JP2015232846A - Front situation determination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a front situation determination device capable of determining whether a driver needs to take precautions against another mobile body present in an area in which a closest leading vehicle hinders driver's vision, without dependence on a device mounted in the other vehicle.SOLUTION: A front situation determination device comprises: an image analysis unit 51 analyzing an analysis target area that is an area contained in an image of an analysis target object selected from among objects free from obstacles between a camera 10 and the objects in a predetermined condition, an image of a mobile body present in an area in which a closest leading vehicle that is a leading vehicle free from obstacles between the camera 10 and the leading vehicle hinders driver's vision may be seen in the analysis target area; and a determination unit 52 determining whether there is a shielded area alert situation in which the driver needs to take precautions against the mobile body present in the area in which the closest leading vehicle hinders driver's vision on the basis of an analysis result of the image analysis unit 51.

Description

  The present invention relates to a forward situation determination device that is used in a vehicle and determines a situation in front of the vehicle.

  A technique is known in which, when traffic jam information is acquired, both the left and right direction indicator lights blink to alert the driver of the following vehicle (Patent Document 1). According to this technique, the driver of the host vehicle travels by observing that both the left and right direction indicator lights of the preceding vehicle, which is the nearest front vehicle traveling in the same lane as the host vehicle, blink. It is possible to know that there is a situation that requires attention in a region where the visibility is obstructed by the preceding vehicle in the lane in which the vehicle is running.

JP 2005-148998 A

  However, the technique of Patent Document 1 cannot alert the driver of the host vehicle unless the device described in Patent Document 1 is installed in the preceding vehicle.

  In addition, even if the detection unit such as a camera mounted on the own vehicle automatically detects blinking of the left and right direction indicator lights of the preceding vehicle and performs vehicle control such as speed control, the device described in Patent Document 1 is used for the preceding vehicle. If is not installed, the vehicle cannot be controlled.

  In order to alert the driver of the host vehicle or control the host vehicle, it is determined whether or not the situation is that the other vehicle in the area where visibility is obstructed by the preceding vehicle should be watched. There is a need to.

  In the technique of Patent Document 1, the apparatus of Patent Document 1 is mounted on another vehicle in order to determine whether or not the other vehicle existing in an area where visibility is obstructed in the preceding vehicle should be watched. It was necessary to have. However, there is a problem that the device of Patent Document 1 is not always mounted on other vehicles.

  The present invention has been made based on this situation, and the object of the present invention is to exist in a region where the field of view is obstructed by the vehicle ahead, without depending on the device mounted on the other vehicle. It is an object of the present invention to provide a forward situation determination device that can determine whether or not it is a situation that requires attention to another moving body that is doing.

  The above object is achieved by a combination of features described in the independent claims, and the subclaims define further advantageous embodiments of the invention. Reference numerals in parentheses described in the claims indicate a correspondence relationship with specific means described in the embodiments described later as one aspect, and do not limit the technical scope of the present invention. .

  In order to achieve the above object, the present invention is used in a vehicle, and a camera (10) that images the front of the vehicle and an image captured by the camera from an object that does not have an obstacle between the camera and a predetermined condition. This is an area included in the image of the selected object to be analyzed, and it shows an image of a moving body that exists in the area where the field of vision is obstructed in the nearest forward car that is the front car without an obstacle between the camera and the camera. Based on the analysis result of the image analysis unit (51) that analyzes a possible analysis target region and the analysis result of the image analysis unit, in a situation that requires attention to a moving body that exists in a region where the field of view is obstructed in the most recent forward vehicle It is a forward situation judgment device provided with a judgment part (52) which judges whether it is a certain shielded area attention situation.

  In the image captured by the camera in front of the vehicle, there is an image of a moving object that exists in the area where the field of view is obstructed by the vehicle ahead, in the glass part of another object or in the mirror part of another object. Sometimes.

  Therefore, in the present invention, the image captured by the camera is an area included in the image of the object to be analyzed, and an image of a moving body that exists in the area where the field of view is obstructed in the vehicle in front is possible. Analyze the analysis target area, which is a characteristic area. Based on the analysis result, the determination unit determines whether or not it is a situation that requires attention to another moving body that exists in the area where the field of view is obstructed in the most recent forward vehicle.

  Therefore, it is determined whether or not it is a situation that requires attention to another moving body that exists in an area in which the field of vision is obstructed in the vehicle immediately ahead without depending on the device mounted on the vehicle immediately ahead. be able to.

It is a lineblock diagram of forward situation judgment device 1 of an embodiment. It is a flowchart which shows the process which ECU50 of FIG. 1 performs. It is a flowchart which shows the process of step S2 of FIG. It is a flowchart which shows the process of step S4 of FIG. It is a figure explaining a camera image. It is a figure explaining the vehicles 120, 130, and 140 seen through the rear window part 112 of the preceding vehicle 110. FIG. It is a flowchart which shows the process of step S5 of FIG. It is a figure explaining the search area | region 160 which searches an analysis object vehicle. It is a flowchart which shows the process of step S6 of FIG.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. A forward situation determination device 1 according to an embodiment of the present invention has the configuration shown in FIG. This forward situation determination device 1 is mounted on a vehicle (not shown). Hereinafter, the vehicle equipped with the front situation determination device 1 is referred to as a host vehicle.

