JP2019087083A - Travel lane recognition device, travel lane recognition method, and program - Google Patents

Abstract

To improve recognition accuracy of a travel lane.SOLUTION: A travel lane recognition device includes: an acquisition unit for acquiring information on the number of travel lanes of a road on which a vehicle travels; a detection unit for detecting travel lane boundary line information from a photographic image shot by an imaging apparatus mounted on the vehicle; and a selection unit for selecting the travel lane boundary line information used for recognition of the travel lane from the detected travel lane boundary line information. The selection unit selects the travel lane boundary line information used for recognition of the travel lane based on comparison results of the number of travel line boundary lines shown by the detected travel lane boundary line information with the number of travel line boundary lines identified in accordance with information on the number of acquired travel lines.SELECTED DRAWING: Figure 5

Description

  The present invention relates to a traffic lane recognition device, a traffic lane recognition method, and a program.

  BACKGROUND Conventionally, there has been known a traveling lane recognition device that detects lane boundaries (lane marks) from a captured image captured by an imaging device mounted on a vehicle and recognizes a lane in which the host vehicle travels.

  By correctly recognizing the traveling lane by the traveling lane recognition device, the various on-vehicle devices can provide various services to the driver regarding safe driving support. For example, in the navigation device, when navigating the host vehicle, it is possible to urge the driver to change lanes at an appropriate timing based on the recognized traveling lane.

JP, 2015-026234, A

  On the other hand, in the conventional traffic lane recognition apparatus, an object other than the detection target (a step on the road shoulder, a guardrail, a lane boundary of the opposite lane, or the like) is erroneously detected as a lane boundary when the traveling lane is recognized. I had something to do. In such a case, in the conventional travel lane recognition device, the decrease in the recognition accuracy of the travel lane is inevitable.

  The present invention has been made in view of the above problems, and an object thereof is to improve the recognition accuracy of a traveling lane.

According to one aspect, the traffic lane recognition device has the following configuration. That is,
An acquisition unit that acquires information on the number of lanes of a road on which the vehicle travels;
A detection unit that detects lane boundary line information from a captured image captured by an imaging device mounted on the vehicle;
And a selection unit that selects lane boundary information to be used for recognizing a traveling lane from the detected lane boundary information.
The selection unit is configured to compare the number of lane boundaries indicated by the detected lane boundary information with the number of lane boundaries identified according to the acquired information on the number of lanes. It is characterized in that lane boundary line information used for recognizing a traveling lane is selected.

  According to the present invention, it is possible to improve the recognition accuracy of the traveling lane.

FIG. 1 is a diagram illustrating an example of a system configuration of an image processing system. It is a figure showing an example of the hardware constitutions of an image processing device. FIG. 2 is a diagram showing an arrangement example of an imaging device and an example of a shooting range of the imaging device. It is a figure for demonstrating the definition of each information used in a traffic lane information output process. It is a 1st figure which shows an example of a function structure of a traffic lane information output part. It is a first flowchart showing a flow of a driving lane information output process. It is a figure which shows the specific example of a lane boundary line detection process. It is a figure which shows the specific example of a lane number information acquisition process. It is a flowchart which shows the detail of a use lane boundary line selection process. It is a figure which shows the specific example of a use lane boundary line selection process. It is a figure which shows the process result of a driving lane information output process. It is a figure which shows the effect of use lane boundary line selection processing. It is a 2nd figure which shows an example of a function structure of a traffic lane information output part. It is a 2nd flowchart which shows the flow of a driving lane information output process. It is a flowchart which shows the detail of reliability calculation processing.

  Hereinafter, each embodiment will be described with reference to the attached drawings. In the present specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals and redundant description will be omitted.

