KR102053246B1 - System and method for detecting road - Google Patents

Abstract

A road detection system and method are disclosed. Road detection system according to an embodiment of the present invention includes a receiving unit for receiving a driving image from the photographing apparatus for photographing the target area; Determining a property of each pixel in the driving image as a road or a non-road based on the RGB value of the driving image, and generating a plurality of superpixels by grouping each pixel according to the color and position of each pixel. And a determiner configured to determine an attribute of each superpixel as a road or a non-road based on the attributes of the pixels in the superpixel.

Description

Road detection system and method {SYSTEM AND METHOD FOR DETECTING ROAD}

Embodiments of the present invention relate to road detection techniques.

Today, as interest in intelligent cars increases, research into the Advanced Driver Assistance System (ADAS) is being actively conducted. ADAS collects driving environment information by using various kinds of sensors and provides necessary information to the driver, thereby increasing the safety and convenience of the driver and the vehicle.

The driving environment is recognized by detecting roads or lanes. A conventional method of detecting a road or lane detects a road area by determining whether each pixel corresponds to a road or a non-road for each pixel existing in a driving image. However, this was a problem that the road area detection is not properly made according to the surrounding environment.

Accordingly, studies for stably detecting roads or lanes in various driving environments have been continuously conducted.

Embodiments of the present invention are to provide a road detection system for stably detecting a road or lane in various driving environments.

According to an exemplary embodiment of the present invention, a receiver for receiving a driving image from a photographing apparatus for photographing a target area; Determining a property of each pixel in the driving image as a road or a non-road based on the RGB value of the driving image, and generating a plurality of superpixels by grouping each pixel according to the color and position of each pixel. A road detection system is provided that includes a determination unit that determines a property of each of the superpixels as a road or a non-road based on the property of the pixels in the superpixel.

The determining unit may define a central lower end of the driving image as a road area and define left and right portions of the driving image as a non-road area, and based on RGB values of the road area and the non-road area, The property can be determined as road or off-road.

The determining unit generates a histogram-based probability model of the road area and the non-road area based on RGB values of the road area and the non-road area, and generates a histogram-based probability model of the road area and the non-road area. The property of each pixel may be determined as road or non-road based on.

The determining unit may compare the probability that the pixel is a road and the probability of the non-road according to the brightness value of each pixel in the driving image based on a histogram-based probability model of the road area and the non-road area. The property can be determined as road or off-road.

The determining unit may determine an attribute of the pixel as a road when the probability that the pixel is a road is equal to or greater than a probability that the road is non-road. Can be determined.

The determining unit may group each of the pixels into a superpixel when the color distance and the location distance of each pixel in the driving image are within a set value.

The determining unit may determine the property of each superpixel as a road or a non-road based on the number of pixels having a road property and the number of pixels having a non-road property in the superpixel.

The determining unit determines the property of the superpixel as a road when the ratio of the pixel having the road property in the superpixel is equal to or greater than the set value, and the superpixel when the ratio of the pixel having the non-road property in the superpixel is equal to or greater than the set value. The property of can be determined as off road.

The determining unit may determine the property of the superpixel as the third property when the ratio of the pixel having the road property in the superpixel is less than the set value and the ratio of the pixel having the non-road property in the superpixel is less than the set value. The superpixel attribute of the third attribute may be determined as a road or a nonroad based on the RGB similarity between the superpixel of the third attribute and the road and the superpixel of the nonroad attribute.

The determining unit generates a histogram-based probability model of the superpixel of the third attribute based on the RGB value of the superpixel of the third attribute, and the histogram-based probability model of the superpixel of the third attribute and the road. The similarity between the histogram-based probabilistic model of the region and the non-road region may determine the property of the superpixel of the third attribute as a road or a non-road.

According to another exemplary embodiment of the present invention, there is provided a road detection method using a road detection system, the method comprising: receiving a driving image from a photographing apparatus photographing the front of a vehicle; Determining an attribute of each pixel in the driving image as a road or a non-road based on an RGB value of the driving image; Generating a plurality of superpixels by grouping each pixel according to a color and a position of each pixel; And determining the property of each superpixel as a road or a non-road based on the property of the respective pixels in the superpixel.

