CN113139975B - Road feature-based pavement segmentation method and device - Google Patents
Road feature-based pavement segmentation method and device Download PDFInfo
- Publication number
- CN113139975B CN113139975B CN202110419236.9A CN202110419236A CN113139975B CN 113139975 B CN113139975 B CN 113139975B CN 202110419236 A CN202110419236 A CN 202110419236A CN 113139975 B CN113139975 B CN 113139975B
- Authority
- CN
- China
- Prior art keywords
- road
- edge
- road surface
- points
- segmentation
- 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
- 230000011218 segmentation Effects 0.000 title claims abstract description 76
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000004590 computer program Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30256—Lane; Road marking
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
- Y02A30/60—Planning or developing urban green infrastructure
Abstract
Embodiments of the present disclosure provide a road surface segmentation method, apparatus, device and computer readable storage medium based on road characteristics. The method comprises the steps of obtaining a road image and selecting a road sample area from the road image; classifying the road sample area to obtain a road sample class data set; dividing the road image according to the road sample class data set to obtain a preliminary road division result; and processing the preliminary road segmentation result to obtain a final road segmentation result. In this way, accurate road segmentation can be performed in a complex road environment.
Description
Technical Field
Embodiments of the present disclosure relate generally to the field of road segmentation and, more particularly, relate to a road surface segmentation method, apparatus, device, and computer-readable storage medium based on road features.
Background
The segmentation of road images is one of the key technologies in the fields of automatic driving, road surface asset management, road surface technical condition management, and the like. The purpose of road segmentation is to distinguish "roads" from "non-roads".
The traditional road segmentation method mainly comprises methods of clustering, threshold segmentation, region growth and the like based on a traditional image processing method, and a road segmentation algorithm based on deep learning. Under a complex road environment (such as a shadow road, a ponding road, a condition that the road and surrounding buildings are connected into a whole), the method/algorithm has the problems of extracting road boundary deficiency, extracting non-road areas with the same material in the road and the like, and further influences the subsequent business application based on the road segmentation result.
Disclosure of Invention
According to an embodiment of the present disclosure, a road surface segmentation scheme based on road characteristics is provided.
In a first aspect of the present disclosure, a road surface segmentation method based on road characteristics is provided. The method comprises the following steps:
acquiring a road image, and selecting a road sample area from the road image;
classifying the road sample area to obtain a road sample class data set;
dividing the road image according to the road sample class data set to obtain a preliminary road division result;
and processing the preliminary road segmentation result to obtain a final road segmentation result.
Further, the classifying the road sample area to obtain a road sample category dataset includes:
establishing an RGB three-dimensional space;
drawing the road sample area in the RGB three-dimensional space according to the RGB three-way channel value of the road sample area;
counting the occurrence frequency of pixels of the road sample area in each square in the RGB three-dimensional space, sequencing the pixels according to the sequence from high to low, and selecting pixels with the occurrence frequency greater than a first threshold value to participate in road sample classification;
and classifying the road samples according to the difference value of the channel value and the Euclidean distance of each road sample to obtain a road sample class data set.
Further, the step of dividing the road image according to the road sample class data set to obtain a preliminary road division result includes:
calculating RGB difference values and Euclidean distances between the road image and the sample class data set;
and carrying out road segmentation according to the RGB difference value and the Euclidean distance to obtain a primary road segmentation result.
Further, the road segmentation is performed according to the RGB difference value and the euclidean distance, and obtaining the preliminary road segmentation result includes:
road segmentation is carried out through pixel-by-pixel operation according to the RGB difference value and the Euclidean distance, and if the RGB difference value is smaller than a difference threshold value and the Euclidean distance is smaller than a distance threshold value, the road segmentation is calibrated as a road surface; and if the RGB difference value is greater than or equal to the difference value threshold value and the Euclidean distance is greater than or equal to the distance threshold value, calibrating as a non-road surface.
