one-dimensional signal information difference all were higher than this predetermined threshold value, then can confirm had vehicle to approach.
Wherein the present invention utilizes the one-dimensional signal quantity of information to judge and approaches the method for position shown in step S04 to S05, mainly be to utilize numerical differentiation estimation mode, try to achieve the graded (I ') of image amount of information, can utilize the particular preset threshold value to come the one-dimensional signal information gap value of comparison one-dimensional signal differential value and adjacent time point to find out the t object position (X that may exist constantly
t) shown in equation 2, wherein the one-dimensional signal information gap value of this one-dimensional signal differential value and adjacent two time points all is to be higher than this predetermined threshold value.
For instance, image at wide 5 meters and long 15 meters of reality, if the use image analytic degree is 320 * 240 camera unit 2, its default a distance values is desirable greatly about about 10 (pixels), if use the camera unit 2 of image analytic degree as 640x480, its default a distance values is desirable greatly about about 15 to 20 (pixels).This shows that camera unit 2 resolutions that use during the invention process are high more, then can the high more measurement spacing of corresponding use, so the user can set up required threshold value on their own according to the individual.For distinguishing the trend of object, in step S06, image process unit 4 of the present invention is with the image amount of information I of adjacent two image gained
tAnd I
T+1Subtract each other in the hope of Δ I
T+1By Δ I
T+1The estimation result can learn, when movement of objects to x
T+1During the position, its Δ I
T+1(x
T+1) be worth for maximum shown in equation 3.
ΔI
t+1(x)＝I
t+1(x)-I
t(x) (3)
The bidimensional image quantity of information is estimated as the one-dimensional signal information of distance axis through the image complexity, the line number of going forward side by side value is differentiated and is used and the one-dimensional signal information gap value of adjacent time point estimation result for the one-dimensional signal differential value of single time point distance axis, can find object in the continuous image if approach phenomenon, it position x occurs
T+1One-dimensional signal differential value I '
T+1(x
T+1) and the one-dimensional signal information gap value Δ I of adjacent time point
T+1(x
T+1) all have obvious crest to exist.
Therefore among one of preferred embodiment of the present invention, the present invention only need utilize relatively one-dimensional signal differential value I ' of a simple threshold values θ
T+1(X
T-1) and the one-dimensional signal information gap value Δ I ' of adjacent time point
T+1(X
T+1) behind the result, can obtain to approach the position identification result of object, if really for object approaches the result who is obtained, the distance of its this object position and position detecting apparatus will increase in time and reduce, getting final product further cognition has during an object approaches really.At last in step S07, if image process unit 4 judges that it approaches distance when reaching the caution scope that the user sets, can further be attached to a caution unit 6, and by giving the user suitable alarm signal as screen display apparatus, light emitting diode, hummer or loudspeaker.
Fig. 3 is one of preferred embodiment of the present invention, mainly is to be used in the existing vehicle-mounted camera chain with the detecting that whether has vehicle or other objects to approach cover of the present invention as discussed previously.As shown in the figure, common vehicle detection orientation is mainly positions such as the place ahead, rear, front side and side, back, therefore image acquisition unit is set respectively in these places of a vehicle also continue to these direction pick-up images as shown in Figure 4; The demand that wherein meets most present vehicle especially with vehicle blind spot caution system, and the present invention is also applicable to the blind spot caution system, it is that image acquisition unit is placed in vehicle two rearview mirror positions, and image acquisition unit is towards the blind spot district, driver visual angle of vehicle two proceeds posterolateral, uses approaching vehicle and caution is provided in the detecting blind spot district.As described in the above-mentioned flow process, institute's picked image can be converted to the one-dimensional signal information of distance axis earlier by image complexity estimation mode afterwards, converts the one-dimensional signal differential value of single time point distance axis again to the numerical differentiation estimation.As shown in Figure 5, because on the characteristic of image information, if only have grey road surface and white lane line to exist, its information magnitude is lower, but if also have on the track automobile storage the time, then may have imagery zones such as body color, vehicle body shade, wheel rim or light modulation and exist, contain higher quantity of information.Therefore the difference of the one-dimensional signal information by calculating the adjacent time point of two-phase at least, the judgement of carrying out the dynamic object position that just can success, and do not using under any radar situation the detecting that realization vehicle object approaches.Be used for any installed video system because image treatment method used in the present invention can overlap, the while is a small amount of digital operating ability of need only, so the present invention can directly apply to vehicle-mounted image platforms such as DSP, FPGA, SoC (SOC (system on a chip)) and Car PC.Utilize the object of image process unit institute construction to approach detecting and caution in addition, its image can be via behind system's adjustment interface setting, take the image information in the different visuals field, and the shadow that or else is subjected to weather and light continues to differentiate the state of alert of this image downwards, and send the caution with warning message give the user, realize a kind of position detecting apparatus and method of round-the-clock object.
