CN102968634B - A kind of parking lot structure extracting method under principal direction constraint - Google Patents
A kind of parking lot structure extracting method under principal direction constraint Download PDFInfo
- Publication number
- CN102968634B CN102968634B CN201210482430.2A CN201210482430A CN102968634B CN 102968634 B CN102968634 B CN 102968634B CN 201210482430 A CN201210482430 A CN 201210482430A CN 102968634 B CN102968634 B CN 102968634B
- Authority
- CN
- China
- Prior art keywords
- parking stall
- line
- stopway
- parking
- line segment
- 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.)
- Expired - Fee Related
Links
Landscapes
- Traffic Control Systems (AREA)
Abstract
Parking lot structure extracting method under the constraint of a kind of principal direction, step includes: for aviation orthography, uses Edison algorithm and Hough transform to carry out initial segment detection, obtains parking stall line principal direction；According to the principal direction obtained, use the line segments extraction method under principal direction constraint, detect accurate parking stall line；According to parking stall line angle degree, parking stall line is carried out marshalling to screen, and use the maximum direction method that intersects to divide stopway；Utilize the parking stall line and the stopway of division extracted, calculate the structural parameters in parking lot；Rebuild the accurate parking stall line in parking lot according to parking lot structure parameter, and generate the cut-off rule of stopway, complete automatically extracting of parking lot structure.The present invention utilizes initial segment detection to obtain the principal direction of parking stall line, carries out parking stall line drawing in this, as constraint, and parking stall line correctness, integrity and the positioning precision of extraction are the highest, it is possible to the preferably extraction for parking lot structure provides foundation.The present invention carries out parking lot structure extraction with single scape aviation orthography for data, and data acquisition is easy, affordable.
Description
Technical field
The present invention relates to a kind of structure extraction method in parking lot, particularly relate to the parking lot structure extracting method under the constraint of a kind of principal direction.
Background technology
Parking lot, the place temporarily rested as a kind of vehicle, the development along with urbanization process is expanded the most rapidly.Nowadays, whether in cities and towns still in suburb, parking lot is seen everywhere, and its scope has been even more than the building itself that they are serviced.Parking lot, as the necessary of automotive society, plays irreplaceable effect in human being's production is lived, and the human-subject test for parking lot is also particularly important.Understanding for parking lot structure and parking stall quantity is directly connected to the decision-making of minimum parking demand (MPRs).MPRs has attracted increasing attention in recent years, becomes the important issue that storing cycle policy is relevant.Reasonably storing cycle policy can not only reduce the layout quantity of parking stall number, and Economization on land utilizes, additionally it is possible to reduces its impact on natural environment, such as Vehicular exhaust and pollutes, tropical island effect, rain water runoff etc..Owing to parking lot quantity is that the basis that associated vehicle parks policy, parking lot structure and quantity are particularly significant.It addition, parking lot structure performer key player in parking tutorial message system (PGIS), PGIS can efficiently solve storing cycle problem.In PGIS, a hot issue is exactly the detection on sky parking stall.For this problem, having had substantial amounts of technological means, it has in common that the use with reference to parking lot structure.Also having and be exactly a bit, the structure in parking lot is the basis of the automatic Pilot of one of study hotspot in robot field.Automatic Pilot main target is so that vehicle, along default rail running, and is finally stopped, and its stop position must be in one accurately parking stall.Such control system requires that autonomous driving is provided along route information and accurately parking lot structure information accurately.
Although parking lot structure has its irreplaceable significance, at present mostly parking lot structure is to use Hand drawing mode, wastes time and energy, inefficiency.Being extracted into for a new problem of large-scale automatic stopping field structure.Automatically the method obtaining parking lot structure currently with remote sensing is the most fewer.1998, Wang, X et al. proposes, in " ApplicationsofComputerVision; 1998.WACV ' 98.Proceedings " meeting, the extraction utilizing stereogram to carry out parking lot structure first, build empty parking stall according to image line segments, build vehicle region parking stall according to the elevation information of stereogram.2009, SeoYW et al. proposes the parking lot structure extracting method of a kind of hierarchical policy at " IntelligentRobotsandSystems; 2009 " and " ProceedingsofInternationalJointConferenceonArtificialInt elligence ", first utilizing the steps such as line segments extraction, interpolation, extrapolation to obtain parking lot structure from image, the method for recycling machine learning carries out the screening on parking stall.The structure Main Basis carrying out parking lot structure is parking position line, the correctness of parking stall line drawing and integrity are directly connected to correctness and the precision that parking lot structure extracts, and existing parking lot structure extracting method is during carrying out parking stall line detection, edge detection method is directly used to carry out parking stall line drawing, the line segment extracted is how uncorrelated with parking stall line, correctness and integrity are the most relatively low, often omit some important parking stall line structure information, directly influence parking lot structure and build result.In addition, original method is for the poor data of the quality of image, and such as violent illumination variation, the shade problem such as substantially, the parking lot structure precision of extraction is poor.
