CN102306274A - Device for identifying parking space state and method - Google Patents
Device for identifying parking space state and method Download PDFInfo
- Publication number
- CN102306274A CN102306274A CN201110165079A CN201110165079A CN102306274A CN 102306274 A CN102306274 A CN 102306274A CN 201110165079 A CN201110165079 A CN 201110165079A CN 201110165079 A CN201110165079 A CN 201110165079A CN 102306274 A CN102306274 A CN 102306274A
- Authority
- CN
- China
- Prior art keywords
- line
- image
- white
- parking stall
- signal
- 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
Abstract
The invention provides a device for identifying a parking space state and a method. A white block diagram is drawn on the ground of a parking space and a vision identifying signal for judging whether the space is occupied or not is provided. The system mainly comprises a singlechip minimum system circuit board, a signal collecting processing module and a communication module. By using the visual advantage of the invention, compared with a traditional parking space monitoring mode, dynamic image threshold computation and self-adaptation adjustment for a specific environment are designed; the device has an efficient judging mode; an idle state of the parking space can be judged within short time; and the invention can be used for enhancing the intelligent degree and the management level of a parking lot and obtaining higher economic benefit and good social benefit.
Description
Technical field
The present invention relates to video monitoring, image processing field, be specifically related to a kind of parking stall status identification means and method.
Background technology
Along with the continuous construction of large parking lot, the intelligent managing system for parking lot that is intended to improve the parking lot operational efficiency also will develop thereupon.For the parking position condition monitoring, existing both at home and abroad multiple implementation method: ultrasound wave parking stall measure method, ground induction coil detection method, infrared detection method and Doppler's microwave method.But these methods all exist limitation: ultrasonic parking stall measure is disturbed by non-car object easily; It is complicated that ground induction coil detects construction, and road pavement causes damage easily; Infra-red ray detection device involves great expense; Doppler's microwave method is difficult to detect the vehicle or the object of static or low speed.
Summary of the invention
To the deficiency of prior art, the present invention provides a kind of parking stall status identification means and method, adopts the CCD camera as vision sensor, through the parking stall is monitored in real time, obtains the seizure condition of parking stall.
Technical scheme of the present invention is: whether occupied on ground, parking stall, be coated with and paint a white segment, provide the parking stall visual signal, the parking stall status identification means comprises: single-chip minimum system circuit board, signal acquisition process module and communication module;
The single-chip minimum system circuit board comprises the input/output port that single-chip minimum system is required with control, and promptly the outside expansion mouth of I/O is supplied power by power module;
The signal acquisition process module comprises CCD camera, vision signal separating chips and AD converter.The vision signal that single-chip microcomputer is gathered CCD through A/D converter is handled and is discerned video image after having gathered a frame information.
Communication module adopts the MAX485 conversion chip, utilizes the SCI interface of single-chip microcomputer, adopts asynchronous communication means to accomplish communication function.The serial protocol signal of output transfers the RS485 agreement to through the MAX485 chip, but teletransmission.
For images acquired information; Single-chip microcomputer need start A/D converter according to row, field sync signal; Particularly:, simultaneously the time of gathering, wait for that then row interrupts and startup A/D next time through timer setting AD converter waiting by the time start AD converter after the line synchronizing signal of CCD output.The CCD camera collection to simulating signal be directly inputted in the single-chip microcomputer through analog port P2.6.Vision signal separating chips output terminal is connected to single-chip microcomputer P2.7 and P0.1 port, from the composite video of CCD, extracts the field system chronizing impulse signal, gets into the control signal of interrupting opening AD converter as single-chip microcomputer.The SCI interface of single-chip microcomputer converts serial protocol into the RS485 agreement through the MAX485 chip; P2.5 and P2.4 are as the enable port of serial communication, and P0.4 and P0.5 are as the serial communication data output.
The method of identification of the present invention parking stall state adopts above-described parking stall status identification means, carries out as follows:
In the said step 1, according to camera output signal sequence, it is following to carry out the collection of video signal flow process:
Step 1.1 is waited for field sync signal, opens row after time-delay a period of time and interrupts the preparation for acquiring data;
Step 1.2 is carried out the AD conversion to this visual signal;
Step 1.3, AD converter constantly write data in the array of memory image, up to linage-counter meter to 40, one field picture data acquisition this moment is accomplished.