The determining of the property of each pixel in the driving image as a road or a non-road is based on the RGB value of the driving image. The center lower portion of the driving image is defined as a road area and the left and right portions of the driving image are non-shown. The road area may be defined as a road area, and an attribute of each pixel may be determined as a road or a non-road based on RGB values of the road area and the non-road area.

The determining of the property of each pixel in the driving image as a road or a non-road based on the RGB value of the driving image may include: the road area and the non-road area based on RGB values of the road area and the non-road area. A histogram based probabilistic model may be generated, and an attribute of each pixel may be determined as a road or a nonroad based on the histogram based probabilistic models of the road area and the non-road area.

The determining of the property of each pixel in the driving image as a road or a non-road is based on the RGB value of the driving image, based on the histogram-based probability model of the road area and the non-road area. By comparing the probability that the pixel is a road and the non-road according to the brightness value of the pixel, an attribute of each pixel may be determined as a road or a non-road.

The determining of the property of each pixel in the driving image as a road or a non-road is based on the RGB value of the driving image. When the probability that the pixel is a road is greater than or equal to the probability of being a non-road, the property of the pixel is defined as a road. If the probability that the pixel is a road is smaller than the probability that the pixel is a non-road, the property of the pixel may be determined as the non-road.

The generating of the plurality of superpixels by grouping the respective pixels according to the color and position of each pixel may include converting each pixel into a superpixel when the color distance and the position distance of each pixel in the driving image are within a set value. Can be grouped

The determining of the property of each superpixel as a road or a non-road based on the property of each of the pixels in the superpixel includes: the number of pixels having a road property and the number of pixels having a non-road property in each superpixel. An attribute of each of the superpixels may be determined as road or non-road based on.

The determining of the property of each of the superpixels as a road or a non-road based on the property of the pixels in the superpixel is performed when the ratio of pixels having a road property in the superpixel is equal to or greater than a set value. May be determined as a road, and when the ratio of pixels having a non-road property in the super pixel is equal to or greater than a set value, the property of the super pixel may be determined as non-road.

Determining the property of each superpixel as a road or a non-road based on the property of each of the pixels in the superpixel is that the ratio of pixels having a road property in the superpixel is less than a set value while the ratio in the superpixel is too high. If the ratio of pixels having a low property is less than a set value, the property of the superpixel is determined as a third property, and based on the RGB similarity between the superpixel of the third property and the superpixel of the road and the non-road property. The superpixel attribute of the third attribute may be determined as road or non-road.

Determining the property of each superpixel as a road or a non-road based on the property of the respective pixels in the superpixel is based on the superpixel of the third property based on the RGB value of the superpixel of the third property. Producing a histogram-based probability model and evaluating the similarity between the histogram-based probability model of the superpixel of the third property and the histogram-based probability model of the road area and the non-road area. The property can be determined as road or off-road.

According to another exemplary embodiment of the present invention, there is provided a road detection method using a road detection system, the method comprising: receiving a driving image from a photographing apparatus photographing the front of a vehicle; Generating a plurality of superpixels by grouping each pixel according to a color and a position of each pixel in the driving image; Determining an attribute of each pixel as a road or a non-road based on an RGB value of the driving image; And determining the property of each superpixel as a road or a non-road based on the property of the respective pixels in the superpixel.

According to another exemplary embodiment of the invention, a computer program stored in a non-transitory computer readable storage medium, the computer program comprising one or more instructions, wherein the instructions are one or more instructions. When executed by a road detection system having processors, the road detection system receives a driving image from a photographing apparatus for photographing the front of the vehicle, and sets the attribute of each pixel in the driving image based on the RGB value of the driving image. Determine as road or non-road, group each pixel according to the color and position of each pixel to generate a plurality of superpixels, and attribute of each superpixel based on the attribute of each pixel in the superpixel Provided with a computer program for determining the road or road .

According to an embodiment of the present invention, the property of each superpixel is determined as a road or a non-road based on the property of each pixel in the superpixel, thereby stably detecting the road or lane in various driving environments by minimizing the influence of noise. can do.

