KR101180621B1 - Apparatus and method for detecting a vehicle - Google Patents

Abstract

The present invention implements a function that can efficiently detect the front and rear vehicles at close range. To this end, the present invention includes a camera for photographing the surroundings of the vehicle, and an ultrasonic sensor for detecting obstacles or vehicles in front and rear, and after detecting a vehicle candidate using an image acquired through the camera, is obtained through the ultrasonic sensor. The obtained distance information is fused to the image of which the vehicle candidate is detected. Then, by excluding vehicle candidates not matching the distance information from the detected vehicle candidates, only valid vehicle positions can be detected. This makes it possible to increase the accuracy of the vehicle detection performance more.

Vehicle, camera, ultrasound

Description

Vehicle detection method and apparatus {APPARATUS AND METHOD FOR DETECTING A VEHICLE}

1 is an internal block diagram of a vehicle detection system according to an embodiment of the present invention;

2 is an exemplary view of a vehicle equipped with an ultrasonic sensor and a camera in the front according to an embodiment of the present invention,

3 is a control flowchart of a vehicle detection system for efficient vehicle detection at a short distance according to an embodiment of the present disclosure;

4 is an exemplary screen illustrating a process of setting a primary vehicle candidate region according to an embodiment of the present invention;

5 is an exemplary screen illustrating a process of setting a secondary vehicle candidate region according to an embodiment of the present invention;

6 is an exemplary view illustrating a symmetry axis and a vehicle width for each vehicle candidate according to an exemplary embodiment of the present invention.

7 is an exemplary view showing a result of vehicle detection according to an embodiment of the present invention.

The present invention relates to a vehicle detection method and apparatus, and more particularly, to a method and apparatus for increasing the vehicle detection performance of the front and rear.

Frequently, a vehicle collides with a vehicle driving in front of a driver due to carelessness of a driver or a poor watch. That is, traffic accidents due to lane departure and collision with driving vehicles due to poor visibility and lack of driver's concentration in long distance driving, rainy weather, and night driving are constantly occurring. In order to prevent such a collision, a driver assistance system that can provide information such as collision warning, collision avoidance, and cruise control between vehicles by grasping the position and speed between vehicles through vehicle detection is provided.

In general, there is a method of detecting a vehicle in front of the vehicle using a radar, a camera, and the like, and a method of detecting an obstacle or a vehicle in a blind spot or rear parking is a method using an ultrasonic sensor.

Among them, the radar method is a method of detecting vehicle distance information in various directions by attaching a plurality of radars in front of a vehicle, and has been applied to a system requiring active control because the exact position of the vehicle can be known. However, the radar method can be used regardless of distance, but it is very expensive to attach to a commercial vehicle and commercialize, and there is a frequency interference problem.

In addition, the method using the ultrasonic sensor detects vehicle distance information by emitting ultrasonic waves rearward from the ultrasonic sensor attached to the vehicle and detecting rear obstacles or reflected waves reflected from the vehicle. Have Therefore, when using this method, vehicle detection can be efficiently performed only in a short range of 10 m or less depending on the performance of the ultrasonic sensor.

On the other hand, in recent years, the vehicle detection method using a camera has been widely applied to the driver assistance system due to the falling price of the camera module and the improvement of image processing performance. However, the vehicle detection method using a camera is effective in detecting a vehicle that is separated by a certain distance, but when the distance between the own vehicle and the relative vehicle gets closer, the characteristic for detecting the vehicle in the image disappears or the image throughput increases, so that the detection performance is improved. Falling problems may occur.

As described above, the conventional vehicle detection methods have a problem in that the vehicle detection performance is affected by the distance between the vehicles and the detection performance is lowered. Therefore, a systematic vehicle detection technique that is more efficient and has high accuracy of vehicle detection performance is required.

Accordingly, the present invention provides a vehicle detection method and apparatus for efficiently increasing the detection performance of front and rear vehicles at a short distance.

According to an aspect of the present invention, there is provided a method of detecting a vehicle, including: photographing a target area using a camera mounted on a vehicle; calculating distance information between vehicles using an ultrasonic sensor mounted on the vehicle; Extracting and verifying a vehicle candidate region from the captured image, converting the calculated distance information into a position in the image, and a difference between the verified vehicle candidate region and the converted position And determining whether the vehicle is less than or equal to the fixed value, and detecting the vehicle candidate area that is less than or equal to the predetermined value as the final vehicle candidate.