First Embodiment
<System Configuration of Image Processing System>
First, an image processing system having an image processing device that functions as a traffic lane recognition device will be described. FIG. 1 is a diagram showing an example of a system configuration of an image processing system. As shown in FIG. 1, the image processing system 100 includes an imaging device 110 and an image processing device 120. The imaging device 110 and the image processing device 120 are communicably connected by wire or wireless in a state of being mounted on a vehicle.

  The imaging device 110 transmits a captured image obtained by capturing the outside of the vehicle to the image processing device 120. A travel lane information output program is installed in the image processing apparatus 120, and the travel lane information output unit 130 is realized in the image processing apparatus 120 by executing the program. Thus, the image processing device 120 functions as a traffic lane recognition device in the image processing system 100.

  The traveling lane information output unit 130 processes the captured image transmitted from the imaging device 110 in frame units, and detects lane boundary information from the captured image. Further, the traveling lane information output unit 130 selects the lane boundary information used for recognizing the traveling lane from the detected lane boundary information. Furthermore, the lane number of the traveling lane is determined by recognizing the traveling lane of the host vehicle using the selected lane boundary information, and the determined lane number is output as traveling lane information.

  When selecting lane boundary information used to recognize a traveling lane, the traveling lane information output unit 130 acquires current position information (latitude, longitude, etc.) of the vehicle from a GPS (Global Positioning System) 150. . Further, the traveling lane information output unit 130 refers to the map information storage unit 140 based on the acquired position information, and acquires information on the number of lanes of the road on which the host vehicle travels. The selection of the lane boundary information is performed using information on the number of lanes (details will be described later).

<Hardware Configuration of Image Processing Device>
Next, the hardware configuration of the image processing apparatus 120 will be described. FIG. 2 is a diagram illustrating an example of the hardware configuration of the image processing apparatus. As shown in FIG. 2, the image processing apparatus 120 includes a central processing unit (CPU) 201, a read only memory (ROM) 202, and a random access memory (RAM) 203. The CPU 201, the ROM 202, and the RAM 203 form a so-called computer. The image processing apparatus 120 further includes an auxiliary storage unit 204, a connection unit 205, and a GPS function unit 206. The components of the image processing apparatus 120 are connected to one another via a bus 207.

  The CPU 201 is a device that executes various programs (for example, a traveling lane information output program and the like) installed in the auxiliary storage unit 204. The ROM 202 is a non-volatile memory, and functions as a main storage device that stores various programs, data, and the like necessary for the CPU 201 to execute various programs installed in the auxiliary storage unit 204. Specifically, the ROM 202 stores a boot program such as a BIOS (Basic Input / Output System) or an EFI (Extensible Firmware Interface). The RAM 203 functions as a main storage device that provides a work area developed when the CPU 201 executes various programs installed in the auxiliary storage unit 204.

  The auxiliary storage unit 204 is an auxiliary storage device that stores various programs and information used when executing the various programs. The map information storage unit 140 described above is realized in the auxiliary storage unit 204.

  The connection unit 205 is a device that is connected to the imaging device 110 and receives a captured image transmitted from the imaging device 110. The GPS function unit 206 is a device for acquiring current position information (latitude, longitude, etc.) of the vehicle from the GPS 150.

<Arrangement Example of Imaging Device and Imaging Range of Imaging Device>
Next, an arrangement example of the imaging device 110 and a shooting range of the imaging device 110 will be described. FIG. 3 is a view showing an arrangement example of the imaging device and an example of the imaging range of the imaging device. As shown in FIG. 3, the imaging device 110 is disposed at the rear center position of the vehicle 300 and captures an image of the rear of the vehicle 300.

  The imaging range of the imaging device 110 is wide as shown in the area 310, and all lane boundaries are included in the captured image, regardless of which lane of the road the vehicle 300 travels in a plurality of lanes. There is.

<Description of Each Information Used in Driving Lane Information Output Processing>
Next, the definition of each information used when the traveling lane information output unit 130 executes the traveling lane information output processing will be described. FIG. 4 is a diagram for explaining the definition of each piece of information used in the travel lane information output process. The example in FIG. 4 shows a state in which a vehicle 300 traveling on a road 400 with four lanes on one side is viewed from above. The definition of each information will be described using FIG.

