JP2013019684A - Vehicle periphery monitoring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle periphery monitoring device capable of improving reliability in a detection of individual objects present in a monitoring area when the object present in the monitoring area around a vehicle is detected by at least a radar device.SOLUTION: The monitoring area around the vehicle to which scanning of a radar wave is executed by a radar device 2 is picked up by an on-vehicle camera 3, and a distribution of reflectance of the radar wave in the monitoring area is estimated by reflectance estimation means 5 on the basis of color information or brightness information on picked-up images. Thresholds to reception intensity (thresholds corresponding to respective local parts in the monitoring area) are set to a reflection wave of the radar wave by reception intensity threshold setting means 6 in accordance with the estimated distribution of reflectance. Object recognition means 7 detects an object present in the monitoring area on the basis of at least a comparison between the reception intensity of the reflection wave and the set threshold.

Description

  The present invention relates to a vehicle periphery monitoring device including a radar device for monitoring the periphery of a vehicle.

  Radar devices (laser radar, millimeter wave radar, etc.) that transmit and scan radar waves and receive their reflected waves for the required monitoring area around the vehicle (for example, the front area of the vehicle) for the purpose of vehicle driving support ) On the vehicle, and based on the reflected wave of the radar radio wave received by the radar device, an object such as another vehicle or a pedestrian existing in the monitoring area around the vehicle is detected, or the detected object Techniques for observing the distance from the host vehicle, the relative speed with respect to the host vehicle, and the like are generally known.

  Furthermore, as seen in Patent Document 1, for example, the detection result of the object by the radar device as described above and the captured image of the monitoring area around the vehicle acquired by the in-vehicle camera are present in the monitoring area. The present applicant has proposed a technique for discriminating the type of object to be performed.

JP 2007-304033 A

  By the way, in order to eliminate the influence of disturbance noise included in the reflected wave to be received, the radar apparatus normally corresponds to the transmission of the radar radio wave in an arbitrary azimuth direction in the monitoring area and exceeds a predetermined threshold value. When a reflected wave having a certain intensity is received, it is detected that an object is present in the azimuth direction.

  However, since the surface portion of the target object such as another vehicle is composed of a plurality of parts, each part of the surface portion of the target object does not always reflect the radar radio wave with substantially the same reflectivity. There are many cases where the variation in the reflectance of the radar radio wave is relatively large depending on the part of the object.

  In such a case, a reflected wave (a reflected wave having an intensity equal to or greater than a predetermined threshold) from a plurality of parts of a single object that are separated from each other is received, while between the plurality of parts. From this part, there may be a situation where a reflected wave having a predetermined threshold value or more is not received.

  In such a case, there is a possibility that a plurality of portions of a single target object may be erroneously recognized as different target objects. Eventually, these objects are regarded as different objects, and there is a risk that notification to the driver, driving control of the vehicle, and the like may be performed.

  The present invention has been made in view of such a background, and when a target existing in a monitoring region around the vehicle is detected by at least a radar device, a single target is erroneously recognized as a plurality of targets, An object of the present invention is to provide a vehicle periphery monitoring device capable of preventing a local part of an object from being recognized as an object and improving the reliability of detection of each object existing in a monitoring area. And

In order to achieve the above object, the vehicle periphery monitoring device of the present invention scans a radar radio wave with respect to a monitoring area around the vehicle, receives a reflected wave of the radar radio wave, and based on the received reflected wave In the vehicle periphery monitoring device including a radar device that detects an object existing in the monitoring region,
An in-vehicle camera that images the monitoring area;
A reflectance distribution estimation means for estimating a reflectance distribution of the radar radio wave in the monitoring region based on color information or luminance information of a captured image of the in-vehicle camera;
A means for setting a threshold value for the reception intensity of the reflected wave in each local area of the monitoring area in accordance with the estimated reflectance distribution, the local area having a relatively low reflectance in the monitoring area; A reception intensity threshold value setting means for setting the threshold value so that the threshold value is relatively higher in a local area where the reflectance is relatively large;
It comprises object recognition means for detecting an object existing in the monitoring area based on comparison of at least the received intensity of the reflected wave and the set threshold (first invention).

  According to the first aspect of the invention, the reflectance distribution estimation means estimates the distribution of the reflectance of the radar radio wave in the monitoring area where the radar radio wave is scanned based on the color information or the luminance information of the captured image of the in-vehicle camera. To do. Furthermore, a threshold for the reception intensity of the reflected wave in each local area of the monitoring area is set by the reception intensity threshold setting means in accordance with the estimated reflectance distribution.

  In this case, the threshold value in each local area of the monitoring region is set so that the local area where the estimated reflectance is relatively large is relatively higher than the local area where the estimated reflectance is relatively small. . Therefore, when there is an object in the monitoring area, if the magnitude of the radar radio wave reflectance of each object has a variation in a part of the object, each location of the monitoring area is adjusted accordingly. The threshold value is set so that a difference occurs in the level of the threshold value.