1 is a block diagram of a road detection system according to an embodiment of the present invention.
2 is an exemplary view of a driving image according to an embodiment of the present invention.
3 is a flowchart illustrating a road detection method according to an embodiment of the present invention.
4 is a block diagram illustrating and describing a computing environment including a computing device suitable for use in example embodiments.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings, and the same or similar components are denoted by the same reference numerals regardless of the reference numerals, and redundant description thereof will be omitted. The suffixes "module" and "unit" for components used in the following description are given or used in consideration of ease of specification, and do not have distinct meanings or roles from each other. In addition, in describing the embodiments disclosed herein, when it is determined that the detailed description of the related known technology may obscure the gist of the embodiments disclosed herein, the detailed description thereof will be omitted. In addition, the accompanying drawings are intended to facilitate understanding of the embodiments disclosed herein, but are not limited to the technical spirit disclosed herein by the accompanying drawings, all changes included in the spirit and scope of the present invention. It should be understood to include equivalents and substitutes.

Terms including ordinal numbers such as first and second may be used to describe various components, but the components are not limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

When a component is referred to as being "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that other components may be present in between. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between.

Singular expressions include plural expressions unless the context clearly indicates otherwise.

In this application, the terms "comprises" or "having" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. It is apparent to those skilled in the art that the present invention can be embodied in other specific forms without departing from the spirit and essential features of the present invention.

1 is a block diagram of a road detection system 100 according to an embodiment of the present invention, Figure 2 is an illustration of a driving image according to an embodiment of the present invention. Referring to FIG. 1, the road detection system 100 according to an embodiment of the present invention includes a receiver 110 and a determiner 120. The above components are described or illustrated as separate components, but these are described or illustrated as separate components in order to describe each functional part separately. Thus, it will be apparent to those skilled in the art that the function of each of the components may be performed in a separate part, as well as the function of each of the components may be performed in the same part. In addition, each configuration may be connected in a wired or wireless form including an electrical connection to each other so as to transmit or receive data.

The receiver 110 receives a driving image from a photographing apparatus photographing a target area. The target area may be an image of the front of the vehicle. However, the target area is not limited thereto. The receiver 110 may include one or more components that enable communication with the photographing apparatus, and may include, for example, a short range communication module, a wired communication module, and a wireless communication module.

The determination unit 120 determines the property of each pixel in the driving image as a road or a non-road based on the RGB value of the driving image, and generates a plurality of superpixels by grouping each pixel according to the color and position of each pixel. Based on the property of each pixel in the superpixel, the property of each superpixel may be determined as a road or a non-road. Hereinafter, look at in detail.

First, the process of determining the properties of each pixel will be described. The determination unit 120 may define a central lower end of the driving image as a road area and define left and right portions of the driving image as a non-road area. Because of the camera's field of view and perspective effects, the lower center of the driving image is usually a road, and the left and right portions of the driving image are non-roads (eg, obstacles such as nearby vehicles, trees, and buildings).

The determiner 120 may generate a histogram-based probability model of the road and non-road areas based on the RGB values of the road and non-road areas. The determiner 120 compares the probability of the pixel being the road and the probability of the non-road according to the brightness value of each pixel in the driving image based on the histogram-based probability models of the road area and the non-road area. Can be determined as road or non-road. For example, the determination unit 120 determines the property of the pixel as the road when the probability that the pixel is the road is greater than or equal to the probability that the road is non-road, and determines the property of the pixel when the probability that the pixel is the road is less than the probability that the road is non-road. You can decide by road. On the other hand, the attribute determination of the pixel may be limited to the area under the horizontal line in the driving image.

On the other hand, before determining the attribute of each pixel, correction of the radiation distortion in the traveling image may be performed.

Next, we will look at the process of creating a superpixel. The determination unit 120 may group each pixel into a superpixel when the color distance and the location distance of each pixel in the driving image are within a set value. The color distance is the difference of RGB values, and the position distance is the difference of coordinates. Superpixelation may be achieved through Simple Linear Interactive Clustering (SLIC). However, the present invention is not limited thereto. On the other hand, super-faxization may be performed by being limited to the area under the horizontal line in the driving image.

