KR101687094B1 - Apparatus for recognizing traffic sign and method thereof - Google Patents

Abstract

According to an embodiment of the present invention, a traffic sign area is detected by detecting a circular pattern on a target image, and a speed sign included in the traffic sign area is recognized. The speed sign is set at a minimum speed A traffic sign recognition device and a traffic sign recognition method which can be recognized as either a sign or a top speed sign.
To this end, the traffic sign recognition apparatus according to an embodiment includes an image acquisition unit for acquiring a target image; And a control unit for detecting a circular pattern on the target image to extract a traffic sign area and recognizing a speed sign included in the traffic sign area, wherein the control unit controls the traffic sign area based on a specific pattern existing in the traffic sign area The speed sign may be recognized as one of the lowest speed sign and the highest speed sign.

Description

FIELD OF THE INVENTION [0001] The present invention relates to a traffic sign recognition apparatus and a traffic sign recognition method,

The present invention relates to a traffic sign recognition apparatus and a traffic sign recognition method. More particularly, the present invention relates to a traffic sign recognition apparatus and a traffic sign recognition method for detecting a traffic sign area included in the traffic sign area, The present invention relates to a traffic sign recognizing apparatus and a traffic sign recognizing method capable of recognizing the speed signboard as one of a lowest speed sign and a highest speed sign based on a specific pattern existing in a region.

In general, due to the improvement of the CPU calculation ability and the development of the integration degree, the image processing of the images photographed in real time becomes possible.

Recently Driver Assistant System provides traffic information to driver in real time automatically through computing power, so that it is convenient for driving and accident prevention.

Among them, TSR (Traffic Sign Recognition) system is a system that notifies the driver of the current speed limit, minimum speed, and other road information by recognizing the traffic sign in front using the camera installed in the windshield inside the vehicle.

In particular, recognition of traffic signs accurately using images is becoming a very important issue in the performance of TSR systems.

The existing system can not distinguish between the maximum speed sign and the lowest speed sign, so there is a problem that the recognition rate and the reliability of the system are lowered due to the erroneous recognition when the minimum speed and the maximum speed simultaneously occur on the expressway.

According to an aspect of the present invention, there is provided a traffic sign area detection method comprising: detecting a circular pattern on a target image to extract a traffic sign area; recognizing a speed sign included in the traffic sign area, And a traffic sign recognizing device and a traffic sign recognizing method capable of recognizing the speed signboard as one of a lowest speed sign and a highest speed sign.

According to an aspect of the present invention, there is provided an apparatus for recognizing a traffic sign, comprising: an image acquiring unit acquiring a target image; And a control unit for detecting a circular pattern on the target image to extract a traffic sign area and recognizing a speed sign included in the traffic sign area, wherein the control unit controls the traffic sign area based on a specific pattern existing in the traffic sign area The speed sign may be recognized as one of the lowest speed sign and the highest speed sign.

According to an embodiment of the present invention, the control unit sets a certain region on the target image as a ROI based on a specific criterion, and detects a circular pattern on the ROI to extract the traffic sign region .

As an example related to the present specification, the specific criterion may be an area corresponding to the upper 60% of the target image.

As an example related to the present specification, the control unit may detect the circular pattern through a Radial Symmetry algorithm.

In one embodiment of the present invention, the control unit extracts target feature points corresponding to the traffic sign area, and determines the target feature points based on the SVM (Supprot Vector Machine) Comparing the reference feature points with the reference feature points, and recognizing the speed sign based on the comparison result.

As an example related to the present specification, the specific pattern may be an under-bar pattern.

In one embodiment of the present invention, the control unit recognizes the speed sign as a minimum speed sign when the under-bar pattern exists in the traffic sign area, and recognizes the speed sign as a speed sign .

As an example related to the present specification, the control unit may extract at least one of the lowest limit speed information and the highest limit speed information from the recognized speed indicator.

In order to achieve the above objects, a Smart Cruise Control (SCC) system according to the present invention is a Smart Cruise Control (SCC) system for controlling a speed of a vehicle, comprising: a communication unit for communicating with a traffic sign recognition apparatus; And a control unit for controlling the communication unit to obtain at least one of the lowest limit speed information and the highest limit speed information from the traffic sign recognizing apparatus and controlling the communication unit based on at least one of the lowest limit speed information and the highest limit speed information The traffic sign recognition device may be a traffic sign recognition device according to the above-described embodiments.

According to another aspect of the present invention, there is provided a traffic sign recognition method comprising: obtaining a target image; Detecting a circular pattern on the target image and extracting a traffic sign area; And recognizing a speed sign included in the traffic sign area as a lowest speed sign and a highest speed sign based on a specific pattern existing in the traffic sign area.

According to one embodiment of the present invention, the step of extracting the traffic sign region includes: setting a certain region on the target image as a ROI based on a specific criterion; and detecting a circular pattern on the ROI And extracting the traffic sign area.

According to the traffic sign recognition apparatus and the traffic sign recognition method according to the embodiment of the present invention, the traffic sign area is extracted by detecting the circular pattern on the target image, the speed sign included in the traffic sign area is recognized, The recognition speed and the reliability of the speed sign can be improved by recognizing the speed sign as any of the lowest speed sign and the highest speed sign based on the specific pattern existing in the sign area.