In the said step 2, the view data that collects is handled and discerned, detailed process is following:
Step 2.1: to camera collection to view data carry out coordinate transform, comprising: the conversion of pixel coordinate is about to image coordinate and changes map or ground coordinate into; Pixel brightness value to after the coordinate transform resamples;
Step 2.2: adopt a kind of dynamic threshold method to come each row of whole field picture is carried out Threshold Segmentation; Detailed process is following: if the black region that lastrow splits is continuous; Calculate the average gray value of target area and background area so, and get its intermediate value, as the current line threshold value.If the black region that lastrow splits is discontinuous, then keep the current line threshold value constant;
Step 2.3: after confirming the dynamic threshold of every field picture, extract image according to threshold value.Concrete process is following:
Step 2.3.1, to camera collection to raw data carry out filtering, judge whether image is the first field data image, if just calculate basic threshold value and then image is carried out binary conversion treatment; If not, just directly image is carried out binary conversion treatment with first threshold value.Under the strong situation about disturbing of white light, at first coloured image to be become gray level image, get the negative film of vision signal then;
Step 2.3.2 notes the white point number that occurs through each row after the binary conversion treatment, writes down white point hop count and center thereof then, and the white point section is for occurring the data segment of white point continuously, for next step graphic searching is prepared;
Step 2.3.3; Find from the nearest delegation that has only one section white point of image initial row; The condition that satisfies when this row: have only one section continuous white data section; And the left and right sides end points of data segment is not all when the left and right edges of image; Think correct delegation; It is capable that this rower is designated as Start, in next line, seeks such one section continuous white line segment then, otherwise look for again from the nearest delegation that has only one section white point of image initial row.If there is one section continuous white line segment in next line; Satisfy such condition simultaneously: its left hand edge is not or not the right side of lastrow white line right hand edge; And its right hand edge is not or not the left side of lastrow white line left hand edge; Continue then to judge again whether next line has so white line segment; When there is not so white line segment in next line; Behavior End is capable for mark; And calculate capable poor of this row and Start; If record by experiment greater than 3(); Then finish this search; Represent the variable of triangle number to add 1; Again the capable nearest delegation that has only one section white point of End is left in search then; If smaller or equal to 3, then the capable nearest delegation that has only one section white point of End is left in search again.Up to having traveled through row all in the image.
Beneficial effect:
The present invention utilizes the advantage of vision, proposes a kind of parking stall status identification means and method, compact conformation, and reliability is high, and is easy for installation, contains much information, and testing result is accurate, can expand other additional functions easily, and production cost is moderate, is suitable for widespread usage.Compare with traditional parking stall monitor mode, have the advantage that image dynamic threshold value calculates and self-adaptation is adjusted, have decision procedure efficiently, through RS485 communication protocol output recognition result.This device can be used in the intelligent parking field system, can obtain higher economic benefit and good social benefit.
Description of drawings
The structured flowchart of Fig. 1 apparatus of the present invention embodiment;
The video capture circuit figure of Fig. 2 apparatus of the present invention embodiment;
The circuit theory diagrams of the power module of Fig. 3 apparatus of the present invention embodiment;
The video signal input interface figure of Fig. 4 apparatus of the present invention embodiment;
The hardware circuit design schematic diagram of Fig. 5 apparatus of the present invention embodiment;
The camera imaging mapping relations figure of Fig. 6 apparatus of the present invention embodiment;
The mapping relations figure that ignores the distortion of axis both sides of Fig. 7 apparatus of the present invention embodiment;
The camera imaging synoptic diagram of Fig. 8 apparatus of the present invention embodiment;
Non-all row of Fig. 9 apparatus of the present invention embodiment are gathered synoptic diagram;
The horizontal direction coordinate transform figure of Figure 10 apparatus of the present invention embodiment;
The parking stall sign rim detection process flow diagram of Figure 11 apparatus of the present invention embodiment;
The synoptic diagram that the single-chip microcomputer of Figure 12 apparatus of the present invention embodiment is communicated by letter with external equipment;
The SCI Interface design figure of Figure 13 apparatus of the present invention embodiment;
The processing result image of Figure 14 apparatus of the present invention embodiment;
The triangle identification process figure of Figure 15 apparatus of the present invention embodiment;
The camera image collecting flowchart figure of Figure 16 apparatus of the present invention embodiment.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is further specified.