Finally, we look at the process of determining the properties of each superpixel. The determination unit 120 may determine the property of each superpixel as a road or a non-road based on the number of pixels having a road property and a non-road property in the superpixel. For example, when the ratio of the pixel having the road property in the superpixel is equal to or greater than a set value (for example, 80%), the determination unit 120 determines the property of the superpixel as the road, and determines the pixel of the non-road property in the superpixel. If the ratio is more than the set value (for example, 80%), the superpixel property can be determined as nonroad.

In addition, the determiner 120 determines that the ratio of the pixels having the road property in the superpixel is less than the set value (eg, 80%) and the ratio of the pixels having the non-road property in the superpixel is set (eg, 80%). If less, the attribute of the superpixel may be determined as a third attribute. Then, the determiner 120 generates a histogram-based probability model of the superpixel of the third attribute based on the RGB value of the superpixel of the third attribute, and the histogram-based probability model and the road of the superpixel of the third attribute. The similarity evaluation of the histogram-based probability model of the region and the non-road region may determine the super pixel attribute of the third attribute as the road or the non-road. Similarity assessment can be done through Bhattacharyya's distance measurement.

3 is a flowchart illustrating a road detection method according to an embodiment of the present invention. In the illustrated flow chart, the method is divided into a plurality of steps, but at least some of the steps may be performed in a reverse order, in combination with other steps, omitted, divided into substeps, or not shown. One or more steps may be added and performed.

The driving image is received (S202). In detail, the driving image may be received from the photographing apparatus photographing the target area.

An attribute of each pixel is determined (S204). In detail, an attribute of each pixel in the driving image may be determined as a road or a non-road based on an RGB value of the driving image.

A superpixel is generated (S206). A plurality of superpixels may be generated by grouping each pixel according to the color and position of each pixel.

The property of each superpixel is determined (S208). An attribute of each superpixel may be determined as a road or a non-road based on an attribute of each pixel in the superpixel.

4 is a block diagram illustrating and describing a computing environment 10 that includes a computing device suitable for use in example embodiments.

In the illustrated embodiment, each component may have different functions and capabilities in addition to those described below, and may include additional components in addition to those described below.

The illustrated computing environment 10 includes a computing device 12. In one embodiment, computing device 12 may be a road detection system (eg, determiner 120). Computing device 12 includes at least one processor 14, computer readable storage medium 16, and communication bus 18. The processor 14 may cause the computing device 12 to operate according to the example embodiments mentioned above. For example, processor 14 may execute one or more programs stored in computer readable storage medium 16. The one or more programs may include one or more computer executable instructions that, when executed by the processor 14, cause the computing device 12 to perform operations in accordance with an exemplary embodiment. Can be.

Computer readable storage medium 16 is configured to store computer executable instructions or program code, program data and / or other suitable forms of information. The program 20 stored in the computer readable storage medium 16 includes a set of instructions executable by the processor 14. In one embodiment, computer readable storage medium 16 includes memory (volatile memory, such as random access memory, nonvolatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash Memory devices, or any other form of storage medium that is accessible by computing device 12 and capable of storing desired information, or a suitable combination thereof.

The communication bus 18 interconnects various other components of the computing device 12, including the processor 14 and the computer readable storage medium 16.

Computing device 12 may also include one or more input / output interfaces 22 and one or more network communication interfaces 26 that provide an interface for one or more input / output devices 24. The input / output interface 22 and the network communication interface 26 are connected to the communication bus 18. The input / output device 24 may be connected to other components of the computing device 12 via the input / output interface 22. Exemplary input / output device 24 may include a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touchpad or touchscreen), a voice or sound input device, various types of sensor devices, and / or imaging devices. Input devices, and / or output devices such as display devices, printers, speakers, and / or network cards. The example input / output device 24 may be included inside the computing device 12 as one component of the computing device 12, and may be connected to the computing device 12 as a separate device from the computing device 12. It may be.