1 is a block diagram of a traffic sign recognition apparatus according to an embodiment of the present invention.
2 is an exemplary diagram illustrating a speed sign recognition process according to an embodiment disclosed herein.
FIGS. 3 to 6 are exemplary views illustrating a preprocessing process according to an embodiment disclosed herein.
7 to 8 are exemplary diagrams illustrating a detection process according to an embodiment disclosed herein.
9 through 17 are diagrams illustrating an example of a recognition process according to an embodiment disclosed herein.
18 is a configuration diagram illustrating an SCC system according to one embodiment disclosed herein.
19 is a flowchart illustrating a traffic sign recognition method according to an embodiment disclosed herein.

The technique disclosed in this specification can be applied to a traffic sign recognition device and a traffic sign recognition method which recognize a traffic sign on a road on which a vehicle is operated.

For example, the traffic sign recognition apparatus may be a TSR (Traffic Sign Recognition) system.

In particular, the techniques disclosed herein can be applied to a recognition apparatus and recognition method for recognizing a speed sign indicating a limit speed related to a minimum speed to a maximum speed.

Further, the technique disclosed in this specification can be applied to a vehicle control system associated with an electronic control unit (ECU) or an electronic control unit to which the traffic sign recognition device or the recognition method is applied.

For example, the techniques disclosed herein may be applied to a Smart Cruise Control (SCC) device or an overspeed prevention system to which the traffic sign recognition device or recognition method is applied.

As described in detail with respect to the vehicle control system associated with the electronic control unit (ECU) or the electronic control unit, an electronic control unit (ECU) used in the vehicle detects from various sensors mounted on the vehicle A control unit for receiving a signal and recognizing a state of a vehicle and operating various actuators, comprising: an input interface for converting a sensing signal input from various sensors; a microcomputer for performing calculation or logical calculation of input data in a predetermined order; And an output interface for converting the result into an operation signal of a driver.

Vehicle control systems associated with the electronic control unit include Electronic Power Steering (EPS), Anti-Lock Braking System (ABS), and Continuous Damper Control (CDC), and the suspension, braking, Thereby improving the safety and ride comfort of the vehicle.

In particular, the ECS (Electronic Control Suspension) is equipped with an electronically controlled hydraulic device according to the driving state of the vehicle, so that the strength and the height of the suspension are automatically controlled by the road condition and the vehicle speed to secure the grounding force and driving force to the road surface. Thereby improving driving safety and ride comfort as a result of the response. In addition, ASV (Advanced Safety Vehicle) improves the safety of vehicles and prevents vehicle accidents by using technology such as drowsy driving alarm system, nighttime obstacle detection system, and danger warning system. Therefore, To be reduced.

In recent years, there has been a growing demand for the stability of the vehicle and the convenience of the driver while driving, and active research on the active safety system and the driver assist system of the vehicle has been actively conducted. The active safety systems include Electronic Stability Control (ESC) and Active Front Steering (AFS). Operator assistance systems include SCC (Smart Cruise Control) or ACC (Adaptive Cruise Control), CA (Collision Avoidance), LKS Keeping Support).

It is noted that the technical terms used herein are used only to describe specific embodiments and are not intended to limit the scope of the technology disclosed herein. Also, the technical terms used herein should be interpreted as being generally understood by those skilled in the art to which the presently disclosed subject matter belongs, unless the context clearly dictates otherwise in this specification, Should not be construed in a broader sense, or interpreted in an oversimplified sense. In addition, when a technical term used in this specification is an erroneous technical term that does not accurately express the concept of the technology disclosed in this specification, it should be understood that technical terms which can be understood by a person skilled in the art are replaced. Also, the general terms used in the present specification should be interpreted in accordance with the predefined or prior context, and should not be construed as being excessively reduced in meaning.

Also, the singular forms "as used herein include plural referents unless the context clearly dictates otherwise. In this specification, the terms "comprising ", or" comprising "and the like should not be construed as necessarily including the various elements or steps described in the specification, Or may be further comprised of additional components or steps.

Furthermore, terms including ordinals such as first, second, etc. used in this specification can be used to describe various elements, but the elements should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, wherein like reference numerals denote like or similar elements, and redundant description thereof will be omitted.

Further, in the description of the technology disclosed in this specification, a detailed description of related arts will be omitted if it is determined that the gist of the technology disclosed in this specification may be obscured. It is to be noted that the attached drawings are only for the purpose of easily understanding the concept of the technology disclosed in the present specification, and should not be construed as limiting the spirit of the technology by the attached drawings.

A traffic sign recognition apparatus according to an embodiment disclosed herein.

A traffic sign recognition apparatus according to an embodiment of the present invention includes an image acquisition unit for acquiring a target image and a traffic sign area by detecting a circular pattern on the target image and displaying a speed sign included in the traffic sign area And the control unit recognizes the speed signboard as one of the lowest speed sign and the highest speed sign based on a specific pattern existing in the traffic sign area.