Further, the processing the preliminary road segmentation result to obtain a final road segmentation result includes:
determining the pavement pattern spots of the road according to the preliminary road segmentation result;
determining four boundary points of the road according to the road surface pattern spots of the road;
determining the edge of the pavement according to the four boundary points of the pavement:
performing edge fitting on edge points in the road surface edge to obtain a final road surface edge;
and obtaining a final road segmentation result according to the final road surface edge.
Further, the determining the road surface edge according to the four boundary points of the road surface comprises:
according to the four boundary points of the pavement, determining a leftmost pixel point set between the two boundary points on the left side and a rightmost pixel point set between the two boundary points on the right side;
obtaining a first road surface edge according to four boundary points, a leftmost pixel point set and a rightmost pixel point set of the road surface;
and slope screening is carried out on edge points in the first pavement edge to obtain the pavement edge.
Further, the slope screening of the edge points in the first road surface edge to obtain the road surface edge includes:
calculating the slope between two adjacent edge points in the leftmost pixel point set to obtain the slope value of the edge point with the highest left slope frequency;
calculating the slope between two adjacent edge points in the rightmost pixel point set to obtain the slope value of the edge point with the highest right slope frequency;
and removing points with slope difference values of edge points with highest left slope frequency larger than a second threshold value from the first pavement edge in the leftmost pixel point set, and points with slope difference values of edge points with highest right slope frequency larger than the second threshold value from the rightmost pixel point set, so as to obtain the pavement edge.
In a second aspect of the present disclosure, a road surface segmentation apparatus based on road characteristics is provided. The device comprises:
the acquisition module is used for acquiring a road image and selecting a road sample area;
the classification module is used for classifying the road sample area to obtain a road sample class data set;
the first segmentation module is used for segmenting the road image according to the road sample class data set to obtain a preliminary road segmentation result;
and the second segmentation module is used for processing the preliminary road segmentation result to obtain a final road segmentation result.
In a third aspect of the present disclosure, an electronic device is provided. The electronic device includes: a memory and a processor, the memory having stored thereon a computer program, the processor implementing the method as described above when executing the program.
In a fourth aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a method as according to the first aspect of the present disclosure.
According to the road feature-based road surface segmentation method provided by the embodiment of the application, a road sample area is selected by acquiring a road image; classifying the road sample area to obtain a road sample class data set; dividing the road image according to the road sample class data set to obtain a preliminary road division result; and processing the preliminary road segmentation result to obtain a final road segmentation result, so that accurate road segmentation can be realized in a complex road environment.
It should be understood that what is described in this summary is not intended to limit the critical or essential features of the embodiments of the disclosure nor to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The above and other features, advantages and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, wherein like or similar reference numerals denote like or similar elements, in which:
FIG. 1 illustrates a flow chart of a road feature-based pavement segmentation method according to an embodiment of the present disclosure;
FIG. 2 shows a schematic diagram of a road image according to an embodiment of the present disclosure;
FIG. 3 illustrates a sample region schematic diagram according to an embodiment of the present disclosure;
FIG. 4 illustrates a preliminary road segmentation result schematic according to an embodiment of the present disclosure;
FIG. 5 illustrates a schematic diagram of edge fitting according to an embodiment of the present disclosure;
FIG. 6 illustrates a final road segmentation result schematic according to an embodiment of the present disclosure;
FIG. 7 illustrates a block diagram of a road surface segmentation method apparatus based on road features, according to an embodiment of the disclosure;
fig. 8 illustrates a block diagram of an exemplary electronic device capable of implementing embodiments of the present disclosure.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some embodiments of the present disclosure, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments in this disclosure without inventive faculty, are intended to be within the scope of this disclosure.
In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.
Fig. 1 shows a flowchart of a road feature-based road surface segmentation method 100 according to an embodiment of the present disclosure, the method 100 comprising:
s110, acquiring a road image, and selecting a road sample area from the road image.