Summary of the invention
Present invention solves the technical problem that and be: overcome the above-mentioned deficiency of prior art, propose the parking lot structure extracting method under the constraint of a kind of principal direction, it can the most intactly extract car bit line information from orthography, thus carries out the structure of parking lot structure.For traditional method, this method can extract more car bit line information more accurately from image, for the data that the quality of image is poor, also can extract car bit line information as much as possible, thus carry out the extraction of parking lot structure more accurately.
In order to solve above-mentioned technical problem, the technical scheme that the present invention proposes is: the parking lot structure extracting method under the constraint of a kind of principal direction, comprises the following steps:
The detection of the first step, initial segment obtains parking stall line principal direction and uses Edison operator and Hough transform to extract the edge line segment of image；Using segmentation and classification algorithm is vacancy area and non-vacancy area by image classification；Utilize classification results to reject the edge line segment being predominantly located in non-vacancy area, obtain initial parking stall line line segment, and carry out cluster marshalling according to the angle between the line line segment of initial parking stall, the edge line segment matching after marshalling is obtained parking stall line principal direction；
Parking stall line drawing use Edison operator under the constraint of second step, parking stall line principal direction carries out rim detection and obtains edge images, under the constraint of the parking stall line principal direction of first step acquisition, carry out Hough transform extract line segment, retain the length line segment more than 1.5m, obtain parking stall line；
3rd step, the maximum stopway intersecting direction divide, according to parking stall line angle degree, parking stall line are carried out marshalling screening, for the parking stall line in each marshalling, use the maximum direction method that intersects to divide stopway；
4th step, the stopway of the parking stall line and division calculating the utilization extraction of parking lot structure parameter calculate the structural parameters in parking lot, including parking stall length, width, direction, stopway position and direction；
5th step, the parking stall line in extraction parking lot structure foundation parking lot structure parameter structure parking lot, and generate the cut-off rule of stopway, complete the extraction of parking lot structure.
The innovative point of the present invention is: first carry out initial segment detection, according to image classification result, parking stall line is carried out screening and obtain parking stall line principal direction, then using parking stall line principal direction as constraints, utilize the Hough transformation under principal direction constraint that image is carried out parking stall line drawing.Being to utilize parking stall line principal direction as constraint owing to extracting line segment, efficiency, the accuracy rate therefore extracted are higher, it is achieved that the full-automatic high precision of parking stall line extracts.Accurately after Extraction of Image to parking stall line, then carry out successively stopway division, parking lot structure parameter calculate, according to parking lot structure parameter acquiring parking lot structure, thus complete whole parking lot structure extract.
Improving further of technique scheme is by the present invention:
One, the concrete grammar of first step extraction parking stall line principal direction is as follows:
1a) use Edison operator to carry out rim detection and obtain edge images；
1b) edge images is carried out Hough transform and extracts edge line segment；
1c) use multi-scale division algorithm that image is split, use K maximum near neighbor method that segmentation dough sheet is classified, image is divided into vacancy area and non-vacancy area；
1d) retain length more than the edge line segment of 1.5m, according to image subarea result, it is judged that every line segment ratio in vacancy area, non-vacancy area, if edge line segment meets following formula condition, reject:In formula, LNon-vacant lotRepresent edge line segment length in non-vacancy area, LVacant lotRepresent edge line segment length in vacancy area；
1e) the edge line segment after screening is organized into groups according to angle, the angle edge line segment less than 3 degree is attributed to same marshalling, if the line segment number in a certain marshalling is less than 5, then reject this marshalling, for each marshalling, use stochastical sampling consistency algorithm that line segment is fitted, obtain parking stall line principal direction.