A kind of parking stall status identification means; Through the parking stall is monitored in real time; Obtain the seizure condition of parking stall; On the ground on parking stall, be coated with and paint a white triangles shape segment; Utilize technology such as video signal collective, picture signal processing and identification; Whether the interior triangle of picture that detects camera collection exists, and promptly whether the parking stall is idle.Device comprises: single-chip minimum system circuit board, signal acquisition process module and communication module, and the apparatus structure block diagram is as shown in Figure 1;
The single-chip minimum system circuit board comprises the input/output port that single-chip minimum system is required with control.Be the outside expansion mouth of I/O, master chip uses the single-chip microcomputer C8051F320 of U.S. Cygnal company, and by the power module power supply, power circuit as shown in Figure 3;
Controlling required input and output mainly is made up of following components:
(1) ccd sensor vision signal input: input end is made up of 3 signal wires, a 5.0V power lead, and a ground wire, a signal wire, circuit theory diagrams are as shown in Figure 4;
(2) asynchronous serial communication input and output: this communication port is used for carrying out radio communication with external other equipment, and to realize the data transmission of single-chip microcomputer and external equipment, general principles figure as shown in Figure 5;
The signal acquisition process module mainly comprises CCD camera, vision signal separating chips and AD converter; The vision signal separating chips adopts LM1881; Through CCD camera collection vision signal; Waiting by the time start AD converter after the line synchronizing signal of CCD output; Set the acquisition time of AD converter simultaneously through timer, wait for that then row interrupts and starts next time.After having gathered a field information picture is handled and discerned;
Communication module adopts the MAX485 chip, utilizes the SCI interface of C8051F320, adopts asynchronous communication means to accomplish communication function.Output RS485 signal can be gone out the recognition result remote transmission.
For images acquired information; Single-chip microcomputer need start A/D converter according to row, field sync signal; Particularly:, simultaneously the time of gathering, wait for that then row interrupts and startup A/D next time through timer setting AD converter waiting by the time start AD converter after the line synchronizing signal of CCD output.Simultaneously, the CCD camera collection to simulating signal also be directly inputted in the single-chip microcomputer through analog port P2.6.Vision signal separating chips output terminal is connected to single-chip microcomputer P2.7 and P0.1 port, from the composite video of CCD, extracts the field system chronizing impulse signal, gets into the control signal of interrupting opening AD converter as single-chip microcomputer.The SCI interface of single-chip microcomputer converts serial protocol into the RS485 agreement through the MAX485 chip; P2.5 and P2.4 are as the enable port of serial communication, and P0.4 and P0.5 are as the serial communication data output.Video capture circuit as shown in Figure 2.
The method of identification of the present invention parking stall state adopts above-described parking stall status identification means, carries out as follows:
In the said step 1,, carry out collection of video signal according to camera output signal sequence, process flow diagram as shown in figure 16, concrete steps are following:
Step 1.1, the isolated parity field synchronizing signal of LM1881 vision signal separating chips produces regularly interrupts, and this signal is distinguished with rising edge and negative edge, and rising edge is strange field sync signal, and negative edge is even field sync signal.According to the PAL standard, after this signal 22 data lines are arranged, about 1408 μ s of transmission time, this transmission time looks the signal lag of LM1881 and decides.Effective for guaranteeing to collect data, single-chip microcomputer timer fixed response time to 23.5 is capable, and transmission data required time is 23.5 * 64 μ s=1504 μ s, and after regularly finishing, the capable interrupt counter of control AD conversion makes zero, and opens row and interrupts the preparation for acquiring data.
Step 1.2 is carried out the AD conversion to this visual signal.