In another aspect, the present invention provides a vehicle detection device, at least one of the front and rear to be photographed in front and rear of the vehicle, the image receiving unit for providing an image according to the target region shooting, and at least on the front and rear of the vehicle At least one ultrasonic signal receiver for emitting distance information and calculating distance information based on a time reflected by another vehicle, and an image processor for detecting and verifying a vehicle candidate region from an image provided from the image receiver; And converting the distance information from the ultrasonic signal receiver into a position in the image, and when the position difference between the converted position and the vehicle candidate extracted from the image is less than or equal to a predetermined value, the distance information of the ultrasonic signal receiver from the image processor. A sensor fusion unit for fusion to a vehicle candidate region in Using the information provided is characterized in that it comprises a vehicle position detection unit that detects a vehicle candidate region in which the predetermined value or less as a final candidate vehicle.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the same elements in the figures are represented by the same reference numerals wherever possible. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

The present invention implements a function that can efficiently detect the front and rear vehicles at close range. To this end, the present invention includes a camera for photographing the surroundings of the vehicle, and an ultrasonic sensor for detecting obstacles or vehicles in front and rear, and after detecting a vehicle candidate using an image acquired through the camera, is obtained through the ultrasonic sensor. The obtained distance information is fused to the image of which the vehicle candidate is detected. Then, by excluding vehicle candidates not matching the distance information from the detected vehicle candidates, only valid vehicle positions can be detected. This makes it possible to increase the accuracy of the vehicle detection performance more.

Referring to FIG. 1 to describe a vehicle detection system in which the above function is implemented. 1 is an internal block diagram of a vehicle detection system according to an exemplary embodiment of the present invention.

Referring to FIG. 1, a vehicle detection system according to the present invention includes a sensor receiver 101, an image processor 104, a sensor fusion unit 105, and a vehicle position detector 106.

First, the sensor receiver 101 is a means for checking the existence of a vehicle around the own vehicle. The sensor receiver 101 includes an image receiver 102 and an ultrasonic signal receiver 103 according to the present invention.

First, the image receiver 102 refers to a camera, and the image receiver 102 photographs an image of a vehicle on the road and provides the captured image to the image processor 104 for image processing. The image receiver 102 acquires an image having a size of 320 × 240 from a camera photographing the surroundings within 45 ° installed in front and rear of the vehicle.

The ultrasonic signal receiver 103 refers to a sonar sensor, which is an ultrasonic sensor, and is configured to radiate ultrasonic waves within the ultrasonic range 10m to measure the time taken for the ultrasonic waves to return to the ultrasonic sensor. In detail, when another vehicle exists around the vehicle, the ultrasonic signal receiver 103 may calculate a distance at which the other vehicle is located based on the time taken for the emitted ultrasonic wave to be reflected by the other vehicle and captured again by the ultrasonic sensor. do.

As such, the image receiver 102 and the ultrasonic signal receiver 103 constituting the sensor receiver 101 may be mounted in the vehicle. 2 illustrates a case where an ultrasonic sensor and a camera are mounted on a front surface of the vehicle to detect a vehicle ahead of the vehicle. In this case, in order to detect a rear vehicle other than the front, an ultrasonic sensor and a camera may be mounted on the rear of the vehicle as described above.

The image processor 104 detects the vehicle by using the shadow of the lower part of the vehicle through the captured image of the vehicle. In detail, the image processing unit 104 uses the shadow of the lower part of the vehicle to set the vehicle candidate region by using the shadow generated at the point where the vehicle meets the road, and the left and right edge components of the set vehicle candidate region to the vehicle. Detecting a vehicle using a template for the width. The detected vehicle is set to the ROI and tracked so that stable detection can be continuously performed.