According to an embodiment, the control unit may set a certain region on the target image as an ROI based on a specific criterion, and detect a circular pattern on the ROI to extract the traffic sign region .

According to an embodiment, the specific criterion may be an area corresponding to the upper 60% of the target image.

According to an embodiment, the control unit may detect the circular pattern through a radial symmetry algorithm.

According to an embodiment of the present invention, the controller extracts target feature points corresponding to the traffic sign area, and based on the SVM (Supprot Vector Machine), searches the target feature points for a reference Comparing the feature points with each other, and recognizing the speed sign based on the comparison result.

Also, according to one embodiment, the specific pattern may be an under-bar pattern.

According to an embodiment of the present invention, when the under-bar pattern exists in the traffic sign area, the control unit recognizes the speed sign as a minimum speed sign, Lt; / RTI >

According to one embodiment, the control unit may extract at least one of the lowest limit speed information and the highest limit speed information from the recognized speed indicator.

Hereinafter, a traffic sign recognition apparatus according to an embodiment disclosed herein will be described in detail with reference to FIG.

However, the traffic sign recognition apparatus disclosed in FIG. 1 is merely an example, and it is apparent to those skilled in the art that the techniques disclosed herein can be applied to various types of traffic sign recognition apparatuses within the scope of the present invention.

The traffic sign recognition apparatus according to an embodiment disclosed herein may be installed, equipped, or mounted on a vehicle.

1 is a block diagram of a traffic sign recognition apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the traffic sign recognition apparatus 100 according to an embodiment of the present invention may include an image acquisition unit 110 and a control unit 120.

The traffic sign recognition apparatus 100 may further include an input unit 130, a display unit 140, and a storage unit 150.

In addition, the traffic sign recognizing apparatus 100 may further include various components for performing the traffic sign recognizing function according to the embodiment disclosed herein.

Hereinafter, the components will be described in order.

The image acquiring unit 110 may acquire an image related to a road on which the vehicle is traveling.

The image related to the road on which the vehicle is traveling is an image to be recognized by the traffic sign recognition device, and may be referred to as a target image or a target image.

According to one embodiment, the image obtaining unit 110 may include an image taking unit (not shown).

The photographing unit may photograph an image of the vehicle or a region surrounding the vehicle.

The image for the area around the vehicle may include an image for a traffic sign (e.g., a speed sign).

The image obtaining unit 110 may output the photographed image to the control unit 120. [

The control unit 120 may operate the photographing unit when a specific key (for example, a camera operation button) is selected by the user.

The photographing unit may be composed of two or more cameras according to the configuration of the vehicle or the traffic sign recognizing apparatus.

For example, the first camera included in the photographing unit may be installed at the center of the front (front) bumper of the vehicle to photograph the front (front) bumper of the vehicle and the surrounding image of the front face thereof.

The second camera included in the photographing unit may be installed at the center of a rear (rear) bumper of the vehicle to photograph a rear bumper of the vehicle and a surrounding image of the rear surface thereof.

The third camera included in the photographing unit may be installed at the center of a left sidemiller of the vehicle to photograph a left side image and a surrounding image of the left side of the vehicle.

The fourth camera included in the photographing unit may be installed at the center of the right side sidemiller of the vehicle to photograph the right side and the surrounding image of the right side of the vehicle.

Here, the installation positions of the first to fourth cameras may be installed at various positions by a designer.

The image obtaining unit 110 may process an image frame such as a still image or a moving image obtained by the photographing unit.

In this case, the processed image frame may be displayed on the display unit 140.

According to another embodiment, the image obtaining unit 110 may obtain the target image from a camera (not shown) connected by wire or wirelessly.

The configuration of the camera connected to the image acquiring unit 110 is similar to that of the camera included in the photographing unit described above, and a detailed description thereof will be omitted.

In this case, the image acquiring unit 110 may include a communication unit (not shown) for receiving an image photographed by the upper-level camera and an image processing unit (not shown) for processing the photographed image.

The communication unit can communicate with the camera via a wired / wireless network.

Here, communication between the image acquisition unit 110 and the camera through the wired / wireless network may be performed using various communication methods or communication protocols.

For example, the image acquiring unit 110 may acquire image information of the camera and a mobile communication network, a transmission control protocol / internet protocol (TCP / IP), a local area network (LAN), a wireless LAN, a Wi- (Wireless Broadband: Wibro), and World Interoperability for Microwave Access (Wimax).

The control unit 120 may control at least one component included in the traffic sign recognition apparatus 100 to provide a traffic sign recognition function according to an embodiment of the present invention.

First, the controller 120 detects a circular pattern on the target image obtained by the image obtaining unit 110, extracts a traffic sign region, and recognizes a speed sign included in the traffic sign region.

In this case, the control unit 120 may recognize the speed signboard as one of the lowest speed signboard and the highest speed signboard based on a specific pattern existing in the traffic sign area.

In order to extract the traffic sign area, the control unit 120 may set a certain area on the target image as a ROI based on a specific criterion.

In this case, the control unit 120 may detect the circular pattern on the ROI to extract the traffic sign area.