According to the standard of PAL, when gathering line data, as rising edge is set when effective; 6 μ s just have real view data appearance after the line synchronizing signal; If it is effective that negative edge is set, be just to have real view data behind the line synchronizing signal 10.3 μ s to occur, and continue 52 μ s to this row end.In order to guarantee that monitoring has better real-time property, it is effective that the present invention is provided with rising edge, in the row of this row interrupts, makes regularly 6 μ s of timer, after regularly finishing, begins to gather line data.More than 280 data line arranged in this data fields; In order to reduce operand; Guarantee the accuracy of identification simultaneously; AD converter is only gathered 40 row wherein, is provided with a capable interrupt counter for this reason, and every execution once; The value of row interrupt counter adds one; When the value of row interrupt counter when certain number in the array that prestores equates, represent that this row should gather, triggering AD changes.
Because AD converter is gathered by sequence, sequence at most can 8 points of continuous acquisition, and it is scan mode that acquisition mode is set.After accomplishing, each sequence collection will produce an AD conversion.For reducing the AD interruption times, it is 8 that conversion sequence length has been set in the AD converter initialization.Open AD with scan mode this moment, and gathering 48 points needs 6 AD conversions, then the data of AD converter is transferred in the array of memory image.Because AD converter will be gathered 6 sequences in the data line, therefore to write down the sequence number of having gathered, an AD conversion of every execution, the recorded sequences number adds one, when sequence number reaches 6, indicates a data line collection completion.The value that to gather linage-counter after a data line collection is accomplished adds one, and should stop AD, waits for row conversion next time.
Step 1.3, AD interrupts constantly in the array of memory image, writing data, and up to linage-counter meter to 40, one field picture data acquisition this moment is accomplished.
In the said step 2, the view data that collects is handled and discerned.
After having gathered a video information image is handled and discerned, comprising: coordinate transform, dynamic threshold confirm that image extracts and identification.
The detailed process of Flame Image Process and identification is following:
Step 2.1: in order to eliminate the geometry deformation in the image that collects; Thereby obtain a new images that meets certain map projection or avatars requirement; To camera collection to view data carry out coordinate transform; Comprise: the conversion of pixel coordinate is about to image coordinate and changes map or ground coordinate into; Pixel brightness value to after the coordinate transform resamples.
Image information storage shooting be one by the process of three-dimensional scenic to equatorial projection, variation has taken place in the relativeness between the object in this process (elephant).Recovering the three-dimensional body visible surface structure by two-dimentional image is the final goal of theory on computer vision.Obviously, this recovery is a very complicated process, has not only comprised some plane coordinate transformations, and has related to the spatial coordinate conversion.And if object in the reality scene is limited on the constant plane, then this process will be simplified greatly, finally deteriorates to a simple coordinate transformation process.
Can know through the camera image-forming principle; Is determined by the geometric model of camera imaging by real surface level to the mapping relations of camera imaging plane; Can contrast this relation from the zone of image that camera becomes and its reflection surface level, camera imaging mapping relations as shown in Figure 6;
Can know that by Fig. 6 the distortion that camera produces can have following three kinds:
1, in vertical direction, camera compresses surface level at a distance, and distance is far away more, and this compression is serious more.
2, in the horizontal direction, camera carries out compression in various degree to the surface level at different distance place, and distance is far away more, compresses serious more.
3, the both sides, axis of image, camera pushes towards center nearby surface level.On hand, this extruding is very slight, a long way off, along with the increase of the point in the surface level apart from axial line distance, pushes more and more serious.
It is thus clear that the mapping relations of image coordinate and surface level true coordinate are still very complicated.Even this relation can use one group of mathematical formulae to express, but will in the C8051F320 single-chip microcomputer, realize remaining unapproachable with program, therefore, the present invention further simplifies this mapping, ignores the third distortion, only considers preceding two kinds.Because the third distortion influence on hand is less, and deals with complex algorithm, so ignore the influence of its generation, only the distortion on vertical and the horizontal direction is proofreaied and correct.The mapping relations of ignoring axis both sides distortion as shown in Figure 7;
At first proofread and correct the distortion on the vertical direction.As shown in Figure 8, h is the height of cam lens apart from ground, a, b be the most nearby with the visual field of farthest, x be in the visual field more arbitrarily, the sight line elevation angle that α, β, θ are respectively a, b, x is corresponding.Can know by geometric relationship and to exist following relation between them:
On photographic plate, d is a photosensitive area length, is equivalent to total line number of image, and r is that x place image is expert at, and has geometric relationship to know:
Can solve the relation of r and x by above two formulas.