In particular, according to an embodiment of the present invention, if an object such as a vehicle is detected within a distance of 3 to 10 m of the ultrasonic signal to increase the vehicle detection performance at a short distance, the sensor fusion unit 105 is provided from the ultrasonic signal receiver 103. The actual distance to the target object is converted into a position in the image captured by the image receiving unit 102 by perspective transformation. As described above, according to the distance between the vehicle and the other vehicle, the vehicle can obtain higher vehicle detection performance by fusing the distance information obtained by the ultrasonic sensor from the captured image at a short distance. Accordingly, the sensor fusion unit 105 determines whether the position difference between the converted position and the vehicle candidate extracted from the image has an error of a predetermined value or more when the distance information obtained by the ultrasonic sensor is converted into the position of the image. If the error is greater than a certain value, the sensor fusion unit 105 does not consider much of one of the information of the ultrasonic signal receiver 103 or the image receiver 102 when detecting the final vehicle position in the image. On the other hand, in the case of an error range, the sensor fusion unit 105 fuses the distance information obtained from the ultrasonic signal receiver 103 to the vehicle candidate region extracted by the image processor 104 and transmits the distance information obtained to the vehicle position detection unit 106. do.

For example, when the vehicle candidate region extracted from the image does not exist at a position corresponding to the distance information obtained from the ultrasonic signal receiver 103, the ultrasonic sensor is considered to be reflected by the ground or malfunctions due to external influences, and thus the distance. Vehicle detection according to the information is ignored. Accordingly, the vehicle position detector 106 detects the final vehicle position only through the image provided by the sensor fusion unit 105. On the other hand, if there is an extracted vehicle candidate region in the image, but there is no distance information received from the ultrasonic signal receiving unit 103, the vehicle position detecting unit 106 considers that an error has occurred in the image processing, and the ultrasonic signal receiving unit 103 Only the distance information obtained will detect the final vehicle position.

As described above, in the present invention, when the ultrasonic signal receiver 103 detects an object within 3 to 10 m, a vehicle candidate is derived by converting the actual distance to a position in the image, and using the feature points of the vehicle in the image. By verifying, the final vehicle position is detected. As a result, the vehicle detection performance using the image is degraded within 3m and the time required for the image processing is deteriorated. Therefore, the final vehicle is obtained using only the information obtained through the ultrasonic signal receiver 103 without considering the vehicle detection method through the image. Detect location.

Hereinafter, an operation of a vehicle detection system for increasing vehicle detection performance in a short range according to an embodiment of the present invention will be described with reference to FIG. 3. 3 is a control flowchart of a vehicle detection system for efficient vehicle detection at a short distance according to an exemplary embodiment of the present invention. Although the distance information obtained through the ultrasonic sensor in FIG. 3 may be an obstacle instead of the distance information with respect to the target vehicle, the following description assumes that the distance between the own vehicle and the target vehicle is assumed.

Referring to FIG. 3, when the user enters the vehicle detection mode by the user, the vehicle detection system starts capturing a vehicle on the road, that is, a target region through a camera, in step 200 and calculates distance information between vehicles through an ultrasonic sensor. . Next, the vehicle detection system determines whether the distance between the vehicles calculated in step 205 is within a detectable distance of the ultrasonic sensor. Here, the detectable distance of the ultrasonic sensor refers to the effective sensing distance of the ultrasonic sensor, and typically means a distance within 10 m.

If it is not within the effective detection distance of the ultrasonic sensor, the vehicle detection system proceeds to step 207 to perform vehicle detection using a camera. In general, a vehicle detection method using a camera may eliminate blind spots by fusing an ultrasonic sensor method to improve the blind spots within a short range of 10m. However, within a distance outside the short distance, for example, within 10m to 50m, a sufficient vehicle detection performance can be expected with a camera alone, and thus a vehicle detection method using a conventional camera is used.

In operation 205, when the vehicle is within the effective sensing distance of the ultrasonic sensor, the vehicle detection system determines whether the distance calculated in operation 210 is within the impossible photographing distance. That is, it is determined whether the vehicle is difficult to detect from the image obtained through the camera, which corresponds to approximately 3m or less. If it is within the impossible shooting distance, it is difficult to detect the vehicle from the captured image and it takes a lot of processing time. Therefore, when the vehicle is within the impossible photographing distance, the vehicle detection system performs vehicle detection using an ultrasonic sensor in step 215. That is, the vehicle position is detected only by the distance information obtained through the ultrasonic sensor. It is difficult to find the feature point of the vehicle in the image within 3m of the impossible shooting distance. do.