  The front situation determination apparatus 1 includes a camera 10, an illuminance sensor 20, a display unit 30, a speaker 40, and an ECU 50.

  The camera 10 images the front of the host vehicle. Assume that this camera 10 is a stereo camera. The installation location of the camera 10 is, for example, the back side of a room mirror in the vehicle interior.

  The illuminance sensor 20 detects the illuminance around the host vehicle and outputs a signal indicating the detected illuminance to the ECU 50. The illuminance sensor 20 is installed on the upper edge of the dashboard in the vehicle cabin.

  The display unit 30 is installed at a position where the driver of the host vehicle can visually recognize the vehicle interior, and includes a display screen. On this display screen, when the ECU 50 determines that the situation is a shielding area caution situation, the ECU 50 displays a character or a figure that means that attention should be paid to the shielding area.

  This shielded area caution situation exists in a region where the field of view is obstructed by the vehicle ahead without an obstacle between the camera 10, that is, the vehicle in front of the vehicle with no obstacle between itself and the vehicle. This means a situation that requires attention to the moving body.

  In addition, the nearest forward vehicle is not limited to the preceding vehicle, which is the nearest forward vehicle traveling in the same lane as the own vehicle, but also between the own vehicle in a direction different from the lane direction in which the own vehicle is traveling. Includes forward vehicles with no objects.

  Therefore, the nearest forward vehicle includes a forward vehicle (hereinafter, a laterally immediate forward vehicle) that has no obstacle between itself and the own vehicle in another lane in the same traveling direction as the lane in which the host vehicle is traveling. In addition, the nearest forward vehicle includes a forward vehicle (hereinafter referred to as the nearest oncoming vehicle) that has no obstacle between itself and the vehicle in a lane opposite to the lane in which the vehicle is traveling.

  The speaker 40 is installed at a position to output sound toward the passenger compartment, and means that when the ECU 50 determines that it is a shielding area attention situation, it should be controlled by the ECU 50 and pay attention to the shielding area. Output sound.

  The ECU 50 includes a known CPU, ROM, and RAM. Moreover, the memory | storage part 55 different from ROM and RAM is also provided. The storage unit 55 is a rewritable ROM, for example, a flash memory. The storage unit 55 or ROM stores programs for the CPU to execute various processes.

  The CPU functions as the image analysis unit 51, the determination unit 52, the illuminance determination unit 53, and the vehicle type determination unit 54 by executing a program stored in the ROM or the storage unit 55 while using the temporary storage function of the RAM. To do. The processes executed by the image analysis unit 51, the determination unit 52, the illuminance determination unit 53, and the vehicle type determination unit 54 will be described later with reference to FIG. Note that some or all of the functions executed by the CPU may be configured by hardware using one or a plurality of ICs.

  The ECU 50 also controls the camera 10, the display unit 30, and the speaker 40. The ECU 50 includes a signal acquisition unit 56. The signal acquisition unit 56 acquires various signals in the host vehicle via the in-vehicle LAN 60. The signal to be acquired includes a signal indicating the turn-on / off state of the direction indicator lamp. Therefore, the signal acquisition unit 56 corresponds to the turn signal detection unit in the claims.

  In addition to the above-described program, the storage unit 55 displays the light emission pattern of the rear window of the preceding vehicle that can be seen from the host vehicle when the brake lamp of the preceding preceding vehicle, which is a vehicle preceding the preceding vehicle, is lit. It is remembered separately. The inclination of the window of the preceding vehicle differs depending on the vehicle type. If the inclination of the window is different, the light scattering pattern when the light from the brake lamp of the preceding preceding vehicle enters the window of the preceding vehicle is different. That is, the light emission pattern of the rear window of the preceding vehicle that can be seen from the own vehicle when the brake lamp of the preceding preceding vehicle is lit differs for each vehicle type. Hereinafter, this light emission pattern is referred to as a brake light emission pattern. The storage unit 55 stores the braking light emission pattern for each vehicle type. This brake light emission pattern is a light emission pattern in a state where both the left and right brake lamps of the preceding preceding vehicle are visible from the host vehicle through the window of the preceding vehicle.

  Similarly, when both the direction indicator lights at the rear of the preceding preceding vehicle are lit, the light emission pattern of the rear window of the preceding vehicle that is visible from the host vehicle is also different. Hereinafter, this light emission pattern is referred to as a hazard light emission pattern. The storage unit 55 also stores this hazard emission pattern for each vehicle type. This hazard light emission pattern is also a light emission pattern in a state where both the left and right direction indicator lights of the preceding preceding vehicle are visible from the host vehicle through the window of the preceding vehicle. The brake lamp and the rear direction indicator lamp are the rear lamps of the claims.

  Further, in order to determine the vehicle type of the preceding vehicle from the rear shape of the preceding vehicle, the storage unit 55 also stores a template image behind the vehicle for each vehicle type.

  Next, processing executed by the ECU 50 will be described with reference to FIG. The ECU 50 repeatedly executes the process shown in FIG. 2 while the host vehicle is traveling forward. Further, the ECU 50 causes the camera 10 to capture a front image of the host vehicle while executing the processing shown in FIG. The image analysis unit 51 executes steps S1 to S3 in FIG. Which element of the ECU 50 executes steps S4 to S6 will be described with reference to FIGS.