Adopt the mode of " non-all row are gathered " to proofread and correct this variation among the present invention.Non-all row are gathered synoptic diagram as shown in Figure 9.So-called non-all row are gathered with all the row collection is corresponding.In all row was gathered, the data line that AD gathered was uniformly distributed in the image (original image) of camera output, and this acquisition mode has guaranteed that the image that collects is relative with original image undistorted.But not all row is gathered and to be meant that then the data line that AD gathered anisotropically is distributed in the original image by certain rule, and this rule is to guarantee that the image that collects is compared with real scenery in the vertical not distort that non-all row are gathered synoptic diagram as shown in Figure 9.
Distortion on the level of corrections direction then.Distortion on the level also can be tried to achieve by geometric transformation, the horizontal direction coordinate transform as shown in figure 10, the distortion of establishing center line is 0, the formula below then choosing in the image delegation arbitrarily corresponding length satisfying in real image:
X---one's own profession is to trapezoidal center live width that this end constituted;
Y---center line is to the distance of one's own profession;
L---one's own profession is to the distance of trapezoidal both sides extending line intersection point.
Then on this journey arbitrarily the side-play amount of any just can in the hope of.Through the coordinate transform on level and the vertical direction, real image can be reduced, make the image that collects more near real image, also lay a good groundwork for the calculating of behindness parameter.
Step 2.2: for the anti-light variation interference performance of enhanced system, native system has adopted a kind of dynamic threshold method to come each row of whole field picture is carried out Threshold Segmentation, if threshold value does not become with the space in time, is called static threshold; If threshold value in time or space and changing, is called and dynamic threshold.Dividing method algorithm based on static threshold is simple, and calculated amount is little, but bad adaptability.The computing method that depend on dynamic threshold based on its complexity of dividing method of dynamic threshold.
Detailed process is following: if the black region that lastrow splits is continuous, calculate the average gray value of target area and background area so, and get its intermediate value, as the current line threshold value.If the black region that lastrow splits is discontinuous, then keep the current line threshold value constant.
Through actual test shows, this threshold value calculation method can very effectively prevent that light from disturbing.
Step 2.3: after confirming the dynamic threshold of every field picture, extract image according to threshold value.
Image extracts and adopts the following limb detection method.Because triangle is a continuous segments, so the marginal point of adjacent two row should be adjacent.Following limb detects and has utilized this characteristic to come searching route just.If searched out the edge of certain row, then just near a last edge, searched next time.This algorithm can be saved computing time greatly, can get rid of disadvantageous interference simultaneously.
In practical application, find, various interference occur through regular meeting in the ground.Some interference are intrinsic on the ground; And some interference produces at random; Like the noise that produces in the camera signals transmission course; Or, camera excessively far sees the place beyond the ground because seeing scope; And the color on ground has the various situation that possibly occur; Or reflective be white, or be entirely black or the like.Therefore, the present invention has adopted binaryzation and following limb detection method way of combining to extract image, and designs corresponding program, gets rid of these and disturbs, and could guarantee the operation of system stability.
The flow process of extracting image according to threshold value as shown in figure 11, concrete process is following:
Step 2.3.1, to camera collection to raw data carry out filtering, think noise when promptly exceeding the data point beyond the normal range, replace with the mean value of the data of two data points before and after this data point.During operation, first field picture of gathering during beginning is good, and first row is the most nearby delegation just, the complete triangle sign that comprised, and use " bimodal method " can correct extraction.Judge whether image is the first field data image, if just calculate basic threshold value and then image is carried out binary conversion treatment; If not, just directly image is carried out binary conversion treatment with first threshold value.Because being the background with picture, thinking of the present invention is split up into two kinds of relatively good color of distinguishing, white and black.Therefore, under the strong situation about disturbing of white light, at first coloured image to be become gray level image, get the negative film of vision signal then, disturb through the eliminating that this way can be very big.
Step 2.3.2 notes the white point number that occurs through each row after the binary conversion treatment, writes down white point hop count and center thereof then, and the white point section is for the data segment of white point occurring, for next step triangle search is prepared continuously.