Two, in the 3rd step, it is a marshalling by the parking stall line angle from each other line segment less than β；Line segment after marshalling is intersected direction method according to maximum and carries out stopway division, specific as follows:
It is 2a) that the center of circle builds semicircle to any bar line segment central point in described marshalling, semicircle is divided into n sector, successively by the two-way extension of centrage of each sector, recording the quantity of other line segments in the described marshalling that each centrage intersects with it, the maximum that direction is this line segment at the centrage place that wherein intersecting lens hop count amount is most intersects direction；
2b) maximum that in each marshalling, in all line segments, intersecting lens hop count amount is most intersects the principal direction that direction is this marshalling；
2c) calculating the distance to described marshalling principal direction of all line segment central points in marshalling, value of adjusting the distance is organized into groups by cluster calculation, and the line segment belonging to same marshalling is i.e. positioned at same stopway.
Three, above-mentioned parameter β is in the range of 3-5 °, and n is in the range of 8-16.
Four, in order to extract parking lot structure, need to solve parking lot parameter, including parking stall length, width, direction, stopway position and direction.Owing to unidirectional parking stall length, width, direction and stopway direction keep consistent substantially, unidirectional stopway only needs individually to calculate stopway position, and other parameters only need to calculate once.Owing to dissimilar stopway calculation method of parameters is slightly different, it is therefore desirable to it is carried out category division, according to stopway direction and parking stall direction relations, it is divided into horizontal stopway, longitudinal stopway and angle-park road.
In 4th step, the method calculating parking lot structure parameter is as follows:
3a) stopway direction and position: will be located in the midpoint fitting a straight line of vehicle axis on same stopway, the slope of straight line is this stopway direction, and intercept is stopway position；
3b) direction, parking stall: using the meansigma methods of the parking stall line angle degree of stopway as the parking stall candidate direction of this stopway, when candidate direction side, parking stall is stopway or edge, parking lot, then this side is to the direction, parking stall for this stopway；
3c) finding the distance that in the line length of parking stall, occurrence number is most interval, in this distance is interval, the meansigma methods of all parking stalls line length is parking stall length；
3d) parking stall width: for vertical and horizontal stopway, finds the distance that in the distance between all adjacent parallel vehicle bit lines, occurrence number is most interval, and in this distance is interval, the meansigma methods of the spacing of all parallel space lines is parking stall width；For angle-park road, parking stall width needs to be converted into its length on stopway direction, i.e.Wherein α is stopway direction and the angle of parking stall line, and D is the parallel distance between the line of parking stall.
Five, in the 5th step, the method building parking position line according to described parking lot structure parameter is specific as follows:
4a) build initial parking stall line from each stopway, select the parking stall line closest to parking stall length, centered by its intersection point with corresponding stopway centrage, along being parallel to the two-way extension in direction, described parking stall so that it is length is equal to described parking stall length, obtain initial parking stall line；
4b) generate parking stall line with initial parking stall line as starting point, along being parallel to stopway direction with parking stall width for being spaced two-way growth, during growth, if the parking stall line of growth and arest neighbors parking stall linear distance are within 0.3m, then move in parallel the parking stall line of growth along stopway direction so that it is the highest with the parking stall line registration extracted；With the parking stall line of growth as reference, continue to grow, until edge, parking lot to both sides.
4c) generation stopway cut-off rule is to each stopway, in its side, direction, parking stall, connects the end points of each parking stall line, forms stopway cut-off rule, complete the extraction of parking lot structure.
Six, in second step, the parking stall line drawing method under principal direction constraint is specific as follows:
5a) edge images is carried out Hough transform, set up accumulative space；
5b) an optional principal direction, searches for maximum in this main direction, obtains the peak value in this principal direction；
5c) obtain length corresponding to the described peak value line segment more than 1.5 meters and the end points of line segment；
The point of these line segments of 5d) conversion composition is to Hough space, and eliminates these contributions to accumulation matrix；
This peak value 5e) is set and adjacent cells accumulation matrix value is 0；
5f) and repeat step 5b)-step 5e), circulate all of principal direction successively, complete parking stall line drawing.
The feature of the present invention is as follows:
(1) classical line segments extraction algorithm based on peak-peak, the line segment extracted is disorderly and unsystematic, the most inconsistent with expectation information, and easily omit useful information, this method uses the parking stall line drawing method of hierarchical policy, utilizes initial segment detection to obtain parking stall line principal direction, carries out the extraction of parking stall line under the constraint of principal direction, the line segment extracted is how relevant to car bit line information, and correctness and integrity are higher.
(2) for the car bit line information extracted, proposing a kind of method intersecting direction based on maximum, it is possible to be effectively realized the division of stopway, accuracy is high.