  In step S <b> 1, an image captured by the camera 10 (hereinafter, camera image) is acquired from the camera 10.

  In step S2, an analysis target vehicle, that is, an analysis target object is determined. An analysis target vehicle is a vehicle in which an image of the analysis target vehicle includes an analysis target region in a camera image. The analysis target area means an area in which an image of a moving body that exists in an area where the field of view is obstructed in the vehicle immediately ahead may be captured. The vehicles to be analyzed include the preceding vehicle, the vehicle that is closest to the side, and the vehicle that is closest to the vehicle. There is an image part and a window image part of the nearest oncoming vehicle.

  There is a possibility that the preceding vehicle appears in the window image portion of the preceding vehicle. There is a possibility that the front side vehicle is reflected in the image part of the side surface of the body of the vehicle in front of the side as a mirror. There is a possibility that a moving body that moves the rear side of the latest oncoming vehicle to the own vehicle side is reflected in the window image portion of the latest oncoming vehicle. Accordingly, the rear window image portion of the preceding vehicle, the image portion of the body side surface of the vehicle immediately before the side vehicle, and the window image portion of the nearest oncoming vehicle are set as analysis target regions.

  The detailed process of step S2 is shown in FIG. In FIG. 3, in step S21, it is determined whether or not the right direction indicator lamp of the host vehicle is blinking. If this judgment is YES, it will progress to Step S22.

  In step S22, it is determined whether the vehicle is traveling in the right lane. This determination is made, for example, by analyzing the camera image acquired in step S1 and determining the distance between the host vehicle and the center line or the center separation band from the position of the center line or the center separation band in the image.

  If the determination in step S22 is NO, the right direction indicator is blinking because the lane is changed. If judgment of step S22 is NO, it will progress to step S23. Moreover, also when the judgment of step S21 is NO, it progresses to step S23.

  In step S23, it is determined whether or not there is a transparent window portion of the preceding vehicle in the camera image. The light-transmitting window means a window through which light is transmitted, and is a window on which a film is attached. A window that has no or almost no scenery beyond the window is not a light-transmitting window. The determination in step S23 is whether or not there is a window that allows the front of the preceding vehicle to be visually recognized.

  The determination in step S23 is, for example, determining the image portion of the preceding vehicle by edge detection or the like, and whether or not there is an area where the position of the same point changes relative to the surrounding portion in the image portion of the preceding vehicle. Judgment by. This is because if there is a translucent window portion, the scenery seen through the window portion changes relative to the body around the window.

  If judgment of step S23 is YES, it will progress to step S24. In step S24, the preceding vehicle is determined as an analysis target vehicle. If judgment of step S23 is NO, it will progress to step S25. In step S25, the vehicle nearest to the side is determined as the vehicle to be analyzed. This is because the window portion of the preceding vehicle cannot be analyzed to determine whether the brake lamp and the direction indicator light of the preceding preceding vehicle are lit, so the image portion of the body side surface of the vehicle that is closest to the side is analyzed.

  On the other hand, if it is determined in step S22 that the vehicle is traveling in the right lane (S22: YES), it is considered that the driver of the host vehicle is about to turn the host vehicle to the right. If judgment of Step S22 is YES, it will progress to Step S26. In step S26, the nearest oncoming vehicle is determined as the vehicle to be analyzed. This is because when turning right, that is, when crossing the oncoming lane, it is necessary to pay attention to the moving body that moves to the host vehicle side behind the nearest oncoming vehicle.

  Returning to FIG. In step S3, it is determined what the vehicle to be analyzed determined in step S2. If the preceding vehicle is the analysis target vehicle, the process proceeds to step S4. If the side nearest vehicle is the analysis target vehicle, the process proceeds to step S5. If the latest oncoming vehicle is the analysis target vehicle, the process proceeds to step S6. move on.

  Details of step S4 are shown in FIG. 4, step S41 is executed by the vehicle type determination unit 54, steps S42 to S44 and S46 to S47 are executed by the image analysis unit 51, step S45 is executed by the illuminance determination unit 53, and steps S48 to S51 are determination units. 52 executes.

  In step S41, a vehicle type determination process is performed to determine the vehicle type of the preceding vehicle. In determining the vehicle type of the preceding vehicle, the degree of coincidence between the image of the preceding vehicle acquired from the camera image and the template image behind the vehicle of the vehicle type stored in the storage unit 55 is used. If the degree of coincidence is equal to or higher than the vehicle type determination value, the vehicle type of the preceding vehicle is assumed to be the vehicle type of the template image used for determining the degree of coincidence. It is assumed that the vehicle type of the preceding vehicle cannot be determined when there is no template image whose coincidence is equal to or higher than the vehicle type determination value. Further, it is assumed that the vehicle type of the preceding vehicle cannot be determined even when there is no preceding vehicle.

  In step S42, it is determined whether or not the vehicle type of the preceding vehicle has been determined in the vehicle type determination process of step S41. When the vehicle type of the preceding vehicle can be determined (S42: YES), the process proceeds to step S43. In step S43, the brake light emission pattern and the hazard light emission pattern of the determined vehicle type are used in the following process among the brake light emission pattern and the hazard light emission pattern stored in the storage unit 55 for each vehicle type. decide.