  As shown in FIG. 4, in the road 400 with four lanes on one side, elements such as the lane boundary 401, the lane boundary 402, the lane boundary 403, and the guardrail 404 are included on the left side with respect to the traveling direction of the vehicle 300. Be On the other hand, on the right side (opposite side) with respect to the traveling direction of the vehicle 300, elements such as a lane boundary 411, a lane boundary 412, and a lane boundary 413 of the opposite lane are included.

  Therefore, the captured image captured by the imaging device 110 mounted on the vehicle 300 includes an information element corresponding to each of these elements, and the image processing apparatus 120 includes information including each of the information elements, Detected as lane boundary information. Among these, lane boundary information detected from the position on the left side with respect to the traveling direction of the vehicle 300 is defined as “left Ms”. On the other hand, lane boundary information detected from the position on the right side with respect to the traveling direction of the vehicle 300 is defined as “right Ms”.

  Further, among the information elements included in the lane boundary information “left Ms”, the information element detected from the position closest to the vehicle 300 is defined as “left Ms. Lane Mark [0]”. In the example of FIG. 4, the information element in the captured image corresponding to the lane boundary 401 is defined as "leftMs.LaneMark [0]". Hereinafter, similarly, each information element is defined as "leftMs.LaneMark [1]" to "leftMs.LaneMark [3]" in order from a position close to the vehicle 300.

  As a result, as shown in FIG. 4, the information element "leftMs.LaneMark [1]" corresponds to the lane boundary 402, and the information element "leftMs.LaneMark [2]" corresponds to the lane boundary 403. become. Also, the information element “leftMs.LaneMark [3]” corresponds to the guardrail 404. In the example of FIG. 4, four information elements are detected from the photographed image as information elements included in the lane boundary information "leftMs", so the number of information elements of lane boundary information (leftMs.numLMarks) is , "4".

  Further, among the information elements included in the lane boundary line information “right Ms”, an information element detected from the position closest to the vehicle 300 is defined as “right Ms. Lane Mark [0]”. In the example of FIG. 4, the information element in the captured image corresponding to the lane boundary 411 is defined as "right Ms. Lane Mark [0]". Hereinafter, similarly, each information element is defined as "rightMs.LaneMark [1]" to "rightMs.LaneMark [2]" in order from a position close to the vehicle 300.

  As a result, as shown in FIG. 4, the information element "rightMs.LaneMark [1]" corresponds to the lane boundary 412, and the information element "rightMs.LaneMark [2]" corresponds to the lane boundary 413 of the opposite lane. It will correspond. In the example of FIG. 4, three information elements are detected from the captured image as information elements included in the lane boundary information "right Ms", so the number of information elements in the lane boundary information (right Ms. num L Marks) is , "3".

  Furthermore, the information acquired by the traveling lane information output unit 130 includes map information and position information, and the traveling lane information output unit 130 detects the road on which the vehicle 300 travels based on the map information and the position information. Get information on the number of lanes. In FIG. 4, the information “numLanes” represents information regarding the number of lanes acquired based on the map information and the position information, and in the example of FIG. 4, “numLanes” = “4”.

  Furthermore, the information acquired by the traveling lane information output unit 130 includes the lane number of each lane defined on the road on which the vehicle 300 travels. In FIG. 4, the information “LaneNo.” Represents the lane number acquired by the traveling lane information output unit 130. In the example of FIG. 4, “LaneNo.” = “From the left side” in the traveling direction of the vehicle 300. It is defined as "0", "1", "2", "3".