In some embodiments, the road image comprises as shown in fig. 2.
Further, the complex road environment includes environments where road surface conditions are complex and variable, such as shadow roads, ponding roads, and/or environments where roads are integrated with surrounding buildings.
In some embodiments, the area in the road image that is intermediate and near the bottom is selected as the road sample area, as shown in fig. 3.
Preferably, in the road image, a sample area is selected from a range of 0.35-0.65 (35% -65%) in the transverse direction and 0-0.1 (0-10%) in the vertical direction.
And S120, classifying the road sample area to obtain a road sample class data set.
In some embodiments, the road sample region is classified using a method of creating an RGB three-dimensional stereo space.
Specifically, an RGB three-dimensional space is created for dividing the sample area and the entire image (road image), providing a three-dimensional space in which the numerical range of the X, Y, Z axis is 1 to 26 and all are integers. Can be selected according to the actual application scene.
And drawing the road sample area in the RGB three-dimensional space according to the RGB three-channel value of the road sample area.
Preferably, in order to reduce the operation amount, the RGB three-channel values of the road sample region may be divided by 10 and rounded, and the road sample region may be drawn in the RGB three-dimensional stereo space according to the rounded RGB three-channel values.
Counting the occurrence frequency of the pixels of the road sample area in each square in the RGB three-dimensional space, sequencing the pixels according to the sequence from high to low, and selecting the pixels with the occurrence frequency greater than a first threshold value to participate in road sample classification.
Preferably, the first threshold is 90%. That is, 90% of the samples before the cumulative sum of the numbers (frequency of occurrence) are selected to participate in the classification operation. Only the pixels with high occurrence frequency are selected as samples for subsequent classification calculation, so that the operation amount can be effectively reduced, and the efficiency is improved.
The road samples are classified, the road sample categories are refined, and roads in different environments in the sample area, such as shadows, earth coverage, and the like, are distinguished.
In some embodiments, the road samples are classified according to the difference in channel values and Euclidean distance of each road sample, resulting in a road sample class data set. If the difference value of the channel values of the road samples (the difference value between RGB channels) is smaller than the difference value threshold value and the Euclidean distance is smaller than the distance threshold value, the road samples are classified into the same type of samples, and other conditions are regarded as different types.
Preferably, the difference threshold is 10; the distance threshold is 2.5.
And S130, dividing the road image according to the road sample class data set to obtain a preliminary road division result.
In some embodiments, the RGB difference and the euclidean distance between the road image and the sample class data set are calculated, and road segmentation is performed according to the RGB difference and the euclidean distance, so as to obtain a preliminary road segmentation result as shown in fig. 4.
Specifically, road segmentation is carried out through pixel-by-pixel operation according to the RGB difference value and the Euclidean distance, and if the RGB difference value is smaller than a difference threshold value and the Euclidean distance is smaller than a distance threshold value, the road is marked as a road surface; and if the RGB difference value is greater than or equal to the difference value threshold value and the Euclidean distance is greater than or equal to the distance threshold value, calibrating as a non-road surface.
Preferably, the difference threshold is 10; the distance threshold is 2.5.
And S140, processing the preliminary road segmentation result to obtain a final road segmentation result.
In some embodiments, the map spot Z with the largest area in the preliminary road segmentation result is selected as the road surface map spot of the road.
In some embodiments, four boundary points of the road surface are determined from the map spot Z.
Specifically, the row farthest from the bottom of the image (preliminary divided image) and the column thereof in the map spot Z are selected as vertices. Searching left and right demarcation points P in the vertex range UL 、P UR ；
Preferably, the distance P is selected UL 、P UR In the 100-line range, the points which are positioned at the leftmost and rightmost positions are respectively used as upper demarcation points at the left side and the right side;
with P UL The first 0 point is the lower left corner point and is denoted as P LL The method comprises the steps of carrying out a first treatment on the surface of the With P UR The first point listed as the image width value is the lower right corner point and is marked as P LR ；
That is, four boundary points of the road surface are determined, respectively:
upper left boundary point P UL ；
Lower left boundary point P LL ；
Upper right boundary point P UR ；
Lower right boundary point P LR 。
In some embodiments, the road surface edge is determined from the four boundary points.