(3) car bit line information is utilized to propose a set of reliable parking lot structure calculation method of parameters.
(4) during parking lot structure extracts, use existing parking stall line as constraint, when the parking stall line of growth and the parking stall linear distance of extraction are nearer, it is moved to extract parking stall line position, can effectively reduce the transmission error in growth course, thus improve the precision that parking lot structure extracts.
Accompanying drawing explanation
Parking lot structure extracting method under principal direction to the present invention retrains below in conjunction with the accompanying drawings is described further.
Fig. 1 embodiment of the present invention parking lot orthography.
Fig. 2 initial edge testing result figure.
Fig. 3 Image Segmentation classification results figure.
Fig. 4 initial segment the selection result figure.
Fig. 5 line segment marshalling result figure.
Line segment detection result figure under the constraint of Fig. 6 principal direction.
Fig. 7 finally extracts the parking stall line obtained.
Fig. 8 stopway division result figure.
Fig. 9 initiates parking stall line schematic diagram.
The parking lot structure figure that Figure 10 extracts.
Detailed description of the invention
Embodiment
Describing the present invention in detail below according to accompanying drawing, the purpose of the present invention and effect will be apparent from.
Data used by example are the orthography of region, a parking lot, and area size is 50m*20m, and resolution is 5cm, as shown in Figure 1.
Parking lot structure extracting method under the constraint of this example principal direction, use ArcGIS9.3, matlab2009, ENVIEx, VisualStudio2008 and open source software EdgeDetectionandImageSegmentationsystem jointly to complete, mainly comprise the steps that
The detection of the first step, initial segment obtains parking stall line principal direction and uses Edison operator and Hough transform to extract the edge line segment of image；Using segmentation and classification algorithm is vacancy area and non-vacancy area by image classification；Utilize classification results to reject the edge line segment being predominantly located in non-vacancy area, obtain initial parking stall line line segment, and carry out cluster marshalling according to the angle between the line line segment of initial parking stall, the edge line segment matching after marshalling is obtained parking stall line principal direction；In this step, the concrete grammar that the described first step extracts parking stall line principal direction is as follows:
1a) the Edison detection algorithm using EdgeDetectionandImageSegmentationsystem to provide carries out edge images detection, edge images is derived with .pgm formatted data；
1b) in matlab, read in .pgm formatted data, use Hough transform to obtain edge line segment from edge images, as shown in Figure 2；
1c) use multi-scale division algorithm that image is split, use K maximum near neighbor method that segmentation dough sheet is classified, image is divided into vacancy area and non-vacancy area；In this example, use FeatureExtraction function that image is carried out segmentation and classification in ENVIEx, arranging multi-scale division division threshold value is 72, merging threshold value is 97.5, selected grader is that the closest point methods of K is classified, obtaining Image Segmentation result as it is shown on figure 3, black represents vacancy area in figure, white represents non-vacancy area；
1d) retain length more than the edge line segment of 1.5m, according to image subarea result, it is judged that every line segment ratio in vacancy area, non-vacancy area, if edge line segment meets following formula condition, reject:In formula, LNon-vacant lotRepresent edge line segment length in non-vacancy area, LVacant lotRepresent edge line segment length in vacancy area, obtain result after screening as shown in Figure 4；
1e) the edge line segment after screening is organized into groups according to angle, the angle edge line segment less than 3 degree is attributed to same marshalling each other, if the line segment number in a certain marshalling is less than 5, then reject this marshalling, for each marshalling, use stochastical sampling consistency algorithm that line segment is fitted, obtain parking stall line principal direction.As it is shown in figure 5, organize into groups result figure for line segment, the parking stall line principal direction obtained in this example is 69 °.
Parking stall line drawing use Edison operator under the constraint of second step, parking stall line principal direction carries out rim detection and obtains edge images, under the constraint of the parking stall line principal direction of first step acquisition, carry out Hough transform extract line segment, retain the length line segment more than 1.5m, obtain parking stall line；
Concrete operations are as follows:
The edge line segment that first step use Edison operator extraction is obtained, its .pgm formatted data is read in matlab, the Hough transform under principal direction constraint is used to obtain edge line segment from edge images, set principal direction threshold space as 68 °-70 °, use the result that the line segments extraction under principal direction constraint obtains as shown in Figure 6, totally 64 line segments.