<Functional Configuration of Driving Lane Information Output Unit>
Next, the functional configuration of the travel lane information output unit 130 will be described. FIG. 5 is a first diagram showing an example of a functional configuration of the traffic lane information output unit. As shown in FIG. 5, the travel lane information output unit 130 includes a captured image acquisition unit 510, a lane boundary detection unit 520, a use lane boundary selection unit 530, a travel lane determination unit 540, and a number of lanes information acquisition unit 550. .

  The captured image acquisition unit 510 acquires a captured image, and notifies the lane boundary detection unit 520 of each frame. The lane boundary detection unit 520 receives the photographed image notified from the photographed image acquisition unit 510 in frame units, and detects lane boundary information (“left Ms”, “right Ms”).

  Here, the lane boundary detection unit 520 directly notifies the traveling lane determination unit 540 of the detected lane boundary information (“left Ms”, “right Ms”) until a predetermined condition is satisfied. The predetermined condition may be, for example, a case where the elapsed time since the traveling lane information output unit 130 starts the traveling lane information output process exceeds a predetermined time threshold. Alternatively, it may be considered that the number of frames for which the travel lane information output unit 130 has executed the travel lane information output process exceeds a predetermined frame number threshold. On the other hand, when it is determined that the predetermined condition is satisfied, the lane boundary detection unit 520 notifies the in-use lane boundary selection unit 530 of the detected lane boundary information ("leftMs", "rightMs"). .

  The use lane boundary selection unit 530 selects an information element to be used for recognizing a traveling lane from each information element of the lane boundary information notified from the lane boundary detection unit 520. Note that among the information elements of the lane boundary information, the information element selected by the in-use lane boundary selection unit 530 is hereinafter referred to as "sel_leftMs.LaneMark [i]" or "sel_rightMs.LaneMark [i]" Define. Here, i represents that the selected information element is the i-th information element counted from the side closer to the vehicle 300. Further, the selected information elements "sel_leftMs.LaneMark [i]" and "sel_rightMs.LaneMark [i]" are collectively defined as "selected lane boundary line information (sel_leftMs, sel_rightMs)" below.

  The use lane boundary selection unit 530 notifies the traveling lane determination unit 540 of the selected lane boundary information ("sel_leftMs", "sel_rightMs"). The use lane boundary selection unit 530 selects lane boundary information according to the following procedure.

  (A) The use lane boundary selection unit 530 acquires, from the number-of-lanes information acquisition unit 550, information ("numLanes") on the number of lanes of the road on which the vehicle 300 travels.

  (B) The use lane boundary selection unit 530, from the travel lane determination unit 540, indicates information on which travel lane the vehicle 300 is traveling from the left side with respect to the traveling direction (lane of travel lane Get the number). The lane number of the traveling lane acquired from the traveling lane determination unit 540 is defined as "LanePosition" below.

  (C) The use lane boundary selection unit 530 specifies the number of lane boundaries present on the left side with respect to the traveling direction of the vehicle 300 based on the acquired lane number ("LanePosition") of the travel lane. For example, in the case of "LanePosition" = "0", since the number of lanes present on the left side with respect to the traveling direction of the vehicle 300 is "0", the number of lane boundaries can be specified as "1". Further, in the case of “LanePosition” = “1”, the number of lanes present on the left side with respect to the traveling direction of the vehicle 300 is “1”, so the number of lane boundaries can be identified as “2”. The number of lane boundaries present on the left side with respect to the traveling direction of the vehicle 300, which is specified based on the acquired lane number of the travel lane, corresponds to each of the lane boundary information notified from the lane boundary detection unit 520. It is nothing but the number of information elements (selection range) to be selected from the information elements. Hereinafter, the selection range is defined as "leftNumUse".