  Then, the object recognition means detects an object existing in the monitoring region based on at least a comparison between the reflected wave reception intensity and the set threshold value.

  In this case, since the threshold value is set as described above, even if the actual single object existing in the monitoring area has a variation in the reflectance of the radar radio wave in the portion, the threshold value is simply set. It is possible to prevent a portion where the received intensity of the reflected wave of the radar radio wave is equal to or higher than the threshold value from being dispersed within the region where one target is present. Therefore, in a region where a single target object actually exists, the reception intensity of the reflected wave of the radar radio wave in the almost entire region is equal to or higher than the threshold value.

  For this reason, it is prevented that the single target object existing in the monitoring area is detected as a plurality of target objects by the target object recognition means.

  Therefore, according to the first invention, it is possible to prevent a single target object from being erroneously recognized as a plurality of target objects, or only a part of the target object from being recognized as a target object. As a result, it is possible to improve the reliability of detection of individual objects existing in the monitoring area.

  In the first invention, for example, the object recognizing unit simply selects each region in the monitoring region in which local areas where the reception intensity of the reflected wave is equal to or higher than the set threshold value are continuously gathered. It is detected as a region where one object exists (second invention).

  According to this second invention, the object recognition means can detect almost the entire area where the single object actually exists in the monitoring area as the existence area of the single object. it can.

  In the first aspect of the present invention, when the object existing in the monitoring area has a portion where the reflectance of the radar radio wave is extremely high, a single object may be erroneously recognized as a plurality of objects. Alternatively, when the radar wave reflectivity of the entire object is extremely low, the reflected wave of the radar wave is reflected in the entire area where the object exists even if the object exists in the monitoring area. May be lower than the threshold.

  Therefore, in the first invention, the object recognizing means includes a condition that the estimated reflectance distribution includes a region where the reflectance is higher than a predetermined upper limit value, and the estimated reflectance. And determining means for determining whether or not at least one of the conditions that the estimated value of the reflectance of each local in the distribution of the rates is lower than a predetermined lower limit value is satisfied. If the determination result of the means is negative, the monitoring region is obtained using the received intensity of the reflected wave of the radar device and the captured image of the in-vehicle camera as main reference information. If the determination result of the determination means is affirmative, the captured image of the received intensity of the reflected wave by the radar device and the captured image of the in-vehicle camera is detected. It may be performed detection of the object in the monitored region using as main reference information (third invention).

  According to the third aspect of the present invention, the estimated reflectance distribution includes a region including a region having a reflectance higher than a predetermined upper limit value, and each local area in the estimated reflectance distribution. Whether or not at least one of the conditions that the estimated value of the reflectance is lower than a predetermined lower limit is satisfied is determined by the determining means.

  In this case, the determination result of the determination means has a meaning as an index indicating whether or not the object can be detected with high reliability based on the reception intensity of the reflected wave of the radar radio wave.

  The object recognizing unit is configured to detect the radar in a case where the determination result of the determining unit is negative, that is, in a situation where the object can be detected with high reliability based on the reception intensity of the reflected wave of the radar radio wave. The object in the monitoring area is detected using the received intensity of the reflected wave of the reflected wave received by the apparatus and the received intensity of the reflected wave of the image taken by the in-vehicle camera as main reference information.

  As a result, when there is an object in the monitoring area that has an appropriate reflectance such that the reflectance of the radar radio wave falls between the upper limit value and the lower limit value, the individual objects can be distinguished from each other with high reliability. It can be detected separately.

  On the other hand, when the determination result of the determination means is affirmative, that is, when there is a region where the reflectance of the radar radio wave is higher than the upper limit in the monitoring region, or When the reflectance of the radar radio wave at each place is lower than the lower limit value, it is difficult to detect the object based on the reception intensity of the reflected wave of the radar radio wave with high reliability. For this reason, in this case, the object recognition means uses the captured image of the received intensity of the reflected wave by the radar device and the captured image of the in-vehicle camera as main reference information in the monitoring area. The object is detected.

  As a result, in a situation where it is difficult to ensure the reliability of detection of the object based on the received intensity of the reflected wave by the radar device, detection of the object is performed using the captured image of the in-vehicle camera as main reference information. This can prevent the detection reliability from being impaired.

  Thus, according to the third aspect, the determination result of the determination unit based on the estimated reflectance distribution is used as an index indicating the level of reliability of detection of an object based on the reception intensity of the reflected wave of the radar radio wave. Can be obtained as Then, according to the determination result, the object is detected using the more reliable one of the reflected wave reception intensity and the captured image of the in-vehicle camera as the main reference information. Can improve the reliability.