On the contrary, when the distance is not within the impossible photographing distance, for example, when the distance is within 3 to 10 m, the vehicle detection system extracts the vehicle candidate region from the image photographed in step 220. Here, the process of extracting the vehicle candidate region will be described in detail with reference to FIGS. 4 and 5. In operation 225, the vehicle detection system verifies the extracted vehicle candidate region using left and right edge components of the vehicle candidate region and the prior information about the vehicle such as a template of the vehicle. Subsequently, the vehicle detection system converts distance information from the ultrasonic sensor into a position in the image by perspective transformation in order to fuse the vehicle candidate region verified in step 230 with the information obtained from the ultrasonic sensor. In step 235, the vehicle detection system compares the vehicle candidate area with the converted position and determines whether the vehicle detection system is below a predetermined value.

If it is less than a predetermined value, the vehicle detection system determines that the vehicle candidate region in the image exists at a position corresponding to the distance information obtained through the ultrasonic sensor, and detects the final vehicle position in step 245. On the other hand, if the predetermined value is greater than or equal to, that is, the vehicle candidate region and the converted position has an error greater than or equal to the predetermined value, the vehicle detection system ignores the region for the vehicle candidate region that is greater than or equal to the predetermined value in step 240, and obtained through the ultrasonic sensor. The final vehicle position is detected by considering only the distance information. In this case, when the vehicle candidate region is detected in the image at the position of the image without the ultrasonic sensor response, it is regarded as an image processing error and the vehicle candidate region is ignored as in step 240. In contrast, if there is an ultrasonic sensor response, that is, if there is no vehicle candidate region detected at a position corresponding to the distance information in the image, the vehicle detection system considers the ultrasonic sensor to be reflected by the ground or malfunction by external influence. Ignore distance information by

As described above, a method of detecting a vehicle position at a short distance may be, for example, that if the vehicle distance obtained through ultrasound is 6m, the image obtained through the camera may be viewed as 6m only by removing perspective. Therefore, in the present invention, the overlapping regions of the vehicle candidate regions obtained by using ultrasound and the vehicle candidate regions obtained by using the camera are extracted as the final vehicle position by removing perspective. At this time, the vehicle detection system performs a position comparison between the vehicle candidate regions obtained by using ultrasound and the vehicle candidate regions obtained by using a camera, and how much the regions overlap with each other. It is determined that the same overlap.

As described above, in the present invention, by providing an improvement method for a vehicle detection method using an ultrasonic sensor and a camera, it is possible to expect superior vehicle detection performance in a near area.

In the following description, a vehicle detection process will be described in detail with reference to FIGS. 4 and 5. 4 is a screen diagram illustrating a process of setting a primary vehicle candidate region according to an embodiment of the present invention, and FIG. 5 is a screen diagram illustrating a process of setting a secondary vehicle candidate region according to an embodiment of the present invention. to be.

First, the vehicle detection system must separate the region 301 for the moving vehicle from the surrounding background region without motion in the entire image 302 as shown in FIG. 4 (a) to set the vehicle candidate region. To this end, the vehicle detection system once removes the surrounding background region from the entire image using the shadow region. Here, the shadow area is an area that appears at the boundary between the road and the vehicle in the image, regardless of the road situation, the shape of the vehicle, or the weather change such as rainy weather. It is a major feature. Such shadow areas may be detected using the feature that the shadow area has a certain intensity darker than the road area.

For effective shadow area detection, the vehicle detection system removes the background part from the entire image 302, removes the background part from the moving area, that is, the region of interest 301, and relatively dark through histogram equalization within the region of interest 301. The normalized image 302 as shown in FIG. 4 (a) and the binarized image 303 as shown in FIG. 4 (b) are generated by increasing the contrast between the pavement and the shadow area having the contrast.

The vehicle detection system then scans with a window 304 having a constant thickness in the direction of the arrow indicated by reference numeral 305 in FIG. 4 (c), i.e., from the bottom to the top of the image. Search for. At this time, the vehicle detection system also moves the window from the lower side to the upper direction in order to scan the entire region of interest 301 and also in the direction of the arrow indicated by the reference numeral 306. By moving the window from left to right while moving from bottom to top as described above, the vehicle detection system scans all areas of the region of interest 301. Similarly, the shadow area 307 is detected and the shadow area 307 is set as a vehicle candidate area. In this way, the primary vehicle candidate region is set using the shadow region.