Step 2.3.3; Find from the nearest delegation that has only one section white point of image initial row; The condition that satisfies when this row: have only one section continuous white data section; And the left and right sides end points of data segment is not all when the left and right edges of image; Think correct delegation; It is capable that this rower is designated as Start, in next line, seeks such one section continuous white line segment then, otherwise look for again from the nearest delegation that has only one section white point of image initial row.If there is one section continuous white line segment in next line; Satisfy such condition simultaneously: its left hand edge is not or not the right side of lastrow white line right hand edge; And its right hand edge is not or not the left side of lastrow white line left hand edge; Continue then to judge again whether next line has so white line segment; When there is not so white line segment in next line; Behavior End is capable for mark; And calculate capable poor of this row and Start; If record by experiment greater than 3(); Then finish this search; Represent the variable of triangle number to add 1; Again the capable nearest delegation that has only one section white point of End is left in search then; If smaller or equal to 3, then the capable nearest delegation that has only one section white point of End is left in search again.Up to having traveled through row all in the image.Result after the Flame Image Process as shown in figure 14.
Said step 3, after handling through RS485 communication interface output the result---whether the parking stall free time.
The synoptic diagram that single-chip microcomputer of the present invention is communicated by letter with external equipment as shown in figure 12, behind Flame Image Process and the end of identification, after can handling through RS485 interface output the result---whether the parking stall free time, send out.
Claims (5)
1. whether occupied a parking stall status identification means is coated with on ground, parking stall and paints a white segment, provide the parking stall visual signal, it is characterized in that this recognition device comprises: single-chip minimum system circuit board, signal acquisition process module and communication module;
The single-chip minimum system circuit board mainly comprises the input/output port that the minimum system of single-chip microcomputer is required with control, and promptly the outside expansion mouth of I/O is supplied power by power module;
The signal acquisition process module comprises CCD camera, vision signal separating chips and AD converter;
Communication module adopts the MAX485 conversion chip, utilizes the SCI interface of single-chip microcomputer, adopts asynchronous communication means to accomplish communication function, and the serial protocol signal of output transfers the RS485 agreement to through the MAX485 chip, but teletransmission.
2. adopt the method for status identification means identification parking stall, the described a kind of parking stall of claim 1 state, it is characterized in that: this method is carried out as follows:
Step 1 uses the CCD camera to keep watch on the parking stall;
Step 2 is handled and is discerned vision signal;
Step 3, recognition result---whether the parking stall is occupied through the output of RS485 agreement.
3. the method for identification according to claim 2 parking stall state is characterized in that: said step 1 is to carry out collection of video signal according to camera output signal sequence, specifically carries out as follows:
Step 1.1 is waited for field sync signal, opens row after time-delay a period of time and interrupts the preparation for acquiring data;
Step 1.2 is carried out the AD conversion to this visual signal;
Step 1.3, AD converter constantly write data in the array of memory image, up to linage-counter meter to 40, one field picture data acquisition this moment is accomplished.
4. the method for identification according to claim 2 parking stall state is characterized in that: the said step 2 pair view data that collects is handled and is discerned, and specifically carries out as follows:
Step 2.1: to camera collection to view data carry out coordinate transform, comprising: the conversion of pixel coordinate resamples to the pixel brightness value after the coordinate transform;
Step 2.2: adopt the dynamic threshold method to come each row of whole field picture is carried out Threshold Segmentation;
Step 2.3: after having confirmed the dynamic threshold of every field picture, extract image according to threshold value.