Specifically, by using the characteristic that the column value of the road edge is the maximum value and the minimum value of the row, searching the leftmost pixel point set between the left edge point PUL and the PLL and the rightmost pixel point set between the right edge point PUR and the PLR by row units to form an optimized edge B1 (first road surface edge). That is, the middle points of the roads in the same row are removed, and only the outermost points remain.
In some embodiments, B will be 1 And deleting pixels with larger medium slope differences.
Specifically, calculating the slope between two adjacent edge points in the leftmost pixel point set to obtain the slope value of the edge point with the highest left slope frequency;
calculating the slope between two adjacent edge points in the rightmost pixel point set to obtain the slope value of the edge point with the highest right slope frequency;
removing points with slope difference larger than a second threshold value from the leftmost pixel point set and the edge point with the highest slope frequency on the left side from the first road surface edge, and points with slope difference larger than a second threshold value from the rightmost pixel point set and the edge point with the highest slope frequency on the right side from the leftmost pixel point set, so as to obtain a new edge B 2 (road surface edge).
Preferably, the second threshold is 0.05.
In some embodiments, for said B 2 Edge fitting is performed on the points in (a) to form an edge B as shown in FIG. 5 3 . Namely, the left and right edges are denoted by B 2 The points in the method are edge points, primary and secondary function fitting is carried out respectively, a function with the minimum residual error after fitting is selected as left and right edge functions, and a new edge B is formed 3 。
As shown in FIG. 6, with edge B 3 And (5) road segmentation is carried out for the road edge to obtain a final road segmentation result.
According to the embodiment of the disclosure, the following technical effects are achieved:
1. the method for establishing the RGB three-dimensional space is used for classifying the sample area, and the accuracy of data is reserved and the dimension of calculation is reduced.
2. Aiming at the complex diversity of road surface environments, two methods of edge point selection and slope screening are adopted, noise points generated by shadows, ponding and the like are removed, and then a fitting function is utilized for edge optimization, so that real road edge points are reserved, and the influence of the surrounding environment of the road is reduced.
In summary, the present disclosure has a good effect on the extraction of edges (roads under complex environments) of shadow roads, ponding roads, damaged roads and roads communicating with surrounding buildings, and can provide a high-efficiency pre-processing method for subsequent road management, road technical condition evaluation, road width, road lane extraction, etc.
It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present disclosure is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present disclosure. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all alternative embodiments, and that the acts and modules referred to are not necessarily required by the present disclosure.
The foregoing is a description of embodiments of the method, and the following further describes embodiments of the present disclosure through examples of apparatus.
Fig. 7 shows a block diagram of a road surface segmentation apparatus 700 based on road characteristics according to an embodiment of the disclosure. As shown in fig. 7, the apparatus 700 includes:
an acquisition module 710 for acquiring a road image from which a road sample area is selected;
the classification module 720 is configured to classify the road sample area to obtain a road sample class data set;
a first segmentation module 730, configured to segment the road image according to the road sample class data set, so as to obtain a preliminary road segmentation result;
and the second segmentation module 740 is configured to process the preliminary road segmentation result to obtain a final road segmentation result.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the described modules may refer to corresponding procedures in the foregoing method embodiments, which are not described herein again.
Fig. 8 shows a schematic block diagram of an electronic device 800 that may be used to implement embodiments of the present disclosure. As shown, the device 800 includes a Central Processing Unit (CPU) 801 that can perform various suitable actions and processes in accordance with computer program instructions stored in a Read Only Memory (ROM) 802 or loaded from a storage unit 808 into a Random Access Memory (RAM) 803. In the RAM803, various programs and data required for the operation of the device 800 can also be stored. The CPU801, ROM802, and RAM803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to the bus 804.