  When the vehicle type of the preceding vehicle cannot be determined (S42: NO), the process proceeds to step S44. In step S44, it is determined to use the default light emission pattern for braking and the light emission pattern for hazard that are set in advance in the following processing.

  If either step S43 or S44 is executed, the process proceeds to step S45. In step S45, a signal indicating illuminance is acquired from the illuminance sensor 20. In step S46, based on the illuminance acquired in step S44, a luminance threshold value used for determining whether or not the lamp of the preceding vehicle is turned on is set. The luminance threshold includes a luminance threshold for determining lighting of the brake lamp and a luminance threshold for determining lighting of the direction indicator lamp. Both threshold values are set to higher values as the illuminance is higher. This is because the brighter the surrounding brightness is, the brighter the rear window image portion in the camera image is even if the lamp of the preceding vehicle is not lit.

  In step S47, the rear window image portion 72 of the preceding vehicle image 70 illustrated in FIG. The rear window image portion 72 of the preceding vehicle image 70 corresponds to an analysis target region in the claims. FIG. 5 is an explanatory diagram of a camera image. When the preceding preceding vehicle exists, the rear image 80 of the preceding preceding vehicle may appear in the rear window image portion 72 of the preceding vehicle image 70. When the brake lamp of the preceding preceding vehicle is lit, the brake lamp of the rear image 80 of the preceding preceding vehicle is also lit. Further, when both direction indicator lights of the preceding preceding vehicle blink, both direction indication lights of the rear image 80 of the preceding preceding vehicle also blink.

  Therefore, in step S47, the rear window image portion 72 of the preceding vehicle image 70 is analyzed. The content of the analysis is to perform processing for performing the determinations of the next steps S48 and S49. That is, in step S47, in the rear window image portion 72 of the preceding vehicle image, the luminance of the portion showing the color of the brake lamp and the luminance of the portion showing the color of the direction indicator light are changed to the respective lamps in step S46. It is determined whether or not the brightness threshold set for is exceeded. Further, the degree of coincidence between the luminance distribution pattern of the portion exceeding the luminance threshold and the light emission pattern determined in step S43 or S44 is determined.

  As a result of the analysis in step S47, if the luminance of the portion exhibiting the color of the brake lamp exceeds the luminance threshold set for the brake lamp in step S46, the determination result in step S48 is YES. In addition, as a result of the analysis in step S47, the determination result in step S48 is also obtained when the luminance of the portion showing the color of the direction indicator lamp exceeds the luminance threshold set for the direction indicator lamp in step S46. Let it be YES. If the decision result in the step S48 is YES, the process proceeds to a step S49, and if NO, the process of FIG. 4 is ended.

  In step S49, whether or not the luminance distribution pattern of the portion exhibiting the color of the brake lamp and exceeding the luminance threshold matches the light emission pattern of the brake lamp determined in step S43 or S44. Is determined from the degree of coincidence determined in step S47.

  Further, in this step S49, the luminance distribution pattern of the portion exhibiting the color of the direction indicator lamp and exceeding the luminance threshold is equal to the light emission pattern of the direction indicator lamp determined in step S43 or S44. It is judged from the degree of coincidence determined in step S47 whether or not it has done.

  If at least one of the coincidence degrees exceeds the coincidence determination threshold value, it is determined that the light emission pattern coincides (S49: YES). If none of the coincidence levels exceed the coincidence determination threshold, it is determined that the light emission pattern does not coincide (S49: NO).

  When the camera 10 is attached in the vicinity of the rearview mirror of the host vehicle 100, as shown in FIG. 6, not only the preceding preceding vehicle 120 but also other vehicles 130, 140 in the adjacent lane through the rear window 112 of the preceding vehicle 110. May be visible. However, as can be seen from FIG. 6, the other vehicles 130 and 140 in the adjacent lane are not located near the center of the rear window 112 when viewed from the host vehicle 100 through the rear window 112 of the preceding vehicle 110. In particular, when the other vehicle 140 on the left side in FIG. 6 is viewed from the own vehicle 100 through the rear window 112 of the preceding vehicle 110, only the right side portion can be seen. Therefore, the rear window of the preceding vehicle image 70 is displayed when the brake lamps and the left and right direction indicator lights of these other vehicles 130 and 140 are turned on and when the brake lamp and the left and right direction indicator lights of the preceding preceding vehicle 120 are turned on. The light emission pattern of the image portion 72 is different.

  On the other hand, in the brake light emission pattern and the hazard light emission pattern determined in step S43 or S44, the left and right brake lamps or the left and right direction indicator lights of the preceding preceding vehicle can be visually recognized from the own vehicle through the window of the preceding vehicle. The light emission pattern in the state. Therefore, when the luminance distribution pattern of the portion exhibiting the color of the brake lamp and exceeding the luminance threshold matches the light emission pattern of the brake lamp determined in step S43 or S44, The brake lamp of the car 120 is turned on.

  In addition, when the luminance distribution pattern of the portion indicating the color of the direction indicator lamp and exceeding the luminance threshold matches the light emission pattern of the direction indicator lamp determined in step S43 or S44, Both direction indicator lights of the preceding vehicle 120 are turned on.