Here, the specific criterion may be an area corresponding to the upper 60% of the target image.

According to one embodiment, the controller 120 may detect the circular pattern through a radial symmetry algorithm.

Here, the radial symmetry algorithm may be a preprocessing algorithm for detecting a circular pattern in a target image.

The Radial Symmetry algorithm is a technology known in the art and will not be described in detail.

In addition, the controller 120 extracts target feature points corresponding to the traffic sign area, and based on the SVM (Supprot Vector Machine), calculates the target feature points corresponding to the plurality of reference signs, And the speed sign can be recognized based on the comparison result.

Here, the support vector machine (SVM) is a learning algorithm, which can be a learning algorithm for analyzing data or recognizing a pattern.

The SVM algorithm is a method that is mainly used in map learning and is a method of finding a hyperplane that is farthest from the data among the hyperplanes separating the data for a given data. This technique can also be applied to nonlinear classifications using kernel tricks.

Similarly, the SVM algorithm is a technology known in the art, and a detailed description will be omitted.

Here, the specific pattern may be an under-bar pattern.

According to one embodiment, when the under-bar pattern exists in the traffic sign area, the controller 120 recognizes the speed sign as a lowest speed sign, and in the opposite case, the speed sign is referred to as a highest speed sign It can be recognition.

Also, according to one embodiment, the controller 120 may extract at least one of the lowest limit speed information and the highest limit speed information from the recognized speed indicator.

The control unit 120 detects a traffic sign area by detecting a circular pattern from the target image and recognizes a speed sign on the traffic sign area, If there is an under-bar pattern on the speed sign, the speed sign is recognized as the lowest speed sign, and if there is no under-bar pattern, the speed sign is recognized as the highest speed sign, Speed or the maximum limit speed.

That is, the traffic sign recognition apparatus 100 according to an embodiment disclosed herein may first determine whether the signboard is the minimum speed or the maximum speed, and finally recognize the internal digits before recognizing the speed signboard digits.

By doing so, a traffic sign recognition apparatus 100 having a high recognition rate and reliability of a speed sign and a remarkably increased image processing speed and sign recognition speed can be provided.

Hereinafter, a traffic sign recognition process by the traffic sign recognition apparatus 100 according to an embodiment of the present invention will be described in detail with reference to FIG. 2 to FIG.

2 is an exemplary diagram illustrating a speed sign recognition process according to an embodiment disclosed herein.

Referring to FIG. 2, the traffic sign recognition apparatus 100 according to an embodiment of the present invention includes a pre-processing unit for receiving an image from a camera (S10) and detecting an area in which a circular object exists using a radial symmetry, (S20).

In addition, the traffic sign recognition apparatus 100 may detect an actual signboard using an adaboost based on Haar-feature learned in off-line in the preprocessed area (S30).

When the actual signboard is detected, the traffic sign recognition apparatus 100 recognizes the current limit speed of the currently detected signboard using the multi-SVM based on the RBF Kernel (S40).

In addition, the traffic sign recognition apparatus 100 may improve reliability of the sign by tracking a plurality of frames with respect to the recognized sign (S50)

Finally, the traffic sign recognition apparatus 100 can determine an attribute of the speed sign based on the tracking result (S60).

The attribute for the speed sign may include information on whether the speed sign corresponds to the sign of the lowest speed sign and the highest speed sign, and information on the speed limit (for example, the lowest limit speed or the maximum limit speed) .

Hereinafter, each step of the speed sign recognition process will be described in detail.

1. Input Image (S10)

As described above, the traffic sign recognition apparatus 100 can acquire an input image (or a target image) from a camera provided therein or a wired or wirelessly connected camera.

2. Pre-processing (S20)

In the preprocessing process of a general object image, since the sign is detected in the entire region of interest (ROI), the computation time is long.

In the technique disclosed in this specification, the Radial-Symmetry algorithm is applied to a region of interest through a preprocessing process to roughly segment an area where a circular object exists in the image.

Since only the segmented area enters the input of the detection step, the execution time can be remarkably shortened.

FIGS. 3 to 6 are exemplary views illustrating a preprocessing process according to an embodiment disclosed herein.

3 to 6, the traffic sign recognition apparatus 100 can set an ROI in which an actual sign is present in an image (S21).

According to one embodiment, the traffic sign recognition apparatus 100 recognizes an area SA10 in which signs of 0.5 m to 5 m exist on the basis of the domestic regulations and the height of a simple sign in an input image (or a target image, I100) It can be divided into an area IA10 where the sign can not exist (see Fig. 4)

In this case, the top 60% of the image can be set to ROI.

In addition, the traffic sign recognition apparatus 100 can acquire a gradient image by applying a Sobel Edge Operator to a set region of interest (Edge detection, S22).

Specifically, the traffic sign recognition apparatus 100 can generate a horizontal gradient image HI10 and a vertical gradient image VI10 by convoluting the input image with a 3 × 3 Sobel Edge operator (see FIG. 5).

The 3x3 Sobel Edge Operator is composed of a 3x3 horizontal sobel operator and a 3x3 vertical sobel operator, and can be expressed by the following Equation 1 and Equation 1, respectively.