Various components in device 800 are connected to I/O interface 805, including: an input unit 806 such as a keyboard, mouse, etc.; an output unit 807 such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, or the like. The communication unit 809 allows the device 800 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The processing unit 801 performs the various methods and processes described above, such as method 100. For example, in some embodiments, the method 100 may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 800 via ROM802 and/or communication unit 809. When the computer program is loaded into RAM803 and executed by CPU801, one or more steps of method 100 described above may be performed. Alternatively, in other embodiments, CPU801 may be configured to perform method 100 by any other suitable means (e.g., by means of firmware).
The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a load programmable logic device (CPLD), etc.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
Moreover, although operations are depicted in a particular order, this should be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are example forms of implementing the claims.
Claims (8)
1. A road surface segmentation method based on road characteristics, comprising:
acquiring a road image, and selecting a road sample area from the road image;
classifying the road sample area to obtain a road sample class data set;
dividing the road image according to the road sample class data set to obtain a preliminary road division result;
processing the preliminary road segmentation result to obtain a final road segmentation result, wherein the processing comprises the following steps:
determining the pavement pattern spots of the road according to the preliminary road segmentation result;
determining four boundary points of the road according to the road surface pattern spots of the road;
determining the edge of the pavement according to the four boundary points of the pavement:
performing edge fitting on edge points in the road surface edge to obtain a final road surface edge;
obtaining a final road segmentation result according to the final road surface edge;
the determining the road surface edge according to the four boundary points of the road surface comprises:
according to the four boundary points of the pavement, determining a leftmost pixel point set between the two boundary points on the left side and a rightmost pixel point set between the two boundary points on the right side;
obtaining a first road surface edge according to four boundary points, a leftmost pixel point set and a rightmost pixel point set of the road surface;
and slope screening is carried out on edge points in the first pavement edge to obtain the pavement edge.
2. The method of claim 1, wherein classifying the road sample region to obtain a road sample class data set comprises:
establishing an RGB three-dimensional space;
drawing the road sample area in the RGB three-dimensional space according to the RGB three-way channel value of the road sample area;
counting the occurrence frequency of pixels of the road sample area in each square in the RGB three-dimensional space, sequencing the pixels according to the sequence from high to low, and selecting pixels with the occurrence frequency greater than a first threshold value to participate in road sample classification;
and classifying the road samples according to the difference value of the channel value and the Euclidean distance of each road sample to obtain a road sample class data set.
3. The method of claim 2, wherein segmenting the road image from the road sample class dataset comprises:
calculating RGB difference values and Euclidean distances between the road image and the sample class data set;
and carrying out road segmentation according to the RGB difference value and the Euclidean distance to obtain a primary road segmentation result.
4. A method according to claim 3, wherein said performing road segmentation based on the RGB differences and euclidean distance to obtain a preliminary road segmentation result comprises:
road segmentation is carried out through pixel-by-pixel operation according to the RGB difference value and the Euclidean distance, and if the RGB difference value is smaller than a difference threshold value and the Euclidean distance is smaller than a distance threshold value, the road segmentation is calibrated as a road surface; and if the RGB difference value is greater than or equal to the difference value threshold value and the Euclidean distance is greater than or equal to the distance threshold value, calibrating as a non-road surface.
5. The method of claim 1, wherein slope screening edge points in the first road surface edge to obtain the road surface edge comprises:
calculating the slope between two adjacent edge points in the leftmost pixel point set to obtain the slope value of the edge point with the highest left slope frequency;
calculating the slope between two adjacent edge points in the rightmost pixel point set to obtain the slope value of the edge point with the highest right slope frequency;
and removing points with slope difference values of edge points with highest left slope frequency larger than a second threshold value from the first pavement edge in the leftmost pixel point set, and points with slope difference values of edge points with highest right slope frequency larger than the second threshold value from the rightmost pixel point set, so as to obtain the pavement edge.