Fig. 6 is another preferred embodiment of the present invention, mainly is the present invention is used in laptop computer or the desktop computer with video signal function.As shown in the figure, present many laptop computers are equipped with a video signal camera lens in screen top or other orientation.If allow computer have the position detecting function, can provide operating system and user more convenience.For instance, the user can allow household PC be set at when the personage approaches, and opens automatically and logins picture, and when the user leaves, enter screen protection pattern or park mode etc. automatically.Method for detecting is then identical, main is exactly by video signal camera lens pick-up image, institute's picked image can be converted to the one-dimensional signal information of distance axis earlier by image complexity estimation mode, convert the one-dimensional signal differential value of single time point distance axis again to the numerical differentiation estimation.The difference of the one-dimensional signal information by calculating the adjacent time point of two-phase at least, just the judgement of carrying out dynamic user position that can be successful and approach state.Same theory also can be overlapped and is used for surveillance, when general surveillance many after the sensing object position usefulness, higher security and expansionary can be provided.For instance, general cash machine is if having the sensing object position function, except can allow the user time near cash machine, automatically open functions such as user's greeting or advertising pronunciations, more can allow the user when withdrawing deposit, whether Auto-Sensing has other people to approach, and drawing person is provided higher security.Because distance detecting algorithm biggest advantage proposed by the invention is used for most image acquisition systems for overlapping, and algorithm also can realize by simple software, in other words, the present invention system that can almost allow all have image capture has the function of sensing object position under the situation that is not equipped with radar.Not only can allow all kinds of electronic systems significantly increase under the cost not needing, obtain the extension on the greater functionality, more can be because of not being subjected to the weather image in detecting, to possess more advantages of excellent stability and accuracy, be a kind of invention in fact with high convenience.
The above is preferred embodiment of the present invention only, is not to be used for limiting scope of the invention process.So be that all equalizations of doing according to the described shape of claim of the present invention, structure, feature and spirit change or modification, all should be included in the claim of the present invention.
Claims (13)
1. sensing object position device comprises:
At least one image acquisition unit captures the image at least one zone; An and image process unit, it is characterized in that, described image process unit is converted to institute's pick-up image the one-dimensional signal information of distance axis, and this one-dimensional signal information carried out numerical differentiation estimation to obtain the one-dimensional signal differential value of single time point distance axis, by this with the position of judgment object;
This image process unit is by image complexity estimation mode institute's pick-up image to be converted to dimensional signal information;
This image complexity estimation mode is an area image entropy estimation mode.
2. sensing object position device according to claim 1 is characterized in that, this image acquisition unit is CCD or CMOS image capture unit.
3. sensing object position device according to claim 1 is characterized in that this image process unit is provided with a threshold value, and according to when this one-dimensional signal differential value during more than or equal to this threshold value, judges the position of object.
4. sensing object position device according to claim 1 is characterized in that, this image process unit is attached to a caution unit, and this caution unit sends caution according to the judgement of this object space to the user.
5. sensing object position device according to claim 4 is characterized in that, this caution unit is warned with proposition by screen display apparatus, light emitting diode, hummer or loudspeaker.
6. sensing object position device comprises:
At least one image acquisition unit captures the image at least one zone; An and image process unit, it is characterized in that, described image process unit is converted to institute's pick-up image the one-dimensional signal information of distance axis, and this one-dimensional signal information is carried out numerical differentiation estimate to obtain the one-dimensional signal differential value of single time point distance axis, calculate this one-dimensional signal information again in the difference of at least two adjacent time points, by this with the state that approaches of judgment object;
This image process unit is by image complexity estimation mode institute's pick-up image to be converted to one-dimensional signal information;
This image complexity estimation mode is an area image entropy estimation mode.
7. sensing object position device according to claim 6 is characterized in that this image process unit is preset with a threshold value, and according to when this one-dimensional signal differential value during more than or equal to this threshold value, judges the position of object.
8. sensing object position device according to claim 7 is characterized in that, this image process unit is judged the state that approaches of object according to when this one-dimensional signal information difference of two adjacent time points during more than or equal to this threshold value.
9. sensing object position device according to claim 7 is characterized in that, this image process unit is attached to a caution unit, and this caution unit sends caution according to the judgement of this object space to the user.
10. a sensing object position method is characterized in that, comprises following steps
Capture the image at least one zone;
Institute's pick-up image is converted to the one-dimensional signal information of distance axis;
This one-dimensional signal information is carried out the numerical differentiation estimation to obtain the one-dimensional signal differential value of single time point distance axis; And
Position by this one-dimensional signal differential value judgment object;
This pick-up image is the one-dimensional signal information that is converted to this distance axis by image complexity estimation mode;
This image complexity estimation mode is an area image entropy estimation mode.
11. sensing object position method according to claim 10 is characterized in that, this one-dimensional signal differential value is judged the position of object during more than or equal at least one predetermined threshold value.
12. sensing object position method according to claim 10 is characterized in that, the step of the position of this judgment object comprises and calculates this one-dimensional signal information in the difference of two adjacent time points.