  Therefore, if the determination in step S49 is yes, the process proceeds to step S50, and it is determined that the lamp of the preceding vehicle has been turned on. When the brake lamp of the preceding preceding vehicle is lit, the preceding preceding vehicle is decelerating, so attention must be paid to the preceding preceding vehicle. In addition, attention must be paid to the preceding vehicle even when both direction indicator lights of the preceding vehicle are blinking.

  Accordingly, the situation in which step S50 is executed is a situation in which attention is required for the preceding vehicle existing in the area where the visibility is obstructed by the preceding vehicle. This situation corresponds to the shielded area attention situation in the claims.

  In step S51, the driver of the host vehicle is notified using one or both of the display unit 30 and the speaker 40 that the lamp of the preceding vehicle has turned on.

  Next, the side closest vehicle analysis process shown in FIG. 7 will be described. In FIG. 7, steps S61 to S63 and S65 to S66 are executed by the image analysis unit 51, step S64 is executed by the illuminance determination unit 53, and steps S67 to S69 are executed by the determination unit 52.

  In step S61, the relative position of the preceding vehicle with respect to the host vehicle is specified. Since the camera 10 in this embodiment is a stereo camera, the relative position of the preceding vehicle is specified by a known calculation method based on the parallax of the same feature point in two camera images. If the preceding vehicle does not exist within a certain distance from the host vehicle, it is determined that there is no preceding vehicle.

  In step S62, a search area 160 for searching for the vehicle to be analyzed is determined. The search area 160 will be described with reference to FIG. In FIG. 8, a preceding vehicle shielding area 150 indicated by a broken line is an area close to the preceding vehicle 110 among the areas in which the driver's view of the host vehicle is obstructed in the preceding vehicle 110. The width of the preceding vehicle shielding area 150 is the same as the width of the preceding vehicle 110, and the length is constant from the front end of the preceding vehicle 110 to the front.

  The fact that at least a part of the situation of the preceding vehicle shielding area 150 is reflected on the side of the body of the vehicle traveling in the adjacent lane is limited to a certain area due to the light reflection law that the incident angle and the reflection angle are equal. This area is the search area 160. As shown in FIG. 8, the search area 160 is determined on each of the left and right sides in the longitudinal direction with respect to the preceding vehicle shielding area 150. In the present embodiment, the search area 160 is set in the adjacent lane. This is because of the limitation due to the angle of view of the camera 10. When the angle of view of the camera 10 is wider, an area that is further away from the host vehicle in the vehicle width direction of the host vehicle than the adjacent lane may be included in the search area.

  In FIG. 8, arrows 171 and 173 conceptually indicate light from the front end and rear end of the preceding vehicle shielding area 150, and arrows 172 and 174 indicate light indicated by arrows 171 and 173, respectively. The reflected light which reflected and produced | generated by the front-end | tip of 173 is shown notionally. As can be seen from these arrows 171, 172, 173, 174, the search area 160 can be determined by geometric calculation when the preceding vehicle shielding area 150 is determined.

  The relative position of the preceding vehicle shielding area 150 with respect to the host vehicle changes depending on the relative position of the preceding vehicle 110. Accordingly, the search area 160 also changes depending on the relative position of the preceding vehicle 110. Therefore, in step S162, the search area 160 is determined based on the relative position to the preceding vehicle 110.

  In step S63, it is determined whether or not a side closest vehicle is present in the search area 160 determined in step S62. In the case where a vehicle nearest to the side is present in the search area 160, a portion of the camera image in which the vehicle nearest to the side is reflected is set as an analysis target area.

  Therefore, the side closest vehicle in the search area 160 is an analysis target vehicle in which the image of the side closest vehicle in the camera image is analyzed. In the example of FIG. 8, a side nearest vehicle 180 that is an analysis target vehicle exists in the search area 160 on the right side. Since the search area 160 is set diagonally in front of the host vehicle, the vehicle 180 closest to the side is traveling diagonally forward of the host vehicle. Whether or not the analysis target vehicle exists in the search area 160 is determined by analyzing the camera image.

  If the determination in step S63 is no, the process in FIG. 7 ends. Even when the search area 160 cannot be determined because there is no preceding vehicle, step S63 is NO.

  If judgment of step S63 is YES, it will progress to step S64. Steps S64 and S65 are the same processes as steps S45 and S46 of FIG. 4, respectively, and a signal indicating illuminance is acquired from the illuminance sensor 20 in step S64. In step S65, based on the illuminance acquired in step S64, a luminance threshold for determining lighting of the brake lamp of the preceding vehicle and a luminance threshold for determining lighting of the direction indicator lamp are set. In addition, when the illuminance is the same, the specific value of these luminance threshold values may be the same value as step S46, but may be a value different from step S46.

  In step S66, the image portion of the body side surface on the own vehicle side of the analysis target vehicle is analyzed in the camera image. The body side image portion on the own vehicle side of the analysis target vehicle corresponds to the analysis target region in the claims, and the analysis target vehicle corresponds to the specular object in the claims. The content of the analysis is to perform a process for performing the determination in the next step S67. That is, in step S47, the luminance of the portion exhibiting the color of the brake lamp and the luminance of the portion exhibiting the color of the direction indicator light in the body side image portion on the own vehicle side of the vehicle to be analyzed are determined in step S65. It is determined whether the brightness threshold set for each lamp is exceeded.