In addition, the vehicle periphery monitoring apparatus of the present invention scans a radar wave beam with respect to a monitoring area around the vehicle, receives a reflected wave of the radar wave, and exists in the monitoring area based on the received reflected wave. In a vehicle periphery monitoring device equipped with a radar device that detects an object to be detected,
An in-vehicle camera that images the monitoring area;
A reflectance distribution estimation means for estimating a reflectance distribution of the radar wave in the monitoring region based on color information or luminance information of a captured image of the in-vehicle camera;
A condition that the estimated reflectance distribution includes a region having a reflectance higher than a predetermined upper limit value, and an estimated value of each local reflectance in the estimated reflectance distribution is A judging means for judging whether or not at least one of the conditions of lower than a set lower limit value is satisfied;
If the determination result of the determining means is negative, the received intensity of the reflected wave of the reflected wave received by the radar device and the captured image of the in-vehicle camera is used as the main reference information. When an object in the monitoring area is detected and the determination result of the determination unit is affirmative, the captured image of the received intensity of the reflected wave by the radar device and the captured image of the in-vehicle camera is detected. It comprises object recognition means for detecting an object in the monitoring area, which is used as main reference information (fourth invention).

  According to the fourth invention, similar to the first invention, the reflectance distribution estimation means is configured to provide the radar in the monitoring region that scans the radar radio wave based on color information or luminance information of a captured image of an in-vehicle camera. Estimate the distribution of radio wave reflectance.

  As in the third aspect of the invention, the estimated reflectance distribution includes a region including a region having a reflectance higher than a predetermined upper limit value, and the estimated reflectance distribution. It is determined by the determination means whether or not at least one of the conditions that the estimated value of each local reflectance is lower than a predetermined lower limit value is satisfied. For this reason, the determination result of the determination means has a meaning as an index indicating whether or not the object can be detected with high reliability based on the reception intensity of the reflected wave of the radar radio wave.

  The object recognizing unit is configured to detect the radar in a case where the determination result of the determining unit is negative, that is, in a situation where the object can be detected with high reliability based on the reception intensity of the reflected wave of the radar radio wave. The object in the monitoring area is detected using the received intensity of the reflected wave of the reflected wave received by the apparatus and the received intensity of the reflected wave of the image taken by the in-vehicle camera as main reference information. As described above, by using the reception intensity of the reflected wave as the main reference information, it is possible to appropriately detect the object.

  In the third invention, when the object is detected using the reception intensity of the reflected wave as the main reference information, the detection may be performed in the same manner as in the first invention or the second invention. The object may be detected by comparing the intensity with a predetermined threshold value.

  On the other hand, when the determination result of the determination means is affirmative, that is, when there is a region where the reflectance of the radar radio wave is higher than the upper limit in the monitoring region, or When the reflectance of the radar radio wave at each place is lower than the lower limit value, it is difficult to detect the object based on the reception intensity of the reflected wave of the radar radio wave with high reliability. Therefore, in this case, the object recognition means mainly refers to the captured image of the received intensity of the reflected wave by the radar device and the captured image of the in-vehicle camera, as in the third invention. The object is detected in the monitoring area using the information.

  As a result, in a situation where it is difficult to ensure the reliability of detection of the object based on the received intensity of the reflected wave by the radar device, detection of the object is performed using the captured image of the in-vehicle camera as main reference information. This can prevent the detection reliability from being impaired.

  As described above, according to the fourth invention, as in the third invention, the judgment result of the judging means based on the estimated reflectance distribution is used to detect the object based on the received intensity of the reflected wave of the radar radio wave. It can be obtained as an index indicating the level of reliability. Then, according to the determination result, the object is detected by using the more reliable one of the reflected wave reception intensity and the picked-up image of the in-vehicle camera as the main reference information. The reliability of detection of individual objects can be increased.

  In the third or fourth invention, the object may be detected using the captured image of the in-vehicle camera as complementary reference information when the determination result of the determination means is negative. Of course, the object may be detected based on the reception intensity of the reflected wave by the radar device without using the image captured by the in-vehicle camera.

  On the contrary, when the determination result of the determination unit is affirmative, it is a matter of course that the object may be detected using the reception intensity of the reflected wave by the radar device as complementary reference information. Instead of using the reception intensity of the reflected wave by the radar device, the object may be detected based only on the image captured by the in-vehicle camera.

The perspective view which shows the vehicle in embodiment of this invention. The block diagram which shows the structure of the vehicle periphery monitoring apparatus with which the vehicle of FIG. 1 was equipped. The figure which shows the scanning form of the radar electromagnetic wave by the radar apparatus with which the vehicle of FIG. 1 was equipped. The flowchart which shows the process in 1st Embodiment of the arithmetic processing unit shown in FIG. The figure for demonstrating the process of STEP1 of FIG. 6A and 6B are diagrams for explaining the processing of STEPs 3 and 4 in FIG. The flowchart which shows the process in 2nd Embodiment of the target object recognition part shown in FIG.

[First Embodiment]
A first embodiment of the present invention will be described below with reference to FIGS.