For the line segment that obtains of extraction, select according to image classification result, reject the ratio value line segment less than 0.2 (if edge line segment meets following formula condition, reject:In formula, LNon-vacant lotRepresent edge line segment length in non-vacancy area, LVacant lotRepresent edge line segment length in vacancy area), screen according to line segment length simultaneously, preseting length threshold value is 1.5m, owing to some line segments span two parking stall lines simultaneously, length approximates the twice of a parking stall line length, selecting these parking stall lines according to length and split, preseting length threshold value is more than 8m.Line segment result after screening and segmentation is as it is shown in fig. 7, totally 38 parking stall lines.
Parking stall line drawing method under principal direction of the present invention constraint is specific as follows:
5a) edge images is carried out Hough transform, set up accumulative space；
5b) an optional principal direction, searches for maximum in this main direction, obtains the peak value in this principal direction；
5c) obtain length corresponding to the described peak value line segment more than 1.5 meters and the end points of line segment；
The point of these line segments of 5d) conversion composition is to Hough space, and eliminates these contributions to accumulation matrix；
This peak value 5e) is set and adjacent cells accumulation matrix value is 0；
5f) and repeat step 5b)-step 5e), circulate all of principal direction successively, complete parking stall line drawing.
3rd step, the maximum stopway intersecting direction divide, according to parking stall line angle degree, parking stall line are carried out marshalling screening, for the parking stall line in each marshalling, use the maximum direction method that intersects to divide stopway；
In this step, it is a marshalling by the parking stall line angle from each other line segment less than β (in this example, β takes 3 °)；Line segment after marshalling is intersected direction method according to maximum and carries out stopway division, specific as follows:
It is 2a) that the center of circle builds semicircle to any bar line segment central point in described marshalling, semicircle is divided into n (in this example, n takes 12) individual sector, successively by the two-way extension of centrage of each sector, recording the quantity of other line segments in the described marshalling that each centrage intersects with it, the maximum that direction is this line segment at the centrage place that wherein intersecting lens hop count amount is most intersects direction；
2b) maximum that in each marshalling, in all line segments, intersecting lens hop count amount is most intersects the principal direction that direction is this marshalling；
2c) calculating the distance to described marshalling principal direction of all line segment central points in marshalling, value of adjusting the distance is organized into groups by cluster calculation, and the line segment belonging to same marshalling is i.e. positioned at same stopway.
Said method is utilized to carry out stopway division, the stopway obtained (in figure, dotted line represents stopway) as shown in Figure 8, totally 6 stopwaies.
4th step, the stopway of the parking stall line and division calculating the utilization extraction of parking lot structure parameter calculate the structural parameters in parking lot, including parking stall length, width, direction, stopway position and direction；
In this step, the method calculating parking lot structure parameter is as follows:
3a) stopway direction and position: will be located in the midpoint fitting a straight line of vehicle axis on same stopway, the slope of straight line is this stopway direction, and intercept is stopway position；Obtaining stopway direction in this example is 21.5 °, and four stopway intercepts (position) are respectively 5729318,5729312,5729301,5729295,5729277,5729282.
3b) direction, parking stall: using the meansigma methods of the parking stall line angle degree of stopway as the parking stall candidate direction of this stopway, when candidate direction side, parking stall is stopway or edge, parking lot, then this side is to the direction, parking stall for this stopway.In this example, direction, six stopway parking stalls is respectively as follows: 111 ° ,-69 °, 111 ° ,-69 °, 111 ° ,-69 °.
3c) parking stall length: finding the distance that in the line length of parking stall, occurrence number is most interval, in this distance is interval, the meansigma methods of all parking stalls line length is parking stall length.In this example, stopway is longitudinal stopway, and the interval width of distance takes 0.3m, distance interval according to (0,0.3], (and 0.3,0.6] ... divide, a length of 4.9m in parking stall that this example calculates.
3d) parking stall width: for vertical and horizontal stopway, finds the distance that in the distance between all adjacent parallel vehicle bit lines, occurrence number is most interval, and in this distance is interval, the meansigma methods of the spacing of all parallel space lines is parking stall width；For angle-park road, parking stall width needs to be converted into its length on stopway direction, i.e.Wherein α is stopway direction and the angle of parking stall line, and D is the parallel distance between the line of parking stall.In this example, the interval width of distance takes 0.3m, distance interval according to (0,0.3], (and 0.3,0.6] ... dividing, the parking stall width that this example calculates is 2.62m.
5th step, the parking stall line in extraction parking lot structure foundation parking lot structure parameter structure parking lot, and generate the cut-off rule of stopway, complete the extraction to parking lot structure.