  In step S67, if the result of the analysis in step S66 shows that the luminance of the portion exhibiting the color of the brake lamp exceeds the luminance threshold set for the brake lamp in step S65, the determination result in step S67 is obtained. Let it be YES. Further, as a result of the analysis in step S66, the determination result in step S67 is also obtained when the luminance of the portion indicating the color of the direction indicator lamp exceeds the luminance threshold set for the direction indicator lamp in step S65. Let it be YES. If the determination result of step S67 is YES, it will progress to step S68, and if it is NO, the process of FIG. 7 will be complete | finished.

  In step S68, it is determined that the lamp of the preceding preceding vehicle has been turned on. As described in step S50 of FIG. 4, the situation in which the lamp of the preceding vehicle is lit is the shielding area attention situation in the claims.

  In step S69, the driver of the host vehicle is notified using one or both of the display unit 30 and the speaker 40 that the lamp of the preceding vehicle has turned on.

  Next, the latest oncoming vehicle analysis process shown in FIG. 9 will be described. In FIG. 9, steps S71 to S72 are executed by the image analysis unit 51, and steps S73 to S74 are executed by the determination unit 52. In the state in which the processing of FIG. 9 is executed, it is determined that the host vehicle makes a right turn, but it is not determined whether or not there is a nearest oncoming vehicle. Therefore, first, in step S71, it is determined whether or not there is a nearest oncoming vehicle. In the present embodiment, it is determined by analyzing the camera image whether or not the nearest oncoming vehicle exists.

  If the decision result in the step S71 is NO, the process of FIG. 9 is terminated, and if the decision result is YES, the process proceeds to a step S72. In step S72, the window image portion of the nearest oncoming vehicle is analyzed in the camera image. When the nearest oncoming vehicle exists, there is a possibility that a moving body that moves the rear side of the nearest oncoming vehicle to the own vehicle side is reflected in the window image portion of the nearest oncoming vehicle. Therefore, the window image portion of the latest oncoming vehicle is analyzed. The window image portion of the latest oncoming vehicle corresponds to the analysis target region in the claims.

  In the analysis, specifically, it is determined whether or not a moving body moving in the direction in which the distance from the host vehicle is shortened in the rear side of the nearest oncoming vehicle is reflected in the window image portion of the nearest oncoming vehicle. This determination is performed by extracting feature points from the window image portion of the nearest oncoming vehicle and determining whether the feature points move relative to the body portion of the nearest oncoming vehicle over time. .

  Further, it is also determined whether the luminance of the portion exhibiting the color of the headlight has exceeded a luminance threshold set in advance for the headlight. This is because the headlight may appear in the window image portion of the most recently oncoming vehicle when a motorcycle is present behind the most recently oncoming vehicle. If the brightness of the headlight-colored part exceeds a preset brightness threshold for the headlight, the determination of whether the object with the headlight is moving is omitted, and the latest It is assumed that there is a moving body that moves to the host vehicle side behind the oncoming vehicle. The reason why the determination of whether the object including the headlight is moving is omitted is because the object including the headlight existing behind the most recently oncoming vehicle normally moves toward the own vehicle.

  In step S73, as a result of the analysis in step S72, whether or not there is a moving body that moves in the direction of the host vehicle, or more precisely, if there is a moving body that moves in the opposite lane in a direction in which the distance from the host vehicle becomes shorter. It is determined whether it has been determined. If it is determined in step S72 that this moving object exists, the determination result in step S73 is YES.

  The situation where there is a moving body that moves in the direction in which the distance from the host vehicle is shortened on the rear side of the nearest oncoming vehicle corresponds to the shielding area caution situation in the claims. If the determination result of step S73 is YES, the process proceeds to step S74, and if NO, the process of FIG. 9 is terminated.

  In step S74, the driver of the host vehicle is notified using one or both of the display unit 30 and the speaker 40 that there is a moving body that moves in the direction of the host vehicle behind the nearest oncoming vehicle. To do.

  As described above, according to the present embodiment described above, the camera 10 that images the front of the host vehicle is provided. In the image captured by the camera 10, the rear window image portion of the preceding vehicle or the nearest oncoming vehicle, or the body side surface portion of the vehicle that is closest to the side, in the area where the visibility is hindered by the preceding vehicle or the nearest facing vehicle. There may be an image of an existing moving object. Therefore, in the camera image, the rear window image portion of the preceding vehicle, the window image portion of the most recent oncoming vehicle, and the body side image portion of the vehicle that is closest to the side are analyzed as analysis target regions (S47, S66, S72). Then, based on the analysis result, it is determined whether or not it is a situation that requires attention to other moving bodies existing in the area where the visibility is obstructed in the preceding vehicle or the nearest oncoming vehicle (S48 to S50, S67 to S48). S68, S73). Therefore, whether or not it is a situation that requires attention to other moving objects existing in the area where visibility is obstructed in the preceding vehicle or the nearest oncoming vehicle without depending on the device mounted on the other vehicle. Can be determined.

  In this embodiment, when it is determined that there is no translucent window portion of the preceding vehicle in the camera image (S23: NO), the vehicle nearest to the side is set as the analysis target vehicle (S25). Therefore, even if there is no transparent window portion of the preceding vehicle in the camera image, whether or not it is a situation that requires attention to other moving objects existing in the area where the visibility is obstructed by the preceding vehicle Can be determined.