6. A road surface segmentation device based on road characteristics, characterized by comprising:
the acquisition module is used for acquiring a road image and selecting a road sample area from the road image;
the classification module is used for classifying the road sample area to obtain a road sample class data set;
the first segmentation module is used for segmenting the road image according to the road sample class data set to obtain a preliminary road segmentation result;
the second segmentation module is configured to process the preliminary road segmentation result to obtain a final road segmentation result, and includes: determining the pavement pattern spots of the road according to the preliminary road segmentation result; determining four boundary points of the road according to the road surface pattern spots of the road; determining the edge of the pavement according to the four boundary points of the pavement: performing edge fitting on edge points in the road surface edge to obtain a final road surface edge; obtaining a final road segmentation result according to the final road surface edge;
the determining the road surface edge according to the four boundary points of the road surface comprises:
according to the four boundary points of the pavement, determining a leftmost pixel point set between the two boundary points on the left side and a rightmost pixel point set between the two boundary points on the right side;
obtaining a first road surface edge according to four boundary points, a leftmost pixel point set and a rightmost pixel point set of the road surface;
and slope screening is carried out on edge points in the first pavement edge to obtain the pavement edge.
7. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, characterized in that the processor, when executing the program, implements the method according to any of claims 1-5.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110419236.9A CN113139975B (en) | 2021-04-19 | 2021-04-19 | Road feature-based pavement segmentation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110419236.9A CN113139975B (en) | 2021-04-19 | 2021-04-19 | Road feature-based pavement segmentation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113139975A CN113139975A (en) | 2021-07-20 |
| CN113139975B true CN113139975B (en) | 2023-11-17 |
Family
ID=76812714
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110419236.9A Active CN113139975B (en) | 2021-04-19 | 2021-04-19 | Road feature-based pavement segmentation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113139975B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4910786A (en) * | 1985-09-30 | 1990-03-20 | Eichel Paul H | Method of detecting intensity edge paths |
| CN102682292A (en) * | 2012-05-10 | 2012-09-19 | 清华大学 | Method based on monocular vision for detecting and roughly positioning edge of road |
| CN104835099A (en) * | 2015-04-29 | 2015-08-12 | 中国科学技术大学 | Urban road network motor vehicle exhaust real-time remote sensing monitoring base address selecting method |
| CN106228138A (en) * | 2016-07-26 | 2016-12-14 | 国网重庆市电力公司电力科学研究院 | A kind of Road Detection algorithm of integration region and marginal information |
| CN106446864A (en) * | 2016-10-12 | 2017-02-22 | 成都快眼科技有限公司 | Method for detecting feasible road |
| CN111046768A (en) * | 2019-12-04 | 2020-04-21 | 武汉大学 | Deep learning method for simultaneously extracting road pavement and center line of remote sensing image |
| CN111439259A (en) * | 2020-03-23 | 2020-07-24 | 成都睿芯行科技有限公司 | Agricultural garden scene lane deviation early warning control method and system based on end-to-end convolutional neural network |
| CN112070037A (en) * | 2020-09-11 | 2020-12-11 | 中国科学院空天信息创新研究院 | Road extraction method, device, medium and equipment based on remote sensing image |
| CN112215852A (en) * | 2020-09-29 | 2021-01-12 | 忻州师范学院 | Digital image segmentation method based on cluster learning device integration |
| CN112580511A (en) * | 2020-12-18 | 2021-03-30 | 广州市城市规划设计所 | Method, device, equipment and storage medium for estimating road area rate |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2511744B (en) * | 2013-03-11 | 2020-05-20 | Reeves Wireline Tech Ltd | Methods of and apparatuses for identifying geological characteristics in boreholes |