Then, in order to set a more precise vehicle candidate region, an operation of setting a vehicle candidate region secondary is performed. Referring to FIG. 5, in FIG. 5, after converting the vehicle candidate regions set as primary to actual coordinates on a plane, the vehicle candidate regions outside the virtual lane on the plane are removed to set a secondary vehicle candidate region. Show the process.

As shown in FIG. 5, the vehicle candidate regions are grouped into at least one or more points, and the vehicle candidate regions are divided into groups 401 and 402 by the virtual lane 403. Among the groups 401 and 402, the group 402 located outside the region of interest of the virtual lane 403 is removed, and the remaining group 401 is set as the secondary vehicle candidate region. As described above, FIG. 5 illustrates a result of processing the vehicle candidate regions extracted from the image by converting them into actual coordinates on a plane. In this case, the coordinate transformation is performed through Equation 1 below.

Equation 1 is a 3x3 matrix representing a transformation between image coordinates and actual coordinates, and the image coordinates ( r _i , c _i ) are reduced to three-dimensional actual coordinates ( x _i , y _i ) on a plane by removing the perspective effect of the image. To).

In the vehicle candidate region extracted first through the above process, invalid vehicle candidate regions are again removed from the vehicle candidate regions extracted first through Equation 1 to generate the vehicle candidate region secondarily.

Hereinafter, a verification process for the secondary vehicle candidate region will be described. The verification of the vehicle candidate area is to determine whether the vehicle is actually located in the area where the vehicle is considered to be. For example, even if a secondary vehicle candidate region is obtained, an obstacle, etc., which is not a vehicle may be detected. Therefore, verification of the vehicle candidate region should be performed.

The secondary vehicle candidate region is verified by the symmetry, left and right vertical edge components, and the width of the vehicle, which are applied prior to the vehicle. Since the longitudinal position of the secondary vehicle candidate is set by the shadow region, it is relatively accurate, but since the lateral position is inaccurate, it is reset to the region having excellent symmetry through the symmetry of the edge as shown in Equation 2 below.

In Equation 2, s is the number of edges having the same angle or opposite sign, and n is the total number of edges.

A process of confirming symmetry of the vehicle candidates through Equation 2 will be described with reference to FIG. 6. 6 is an exemplary diagram illustrating a symmetry axis and a vehicle width for each vehicle candidate according to an exemplary embodiment of the present invention.

FIG. 6 shows a histogram of each vehicle width 502 and the left and right edges of each vehicle in the image. Each histogram shows a peak at the point indicated by reference numeral 501. If the peak appears in the center, it means that the vehicle is located in the front direction. If the peak is biased to the right, the vehicle is located on the right side. It can be seen that. In this way, the vehicle candidate regions are once again verified through the histogram of the left and right edges and the actual vehicle width (1.5 to 4.5m), so that the final vehicle candidate in the image can be output.

The vehicle candidate finally detected through the above process is stably detected in the continuous image through the vehicle tracking process. The vehicle detection system sets a vehicle candidate detected through the above-described process as a real-time vehicle model and updates it every frame through reliability. The vehicle detection system then tracks using the Lucas-Kanade method, which is well known as the tracking method in images, using updated vehicle models every frame. If tracking of a vehicle having high reliability fails, the vehicle is restored by lowering the threshold set by the reliability.

As described above, when a vehicle candidate is determined only by using an image obtained through a camera, a shadow is thickly formed at a distance of 3 to 10 m, and it may include an error in estimating the actual distance between vehicles using an inverse perspective transformation. Therefore, it improves this and uses ultrasonic method and camera method in this distance. When the distance between the vehicles is within 3m, a result as shown in FIG. 7 (a) may be obtained by using only the ultrasonic method, in which the vehicle in the front lane is similarly overlapped in the rectangular final vehicle area. In addition, when the distance between the vehicles is within 3 ~ 10m by using the sensor fusion method can be obtained as shown in Figure 7 (b) that the tracking vehicle is accurately overlapped in the final vehicle area.

As described above, the present invention is capable of detecting vehicles in front and rear regardless of the distance between vehicles by detecting the vehicle by fusion of the information of the ultrasonic sensor and the camera. In addition, in the present invention, the vehicle detection performance can be increased by fusion using an ultrasonic method within an area of 10m included in the camera. In addition, the present invention can be applied to a driver assistance system such as collision alarm, collision avoidance control by overcoming the limitation of vehicle detection when only the ultrasonic method or the camera method is used and increasing the reliability of the distance measurement.