  In the present embodiment, in the situation where the host vehicle is about to turn right (S22: YES), the nearest oncoming vehicle is set as the analysis target vehicle (S26). Therefore, it is possible to determine the situation of the area requiring attention in the case of a right turn that requires attention not to the area where the visibility is obstructed by the preceding vehicle but to the area where the visibility is obstructed by the nearest oncoming vehicle.

  In addition, a luminance threshold is set based on the illuminance around the host vehicle (S46, S65). Therefore, it can be accurately determined whether or not the lamp of the preceding vehicle has been lit.

  Also, not only the magnitude of the brightness value, but also whether the brightness distribution pattern of the part exceeding the brightness threshold matches the light emission pattern at the time of the brake lamp and the light emission pattern at the time of the hazard, whether the brake lamp of the preceding preceding vehicle was lit, Alternatively, it is determined whether both the left and right direction indicator lights are lit (S49). Thereby, it can be determined more accurately whether or not the lamp of the preceding vehicle has been lit.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the above-mentioned embodiment, The following modification is also contained in the technical scope of this invention, Furthermore, the summary other than the following is also included. Various modifications can be made without departing from the scope.

<Modification 1>
For example, in the above-described embodiment, the analysis target object is a vehicle, but an object other than the vehicle shown in the camera image may be used as the analysis target object. Specifically, in the situation where the vehicle nearest to the side is the vehicle to be analyzed, that is, in the situation where Step S5 is executed, the glass of the building existing diagonally forward of the host vehicle instead of the vehicle nearest to the side The surface may be a specular object. This is because the preceding car may be reflected on the glass surface of the building.

<Modification 2>
When there is a transparent window portion of the preceding vehicle, in addition to the preceding vehicle analysis process (S4), the side nearest front vehicle analysis process (S5) may be executed.

<Modification 3>
In the preceding vehicle analysis process (S4), it is determined from the camera image whether or not the lamp of the preceding preceding vehicle has been lit, and when it is determined that the lamp has been lit (S50), it is assumed that the shielding area attention situation has occurred. However, it is not limited to this. When the image of the preceding preceding vehicle shown in the rear window image portion of the preceding vehicle is large, the shielding area attention state may be set.

<Modification 4>
In the above-described embodiment, the luminance threshold is determined based on the illuminance around the host vehicle, but the luminance threshold may be a constant value.

<Modification 5>
In addition, when there is no analysis target vehicle in the camera image, it may be displayed on the display unit 30 that it is in a state where it cannot be determined whether or not it is a shielding area attention situation from the camera image.

<Modification 6>
In addition, when the brightness of the image of the rear window of the preceding vehicle is the brightness indicating that sunlight is reflected, even if the preceding vehicle has a translucent window portion, the vehicle to be analyzed is the vehicle that is closest to the side. It is good. If the rear window of the preceding vehicle reflects sunlight, analyze the image of the rear window of the preceding vehicle to accurately detect the lamp lighting of the preceding preceding vehicle and the size of the preceding preceding vehicle. This is because it is considered difficult.

<Modification 7>
A radar that detects an object ahead may be provided, and the position determination of the preceding vehicle (S61) and the presence / absence determination of the nearest oncoming vehicle (S71) may be performed using the radar.

<Modification 8>
In addition to determining that the brake lamp of the preceding vehicle has been lit, it may be determined that the situation is a shielded area attention situation when the brake lamp is continuously lit for a certain time or longer. .

<Modifications 9 to 11>
You may acquire the vehicle type of a preceding vehicle by vehicle-to-vehicle communication (modification 9). Instead of determining whether the vehicle is driving in the right lane, or in addition to determining whether the vehicle is traveling at the right speed An oncoming vehicle may be an analysis target vehicle (Modification 10). When it is determined that the situation is a shielding area attention situation, the behavior of the host vehicle may be controlled instead of the notification or in addition to the notification, such as deceleration control (Modification 11).

<Modification 12>
Further, in the latest oncoming vehicle analysis process of FIG. 9, as in the preceding vehicle analysis process of FIG. 4, the vehicle type of the vehicle to be analyzed may be considered when analyzing the luminance of the window portion. In this case, the light emission pattern of the window of the most recent oncoming vehicle that can be seen from the host vehicle is stored in the storage unit 55 in a state where the moving body that is present behind the most recent oncoming vehicle lights the headlight. . In addition, in order to determine the vehicle type of the most recent oncoming vehicle from the front shape of the most recent oncoming vehicle, a template image in front of the vehicle is also stored for each vehicle type.

  Then, if it is determined in step S71 of FIG. 9 that there is a nearest oncoming vehicle (S71: YES), before executing step S72, a template image in front of the vehicle stored in the vehicle type is used. Determine the type of car. In step S72, in addition to determining whether the luminance of the headlight color portion has exceeded a luminance threshold value set in advance for the headlight, a light emission pattern corresponding to the vehicle type of the vehicle ahead Analysis using is also performed. That is, it is also determined whether the light emission pattern of the headlight color in the window image portion of the latest front vehicle matches the light emission pattern corresponding to the vehicle type of the latest front vehicle. By doing in this way, it can be judged accurately whether the mobile body which moves the back side of the nearest oncoming vehicle to the own vehicle side exists.