The traffic sign recognition apparatus 100 can generate an S-map by applying a radial symmetry to a gradient image (S23).

That is, the traffic sign recognition apparatus 100 can generate an S-map by applying a radial symmetry algorithm to the generated Gradient Image as described above.

The S-map is a probability space for a circular object. The closer it is to the center of the circular object, the brighter the brightness.

The traffic sign recognizing apparatus 100 can remove only points below the threshold value in the S-map and leave only points that can be candidates for the actual circular object.

The traffic sign recognition apparatus 100 may perform clustering of the same objects by applying labeling to the remaining points (S24).

Specifically, in the signing step S24, the traffic sign recognition apparatus 100 can binarize the image resulting from the radial symmetry to [0, 1].

For example, the traffic sign recognizing apparatus 100 makes a point of 30% or more of the maximum value of the Raidal symmetry in the image to be less than 1% and 0 in the image, (Or graphical object).

Finally, the traffic sign recognition apparatus 100 can finally generate a box having a width and an area of 50 based on the center point of the labeled object (S25).

As shown in FIG. 6, the segmented image SIlO may be the input image of the detection.

3. Detection process (Detection, S30)

7 to 8 are exemplary diagrams illustrating a detection process according to an embodiment disclosed herein.

7 to 8, the traffic sign recognition apparatus 100 can learn two horizontal / vertical detectors based on a Haar-feature in an off-line learning process.

In addition, the traffic sign recognition apparatus 100 may apply two learning units to the segmented region DI10 as a result of preprocessing, so that an overlap region in which two detectors are commonly overlapped can be regarded as a region in which an actual sign exists (See Fig. 7)

FIG. 8 shows a detection process and a detection result by the traffic sign recognition apparatus 100. FIG.

4. Recognition (S40)

9 through 17 are diagrams illustrating an example of a recognition process according to an embodiment disclosed herein.

Referring to FIGS. 9 to 17, the traffic sign recognition apparatus 100 can recognize recognition of the actual sign content with respect to the area detected by the detection result sign.

Recognition can be divided into two stages: feature point detection and SVM recognition.

First, the traffic sign recognition apparatus 100 can extract a normalized gray value from a specific region in the detected sign.

Next, the traffic sign recognition apparatus 100 can finally recognize the contents of the actual signboard using the SVM learned beforehand on the off-line of the extracted feature points.

The traffic sign recognizing apparatus 100 can determine the attribute or category of the sign before recognizing the actual internal digits, which is information related to the speed limit.

That is, the traffic sign recognition apparatus 100 can determine whether the detected speed sign is the lowest speed sign or the maximum speed sign before recognizing the number in the sign associated with the speed limit.

The lowest speed sign may have a long rectangular pattern (under-bar pattern, see Fig. 9), which is a specific pattern at the lower end, unlike the maximum speed.

The traffic sign recognition apparatus 100 recognizes a rectangular pattern and can determine the attribute of the speed sign by using the minimum speed sign if the pattern exists, and the maximum speed sign if the pattern exists.

Specifically, as shown in FIG. 9, the traffic sign recognition apparatus 100 can normalize the sign image resulting from the detection to 50X50.

Since the under-bar pattern such as '-' exists in the image of the lowest speed sign, the operation area can be reduced by ROI of the area UB10 at the lower end of the image (ROI selection).

According to one embodiment, the traffic sign recognition apparatus 100 convolutes a 3 × 3 sobel operator with an input image (or a target image) to calculate Gradient (Gx, Gy) for each of x and y directions using Equation (Detection of gradient).

Here, Ex and Ey represent the 3 × 3 horizontal sobel operator and the 3 × 3 vertical sobel operator, respectively, and I (i, j) represents the intensity value at each pixel or point of the target image.

Next, the traffic sign recognition apparatus 100 may binarize the Gx and Gy images to 0 or 1 to generate binary functions Bx and By (binarization).

Specifically, the traffic sign recognition apparatus 100 can binarize the point where the Gradient value of the Gx and Gy points is greater than or equal to the threshold value to 1, and the point that is not Gradient value to 0 (see FIGS. 10 to 11).

10, G (i, j) denotes Gx (i, j) or Gy (i, j) and B (i, j) denotes Bx (i, j) or By (i, j).

Next, the traffic sign recognition apparatus 100 can calculate the profile functions Pbex and Pbey of the binarized images Bx and By.

Specifically, the traffic sign recognition apparatus 100 may normalize each profile to [0 1].

Next, the traffic sign recognition apparatus 100 can extract peak points px and py of a threshold value or more from each profile.

FIG. 12 shows the extraction process of the peak points px and py, and FIGS. 13 to 14 show profile functions Pbex and Pbey of Bx and By and extracted peak points px and py (Peak Point Detection).

The calculation formula for calculating the profile functions Pbex and Pbey of Bx and By is shown in Equation 4 below.

Next, the traffic sign recognition apparatus 100 may perform Rectangle Detection at the end of calculation of Px and Py. That is, the traffic sign recognition apparatus 100 can calculate Rectangle (Pxl, Pyl, Pxr, Pyr) from the calculated Px and Py (Rectangle Detection).