  (D) The use lane boundary selection unit 530 determines the traveling direction of the vehicle 300 based on the acquired information on the number of lanes ("numLanes") and the acquired lane number of the travel lane ("LanePosition"). Identify the number of lane boundaries that exist on the right. For example, in the case of "LanePosition" = "0" and "numLanes" = "4", the number of lane boundaries is "3" because the number of lanes present on the right side with respect to the traveling direction of the vehicle 300 is "3". It can be identified as 4 ". Further, in the case of “LanePosition” = “1”, the number of lanes present on the right side with respect to the traveling direction of the vehicle 300 is “2”, so the number of lane boundaries can be identified as “3”. Note that the number of lane boundaries present on the right side with respect to the traveling direction of the vehicle 300 identified based on the acquired lane number of the traveling lane is the respective lane boundary information notified from the lane boundary detection unit 520. It is nothing but the number of information elements (selection range) to be selected from the information elements. Hereinafter, the selection range is defined as "rightNumUse".

  (E) The use lane boundary selection unit 530 detects lane boundary information ("left Ms") detected from the left side with respect to the traveling direction of the vehicle 300 and a selection range on the left side with respect to the traveling direction of the vehicle 300 ("left NumUse Compare with "). In addition, the use lane boundary selection unit 530 selects the number of information elements ("leftMs.LaneMark [i]") included in the selection range ("leftNumUse") among the information elements of the lane boundary information ("leftMs"). Choose).

  (F) The use lane boundary selection unit 530 detects a lane boundary ("right Ms") detected from the right side with respect to the traveling direction of the vehicle 300 and a selection range ("right NumUse") with respect to the traveling direction of the vehicle 300. Compare with). In addition, the use lane boundary selection unit 530 selects the number of information elements ("rightMs.LaneMark [i]") included in the selection range ("rightNumUse") among the information elements of the lane boundary information ("rightMs"). Choose).

  By the above procedure ((a) to (f)), the use lane boundary selection unit 530 can select lane boundary information.

  The travel lane determination unit 540 recognizes the current travel lane of the vehicle 300 and determines the lane number ("LanePosition"). For example, it is assumed that the traveling lane determination unit 540 receives the detected lane boundary information ("left Ms", "right Ms") from the lane boundary detection unit 520. In this case, the traveling lane determination unit 540 recognizes the traveling lane based on the detected lane boundary information (“left Ms”, “right Ms”).

  Further, for example, it is assumed that the traveling lane determination unit 540 receives the selected lane boundary information ("sel_leftMs", "sel_rightMs") from the use lane boundary selection unit 530. In this case, the traveling lane determination unit 540 recognizes the traveling lane based on the selected lane boundary information ("sel_leftMs", "sel_rightMs"). In addition, as a recognition method for recognizing a traveling lane based on lane boundary line information, any conventional method can be applied, and the description thereof is omitted here.

  Further, the traveling lane determination unit 540 outputs the lane number ("LanePosition") determined based on the recognized traveling lane as traveling lane information and notifies the use lane boundary selection unit 530.

  The lane number information acquisition unit 550 acquires map information from the map information storage unit 140. Further, the lane number information acquisition unit 550 acquires the current position information of the vehicle 300 received by the GPS function unit 206. Furthermore, the number-of-lanes information acquisition unit 550 identifies the road on which the vehicle 300 travels, and notifies the in-use lane boundary selection unit 530 of information ("numLanes") related to the number of lanes of the identified road.

<Flow of traffic lane information output process and specific example>
Next, the flow and specific example of the travel lane information output processing by the travel lane information output unit 130 will be described. FIG. 6 is a first flow chart showing the flow of the travel lane information output process. When the imaging device 110 starts shooting, the traveling lane information output unit 130 executes the traveling lane information output process shown in FIG.

  In step S601, the captured image acquisition unit 510 executes a “captured image acquisition process” to acquire a captured image transmitted from the imaging device 110. In step S602, the lane boundary detection unit 520 executes “lane boundary information detection processing” to detect lane boundary information from the captured image.