Claims (12)

In the vehicle detection method, Photographing a target area using a camera mounted on a vehicle, Calculating distance information between vehicles using an ultrasonic sensor mounted on the vehicle; Extracting and verifying a vehicle candidate region from the captured image; Converting the calculated distance information into a position in the image; Determining whether a difference between the verified vehicle candidate region and the converted position is equal to or less than a predetermined value; And detecting a vehicle candidate region that is less than or equal to the predetermined value as a final vehicle candidate. The method of claim 1, wherein the detecting of the final vehicle candidate is performed. And removing the vehicle candidate region that is greater than or equal to the predetermined value from the one or more vehicle candidate regions when the verified vehicle candidate region is one or more. The method of claim 1, The sensing distance of the ultrasonic sensor is within 10m, and the non-capable distance of the camera is within 3m. The process of claim 1, wherein the extracting and verifying the vehicle candidate region comprises: Detecting a vehicle candidate region by using a shadow region generated between the vehicle and the road in the image; And verifying the vehicle candidate region by applying prior knowledge of the left and right edge components and the vehicle width to the detected vehicle candidate region. The method of claim 4, wherein the detecting of the vehicle candidate area comprises: Setting a region of interest from which the background part is removed from the image by using a shadow pattern generated between the vehicle and the road; Detecting a shadow area in the ROI and setting the detected area as a vehicle candidate area primarily; Converting image coordinates of the primary vehicle candidate region into three-dimensional coordinates on a plane; And detecting a vehicle candidate region secondarily by removing a group outside the region of interest from among the vehicle candidate regions crowded by the virtual lane on the plane. In the vehicle detection device, An image receiving unit installed on at least one of the front and rear sides of the vehicle so as to photograph front and rear, and providing an image according to the target region shooting; An ultrasonic signal receiver installed in at least one of the front and rear surfaces of the vehicle and calculating distance information based on a time when the ultrasonic signal is reflected and returned to another vehicle; An image processor for detecting and verifying a vehicle candidate region from an image provided from the image receiver; The distance information from the ultrasonic signal receiver is converted into a position in the image, and when the position difference between the converted position and the vehicle candidate extracted from the image is equal to or less than a predetermined value, the distance information of the ultrasonic signal receiver is converted by the image processor. A sensor fusion unit fused to a vehicle candidate region of And a vehicle position detector that detects a vehicle candidate region that is less than or equal to the predetermined value as a final vehicle candidate by using a result of the fusion from the sensor fusion unit. The method of claim 6, wherein the sensor fusion unit Distance information from the ultrasonic signal receiver is converted to a position in the image by perspective transformation, and when the position difference between the converted position and the vehicle candidate extracted from the image is a predetermined value or more, distance information from the ultrasonic signal receiver. Or does not consider any one of the vehicle candidate regions extracted by the image processing unit. The method according to claim 6, The sensing distance of the ultrasonic sensor is within 10m, and the non-capable distance of the image receiving unit is within 3m. The method of claim 6, wherein the image processing unit In the image, the vehicle candidate region is detected using a shadow region generated between the vehicle and the road, and the vehicle candidate region is verified by applying prior knowledge of left and right edge components and the vehicle width to the detected vehicle candidate region. Vehicle detection apparatus, characterized in that. The method of claim 9, wherein the image processing unit The ROI is set by removing a background part from the image by using a shadow pattern generated between the vehicle and the road. The ROI is detected from the ROI, and the detected area is primarily set as a vehicle candidate area. And converting the image coordinates of the candidate area into three-dimensional coordinates on the plane, and removing the group outside the ROI among the vehicle candidate areas crowded by the virtual lane on the plane to detect the vehicle candidate area secondarily. Vehicle detection device. The process of claim 1, wherein the extracting and verifying a vehicle candidate region from the captured image comprises: And extracting a vehicle candidate region from the photographed image when the calculated distance information is within a detectable distance of the ultrasonic sensor and not within a non-capable distance of the camera. The method of claim 6, wherein the sensor fusion unit, And converting the calculated distance information into a position in the image when the calculated distance information is within a detectable distance of the ultrasonic sensor and not within a non-capable distance of the camera.

Images