The concrete Rectangle Detection process is as follows.

First, candidate points Px1 and Pxr are detected using the following equation (5).

Next, the final Pxmin and Pxmax are determined using Equation (6) below.

Similarly, for Pyl and Pyr, the final Pymin and Pymax can be determined using Equations (7) and (8) below.

15 shows the final recognition result of Rectangle Detection.

Next, the traffic sign recognition apparatus 100 can normalize the area resulting from the detection to 50x50.

Next, the traffic sign recognition apparatus 100 can detect feature points as shown in FIG. 16 in the normalized region (Feature Extraction).

For example, the traffic sign recognition apparatus 100 can extract feature points from the 'hundred' region BF10, the 'ten' region RF10 and the 'lower' region YF10 as shown in FIG.

In the technique disclosed in the present specification, the normalized gray value is used as a feature point to strongly respond to illumination change.

Next, the traffic sign recognition apparatus 100 can detect the feature points by preparing samples for the speed sign boards 30 to 110 in advance on the off-line.

For example, the traffic sign recognition apparatus 100 can learn minutiae points using RBF-Kernel based SVM (RBF Kernel Multi Class SVM).

The traffic sign recognition apparatus 100 can recognize the contents of the final sign by using the support vector of the SVM from the learning result.

17 shows recognition results using the SVM based on RBF-Kernel.

5. Tracking process (Tracking, S50)

The traffic sign recognition apparatus 100 can improve the reliability of the sign by tracking the recognized sign using a Kalman filter for a plurality of frames.

6. Decision (S60)

The traffic sign recognition apparatus 100 can finally determine a sign corresponding to an actual vehicle using a yaw rate of a vehicle, a lane recognition result, and the like.

Referring again to FIG. 1, the input unit 130 may receive a user input for controlling the traffic sign recognition apparatus 100.

The input unit 130 may generate input data for controlling the operation of the traffic sign recognition apparatus 100 by the user.

The input unit 130 may include a key pad, a dome switch, a touch pad (static / static), a jog wheel, a jog switch, and the like. In particular, when the touch pad has a mutual layer structure with the display unit 140, it may be referred to as a touch screen.

According to one embodiment, the input unit 130 may receive (or receive) the various user inputs described above from the user.

For example, the input unit 130 may receive a control input related to a traffic sign recognition function input on a control screen for controlling the traffic sign recognition apparatus 100 displayed by the display unit 140 .

The display unit 140 can basically display the information processed or generated by the control unit 120 or the image or information obtained by the image obtaining unit 110 on the screen.

Also, the display unit 140 may display a control screen for performing the traffic sign recognition function by the traffic sign recognition apparatus 100.

Also, the display unit 140 may display a UI (User Interface) or a GUI (Graphic User Interface) related to the function provided by the traffic sign recognition apparatus 100.

Also, the display unit 140 may display the processing result of the control unit 110 on the control screen.

For example, the display unit 140 displays a traffic sign area extracted from the target image, which is information on the target image and the target image, information on the recognized speed sign (for example, Binary information, limit speed information, etc.) can be displayed on the control screen.

The display unit 140 may be a liquid crystal display, a thin film transistor-liquid crystal display, an organic light-emitting diode, a flexible display, a three-dimensional display (3D display).

In addition, there may be two or more display units 140 according to the embodiment of the traffic sign recognition apparatus 100. [ For example, the traffic sign recognition apparatus 100 may include an external display unit (not shown) and an internal display unit (not shown) at the same time.

(Hereinafter, referred to as 'touch screen') having a mutual layer structure with a sensor (hereinafter, referred to as 'touch sensor') for sensing a touch operation with the display unit 140, It can also be used as an input device in addition to the device. The touch sensor may have the form of, for example, a touch film, a touch sheet, a touch pad, or the like.

The storage unit 150 may store the information received, processed, and generated by the traffic sign recognition apparatus 100.

For example, the storage unit 150 may store a program for processing and controlling the controller 120, and may store input / output data (for example, user input through a user interface, control processing A process or a control result, and the like).

Also, the storage unit 150 may store a user interface screen (for example, the control screen).

The storage unit 150 may be a flash memory type, a hard disk type, an SSD (solid state disk or solid state drive), a multimedia card micro type, a card type (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM) A programmable read-only memory (PROM) magnetic memory, a magnetic disk, and an optical disk.

The SCC (Smart Cruise Control) system to which the traffic sign recognition function according to the embodiment disclosed herein is applied

The SCC system according to an embodiment disclosed herein below may be implemented as a part or a combination of the configurations or steps included in the embodiments described above or may be implemented as a combination of the embodiments, The overlapping portions can be omitted for clarity of the embodiment of the present invention.

Therefore, description of the above-described traffic sign recognition apparatus and description of the overlapping SCC system can be omitted.

The technique disclosed in this specification can be applied to an SCC system capable of controlling the vehicle speed of a vehicle by applying the above-described traffic sign recognition function.