5. the method for identification according to claim 4 parking stall state is characterized in that: said step 2.3 is after the dynamic threshold of having confirmed every field picture, extracts image according to threshold value, specifically carries out as follows:
Step 2.3.1, to camera collection to raw data carry out filtering, judge whether image is the first field data image, if just calculate basic threshold value and then image is carried out binary conversion treatment; If not, just directly image is carried out binary conversion treatment with first threshold value; Under the strong situation about disturbing of white light, at first coloured image to be become gray level image, get the negative film of vision signal then;
Step 2.3.2 notes the white point number that occurs through each row after the binary conversion treatment, writes down white point hop count and center thereof then, and the white point section is for the data segment of white point occurring, for next step triangle search is prepared continuously;
Step 2.3.3, the image from the starting line to find the nearest point only a white line, when the line meets the conditions: only one continuous white data segment and the data segment endpoints are not about left and right edges of the image when considered the correct line to the line labeled Start line, and then the next line looking for such a continuous white line, or to re-find images from the starting line nearest the point that only a white line; next line if there is a continuous white line, Such conditions are satisfied: the left edge of the white line is not a line on the right edge of the right side, and the right edge of the white line is not a line on the left edge of the left side, then continue to the next line and then determine whether there is such a white line until the next there is no such line when the white lines, marking the act End row and calculate the difference between the Bank and the Start row, if greater than 3, then the end of this search, the triangle represents the number of variables plus one, and then search again from the End Line Recent only a white line, if less than or equal to 3, then search again from the End of the line nearest the point that only a white line traversing the image until all rows.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110165079A CN102306274A (en) | 2011-06-17 | 2011-06-17 | Device for identifying parking space state and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110165079A CN102306274A (en) | 2011-06-17 | 2011-06-17 | Device for identifying parking space state and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102306274A true CN102306274A (en) | 2012-01-04 |
Family
ID=45380134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110165079A Pending CN102306274A (en) | 2011-06-17 | 2011-06-17 | Device for identifying parking space state and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102306274A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102592472A (en) * | 2012-02-21 | 2012-07-18 | 北京工业大学 | Parking-space detection method based on JPEG (joint photographic experts group) compressed image |
| WO2014114067A1 (en) * | 2013-01-28 | 2014-07-31 | Zhang Guobiao | Large-area parking space monitoring system |
| CN104574354A (en) * | 2013-09-26 | 2015-04-29 | 张国飙 | Parking space monitoring method based on edge detection |
| CN106289335A (en) * | 2016-07-19 | 2017-01-04 | 湖北航天技术研究院总体设计所 | A kind of object type and location status recognition methods |
| CN107886080A (en) * | 2017-11-23 | 2018-04-06 | 同济大学 | One kind is parked position detecting method |
| CN108847044A (en) * | 2018-06-22 | 2018-11-20 | 安徽尼古拉电子科技有限公司 | A kind of vehicle management system based on Beidou satellite alignment system |
| CN109389013A (en) * | 2017-08-10 | 2019-02-26 | 纵目科技（上海）股份有限公司 | Parking stall combinational algorithm and medium based on parking stall principal direction and template response point |
| CN109686110A (en) * | 2019-01-17 | 2019-04-26 | 蜂寻（上海）信息科技有限公司 | Parking stall sky expires condition discrimination method and apparatus |
| CN111325989A (en) * | 2020-03-16 | 2020-06-23 | 重庆第二师范学院 | Artificial intelligence robot based on thing networking |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251952A (en) * | 2008-04-03 | 2008-08-27 | 王文彪 | Underground garage intelligent managing system and method |
| CN101807352A (en) * | 2010-03-12 | 2010-08-18 | 北京工业大学 | Method for detecting parking stalls on basis of fuzzy pattern recognition |
| CN102005133A (en) * | 2010-11-04 | 2011-04-06 | 任杰 | Recognizable mark-based parking position detecting method |
-
2011
- 2011-06-17 CN CN201110165079A patent/CN102306274A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251952A (en) * | 2008-04-03 | 2008-08-27 | 王文彪 | Underground garage intelligent managing system and method |