In this step, in ArcGIS, the method building parking position line according to described parking lot structure parameter is specific as follows:
4a) build initial parking stall line from each stopway, select the parking stall line closest to parking stall length, centered by its intersection point with corresponding stopway centrage, along being parallel to the two-way extension in direction, described parking stall so that it is length is equal to described parking stall length, obtain initial parking stall line；As it is shown in figure 9, white line wider in figure is initial parking stall line.
4b) generate parking stall line with initial parking stall line as starting point, along being parallel to stopway direction with parking stall width for being spaced two-way growth, during growth, if the parking stall line of growth and arest neighbors parking stall linear distance are within 0.3m, then move in parallel the parking stall line of growth along stopway direction so that it is the highest with the parking stall line registration extracted；With the parking stall line of growth as reference, continue to grow, until edge, parking lot to both sides.
4c) generation stopway cut-off rule is to each stopway, in its side, direction, parking stall, connects the end points of each parking stall line, forms stopway cut-off rule, complete the extraction of parking lot structure.As shown in Figure 10.
Present invention parking lot structure based on orthography extracting method is not limited to the concrete technical scheme described in above-described embodiment, and the technical scheme that all employing equivalents are formed is the protection domain of application claims.
Claims (5)
1. the parking lot structure extracting method under principal direction constraint, comprises the following steps:
The detection of the first step, initial segment obtains parking stall line principal direction and uses Edison operator and Hough transform to extract the edge line segment of orthography；Use segmentation and classification algorithm that orthography is categorized as vacancy area and non-vacancy area；Utilize classification results to reject the edge line segment being predominantly located in non-vacancy area, obtain initial parking stall line line segment, and carry out cluster marshalling according to the angle between the line line segment of initial parking stall, the edge line segment matching after marshalling is obtained parking stall line principal direction；
Parking stall line drawing use Edison operator under the constraint of second step, principal direction carries out rim detection and obtains edge images, under the constraint of the parking stall line principal direction of first step acquisition, carry out Hough transform extract line segment, retain the length line segment more than 1.5m, obtain parking stall line；
3rd step, the maximum stopway intersecting direction divide, according to parking stall line angle degree, parking stall line are carried out marshalling screening, for the parking stall line in each marshalling, use the maximum direction method that intersects to divide stopway；
4th step, the stopway of the parking stall line and division calculating the utilization extraction of parking lot structure parameter calculate the structural parameters in parking lot, including parking stall length, width, direction, stopway position and direction；
5th step, the parking stall line in extraction parking lot structure foundation parking lot structure parameter structure parking lot, and generate the cut-off rule of stopway, complete the extraction of parking lot structure；
In described 3rd step, it is a marshalling by mutual for the parking stall line angle line segment less than β；Line segment after marshalling is intersected direction method according to maximum and carries out stopway division, specific as follows:
It is 2a) that the center of circle builds semicircle to any bar line segment central point in described marshalling, semicircle is divided into n sector, successively by the two-way extension of centrage of each sector, recording the quantity of other line segments in the described marshalling that each centrage intersects with it, the maximum that direction is this line segment at the centrage place that wherein intersecting lens hop count amount is most intersects direction；
2b) maximum that in each marshalling, in all line segments, intersecting lens hop count amount is most intersects the principal direction that direction is this marshalling；
2c) calculating the distance to described marshalling principal direction of all line segment central points in marshalling, value of adjusting the distance is organized into groups by cluster calculation, and the line segment belonging to same marshalling is i.e. positioned at same stopway.
Parking lot structure extracting method under principal direction the most according to claim 1 constraint, it is characterised in that: the concrete grammar that the described first step extracts parking stall line principal direction is as follows:
1a) use Edison operator to carry out rim detection and obtain edge images；
1b) edge images is carried out Hough transform and extracts edge line segment；
1c) use multi-scale division algorithm that orthography is split, use K maximum near neighbor method that segmentation dough sheet is classified, orthography is divided into vacancy area and non-vacancy area；
1d) retain length more than the edge line segment of 1.5m, according to orthography division result, it is judged that every line segment ratio in vacancy area, non-vacancy area, if edge line segment meets following formula condition, reject:In formula, LNon-vacant lotRepresent edge line segment length in non-vacancy area, LVacant lotRepresent edge line segment length in vacancy area；
1e) the edge line segment after screening is organized into groups according to angle, the angle edge line segment less than 3 degree is attributed to same marshalling each other, if the line segment number in a certain marshalling is less than 5, then reject this marshalling, for each marshalling, use stochastical sampling consistency algorithm that line segment is fitted, obtain parking stall line principal direction.