  In step S603, the lane boundary detection unit 520 determines whether a predetermined condition is satisfied. If it is determined in step S603 that the predetermined condition is not satisfied (in the case of No in step S603), the process proceeds to step S607. In step S 607, the lane boundary detection unit 520 notifies the traveling lane determination unit 540 of the detected lane boundary information.

  On the other hand, when it is determined in step S603 that the predetermined condition is satisfied (in the case of Yes in step S603), the process proceeds to step S604. In step S604, the number-of-lanes information acquisition unit 550 executes a "number-of-lanes information acquisition process" to acquire information on the number of lanes.

  In step S605, the use lane boundary selection unit 530 executes the “use lane boundary selection process”, and selects an information element to be used for recognition of the traveling lane from each information element of the detected lane boundary information. In step S606, the use lane boundary selection unit 530 notifies the traveling lane determination unit 540 of the selected lane boundary information.

  In step S608, the traveling lane determination unit 540 executes "the traveling lane recognition process", recognizes the traveling lane of the vehicle 300 based on the detected lane boundary information or the selected lane boundary information, and recognizes the traveling. Determine the lane number of the lane. In step S609, the captured image acquisition unit 510 determines whether to end the travel lane information output process. If it is determined in step S609 that the process is not ended (in the case of No in step S609), the process returns to step S601. On the other hand, when it is determined in step S609 that the process is ended (in the case of Yes in step S609), the traveling lane information output process is ended.

  Subsequently, a specific example or details of each step of the traveling lane information output process shown in FIG. 6 will be described. Here, in particular, specific examples or details of steps S602, S604, and S605 will be described.

(1) Specific Example of Step S602 (Lane Boundary Line Information Detection Process) FIG. 7 is a diagram showing a specific example of the lane boundary line information detection process. As shown in FIG. 7, the captured image 700 includes information elements corresponding to the lane boundary 401, the lane boundary 402, the lane boundary 403, and the guardrail 404. Furthermore, the captured image 700 includes information elements corresponding to the lane boundary 411, the lane boundary 412, and the lane boundary 413 of the opposite lane.

  Photographed image 710 shows that lane boundary detection unit 520 executes lane boundary line information detection processing for the photographed image 700 and detects lane boundary line information including each information element shown by a dotted line. .

  The lane boundary line information 720 shows a specific example of the lane boundary line information “leftMs” including the information elements positioned on the left side with respect to the traveling direction of the vehicle 300 among the detected information elements. In the lane boundary line information 720, "a" and "b" are coefficients in the case where each information element is represented by a linear equation Y = aX + b. Also, "width" represents the line width of each information element. Furthermore, “type” represents the line type of each information element, “0” represents a broken line, and “1” represents a solid line. In the case of the lane boundary line information 720, the number of information elements ("left Ms. num L Marks") positioned on the left side with respect to the traveling direction of the vehicle 300 is "4".

  Similarly, the lane boundary information 730 indicates a specific example of the lane boundary information “right Ms” that includes the information elements positioned to the right of the traveling direction of the vehicle 300 among the detected information elements. Similarly to the lane boundary line information 720, "a" and "b" are coefficients when each information element is represented by a linear equation Y = aX + b. Also, "width" represents the line width of each information element. Furthermore, “type” represents the line type of each information element, “0” represents a broken line, and “1” represents a solid line. In the case of the lane boundary line information 730, the number of information elements ("right Ms. num L Marks") positioned on the right side of the traveling direction of the vehicle 300 is "3".

(2) Specific Example of Step S604 (Lane Number Information Acquisition Processing) FIG. 8 is a diagram showing a specific example of the lane number information acquisition processing. In FIG. 8, the map information 800 is an example of the map information read by the lane number information acquisition unit 550 from the map information storage unit 140. Further, in FIG. 8, the map information 810 shows that the current position information of the vehicle 300 is reflected on the map information 800. As shown in the map information 810, by reflecting the current position information of the acquired vehicle 300 from the GPS function unit 206, the lane number information acquisition unit 550 can specify the road 400 on which the vehicle 300 travels. .