  As shown in FIGS. 1 and 2, the vehicle periphery monitoring device of the present embodiment includes a radar device 2, an in-vehicle camera 3, and an arithmetic processing unit 4 that are mounted on the vehicle 1. A monitoring region around the vehicle 1 by the vehicle periphery monitoring device is a front region of the vehicle 1. For this reason, the radar device 2 and the in-vehicle camera 3 are mounted on the front portion of the vehicle 1.

  In the present embodiment, the radar apparatus 2 is a laser radar that uses, for example, laser light having a wavelength in the near infrared region as a radar radio wave. As shown in FIG. 3, the radar apparatus 2 transmits a beam BM of a radar radio wave (laser light) to various places in a monitoring area in front of the vehicle 1 (area between straight lines L1 and L2 in the figure). A radar wave reflected by an object existing in the transmission direction of the beam BM is received.

  In this case, the radar apparatus 2 is a scanning type radar, and changes the transmission direction (azimuth angle) of the beam BM of the radar radio wave by a certain angle in the direction around the yaw axis so that the radar radio wave is transmitted to the monitoring area. The beam BM is scanned on a scanning plane substantially parallel to the traveling road surface of the vehicle 1. The radar apparatus 2 receives the reflected wave of the transmitted radar radio wave for each transmission direction of the beam BM of the radar radio wave.

  The in-vehicle camera 3 is an imaging camera composed of a CCD camera, a CMOS camera, or the like, and is mounted on the vehicle 1 so as to capture an image of an imaging area including the monitoring area. In this embodiment, the captured image of the in-vehicle camera 3 is a color image.

  The arithmetic processing unit 4 is an electronic circuit unit configured by a CPU, a RAM, a ROM, an interface circuit, and the like. The arithmetic processing unit 4 receives the received signal of the reflected wave received by the radar device 2 and the video signal of the captured image of the in-vehicle camera, and these signals are A / D converted and taken into the arithmetic processing unit 4. It is.

  Then, the arithmetic processing unit 4 recognizes a predetermined type of object existing ahead of the vehicle 1 based on the input data, and the distance from the own vehicle 1 to the object or the object with respect to the own vehicle 1. Estimate the relative speed up to. Further, the arithmetic processing unit 4 determines the possibility of contact between the recognized object and the own vehicle 1 based on the distance from the own vehicle 1 to the object, the relative speed to the object with respect to the own vehicle 1, and the like. The warning or notification to the driver is performed by audio information or visual information, or braking control of the host vehicle 1 is performed.

  Hereinafter, the object recognition processing executed by the arithmetic processing unit 4 will be described by taking, for example, the case where the arithmetic processing unit 4 recognizes another vehicle (an oncoming vehicle or a preceding vehicle) as an object.

  The arithmetic processing unit 4 includes a reflectance estimation unit 5, a reception intensity threshold setting unit 6, and an object recognition unit 7 as functions related to the present invention. Each of these functional units is a function realized by an installed program or the like, and corresponds to the reflectance distribution estimation unit, the reception intensity threshold setting unit, and the object recognition unit in the present invention, respectively.

  Below, the process in the case of recognizing an object (here, another vehicle) by the arithmetic processing unit 4 will be described, including details of the processing of each functional object.

  The arithmetic processing unit 4 executes the process shown in the flowchart of FIG. 4 when recognizing another vehicle as an object.

  Specifically, the arithmetic processing unit 4 first executes the processing of STEP1. In STEP 1, the arithmetic processing unit 4 has an object (an object that reflects the transmitted radar radio wave) in each transmission direction of the radar radio wave beam BM based on the received signal of the radar radio wave reflected by the radar device 2. And the relative speed of the object with respect to the host vehicle 1 is estimated in each transmission direction in which the presence of the object is detected. Then, the arithmetic processing unit 4 determines whether or not an object existing in each transmission direction of the beam BM of the radar radio wave is another vehicle as a predetermined type of object based on the estimated relative speed.

  Here, whether or not an object exists in each transmission direction of the beam BM of the radar radio wave is detected depending on whether or not the reception intensity of the reflected wave of the radar radio wave in the transmission direction is a predetermined value or more. In addition, the relative velocity of an object existing in each transmission direction of the beam BM of the radar radio wave is estimated by a known method based on the Doppler shift between the transmitted laser radio wave and the received reflected wave. The relative speed estimation process may be performed by the radar device 2.

  Then, when the estimated relative speed is “0” or a speed in the vicinity thereof for each transmission direction of the beam BM of the radar radio wave, the arithmetic processing unit 4 performs a preceding vehicle as another vehicle in the transmission direction. Is determined to exist. The arithmetic processing unit 4 determines that there is an oncoming vehicle as another vehicle in the transmission direction when the estimated relative speed is twice the vehicle speed of the host vehicle 1 or a speed in the vicinity thereof.