The smart cruise control device or system is a device for controlling the vehicle so as to maintain the inter-vehicle distance by measuring the inter-vehicle distance from the preceding vehicle using a radar.

here. Cruise control can mean cruise control or automatic cruise control to keep the vehicle speed constant.

That is, the SCC system detects a preceding vehicle and adjusts the speed according to the distance to the preceding vehicle, thereby preventing collision with the preceding vehicle. For this purpose, the SCC system can use a radar, which may be a sensor for measuring the distance to the preceding vehicle and the relative speed.

The SCC system capable of controlling the vehicle speed of the vehicle implemented by applying the traffic sign recognition function can be configured such that when the information on the lowest limit speed or the highest limit speed is detected through the speed signs recognized according to the above- To exceed the minimum limit speed or to be less than the maximum limit speed.

Further, the technique disclosed in this specification can be applied to an overspeed prevention system that provides the overspeed protection function for a vehicle by applying the above-described traffic sign recognition function.

The overspeed prevention system may be configured such that when the information on the lowest limit speed or the highest limit speed is detected through the speed indicator recognized according to the above embodiments, the speed of the vehicle is less than the lowest limit speed or the maximum limit speed is exceeded A warning message or the like may be output to the driver of the vehicle.

In the following, details of a Smart Cruise Control (SCC) system for controlling the speed of a vehicle to which the techniques disclosed herein are applied will be described in detail.

It will also be apparent to those skilled in the art that the contents of the Smart Cruise Control (SCC) system to be described below may be applied to the overspeed prevention system to the extent that the techniques disclosed herein can be applied.

According to an embodiment disclosed herein, a Smart Cruise Control (SCC) system for controlling the speed of a vehicle includes a communication unit that communicates with a traffic sign recognition apparatus, and a communication unit that receives the traffic information from the traffic sign recognition apparatus, And an SCC control unit for controlling the communication unit to acquire at least one of information about the vehicle speed and the vehicle speed based on at least one of the lowest limit speed information and the maximum limit speed information, The traffic sign recognizing apparatus according to the above-described embodiments.

18 is a configuration diagram illustrating an SCC system according to one embodiment disclosed herein.

Referring to FIG. 18, an SCC system 200 according to an embodiment disclosed herein may include a communication unit 210 and an SCC control unit 220.

The SCC system 200 may further include an input unit 230, a display unit 240, and a storage unit 250.

Hereinafter, the components will be described in order.

The communication unit 210 may perform communication with the traffic sign recognition apparatus 100 according to the above-described embodiments.

The communication unit 210 can communicate with the traffic sign recognition apparatus 100 through a wired / wireless network.

Here, communication between the communication unit 210 and the traffic sign recognition apparatus 100 through the wired / wireless network may be performed using various communication methods or communication protocols.

For example, the communication unit 210 may communicate with the traffic sign recognition apparatus 100 through a mobile communication network, a TCP / IP (Transmission Control Protocol / Internet Protocol), a LAN (Local Area Network), a Wireless LAN, Fi, a wireless broadband (Wibro), and a World Interoperability for Microwave Access (Wimax).

The SCC control unit 220 may control the communication unit to obtain at least one of the lowest limit speed information and the highest limit speed information from the traffic sign recognition apparatus 100. [

In addition, the SCC control unit 220 may control the speed of the vehicle based on at least one of the lowest limit speed information and the maximum limit speed information.

For example, when the speed sign recognized by the traffic sign recognition apparatus 100 includes information indicating a maximum speed limit of '100 km / h', the SCC control unit 220 determines that the speed of the vehicle is the speed The vehicle can be controlled to be less than 100 km.

The input unit 230 may receive a user input for controlling the SCC system 200.

The input unit 230 may generate input data for controlling the operation of the SCC system 200 by a user.

The input unit 230 may include a key pad, a dome switch, a touch pad (static / static), a jog wheel, a jog switch, and the like. Particularly, when the touch pad has a mutual layer structure with the display unit 240, it can be called a touch screen.

According to one embodiment, the input unit 230 may receive (or receive) various user inputs as described above from a user.

For example, the input unit 230 may receive a control input associated with a traffic sign recognition function, which is input on a control screen for controlling the SCC system 200 displayed by the display unit 240.

Basically, the display unit 240 may display information processed or generated by the SCC control unit 220 or images or information obtained by the communication unit 210 on the screen.

In addition, the display unit 240 may display a control screen for performing a vehicle control function by the SCC system 200.

The display unit 240 may display a UI (User Interface) or a GUI (Graphic User Interface) associated with the functions provided by the SCC system 200.

Also, the display unit 240 may display the processing result of the SCC control unit 210 on the control screen.

For example, the display unit 240 may display the speed limit information recognized by the traffic sign recognizing apparatus 100, the state of the vehicle (speed information, etc.) controlled by the speed limit information on the control screen Can be displayed.

The display unit 240 may be a liquid crystal display, a thin film transistor-liquid crystal display, an organic light-emitting diode, a flexible display, a three-dimensional display (3D display).

In addition, there may be two or more display units 240 according to the implementation of the SCC system 200. For example, the SCC system 200 may include an external display unit (not shown) and an internal display unit (not shown) at the same time.