Parking lot structure extracting method under principal direction the most according to claim 2 constraint, it is characterised in that: β is in the range of 3-5 °, and n is in the range of 8-16.
Parking lot structure extracting method under principal direction the most according to claim 3 constraint, it is characterised in that in the 4th step, the method calculating parking lot structure parameter is as follows:
3a) stopway direction and position: will be located in the midpoint fitting a straight line of vehicle axis on same stopway, the slope of straight line is this stopway direction, and intercept is stopway position；
3b) direction, parking stall: using the meansigma methods of the parking stall line angle degree of stopway as the parking stall candidate direction of this stopway, when candidate direction side, parking stall is stopway or edge, parking lot, then this side is to the direction, parking stall for this stopway；
3c) finding the distance that in the line length of parking stall, occurrence number is most interval, in this distance is interval, the meansigma methods of all parking stalls line length is parking stall length；
3d) parking stall width: for vertical and horizontal stopway, finds the distance that in the distance between all adjacent parallel vehicle bit lines, occurrence number is most interval, and in this distance is interval, the meansigma methods of the spacing of all parallel space lines is parking stall width；For angle-park road, parking stall width needs to be converted into its length on stopway direction, i.e.Wherein α is stopway direction and the angle of parking stall line, and D is the parallel distance between the line of parking stall.
Parking lot structure extracting method under principal direction the most according to claim 4 constraint, it is characterised in that in the 5th step, the method building parking position line according to described parking lot structure parameter is specific as follows:
4a) build initial parking stall line from each stopway, select the parking stall line closest to parking stall length, centered by its intersection point with corresponding stopway centrage, along being parallel to the two-way extension in direction, described parking stall so that it is length is equal to described parking stall length, obtain initial parking stall line；
4b) generate parking stall line with initial parking stall line as starting point, along being parallel to stopway direction with parking stall width for being spaced two-way growth, during growth, if the parking stall line of growth and arest neighbors parking stall linear distance are within 0.3m, then move in parallel the parking stall line of growth along stopway direction so that it is the highest with the parking stall line registration extracted；With the parking stall line of growth as reference, continue to grow, until edge, parking lot to both sides；
4c) generation stopway cut-off rule is to each stopway, in its side, direction, parking stall, connects the end points of each parking stall line, forms stopway cut-off rule, complete the extraction of parking lot structure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210482430.2A CN102968634B (en) | 2012-11-23 | 2012-11-23 | A kind of parking lot structure extracting method under principal direction constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210482430.2A CN102968634B (en) | 2012-11-23 | 2012-11-23 | A kind of parking lot structure extracting method under principal direction constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102968634A CN102968634A (en) | 2013-03-13 |
| CN102968634B true CN102968634B (en) | 2016-08-03 |
Family
ID=47798765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210482430.2A Expired - Fee Related CN102968634B (en) | 2012-11-23 | 2012-11-23 | A kind of parking lot structure extracting method under principal direction constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102968634B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107993488A (en) * | 2017-12-13 | 2018-05-04 | 深圳市航盛电子股份有限公司 | A kind of parking stall recognition methods, system and medium based on fisheye camera |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106548479B (en) * | 2016-12-06 | 2019-01-18 | 武汉大学 | A kind of multi-level laser point cloud building boundary rule method |
| CN108609010A (en) * | 2016-12-13 | 2018-10-02 | 奥迪股份公司 | Parking aid, vehicle and parking assist method |
| CN107316492B (en) * | 2017-07-25 | 2020-10-23 | 纵目科技（上海）股份有限公司 | Method and system for positioning parking space in image |
| CN109389643B (en) * | 2017-08-10 | 2024-02-06 | 纵目科技（上海）股份有限公司 | Parking space main direction judging method, system and storage medium |
| CN110309694B (en) * | 2018-08-09 | 2021-03-26 | 中国人民解放军战略支援部队信息工程大学 | Method and device for determining main direction of remote sensing image |
| CN109800658B (en) * | 2018-12-26 | 2023-05-26 | 中汽研（天津）汽车工程研究院有限公司 | Parking space type online identification and positioning system and method based on neural network |