| CN101807352A (en) * | 2010-03-12 | 2010-08-18 | 北京工业大学 | Method for detecting parking stalls on basis of fuzzy pattern recognition |
| CN102005133A (en) * | 2010-11-04 | 2011-04-06 | 任杰 | Recognizable mark-based parking position detecting method |
Non-Patent Citations (1)
| Title |
|---|
| 马铭: "基于视频的停车场车位状态监控识别系统", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102592472A (en) * | 2012-02-21 | 2012-07-18 | 北京工业大学 | Parking-space detection method based on JPEG (joint photographic experts group) compressed image |
| CN104169989B (en) * | 2013-01-28 | 2016-03-30 | 杭州海存信息技术有限公司 | Large area parking stall supervisory system |
| WO2014114067A1 (en) * | 2013-01-28 | 2014-07-31 | Zhang Guobiao | Large-area parking space monitoring system |
| CN104169989A (en) * | 2013-01-28 | 2014-11-26 | 杭州海存信息技术有限公司 | Large-area parking space monitoring system |
| CN104574354B (en) * | 2013-09-26 | 2017-10-10 | 成都海存艾匹科技有限公司 | Parking position monitoring method based on edge detection |
| CN104574354A (en) * | 2013-09-26 | 2015-04-29 | 张国飙 | Parking space monitoring method based on edge detection |
| CN106289335A (en) * | 2016-07-19 | 2017-01-04 | 湖北航天技术研究院总体设计所 | A kind of object type and location status recognition methods |
| CN106289335B (en) * | 2016-07-19 | 2019-04-30 | 湖北航天技术研究院总体设计所 | A kind of object type and location status recognition methods |
| CN109389013A (en) * | 2017-08-10 | 2019-02-26 | 纵目科技（上海）股份有限公司 | Parking stall combinational algorithm and medium based on parking stall principal direction and template response point |
| CN109389013B (en) * | 2017-08-10 | 2024-02-09 | 纵目科技（上海）股份有限公司 | Parking space combination algorithm and medium based on parking space main direction and template response points |
| CN107886080A (en) * | 2017-11-23 | 2018-04-06 | 同济大学 | One kind is parked position detecting method |
| CN108847044A (en) * | 2018-06-22 | 2018-11-20 | 安徽尼古拉电子科技有限公司 | A kind of vehicle management system based on Beidou satellite alignment system |
| CN109686110A (en) * | 2019-01-17 | 2019-04-26 | 蜂寻（上海）信息科技有限公司 | Parking stall sky expires condition discrimination method and apparatus |
| CN111325989A (en) * | 2020-03-16 | 2020-06-23 | 重庆第二师范学院 | Artificial intelligence robot based on thing networking |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102306274A (en) | Device for identifying parking space state and method | |
| WO2021093283A1 (en) | Sea surface small-area oil spill region detection system and detection method based on multi-sensing fusion | |
| CN111368706A (en) | Data fusion dynamic vehicle detection method based on millimeter wave radar and machine vision | |
| CN103500322B (en) | Automatic lane line identification method based on low latitude Aerial Images | |
| CN103017869B (en) | A kind of water stage measurement system and method based on Digital Image Processing | |
| CN102110376B (en) | Roadside parking space detection device based on computer vision | |
| CN102867414B (en) | Vehicle queue length measurement method based on PTZ (Pan/Tilt/Zoom) camera fast calibration | |
| CN110766979A (en) | Parking space detection method for automatic driving vehicle | |
| CN102426801A (en) | Automatic parking lot berth management method and system based on visual analysis | |
| CN105611244A (en) | Method for detecting airport foreign object debris based on monitoring video of dome camera | |
| CN202676156U (en) | Electric transmission line condition monitoring system based on target image identification and benchmark measurement | |
| CN103069796A (en) | Method for counting objects and apparatus using a plurality of sensors | |
| CN202393715U (en) | Automatic detecting device for grey cloth defect spot | |
| CN106910198A (en) | A kind of boundary determining method of hay mower without electric wire fence | |
| CN104376732A (en) | Intelligent traffic light based on signal control processing technology and signal control method | |
| CN107976685B (en) | Indoor human target tracking system of infrared sensor based on thing networking | |
| CN110674672B (en) | Multi-scene people counting method based on tof camera | |
| CN112016492A (en) | Teaching attention monitoring system and method based on vision | |
| CN111950491A (en) | Personnel density monitoring method and device and computer readable storage medium | |
| CN103390259A (en) | Ground image processing method in visual guidance AGV | |
| CN202205350U (en) | Device for shooting vehicles against traffic regulations during bus lane restriction period | |
| CN101980299B (en) | Chessboard calibration-based camera mapping method | |
| CN107918935B (en) | Image matching cutting device and cutting method of embedded system | |
| CN109657639A (en) | A kind of Situation Awareness System and method based on panoramic vision | |
| CN106997685A (en) | A kind of roadside parking space detection device based on microcomputerized visual |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120104 |