<Modification 13>
The luminance threshold value used in step S72 of FIG. 9 may be changed according to the illuminance around the host vehicle as in step S46 of FIG.

DESCRIPTION OF SYMBOLS 1: Front situation determination apparatus, 10: Camera, 20: Illuminance sensor, 30: Display part, 40: Speaker, 50: ECU, 51: Image analysis part, 52: Determination part, 53: Illuminance determination part, 54: Car model determination , 55: storage unit, 56: signal acquisition unit, 60: in-vehicle LAN, 70: preceding vehicle image, 72: rear window image portion, 80: rear image, 100: own vehicle, 110: preceding vehicle, 112: rear window , 120: preceding preceding vehicle, 130: other vehicle, 140: other vehicle, 150: preceding vehicle shielding region, 160: search region, 180: vehicle in front of side

Claims (14)

Used in vehicles,
A camera (10) for imaging the front of the vehicle;
In the image captured by the camera, an area included in the image of the analysis target object selected according to a predetermined condition from an object that does not have an obstacle between the camera and the front without an obstacle between the camera An image analysis unit (51) that analyzes an analysis target region in which an image of a moving body that exists in a region where visibility is obstructed in a vehicle immediately ahead that is a car may be reflected;
Based on the analysis result of the image analysis unit, a determination unit (52) that determines whether or not there is a shielded region caution situation that is a situation that requires attention to the moving object that exists in a region where the field of view is obstructed in the latest forward vehicle. And a forward situation determination device. In claim 1,
The forward situation determination device, wherein the analysis target object is an analysis target vehicle selected from the latest forward vehicles. In claim 2,
The image analysis unit sets, as the analysis target region, a region in which a rear window of a preceding vehicle that is the closest front vehicle traveling in the same lane as the host vehicle that is the vehicle in which the front situation determination device is used is reflected. A forward situation determination device comprising: In claim 3,
The image analysis unit detects that a rear lamp of a preceding preceding vehicle, which is a vehicle traveling immediately before the preceding vehicle, is turned on based on a luminance value of a rear window image portion of the preceding vehicle. ,
The determination unit determines that the shielding area attention state is based on the fact that the image analysis unit detects that a rear lamp of the preceding preceding vehicle has been turned on. Judgment device. In claim 4,
An illuminance determining unit (53) for determining the illuminance around the host vehicle;
The said image analysis part determines the brightness | luminance threshold value which determines with the rear lamp | ramp of the preceding preceding vehicle having turned on based on the illumination intensity determined by the said illumination intensity determination part, The front condition determination apparatus characterized by the above-mentioned. In claim 4 or 5,
In addition to the luminance value of the rear window image portion of the preceding vehicle, the image analysis unit is configured to generate a light emission pattern of the rear window image portion based on a preset light emission pattern. A front situation determination device that determines that a rear lamp is in a lighting state. In claim 6,
A vehicle type determination unit (54) for determining a vehicle type of the vehicle to be analyzed;
The said image analysis part determines the said light emission pattern based on the vehicle type which the said vehicle type determination part determined, The front condition determination apparatus characterized by the above-mentioned. In any one of Claims 1-7,
The image analysis unit includes, in the analysis target region, a region in which a specular object that exists obliquely forward of the host vehicle that is the vehicle in which the device for determining a front state is used is included in the analysis target region. Judgment device. In claim 8,
The image analysis unit uses the side nearest front vehicle that is traveling in front of the host vehicle obliquely in front of the vehicle as the mirror object, and analyzes the region in which the side surface of the side nearest front vehicle is reflected. A forward situation determination apparatus characterized by being included in a target area. In claim 8 or 9,
The image analysis unit determines that there is no window on the rear surface of the preceding vehicle that is the nearest front vehicle traveling in the same lane as the host vehicle, and the region where the specular object is reflected is A forward situation determination device characterized in that the condition is included in an analysis target region. In any one of Claims 1-10,
The image analysis unit includes, in the analysis target region, a region in which a window of the latest front vehicle located in a lane opposite to the host vehicle that is the vehicle in which the front situation determination device is used is reflected. A forward situation determination device. In claim 11,
A direction indicator lamp detection unit (56) for detecting a lighting state of the vehicle direction indicator lamp;
The image analysis unit detects that the direction indicator lamp detection unit has detected that the direction indicator lamp that is turned on when the vehicle crosses the oncoming lane is blinking, in the opposite lane to the host vehicle. A front situation determination device characterized in that a region in which a window of the nearest forward vehicle is located is included in the analysis target region. In claim 11 or 12,
The image analysis unit analyzes whether there is a moving body that moves in the direction of the host vehicle on the rear side of the nearest oncoming vehicle based on the luminance value of the window image portion of the latest forward vehicle. ,
The determination unit determines that it is the shielding area attention situation based on the analysis result of the image analysis unit being an analysis result that there is a moving body moving in the direction of the host vehicle. A forward situation determination device as a feature. In claim 13,
The image analysis unit
In addition to the luminance value of the window image portion of the latest front car, the luminance distribution pattern of the window image portion of the latest front car is a light emission pattern that is set in advance. Determined that there is a moving body that moves in the direction of the host vehicle,
And the said light emission pattern is determined based on the vehicle type of the said nearest front vehicle, The front condition determination apparatus characterized by the above-mentioned.

Images