| CN112017238A (en) * | 2019-05-30 | 2020-12-01 | 北京初速度科技有限公司 | Method and device for determining spatial position information of linear object |
| CN110472580B (en) * | 2019-08-16 | 2022-04-05 | 北京茵沃汽车科技有限公司 | Method, device and storage medium for detecting parking stall based on panoramic image |
| CN110705595B (en) * | 2019-09-04 | 2022-02-25 | 苏州工业园区测绘地理信息有限公司 | Parking space automatic extraction method based on knapsack type three-dimensional laser point cloud data |
| CN113063375B (en) * | 2021-03-16 | 2022-04-08 | 成都理工大学 | Unmanned aerial vehicle remote sensing extraction method for linear farming ridges |
| CN115995161A (en) * | 2023-02-01 | 2023-04-21 | 华人运通(上海)自动驾驶科技有限公司 | Method and electronic device for determining parking position based on projection |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1387160A (en) * | 2001-05-22 | 2002-12-25 | 苏庚癸 | Display and management apparatus for vehicle position in parking lot |
| CN102005132A (en) * | 2010-11-04 | 2011-04-06 | 任杰 | Parking berth detection system based on berth mark identification |
| CN102243813A (en) * | 2010-05-11 | 2011-11-16 | 岳洋有限公司 | Parking lot guidance system, picking up vehicle guidance method and parking guidance method applying same |
-
2012
- 2012-11-23 CN CN201210482430.2A patent/CN102968634B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1387160A (en) * | 2001-05-22 | 2002-12-25 | 苏庚癸 | Display and management apparatus for vehicle position in parking lot |
| CN102243813A (en) * | 2010-05-11 | 2011-11-16 | 岳洋有限公司 | Parking lot guidance system, picking up vehicle guidance method and parking guidance method applying same |
| CN102005132A (en) * | 2010-11-04 | 2011-04-06 | 任杰 | Parking berth detection system based on berth mark identification |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107993488A (en) * | 2017-12-13 | 2018-05-04 | 深圳市航盛电子股份有限公司 | A kind of parking stall recognition methods, system and medium based on fisheye camera |
| CN107993488B (en) * | 2017-12-13 | 2021-07-06 | 深圳市航盛电子股份有限公司 | Parking space identification method, system and medium based on fisheye camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102968634A (en) | 2013-03-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102968634B (en) | A kind of parking lot structure extracting method under principal direction constraint | |
| CN102938064B (en) | Park structure extraction method based on LiDAR data and ortho-images | |
| CN101714211B (en) | Detection method of high-resolution remote sensing image street center line | |
| CN102564335B (en) | Method for measuring deformation of large-scale tunnel | |
| CN110345952A (en) | A kind of serializing lane line map constructing method and building system | |
| CN103914544A (en) | Method for quickly matching Chinese addresses in multi-level manner on basis of address feature words | |
| CN104866670B (en) | Based on road network topology change automatic testing method and the system of GPS space-time track | |
| CN110532720A (en) | A kind of rapid automatized modeling method of Urban Buried Pipeline Network BIM | |
| CN110427993B (en) | High-speed train navigation blind area positioning method based on meteorological parameters | |
| CN107657636A (en) | A kind of method that route topography figure elevational point is automatically extracted based on mobile lidar data | |
| CN113628291B (en) | Multi-shape target grid data vectorization method based on boundary extraction and combination | |
| CN106840175A (en) | A kind of vehicle driving trace matches the method and device of road network | |
| CN101929858B (en) | Precise surveying and mapping method of 0.25m contour interval of flat ground with large scale of 1:500 to 1:200 | |
| CN103680185B (en) | A kind of vehicle travel level precise division method | |
| CN105427287A (en) | Projection transformation-based connected region marking method | |
| CN109344774A (en) | Thermal power station target identification method in remote sensing image | |
| Wang et al. | Construction and application of artificial intelligence crowdsourcing map based on multi-track GPS data | |
| CN105469410A (en) | Landscape vision field analysis method based on GIS | |
| CN110162903B (en) | Urban building windward surface density calculation method and system based on grid parallelism | |
| CN101833665B (en) | Method for extracting roads from remote sensing map image | |
| De Marco et al. | The recognition of Heritage qualities from feature-based digital procedures in the analysis of historical urban contexts | |
| CN106199604A (en) | A kind of typhoon motion method for tracing based on correlation analysis | |
| CN103335626B (en) | The sample point optimum option method of three coordinate measuring engine measurement flatness | |
| CN1299220C (en) | Automatic splicing method for digital road map | |
| CN111462478B (en) | Method and device for dividing urban road network signal control subareas |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160803 Termination date: 20161123 |