13. sensing object position method according to claim 12 is characterized in that, the step of the position of this judgment object comprises establishes a threshold value, when this difference during more than or equal to this threshold value, judges the state that approaches of object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101618082A CN101685014B (en) | 2008-09-22 | 2008-09-22 | Device and method for sensing object position |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101618082A CN101685014B (en) | 2008-09-22 | 2008-09-22 | Device and method for sensing object position |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101685014A CN101685014A (en) | 2010-03-31 |
| CN101685014B true CN101685014B (en) | 2011-06-29 |
Family
ID=42048289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101618082A Active CN101685014B (en) | 2008-09-22 | 2008-09-22 | Device and method for sensing object position |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101685014B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2015226202B2 (en) * | 2014-03-03 | 2019-07-25 | Vsk Electronics Nv | Intrusion detection with motion sensing |
| TWI618647B (en) * | 2016-02-02 | 2018-03-21 | 財團法人資訊工業策進會 | System and method of detection, tracking and identification of evolutionary adaptation of vehicle lamp |
| CN108873100A (en) * | 2018-08-15 | 2018-11-23 | 深圳市枫芒科技有限公司 | A kind of seat occupancy detection device and method |
| TWI730861B (en) * | 2020-07-31 | 2021-06-11 | 國立勤益科技大學 | Warning method of social distance violation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1488224A (en) * | 2001-11-19 | 2004-04-07 | ������������ʽ���� | Feature extraction and detection of events and temporal variations in activity in video sequences |
| CN101161524A (en) * | 2006-10-12 | 2008-04-16 | 财团法人车辆研究测试中心 | Method and apparatus for detecting vehicle distance |
| CN101236657A (en) * | 2008-03-03 | 2008-08-06 | 吉林大学 | Single movement target track tracking and recording method |
| CN101266132A (en) * | 2008-04-30 | 2008-09-17 | 西安工业大学 | Running disorder detection method based on MPFG movement vector |
-
2008
- 2008-09-22 CN CN2008101618082A patent/CN101685014B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1488224A (en) * | 2001-11-19 | 2004-04-07 | ������������ʽ���� | Feature extraction and detection of events and temporal variations in activity in video sequences |
| CN101161524A (en) * | 2006-10-12 | 2008-04-16 | 财团法人车辆研究测试中心 | Method and apparatus for detecting vehicle distance |
| CN101236657A (en) * | 2008-03-03 | 2008-08-06 | 吉林大学 | Single movement target track tracking and recording method |
| CN101266132A (en) * | 2008-04-30 | 2008-09-17 | 西安工业大学 | Running disorder detection method based on MPFG movement vector |
Non-Patent Citations (3)
| Title |
|---|
| JP特开平7-301523A 1995.11.14 |
| JP特开平8-241498A 1996.09.17 |
| JP特开平8-30897A 1996.02.02 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101685014A (en) | 2010-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Discant et al. | Sensors for obstacle detection-a survey | |
| CN102765365B (en) | Pedestrian detection method based on machine vision and pedestrian anti-collision warning system based on machine vision | |
| US8077198B2 (en) | Object position detection device and method thereof | |
| CN103852067B (en) | The method for adjusting the operating parameter of flight time (TOF) measuring system | |
| CN103852754B (en) | The method of AF panel in flight time (TOF) measuring system | |
| US8908038B2 (en) | Vehicle detection device and vehicle detection method | |
| EP2602640B1 (en) | Vehicle occupancy detection using time-of-flight sensor | |
| US7672514B2 (en) | Method and apparatus for differentiating pedestrians, vehicles, and other objects | |
| US20050232463A1 (en) | Method and apparatus for detecting a presence prior to collision | |
| Lee et al. | Stereo vision–based vehicle detection using a road feature and disparity histogram | |
| CN107662872A (en) | The monitoring system and its monitoring method of passenger conveyor | |
| US20050270286A1 (en) | Method and apparatus for classifying an object | |
| Arenado et al. | Monovision‐based vehicle detection, distance and relative speed measurement in urban traffic | |
| US11532233B2 (en) | Vehicle vision system with cross traffic detection | |
| CN105574552A (en) | Vehicle ranging and collision early warning method based on monocular vision | |
| KR20160062880A (en) | road traffic information management system for g using camera and radar | |
| JP2007255979A (en) | Object detection method and object detector | |
| CN101685014B (en) | Device and method for sensing object position | |
| Chen et al. | Real-time approaching vehicle detection in blind-spot area | |
| WO2010007718A1 (en) | Vehicle vicinity monitoring device | |
| Brauckmann et al. | Towards all around automatic visual obstacle sensing for cars | |
| JP2003028635A (en) | Image range finder | |
| Atiq et al. | Vehicle detection and shape recognition using optical sensors: a review | |
| Chiu et al. | Real-Time Front Vehicle Detection Algorithm for an Asynchronous Binocular System. | |
| Meinecke et al. | Approach for protection of vulnerable road users using sensor fusion techniques |
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 |