The above detailed description should not be construed as limiting in all respects but should be considered as illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

100: road detection system
110: receiver
120: detector

Claims (22)

A receiver configured to receive a driving image from a photographing apparatus photographing a target area; And
Define a lower center portion of the driving image as a road area and define left and right portions of the driving image as a non-road area,
Generate a histogram based probabilistic model of the road area and the non-road area based on the RGB values of the road area and the non-road area,
Based on the histogram-based probability models of the road area and the non-road area, the probability that the pixel is the road is compared with the probability that the pixel is the road and the non-road according to the brightness value of each pixel in the driving image. If the property is greater than or equal to the probability of being nonroad, the pixel property is determined as a road. If the probability that the pixel is a road is less than the probability of being nonroad, the property of the pixel is determined as a nonroad,
When the color distance and the position distance of each pixel are within a set value, the pixels are grouped to generate a plurality of superpixels.
If the ratio of pixels having a road property in the superpixel is equal to or greater than a set value, the property of the superpixel is determined as a road,
If the ratio of the pixel having the non-road property in the super pixel is more than the set value, the property of the super pixel is determined as the non-road,
When the ratio of the pixel having the road property in the superpixel is less than the set value and the ratio of the pixel having the non-road property in the superpixel is less than the set value, the property of the superpixel is determined as the third property, and the third property is determined. Generate a histogram-based probability model of the superpixel of the third attribute based on the RGB value of the superpixel of the attribute, and generate the histogram-based probability model of the superpixel of the third attribute and the And a determining unit determining a property of the superpixel of the third property as a road or a non-road through similarity evaluation of a histogram-based probability model.
delete delete delete delete delete delete delete delete delete In the road detection method using a road detection system,
Receiving a driving image from a photographing apparatus photographing the front of the vehicle;
A lower center portion of the driving image is defined as a road area, and left and right portions of the driving image are defined as a non-road area, and the road area and the non-road area are based on RGB values of the road area and the non-road area. Probability model based on the histogram-based probabilistic model of the road area and the non-road area is generated and a probability that the pixel is a road and a non-road is based on the brightness value of each pixel in the driving image. By comparison, the property of the pixel is determined as a road when the probability that the pixel is a road is equal to or greater than the probability that the road is non-road. Doing;
Generating a plurality of superpixels by grouping each pixel when the color distance and the position distance of each pixel are within a set value; And
If the ratio of the pixel having the road property in the superpixel is equal to or greater than the set value, the property of the superpixel is determined as the road. If the ratio of the pixel having the road property in the superpixel is less than the set value and the ratio of the pixel having the non-road property in the superpixel is less than the set value, the property of the superpixel is determined as the third property. Generate a histogram-based probability model of the superpixel of the third attribute based on the RGB value of the superpixel of the third attribute, generate a histogram-based probability model of the superpixel of the third attribute, the road region, The property of the superpixel of the third property is determined by the similarity evaluation of the histogram-based probability model of the non-road area. , Road detecting comprises determining in a non-.
delete delete delete delete delete delete delete delete delete delete A computer program stored on a non-transitory computer readable storage medium,
The computer program includes one or more instructions that, when executed by a road detection system having one or more processors, cause the road detection system to:
Receive a driving image from the photographing device for photographing the front of the vehicle,
A lower center portion of the driving image is defined as a road area, and left and right portions of the driving image are defined as a non-road area, and the road area and the non-road area are based on RGB values of the road area and the non-road area. Probability model based on the histogram-based probabilistic model of the road area and the non-road area is generated and a probability that the pixel is a road and a non-road is based on the brightness value of each pixel in the driving image. In comparison, when the probability that the pixel is a road is greater than or equal to the probability that the road is non-road, the attribute of the pixel is determined as the road, and when the probability that the pixel is the road is less than the probability that the road is non-road, the attribute of the pixel is determined as the non-road and,
Generating a plurality of superpixels by grouping each pixel when the color distance and the position distance of each pixel are within a set value, and
If the ratio of the pixel having the road property in the superpixel is equal to or greater than the set value, the property of the superpixel is determined as the road. If the ratio of the pixel having the road property in the superpixel is less than the set value and the ratio of the pixel having the non-road property in the superpixel is less than the set value, the property of the superpixel is determined as a third property Generate a histogram-based probability model of the superpixel of the third attribute based on the RGB value of the superpixel of the third attribute, generate a histogram-based probability model of the superpixel of the third attribute, the road region, The property of the superpixel of the third property is determined by the similarity evaluation of the histogram-based probability model of the non-road area. A computer program stored on the storage medium to be determined in a non-.

Images