When the display unit 240 and the sensor (hereinafter, referred to as 'touch sensor') for sensing the touch operation have a mutual layer structure (hereinafter referred to as 'touch screen'), the display unit 130 outputs It can also be used as an input device in addition to the device. The touch sensor may have the form of, for example, a touch film, a touch sheet, a touch pad, or the like.

The storage unit 250 may store information received, processed, and generated by the SCC system 200.

For example, the storage unit 250 may store a program for processing and controlling the SCC controller 220, and may store input / output data (for example, user input through a user interface, control Processing result, control result, and the like).

Also, the storage unit 250 may store a user interface screen (for example, the control screen).

The storage unit 250 may be a flash memory type, a hard disk type, an SSD (Solid State Disk or Solid State Drive), a multimedia card micro type, a card type (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM) A programmable read-only memory (PROM) magnetic memory, a magnetic disk, and an optical disk.

The traffic sign recognition method according to the embodiment disclosed herein

The traffic sign recognition method according to an embodiment disclosed below may be implemented as a part or a combination of the configurations or steps included in the embodiments described above or a combination of the embodiments, The overlapping part can be omitted for the clear representation of the traffic sign recognition method.

19 is a flowchart illustrating a traffic sign recognition method according to an embodiment disclosed herein.

Referring to FIG. 19, a traffic sign recognition method according to an embodiment disclosed herein may be performed as follows.

First, a target image can be acquired (S110).

Next, the traffic pattern area may be extracted by detecting the circular pattern on the target image (S120).

Next, the speed sign included in the traffic sign area may be recognized as one of the lowest speed sign and the highest speed sign based on the specific pattern existing in the traffic sign area (S130).

According to one embodiment, the step of extracting the traffic sign region may include the steps of setting a certain region on the target image as a ROI based on a specific criterion, detecting a circular pattern on the region of interest, And extracting the region.

According to an embodiment, the specific criterion may be an area corresponding to the upper 60% of the target image.

Also, according to one embodiment, the specific pattern may be an under-bar pattern.

According to another aspect of the present invention, there is provided a traffic sign recognition apparatus and a traffic sign recognition method, comprising the steps of: detecting a circular pattern on a target image to extract a traffic sign area; recognizing a speed sign included in the traffic sign area, The recognition rate and the reliability of the speed sign can be improved by recognizing the speed sign as any of the lowest speed sign and the highest speed sign based on the specific pattern existing in the traffic sign area.

The scope of the present invention is not limited to the embodiments disclosed herein, and the present invention can be modified, changed, or improved in various forms within the scope of the present invention and the claims.

100: Traffic sign recognition device 200: SCC system

Claims (10)

An image acquiring unit acquiring a target image; And
A controller for detecting a circular pattern on the target image to extract a traffic sign area and recognizing a speed sign included in the traffic sign area,
Wherein,
(ROI) based on an upper end of the target image, extracting the traffic sign area by detecting a circular pattern on the ROI, and generating an under-bar pattern in the traffic sign area and if the under-bar pattern exists in the traffic sign area, the speed sign is recognized as the lowest speed sign, and if there is no under-bar pattern in the traffic sign area, Recognized as a top speed sign,
Wherein the ROI is normalized to 50 x 50 and feature points are detected in the normalized region to recognize the lowest speed sign or the highest speed sign. delete The method according to claim 1,
And sets an area corresponding to the upper 60% of the upper part of the target image as the area of interest. The apparatus of claim 1,
Wherein the circular pattern is detected through a radial symmetry algorithm. The apparatus of claim 1,
Extracting target feature points corresponding to the traffic sign area,
The target feature point is compared with reference feature points corresponding to each of a plurality of reference signs previously learned based on an SVM (Supprot Vector Machine)
And recognizes the speed sign based on the comparison result. delete delete The apparatus of claim 1,
And extracts at least one of the lowest limit speed information and the maximum limit speed information from the recognized speed indicator. 1. A Smart Cruise Control (SCC) system for controlling a speed of a vehicle,
A communication unit for communicating with the traffic sign recognition apparatus; And
Controlling the communication unit to acquire at least one of the lowest limit speed information and the highest limit speed information from the traffic sign recognizing device,
And an SCC control unit for controlling the speed of the vehicle based on at least one of the lowest limit speed information and the maximum limit speed information,
The traffic sign recognition apparatus includes:
The SCC system according to any one of claims 1, 3, 4, 5 and 8, characterized in that it is a traffic sign recognition device. Obtaining a target image;
Setting a certain region as a region of interest (ROI) based on an upper end of the target image;
Detecting a circular pattern in the area of interest and extracting a traffic sign area;
Checking whether there is an under-bar pattern in the traffic sign area; And
Recognizing the sign as a lowest speed sign if the under-bar pattern exists in the traffic sign area and recognizing the speed sign as a highest speed sign if the under-bar pattern does not exist in the traffic sign area; Including,
Wherein the method further comprises normalizing the ROI to 50 x 50, detecting feature points in the normalized region, and recognizing numbers in the lowest speed sign or the highest speed sign.

Images