  As a result, for each transmission direction of the radar radio wave, it is recognized whether or not there is another vehicle as an object in the transmission direction. For example, in the situation shown in FIG. 5, it is recognized that there is another vehicle in the transmission direction of the radar wave from the own vehicle 1 toward the preceding vehicle and the transmission direction of the radar wave from the own vehicle 1 toward the oncoming vehicle. The Rukoto.

  In the processing of STEP1, it is not determined whether other vehicles existing in different transmission directions of radar radio waves are the same other vehicles. Also, in the processing of STEP 1, the reception intensity (digital data after A / D conversion) of the reflected wave for each transmission direction of the beam BM of the radar radio wave is stored and held in the storage device of the arithmetic processing unit 4.

  Next, the arithmetic processing unit 4 executes the processing of STEP2. In STEP 2, the arithmetic processing unit 4 acquires a captured image of the in-vehicle camera 3 (a captured image at a time when the radar apparatus 2 scans the radar radio wave or a time in the vicinity thereof). The captured image is stored and held in the storage device of the arithmetic processing unit 4.

  Next, the arithmetic processing unit 4 performs the processing of STEP 3 by the reflectance estimation unit 5. In STEP 3, the reflectivity estimation unit 5 estimates the distribution of the reflectivity of the radar radio wave in the object existing in front of the vehicle 1 in the monitoring area where the radar radio wave is scanned based on the captured image acquired in STEP 2. More specifically, the reflectance distribution is a distribution in the scanning direction of the radar radio wave (lateral direction of the captured image).

  Here, in the present embodiment, the radar radio wave is a laser beam having a wavelength in the near infrared region, and the reflectance of the radar radio wave in various objects is generally approximate to the reflectance of visible light. That is, the reflectivity of radar radio waves, which are laser light with a wavelength in the near-infrared region, tends to change according to the color of the object, and when the color of the object is close to white than when it is close to black The reflectance of radar radio waves at the object tends to be relatively high.

  Moreover, since the object which exists in the monitoring area | region is projected as a color image on the picked-up image of the vehicle-mounted camera 3, each local color of a monitoring area | region can be recognized with this picked-up image.

  Therefore, in the present embodiment, the reflectance estimation unit 5 determines the distribution of the reflectance of the radar radio wave in the monitoring area as a horizontal line of the captured image corresponding to the scanning surface of the radar radio wave (hereinafter referred to as the scanning line on the image). In the above, based on the image color, the reflectance distribution of the radar radio wave is estimated.

  In this case, the estimated value of the reflectance of the radar radio wave at each position on the scanning line on the image is determined in advance according to the average color of the image in the local area in the vicinity of each position. It is determined using a data table showing the relationship between reflectance and color. Note that the estimated value of the radar radio wave reflectance at each position on the scanning line on the image is, in real space, from each optical center of the in-vehicle camera 3 to each point on the scanning line on the image on the imaging surface of the captured image. This corresponds to an estimated value of the reflectance of an object existing in front of the vehicle 1 in the direction of the straight line facing.

  A specific example of the reflectance distribution estimated in this way will be described with reference to FIGS. FIGS. 6 (a) and 6 (b) each assume a situation in which another vehicle exists in the monitoring area, and each of the uppermost figures schematically shows a captured image of the other vehicle. In this case, it is assumed that the other vehicle shown in FIG. 6A is a vehicle having a color in which radar radio waves (near-infrared laser light) are difficult to be reflected, for example, a black body color. Further, the other vehicle shown in FIG. 6B is assumed to be a vehicle having a color that easily reflects radar radio waves, for example, a white body color. Note that a dashed line La in the uppermost drawing is a scanning line on the radar radio wave image.

  In the example of FIG. 6A, the distribution of the estimated value of the reflectance of the radar radio wave on the on-image scanning line La is, for example, a pattern distribution as shown in the middle graph of FIG. . In this case, the areas A1, A3, and A5 are areas corresponding to the black body color that hardly reflects radar radio waves. The estimated values of reflectivity corresponding to these regions A1, A3, and A5 are set to low values.

  The areas A2 and A4 are areas corresponding to lamps (tail lamps) of other vehicles, for example, red color areas. And the estimated value of the reflectance corresponding to these area | regions A2 and A4 is made into a high value.

  On the other hand, in the example of FIG. 6B, the distribution of the estimated value of the reflectance of the radar radio wave on the on-image scanning line La is, for example, a pattern distribution as shown in the middle graph of FIG. Is done. In this case, the regions B1, B3, and B5 are regions corresponding to the white body color that relatively easily reflects radar radio waves. The estimated values of reflectance corresponding to these regions B1, B3, and B5 are relatively high values (values higher than the regions A1, A3, and A5 in FIG. 6A).

  The areas B2 and B4 are areas corresponding to the lamps (tail lamps) of other vehicles, and are, for example, red-colored areas similar to the areas A2 and A4 in FIG. Therefore, the estimated reflectance values corresponding to these regions B2 and B4 are set to high values as in the regions A2 and A4 in FIG. In the present embodiment, the estimated reflectance values corresponding to the regions B2 and B4 are higher than the regions B1, B3, and B5.

  Next, the arithmetic processing unit 4 executes the processing of STEP 4 by the reception intensity threshold setting unit 6. In this STEP 4, the reception intensity threshold value setting unit 6 defines a data table or an arithmetic expression (relationship between radar wave reflectivity and threshold value) set in advance according to the radar wave reflectivity distribution estimated in STEP 3. (Data table or calculation formula) is used for each local area of the monitoring area in front of the vehicle 1 (local area corresponding to each transmission direction of the beam BM of the radar radio wave) with respect to the reception intensity of the reflected wave received by the radar device 2. A threshold (hereinafter referred to as a reception strength threshold) is set.

  In this case, the reception intensity threshold value at each position in the scanning direction of the radar radio wave is set so that the reception intensity threshold value increases as the estimated reflectance value at that position increases. For example, in the example shown in FIG. 6A, as shown by the broken line in the lowermost diagram in FIG. 6A, the distribution pattern of the estimated value of the reflectance of the radar radio wave (the middle diagram in FIG. 6A) is the same. The reception intensity threshold value corresponding to each position in the scanning direction of the radar radio wave (each transmission direction of the beam BM) is set.

  In the example shown in FIG. 6B, as shown by the broken line in the lowermost diagram in FIG. 6B, the distribution pattern of the estimated value of the radar radio wave reflectivity (the middle diagram in FIG. 6B) and A reception intensity threshold corresponding to each position in the scanning direction of the radar radio wave (each transmission direction of the beam BM) is set with a similar distribution pattern.

  Accordingly, the reception intensity threshold values set in the areas A1, A3, A5 (area corresponding to the black body color) in FIG. 6A are the areas B1, B3, B5 (white body color in FIG. 6B). The value is smaller than the reception intensity threshold value set in the area corresponding to

  Next, the arithmetic processing unit 4 executes the processing of STEP 5 by the object recognition unit 7. In STEP 5, the object recognition unit 7 stores and holds the actual received intensity (STEP 1) of the reflected wave at each position (position in the scanning direction) of the monitoring area corresponding to each transmission direction of the beam BM of the radar radio wave. The received intensity is compared with the received intensity threshold set in STEP 4, and a local area where the actual received intensity of the reflected wave is equal to or higher than the received intensity threshold is extracted in the monitoring area in front of the vehicle 1.

  Then, the object recognizing unit 7 is an area in which the local area where the actual reception intensity of the reflected wave is equal to or higher than the reception intensity threshold is continuously gathered in the scanning direction of the radar radio wave (however, in the scanning direction). The object is determined by determining a region having a width equal to or greater than a predetermined lower limit width as a region where a single object (another vehicle) is present. Here, the area where the local area where the actual reception intensity is equal to or higher than the reception intensity threshold is continuously gathered in the scanning direction of the radar radio wave is that the actual reception intensity in each area (each area in the scanning direction) This is an area that is equal to or greater than the reception intensity threshold.

  For example, in the situation shown in FIGS. 6 (a) and 6 (b), there is a single other vehicle in each of the object existence regions (regions having substantially the same width as other vehicles) shown in the drawings. It is determined as the area to be.

  Here, if the reception intensity threshold value is set to a predetermined constant value, the object to be monitored is simply an object when the reflectance of the radar radio wave is relatively different in a part. One object may be erroneously recognized as a plurality of objects. For example, when the reception intensity threshold is set as shown by the two-dot chain line S in the lowermost diagram of FIGS. 6A and 6B, the actual reception intensity is obtained in the other vehicle shown in FIG. Since the region where is equal to or greater than the reception intensity threshold is separated into two, a single object (another vehicle) is erroneously recognized as two objects.

  On the other hand, in the present embodiment, the reception intensity threshold is changed according to the reflectance estimation value, so that a single target object (another vehicle) to be monitored is erroneously recognized as a plurality of target objects. Therefore, it can be correctly recognized as a single object.

  In the processing of STEP5, not only the actual reception intensity of the reflected wave of the radar radio wave but also the object (other vehicle) existing in the monitoring area using the captured image of the in-vehicle camera 3 as complementary reference information as necessary. ) May be detected.

  In the present embodiment, the arithmetic processing unit 4 determines the distance of the target object from the host vehicle 1 for each target object (other vehicle) recognized as described above, based on the received signal of the reflected wave from the target object. presume. The distance estimation is performed by a known method based on, for example, a time difference between transmission and reception of radar radio waves. Based on the relative speed of the object with respect to the own vehicle 1 and the distance from the own vehicle 1, the possibility of contact between the object and the own vehicle 1 is predicted, and an alarm or notification corresponding to the prediction result is given. This is performed for the driver, or the control of the braking device of the vehicle 1 is performed.

  According to this embodiment, even when the reflectance of the radar radio wave at each target varies in a partial portion, a single target is erroneously recognized as a plurality of targets. While preventing, it is possible to recognize each object existing in the monitoring area in front of the vehicle 1 based on the reception intensity of the reflected wave of the radar radio wave.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIG. In this embodiment, only the processing of the object recognition unit 7 of the arithmetic processing unit 4 is different from that of the first embodiment. Therefore, the difference will be mainly described, and the same matters as those of the first embodiment will be described. Is omitted.

  When an object existing in the transmission direction of the radar radio wave beam BM is an object having a very high reflectance of the radar radio wave, even if the object is a single object, even if the object is a single object, scattering of the reflected wave of the radar radio wave is caused. As a result, it may be recognized that there are a plurality of objects.

  Further, when an object existing in the transmission direction of the beam BM of the radar radio wave is an object having a very low reflectivity of the radar radio wave, the reception intensity of the reflected wave by the radar device 2 becomes low. It may be difficult to recognize the presence of an object based on intensity with high reliability.

  Therefore, in the present embodiment, the object recognition unit 7 of the arithmetic processing unit 4 executes the process of FIG. That is, the object recognition unit 7 first executes the determination processes of STEPs 11 and 12. In STEP 11, the object recognition unit 7 includes a region in which the estimated reflectance value is higher than a predetermined upper limit value in the reflectance distribution estimated by the reflectance estimating unit 5 in STEP 3. Determine whether or not.

  In STEP 12, the object recognition unit 7 reflects the reflectance at each position (position corresponding to each transmission direction of the beam BM of the radar radio wave) in the reflectance distribution estimated by the reflectance estimation unit 5 in STEP 3. Whether or not the estimated value is lower than a predetermined lower limit value (in other words, whether or not all the estimated reflectance values in the estimated reflectance distribution are lower than the predetermined lower limit value) Judging.

  Note that the determination processes in STEPs 11 and 12 correspond to determination means in the present invention.

  And when the judgment result of both STEP11 and 12 becomes negative, the target object recognition part 7 is the same as the process of the target object recognition part 7 in 1st Embodiment similarly to 1st Embodiment. Based on the received intensity of the reflected wave of the radar radio wave, each object (other vehicle) existing in the monitoring area in front of the vehicle 1 is recognized.

  Therefore, when the distribution state of the reflectance of the radar radio wave in the monitoring area is such a distribution state that the detection of the object based on the reception intensity of the reflected wave of the radar radio wave can be performed with high reliability, the first embodiment. In the same manner as the processing of the object recognition unit 7 in, individual objects (other vehicles) existing in the monitoring area are recognized separately from each other based on the reception intensity of the reflected wave of the radar radio wave.

  On the other hand, if the determination result of any one of STEPs 11 and 12 is affirmative, it is difficult to recognize the object based on the reception intensity of the reflected wave with high reliability. Recognizes each object (another vehicle) present on the on-image scanning line La based on a captured image of the in-vehicle camera 3 (captured image acquired in STEP 2). In this case, the object recognition unit 7 extracts, for example, a shape portion characteristic of the object (another vehicle) from the captured image as a region where the object exists.

  The above is the processing of the object recognition unit 7 in the present embodiment. The present embodiment is the same as the first embodiment except for the matters described above.

  According to this embodiment, since the object is recognized using the captured image when the determination result of any one of STEPs 11 and 12 is positive, the object based on the received intensity of the reflected wave of the radar radio wave is used. Even in a situation where it is difficult to recognize an object with high reliability, it is possible to appropriately recognize each object.

  Further, when the determination results of both STEPs 11 and 12 are negative, based on the reception intensity of the reflected wave of the radar radio wave, the individual objects are correctly distinguished from each other in the monitoring area in front of the vehicle 1. Each existing object can be recognized.

  In the second embodiment, when both the determination results of STEP 11 and 12 are negative, the reception intensity threshold set according to the distribution of the estimated value of the reflectance of the radar radio wave is present in the monitoring area. The target that exists in the monitoring region is obtained by omitting the process of the reception intensity threshold setting unit 6 (the process of STEP 4) and comparing the reception intensity of the reflected wave with a predetermined threshold. You may make it recognize a thing.

  Further, in the second embodiment, the determination process of any one of STEPs 11 and 12 may be omitted.

  Further, in the processing of STEP 13 in the second embodiment, not only the received intensity of the reflected wave of the radar radio wave but also the captured image of the in-vehicle camera 3 is used as complementary reference information as necessary, and exists in the monitoring area. You may make it recognize the target object. Similarly, in the processing of STEP14, not only the captured image of the in-vehicle camera 3, but also the received intensity of the reflected wave of the radar radio wave is used as complementary reference information as necessary to recognize an object existing in the monitoring area. You may make it do.

  Further, in each of the above embodiments, the case where another vehicle existing in the monitoring area is recognized as an object has been described as an example. However, for example, a pedestrian may be recognized as an object, or You may make it recognize both with a vehicle as a target object.

  In each of the above embodiments, the case where the radar radio wave is laser light having a wavelength in the near infrared region has been described as an example. However, the radar apparatus 2 uses an electromagnetic wave in a wavelength range such as a millimeter wave as a radar radio wave. It may be.

  In the embodiment, the distribution of the reflectance of the radar radio wave in the monitoring area is estimated based on the color of the captured image. However, based on the brightness of the captured image, or on both the color and the brightness. Based on this, the distribution of the reflectance of the radar radio wave in the monitoring area may be estimated.

  In particular, when the in-vehicle camera is a camera having sensitivity in the near infrared region in the same manner as the radar radio wave in each of the above embodiments, the luminance distribution of the captured image in the monitoring area and the reflectance of the radar radio wave in the monitoring area Correlation with the distribution of increases. Therefore, it is possible to estimate the distribution of the radar radio wave reflectance in the monitoring area with high reliability based on the brightness of image capturing.

  In each of the above embodiments, when the presence of an object is recognized in an area where the reflectance of the estimated radar radio wave is relatively high or low in the monitoring area, only a part of the single object is detected. May be recognized as an object, it may be determined that the reliability of the recognition of the object in the area in the forward direction of the object is low. Then, depending on the level of recognition of the object, the manner of notification or warning to the driver, the manner of controlling the vehicle braking device, or the like may be changed.

  DESCRIPTION OF SYMBOLS 1 ... Vehicle, 2 ... Radar apparatus, 3 ... In-vehicle camera, 5 ... Reflectance estimation part (reflectance estimation means), 6 ... Reception intensity threshold setting part (reception intensity threshold setting means), 7 ... Object recognition part (object Object recognizing means), STEP11, 12, ... judging means.

Claims (4)

A vehicle periphery including a radar device that scans a radar radio wave with respect to a monitoring area around the vehicle, receives a reflected wave of the radar radio wave, and detects an object existing in the monitoring area based on the received reflected wave In the monitoring device,
An in-vehicle camera that images the monitoring area;
A reflectance distribution estimation means for estimating a reflectance distribution of the radar radio wave in the monitoring region based on color information or luminance information of a captured image of the in-vehicle camera;
A means for setting a threshold value for the reception intensity of the reflected wave in each local area of the monitoring area in accordance with the estimated reflectance distribution, the local area having a relatively low reflectance in the monitoring area; A reception intensity threshold value setting means for setting the threshold value so that the threshold value is relatively higher in a local area where the reflectance is relatively large;
A vehicle periphery monitoring device comprising: object recognition means for detecting an object existing in the monitoring area based on a comparison between at least the reception intensity of the reflected wave and the set threshold value. The vehicle periphery monitoring device according to claim 1,
The object recognizing means uses, as the existence area of a single object, each of the areas where the local areas where the reflected wave reception intensity is equal to or greater than the set threshold are continuously gathered in the monitoring area. A vehicle periphery monitoring device characterized by detecting. The vehicle periphery monitoring device according to claim 1,
The object recognizing means includes a condition that the estimated reflectance distribution includes a region having a reflectance higher than a predetermined upper limit value, and each local in the estimated reflectance distribution. A judgment means for judging whether or not at least one of the conditions that the estimated value of the reflectance is lower than a predetermined lower limit is satisfied, and the judgment result of the judgment means is negative; In this case, detection of an object in the monitoring area is performed by using, as main reference information, the reception intensity of the reflected wave from the reception intensity of the reflected wave by the radar device and the image captured by the in-vehicle camera. When the determination result of the determination unit is affirmative, the captured image of the received intensity of the reflected wave by the radar device and the captured image of the in-vehicle camera is used as main reference information. Vehicle periphery monitoring device which is characterized in that the detection of the object in the monitoring area. A radar device that scans a radar wave beam to a monitoring area around the vehicle, receives a reflected wave of the radar wave, and detects an object existing in the monitoring area based on the received reflected wave is provided. In the vehicle periphery monitoring device,
An in-vehicle camera that images the monitoring area;
A reflectance distribution estimation means for estimating a reflectance distribution of the radar wave in the monitoring region based on color information or luminance information of a captured image of the in-vehicle camera;
A condition that the estimated reflectance distribution includes a region having a reflectance higher than a predetermined upper limit value, and an estimated value of each local reflectance in the estimated reflectance distribution is A judging means for judging whether or not at least one of the conditions of lower than a set lower limit value is satisfied;
If the determination result of the determining means is negative, the received intensity of the reflected wave of the reflected wave received by the radar device and the captured image of the in-vehicle camera is used as the main reference information. When an object in the monitoring area is detected and the determination result of the determination unit is affirmative, the captured image of the received intensity of the reflected wave by the radar device and the captured image of the in-vehicle camera is detected. A vehicle periphery monitoring device comprising: object recognition means for detecting an object in the monitoring area using main reference information.

Images