| CN105184852B (en) * | 2015-08-04 | 2018-01-30 | 百度在线网络技术（北京）有限公司 | A kind of urban road recognition methods and device based on laser point cloud |
-
2021
- 2021-04-19 CN CN202110419236.9A patent/CN113139975B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4910786A (en) * | 1985-09-30 | 1990-03-20 | Eichel Paul H | Method of detecting intensity edge paths |
| CN102682292A (en) * | 2012-05-10 | 2012-09-19 | 清华大学 | Method based on monocular vision for detecting and roughly positioning edge of road |
| CN104835099A (en) * | 2015-04-29 | 2015-08-12 | 中国科学技术大学 | Urban road network motor vehicle exhaust real-time remote sensing monitoring base address selecting method |
| CN106228138A (en) * | 2016-07-26 | 2016-12-14 | 国网重庆市电力公司电力科学研究院 | A kind of Road Detection algorithm of integration region and marginal information |
| CN106446864A (en) * | 2016-10-12 | 2017-02-22 | 成都快眼科技有限公司 | Method for detecting feasible road |
| CN111046768A (en) * | 2019-12-04 | 2020-04-21 | 武汉大学 | Deep learning method for simultaneously extracting road pavement and center line of remote sensing image |
| CN111439259A (en) * | 2020-03-23 | 2020-07-24 | 成都睿芯行科技有限公司 | Agricultural garden scene lane deviation early warning control method and system based on end-to-end convolutional neural network |
| CN112070037A (en) * | 2020-09-11 | 2020-12-11 | 中国科学院空天信息创新研究院 | Road extraction method, device, medium and equipment based on remote sensing image |
| CN112215852A (en) * | 2020-09-29 | 2021-01-12 | 忻州师范学院 | Digital image segmentation method based on cluster learning device integration |
| CN112580511A (en) * | 2020-12-18 | 2021-03-30 | 广州市城市规划设计所 | Method, device, equipment and storage medium for estimating road area rate |
Non-Patent Citations (1)
| Title |
|---|
| 一种快速准确非结构化道路检测方法研究;武历颖;余强;计算机仿真(第009期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113139975A (en) | 2021-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113781402B (en) | Method and device for detecting scratch defects on chip surface and computer equipment | |
| US9483835B2 (en) | Depth value restoration method and system | |
| CN111709420B (en) | Text detection method, electronic device and computer readable medium | |
| CN111415363B (en) | Image edge identification method | |
| CN108537239B (en) | Method for detecting image saliency target | |
| CN110197153B (en) | Automatic wall identification method in house type graph | |
| US8744177B2 (en) | Image processing method and medium to extract a building region from an image | |
| US20170178341A1 (en) | Single Parameter Segmentation of Images | |
| EP3073443B1 (en) | 3d saliency map | |
| KR20220139292A (en) | Character segmentation method, apparatus and computer readable storage medium | |
| CN115661173B (en) | Remote sensing image-based land parcel segmentation method | |
| CN113298809B (en) | Composite material ultrasonic image defect detection method based on deep learning and superpixel segmentation | |
| CN110852207A (en) | Blue roof building extraction method based on object-oriented image classification technology | |
| CN111222511B (en) | Infrared unmanned aerial vehicle target detection method and system | |
| CN112183301A (en) | Building floor intelligent identification method and device | |
| CN111161288B (en) | Image processing method and device | |
| CN112561919A (en) | Image segmentation method, device and computer readable storage medium | |
| CN112614074B (en) | Robust vanishing point detection method and device based on response graph and clustering | |
| CN113139975B (en) | Road feature-based pavement segmentation method and device | |
| JP4967045B2 (en) | Background discriminating apparatus, method and program | |
| CN111985507A (en) | Rock three-dimensional point cloud joint trace extraction method | |
| CN114511862B (en) | Form identification method and device and electronic equipment | |
| CN111047614A (en) | Feature extraction-based method for extracting target corner of complex scene image | |
| Mahmoudpour et al. | Superpixel-based depth map estimation using defocus blur | |
| CN115880228A (en) | Multi-defect merging method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |