KR20130054637A - Apparatus and method detecting object in automatic parking system - Google Patents

Abstract

PURPOSE: An object detection apparatus in an automatic parking system and a method thereof are provided to enhance the performance of the automatic parking system by intellectually detecting obstacle movement by using obstacle characteristics. CONSTITUTION: An object detection apparatus in an automatic parking system comprises an obstacle detecting unit(140), an obstacle movement detecting unit(150), and an obstacle movement alerting unit(160). The obstacle detecting unit detects an obstacle based on location information of a recognized vehicle, location information of the obstacle, and photographed image data. If an obstacle is detected by the obstacle detecting unit, the obstacle movement detecting unit detects obstacle movement in near and far distances by using a SVM(Support Vector Machine) algorithm. The obstacle movement alerting unit gives notices of a detection result of obstacle movement which is detected by the obstacle movement detecting unit to a driver. [Reference numerals] (110) Vehicle position recogition unit; (120) Sensor unit; (130) Camera image unit; (140) Obstacle detecting unit; (150) Obstacle movement detecting unit; (160) Obstacle movement alerting unit; (AA,CC) Driver; (BB,DD) Vehicle

Description

Apparatus And Method Detecting Object in Automatic Parking System

The present invention relates to an apparatus and a method for detecting an object in an automatic parking system, and more particularly, to an apparatus and a method for detecting an object in an automatic parking system to assist a driver's parking.

Existing automatic parking system is divided into Application software and Base software, as shown in Figure 1, Application software includes a system management unit, the system management unit automatic location recognition unit, parking space determination unit, parking path generation unit and A parking path follower.

The automatic position recognition unit determines the state and position of the vehicle based on the vehicle information (Steering Angle, Wheel Pulse, Yaw-rate, Temperature, etc.) and transmits it to the trajectory calculation / control module.

The parking space determination unit determines a parking space using the information transmitted from the sensor, calculates a parking target point of the vehicle, and transmits the parking target point to the parking path generation unit.

The parking path generation unit calculates the parking guidance trajectory based on the parking space calculation result transmitted from the parking space determination unit.

The parking path tracking unit induces the parking space tracking using the obstacle information around the vehicle transmitted from the parking space determination unit and the parking trajectory transmitted from the parking path generation unit.

The system manager integrally manages each block constituting the automatic parking system.

On the other hand, the parking space determination unit searches for a parking available space from the distance of the ultrasonic sensor, and determines whether it is possible to add based on the search result.

For example, the parking space determiner determines the current vehicle location by using the distance measured by the ultrasonic sensor and the location information of the vehicle location recognizer in association with the parking method and the path generation method when determining whether parking is possible, and uses both information. A map is constructed and parking availability is determined according to the route planning method to be applied.

However, since the parking space may vary according to the movement of the obstacle, an obstacle movement detection system is required for accurate parking space determination.

Obstacle motion detection system has recently become increasingly important as a variety of applications such as human computer interface, video surveillance system, image retrieval using obstacles, but the detection reliability of the algorithm is low and detection speed for real life application It is slow and does not recognize the movement of obstacles accurately.

On the other hand, studies on the collision-preparation system of the way to detect the obstacles likely to collide near the vehicle by using a camera mounted on the vehicle to inform the driver in advance.

Obstacle detection device mounted on the vehicle detects the obstacles in the front, rear, left and right areas of the vehicle in a blind spot in the driver's seat and notifies the driver. Obstacles are detected by attaching two sensors to the vehicle's front bumper and two sensors to the rear bumper. The sensor detection zone is a hemispherical shape centered around the sensor and is designed to sound the buzzer at regular intervals through the control unit when an obstacle is detected.

In the obstacle recognition method of these systems, the candidate area of the obstacle is secured from the camera image, and then the existence of the obstacle is determined to determine whether the obstacle exists.

To this end, a decision boundary is learned from obstacle sample patterns and applied to obstacle recognition. The representative method is MLP (Multi Layer Perceptron). The learning method through MLP is a technique for minimizing only errors from given sample data.

This obstacle detection system has the following problems.

First, the error rate is high in obstacle detection. That is, even in the case of a driver who is experienced in driving, when the vehicle is parked in a narrow place, the vehicle may be damaged by colliding with a vehicle of another person or the wall located in front of the vehicle due to an obstacle recognition error. In addition, even when there is room in the parking space, the lower part of the vehicle may be damaged during parking because a low object such as a stone is not recognized by the system.

In this way, simply mounting the sensor allows only the distance measurement with the obstacle, there is a problem that can not detect the obstacle in the direction of the vehicle blind spot.

In addition, the detection algorithm is incorrect. For example, the MLP method minimizes only errors from a given sample data, ensuring high accuracy in the training data, but very low in non-learning data. Considering an environment where various changes such as brightness, position, and shape of light occur, such as obstacle image detection, reliability is difficult to secure unless the number of sample data is large.

Finally, there is no perception of obstacle movement. For example, the conventional technology warns the driver by measuring only the distance to an obstacle in proximity to the vehicle by a distance sensor mounted on the front and rear bumpers of the vehicle, thereby failing to determine the movement of the obstacle.

The present invention has been made in view of the above problems, by mounting four ultrasonic sensors on the front bumper of the vehicle, four ultrasonic sensors on the rear bumper, and two sensors on the side to expand the obstacle movement detection area. An object of the present invention is to provide an apparatus and method for detecting an object in an automatic parking system capable of accurately detecting obstacle movement.

In order to achieve the above object, in the automatic parking system according to an aspect of the present invention, an apparatus for detecting an object when a vehicle is parked detects an obstacle based on the recognized position information of the vehicle, position information of the obstacle, and photographed image data. Obstacle detection unit; An obstacle motion detection unit detecting an obstacle motion in a short distance and a long distance by using a machine learning algorithm (SVM) when the obstacle is detected by the obstacle detection unit; And an obstacle motion alarm unit which notifies the driver of a detection result of the obstacle motion detected by the obstacle motion detector.

According to another aspect of the present invention, a method for detecting an object when a vehicle is parked includes detecting an obstacle based on recognized vehicle position information, position information of an obstacle, and photographed image data; If an obstacle is detected, detecting obstacle movement in a short distance and in a distance using a support vector machine (SVM); And informing the driver of the detection result of the obstacle movement.

According to the present invention, the performance of the automatic parking system can be improved by using the principal component analysis method and the obstacle image feature value for obstacle detection.

The system automatically detects obstacles and detects movement, eliminating the need for operator intervention.

The system change can be minimized by adding an obstacle motion detection system to the existing automatic parking system.

Complexity can be reduced by extracting feature vectors of data and classifying them using machine learning algorithms (SVMs).

By detecting obstacle movement using the obstacle image, other obstacles such as lighting or pose can be accurately detected.

Obstacle features can be used to intelligently detect obstacle movements, improving the performance of the automatic parking system, and storing only the necessary images later, enabling efficient use of video storage capacity.

In addition, the computational amount can be reduced by detecting an obstacle region only for an image having a change in brightness without finding an obstacle in the entire image.

By using SVM, it is possible to effectively represent decision boundary with a small number of support vectors, which greatly reduces the amount of computation, improves the speed of obstacle detection classification, and ensures the reliability of obstacle detection by obtaining a larger margin.

1 is a view for explaining a conventional technology.
2 is a block diagram illustrating an object detecting apparatus in an automatic parking system according to an embodiment of the present invention.
3 is a view for explaining FIG. 2 in more detail.
4 is a diagram for explaining an obstacle motion detector.
5 is a flowchart illustrating an object detection method in an automatic parking system according to an embodiment of the present invention.

Advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. And is intended to enable a person skilled in the art to readily understand the scope of the invention, and the invention is defined by the claims. It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. It is noted that " comprises, " or "comprising," as used herein, means the presence or absence of one or more other components, steps, operations, and / Do not exclude the addition.

Hereinafter, an object detecting apparatus in an automatic parking system according to an embodiment of the present invention will be described with reference to FIGS. 2 to 4. FIG. 2 is a block diagram illustrating an object detecting apparatus in an automatic parking system according to an embodiment of the present invention. FIG. 3 is a view for explaining FIG. 2 in more detail. FIG. 4 is for explaining an obstacle motion detector. It is for the drawing.

As shown in FIG. 2, the object detecting apparatus of the present invention includes a vehicle position recognizing unit 110, a sensor unit 120, a camera imaging unit 130, an obstacle detecting unit 140, an obstacle movement detecting unit 150, and Obstacle movement warning unit 160 is included.

The obstacle detecting unit 140 obtains sensing values from the vehicle position recognizing unit 110 and the sensor unit 120 and image data from the camera image unit 130, and detects obstacles based on the obtained sensing values and image data. If so, the transfer to the obstacle motion detection unit 150.

When the obstacle detecting unit 150 detects an obstacle by the obstacle detecting unit 140, the obstacle movement detecting unit 150 detects obstacle movements in a short distance and a long distance by using a machine learning technique.

The obstacle motion detection unit 150 may not only accurately detect an obstacle through the above-described bar, but also greatly improve the processing speed for obstacle motion detection, thereby enabling real-time obstacle detection and motion detection.

The obstacle movement warning unit 160 notifies the driver of the obstacle movement detection result.

Hereinafter, the object detecting apparatus of the present invention will be described in detail with reference to FIG. 3. As shown in FIG. 3, the vehicle position recognizing unit 110 recognizes a steering angle of a vehicle and a wheel pulse. Pulse), Yawrate (Raw Speed), Temperature (temperature) signals are transmitted to the obstacle detection unit 140, the sensor unit 120 includes an Ultrasonic sensor (ultrasound signal sensor) for the obstacle detection, to the ultrasonic signal sensor The measured value is transmitted to the obstacle detector 140, and the camera image unit 130 transmits the image captured by the camera to the obstacle detector 140.

Obstacle detector 140 captures the steering angle of the vehicle, the wheel pulse, the yawrate, the temperature signal, the measured value by the ultrasonic sensor, and the camera. The feature vector is extracted from the captured image.

Since the steering wheel sensor unit (not shown) is included in the vehicle position recognition unit 110 and measures a signal based on the steering angle of the steering wheel, the obstacle detecting unit 140 may determine the vehicle through the value measured by the steering wheel sensor unit. Recognize the position of.

Since the ultrasonic signal sensor (not shown) detects side, front and rear obstacles of the vehicle, the obstacle detection unit 140 determines whether the obstacle is based on the measurement value by the ultrasonic signal sensor.

The camera image unit 130 obtains an image data value from the camera and transmits the image data value to the obstacle detector 140, and the obstacle detector 140 receives the image data and transmits it to the obstacle motion detector 150, and the obstacle motion detector 150. Uses image data as learning data.

For example, when the obstacle motion detector 150 receives an image signal from the obstacle detector 140, the obstacle motion detector 150 checks whether there is a brightness change in the transmitted image signal, and when there is a brightness change, distinguishes an obstacle area from a non-obstacle area. The motion detection signal is output after detecting the obstacle area by using the decision boundary of the SVM.

When the motion detection signal is transmitted from the obstacle motion detection unit 150, the obstacle detection warning unit 160 generates a stop signal to notify the driver that the obstacle is detected.

For example, the obstacle detection alarm unit 160 controls the control unit so that the indicator flashes and the buzzer keeps ringing when the distance to the obstacle is about 50 cm, and the closer the vehicle is to the obstacle, the faster the cycle and the obstacle. When the distance is within 20cm, the indicator lights up continuously and the buzzer rings continuously.

Hereinafter, the operation of the obstacle motion detection unit will be described in more detail with reference to FIG. 4. As shown in FIG. 4, the obstacle motion detection unit 150 using the obstacle feature includes a frame memory 151 and a feature value extraction unit 152. ), A first threshold setting unit 153, a comparator 154, a binarizer 155, an adder 156, a motion discriminator 157, a motion detector 158, and a subtractor 159.

The frame memory 151 stores the image acquired from the camera image unit 130.

The subtractor 159 obtains a difference image of a brightness value between the stored previous image and the currently input image.

The feature value extractor 152 extracts a feature value for obstacle movement detection.

Feature values here include mean, standard deviation, normal distribution, histogram, and so on.

The first threshold value setting unit 153 sets a pixel boundary value that assumes that there is an obstacle movement from the feature value extraction unit 152.

The comparator 154 compares the difference image of the brightness value of the previous image and the current input image obtained by the subtractor 159 with the first threshold value.

For example, the comparison unit 154 determines that the difference image is greater than the first threshold value when the difference image is greater than the first threshold value, and that the difference image is less than the first threshold value, the pixel having no obstruction movement Judging by.

The binarizing unit 155 allocates +1 and -1 values depending on whether the brightness value changes.

For example, the binarization unit 155 binarizes an image by assigning a +1 value to a pixel having a change in brightness and a -1 value to a pixel having no change.

The adder 156 obtains the total sum of pixels with obstacle movement in the binarized image.

The motion determining unit 157 calculates whether there is a change in brightness value in the entire input image and then determines whether there is an obstacle movement area according to the change.

For example, the motion determining unit 157 determines that there is an obstacle movement in the current input image when the total sum of the moving pixels is greater than the threshold value.

When the motion detector 158 determines that there is an obstacle movement area, the motion detection unit 158 determines whether there is an obstacle characteristic in the portion having the movement area, and if there is an obstacle, outputs an obstacle motion detection signal, and detects the obstacle image from the primarily determined image. If there is an obstacle image, the obstacle movement detection signal is output because there is an obstacle movement finally.

The motion detector 158 includes an M-grid Gabor Wavelet-based candidate region detector (not shown), a low-resolution SVM-based candidate region detector (not shown), and a high-resolution SVM-based candidate region detector (not shown).

The M-grid Gabor Wavelet-based candidate region detection unit (not shown) detects a candidate region in which an obstacle may exist by applying a specific pattern to the region determined to have obstacle movement.

For example, when a camera image sequence is input in units of frames, the M-grid Gabor Wavelet-based candidate region detector (not shown) matches a M like shape grid at all possible positions, extracts a feature vector, and calculates a distance from the average vector. If the distance to the average vector is smaller than the maximum distance obtained by the previous learning, it is determined as an obstacle area and a candidate area in which an obstacle exists is detected.

The low-resolution SVM-based candidate region detector (not shown) performs principal component analysis (PCA, PrincipleComponentAnalysis) on a plurality of obstacle images (NXN pixel sizes) normalized in the previous learning stage to extract feature vectors using N eigen vectors. do.

For example, a low-resolution SVM-based candidate region detector (not shown) reduces the dimension by linearly projecting the feature space in a direction to maximize the variance represented by the image feature vector using the PCA, and obtains the eigenvalue and the feature vector by using the variance. Then, the feature vectors of the desired dimension are extracted by sequential ordering of the eigenvalues, and k is selected whose cumulative contribution (cumulative variance) makes up 99% of the sum of the eigenvalues.

The low-resolution SVM-based candidate region detector (not shown) applies extracted obstacle feature vectors and non-obstacle feature vectors to the SVM to obtain a decision boundary that can distinguish the obstacle class from the non-obstacle class. The grid gabor wavelet-based candidate region detection unit (not shown) searches all possible partial images around the candidate points determined as obstacles and finds the obstacle candidate regions by the previously learned decision boundary.

The high-resolution SVM-based final obstacle detector (not shown) uses a 2N X 2N pixel-size obstacle image in the learning step in a similar manner to the learning and detection step of the low resolution SVM-based candidate area detector (not shown), and uses 2N eigen vectors. In the detection step, a low resolution SVM-based candidate region detector (not shown) is used to detect an obstacle finally using a high resolution SVM around the candidate region determined to be a detected obstacle.

As described above, according to the present invention, it is possible to improve the automatic parking system performance by using the principal component analysis method and the obstacle image feature value for obstacle detection, and the system automatically detects the obstacle and detects the movement so that the driver can operate it separately. It is convenient because there is no need, and the system change can be minimized by adding the obstacle motion detection system to the existing automatic parking system, and the complexity can be reduced by extracting the feature vector of data and classifying it through machine learning algorithm (SVM). have.

The object detection apparatus in the automatic parking system according to the embodiment of the present invention has been described above with reference to FIGS. 2 to 4, and the object detection in the automatic parking system according to the embodiment of the present invention is described below with reference to FIG. 5. Explain how. 5 is a flowchart illustrating an object detection method in an automatic parking system according to an embodiment of the present invention.

As shown in FIG. 5, when a vehicle position value, an obstacle position value, and obstacle image data are input (S500), it is determined whether an obstacle exists based on the input information (S501).

As a result of the determination, if there is an obstacle, a difference image of the brightness values of the pre-stored image and the input image is obtained (S502), the brightness value change is calculated from the entire input image, and the movement of the obstacle based on the calculated brightness value change. It is determined whether an area exists (S503).

For example, after calculating whether there is a change in brightness value in the entire input image, it is determined whether there is an obstacle movement area according to the change, and if the total sum of the moving pixels is larger than the threshold, there is an obstacle movement in the current input image. Determine.

As a result of the determination, when there is a movement region of the obstacle, it is determined whether there is a characteristic of the obstacle in the entire input image in which the movement region exists, and when it is determined that there is an obstacle characteristic, the movement of the obstacle is detected (S504).

When the distance to the detected obstacle is about 50cm, the indicator light blinks and the control unit is controlled so that the buzzer keeps ringing continuously.The closer the vehicle is to the obstacle, the faster the cycle and the distance from the obstacle is within 20cm. When the indicator lights continuously and at the same time the buzzer rings continuously to inform the driver of the movement of the obstacle (S505).

However, as a result of the determination in step S503, when there is no movement area of the obstacle, it is determined that the fixed obstacle, and the alarm according to the position of the fixed obstacle (S506).

Although the configuration of the present invention has been described in detail with reference to the preferred embodiments and the accompanying drawings, this is only an example, and various modifications are possible within the scope without departing from the spirit of the present invention. Therefore, the scope of the present invention should not be limited by the illustrated embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

110: vehicle position recognition unit 120: sensor unit
130: camera image unit 140: obstacle detection unit
150: obstacle movement detection unit 160: obstacle movement alarm

Claims (6)

An apparatus for detecting an object during vehicle parking in an automatic parking system,
An obstacle detecting unit detecting an obstacle based on the recognized position information of the vehicle, position information of the obstacle, and photographed image data;
An obstacle motion detection unit detecting an obstacle motion in a short distance and a long distance by using a machine learning algorithm (SVM) when the obstacle is detected by the obstacle detection unit; And
Obstacle motion alarm unit for informing the driver of the detection result of the obstacle motion detected by the obstacle motion detection unit
Object detection device in an automatic parking system comprising a. The method of claim 1,
The obstacle motion detector determines whether there is a brightness change in the image data transmitted through the obstacle detector, and if there is a brightness change, a decision boundary of the machine learning algorithm for distinguishing an obstacle area from a non-obstacle area. Detecting obstacle areas using
Object detection device in automatic parking system. The method of claim 1,
The obstacle motion detector calculates a change in brightness value from the image data transmitted through the obstacle detector, and if the total sum of pixels moving in the obstacle motion area determined based on the calculated brightness value change is greater than a predetermined threshold value. And determining that there is an obstacle movement in the transmitted image data, and outputting an obstacle movement detection signal when an obstacle characteristic exists in an area where the obstacle movement exists.
When the obstacle motion detection signal is input from the obstacle motion detection unit, the obstacle detection alarm unit blinks an indicator light and sounds a buzzer every predetermined period when the distance between the vehicle and the obstacle is greater than a predetermined value. As the distance between them decreases, the blinking period of the indicator light and the ringing period of the buzzer are quickly set. If the distance between the vehicle and the obstacle is less than a predetermined value, the indicator is continuously turned on and the buzzer rings continuously so that Informing the driver of the movement
Object detection device in automatic parking system. In a method for detecting an object when parking a vehicle in an automatic parking system,
Detecting the obstacle based on the recognized position information of the vehicle, the position information of the obstacle, and the captured image data;
If an obstacle is detected, detecting obstacle movement in a short distance and in a distance using a support vector machine (SVM); And
Informing the driver of the detection result of the obstacle movement
Object detection method in an automatic parking system comprising a. The method of claim 4, wherein the detecting comprises:
Checking whether there is a brightness change in the transmitted image data; And
Detecting an obstacle region by using a decision boundary of the machine learning algorithm that distinguishes the obstacle region from the non-obstacle region when the brightness is changed.
Object detection method in automatic parking system. 5. The method of claim 4,
Wherein the detecting comprises:
Calculating a change in brightness value from the transmitted image data and determining an obstacle movement area based on the calculated change in brightness value;
Determining that there is an obstacle movement in the transferred image data when the total sum of the moving pixels in the determined obstacle movement area is larger than a predetermined threshold value; And
Outputting an obstacle motion detection signal when an obstacle characteristic exists in the region where the obstacle motion is determined;
Informing the step,
When the obstacle motion detection signal is input, when the distance between the vehicle and the obstacle is equal to or greater than a predetermined value, blinking an indicator light and causing the buzzer to sound every predetermined period; And
Quickly setting a light blinking period and a buzzer ringing period as the distance between the vehicle and the obstacle decreases; And
If the distance between the vehicle and the obstacle is less than a predetermined value, continuously lighting a light and causing the buzzer to ring continuously to notify the driver of the movement of the detected obstacle.
Object detection method in automatic parking system.

Images