  Here, in the map information 810, information on the number of lanes is associated with each road. Therefore, according to the example of FIG. 8, the number-of-lanes information acquisition unit 550 acquires “4” as information (“numLanes”) related to the number of lanes associated with the road 400.

(3) Details of Step S605 (Used Lane Boundary Line Selection Process) FIG. 9 is a flowchart showing details of the used lane boundary line selection process. In step S901, the use lane boundary selection unit 530 acquires information ("numLanes" = "4") regarding the number of lanes of the road 400 on which the vehicle 300 travels.

  In step S902, the use lane boundary selection unit 530 acquires the lane number ("LanePosition") of the travel lane of the vehicle 300 from the travel lane determination unit 540, and calculates the selection range ("leftNumUse"). Here, it is assumed that traveling lane determination unit 540 determines that the lane number ("LanePosition") of the current traveling lane of vehicle 300 is "2". Therefore, the use lane boundary selection unit 530 adds 1 to the travel lane information ("LanePosition" = "2") to calculate the selection range ("leftNumUse") = "3".

  In step S903, the use lane boundary selection unit 530 determines the calculated selection range ("leftNumUse") from the number of information elements ("leftMs.numLMarks") included in the detected lane boundary information ("leftMs"). It is determined whether or not it is large. If it is determined in step S903 that the calculated selection range is larger than the number of information elements included in the detected lane boundary information (in the case of Yes in step S903), the process proceeds to step S904.

  In step S904, the in-use lane boundary selection unit 530 sets the number of information elements ("leftMs.numLMarks") included in the detected lane boundary information ("leftMs") as a selection range ("leftNumUse").

  On the other hand, when it is determined in step S903 that the calculated selection range is equal to or less than the number of information elements included in the detected lane boundary information (in the case of No in step S903), the process proceeds to step S905. Here, the calculated selection range ("leftNumUse" = "3") is equal to or less than the number of information elements ("leftMs.numLMarks" = "4") included in the detected lane boundary information ("leftMs"). . Therefore, the use lane boundary selection unit 530 proceeds to step S905 without changing the selection range calculated in step S902.

  In step S 905, the use lane boundary selection unit 530 selects a selection range (“right NumUse”) based on the acquired lane number (“LanePosition”) of the travel lane of the vehicle 300 and the acquired information on the number of lanes (“numLanes”). Calculate "). Specifically, the use lane boundary selection unit 530 subtracts the lane number ("LanePosition" = "2") of the traffic lane from the information on the number of lanes ("numLanes" = "4"), thereby selecting the selection range ( Calculate "rightNumUse") = "2".

  In step S 906, the in-use lane boundary selection unit 530 determines the calculated selection range (“rightNumUse”) from the number of information elements (“rightMs.numLMarks”) included in the detected lane boundary information (“rightMs”). It is determined whether or not it is large. If it is determined in step S906 that the calculated selection range is larger than the number of information elements included in the detected lane boundary information (in the case of Yes in step S906), the process proceeds to step S907.

  In step S 907, the in-use lane boundary selection unit 530 sets the number of information elements (“right Ms. num LMarks”) included in the detected lane boundary (“right Ms”) as a selection range (“right NumUse”).

  On the other hand, if it is determined in step S906 that the calculated selection range is equal to or less than the number of information elements included in the detected lane boundary (in the case of No in step S906), the process proceeds to step S606 in FIG. Return. Here, the calculated selection range ("rightNumUse" = "2") is equal to or less than the number of information elements ("rightMs.numLMarks" = "3") included in the detected lane boundary information ("rightMs"). . Therefore, the use lane boundary selection unit 530 returns to step S606 in FIG. 6 without changing the selection range calculated in step S905.

(4) Specific Example of Step S605 (Used Lane Boundary Line Selection Processing) FIG. 10 is a diagram showing a specific example of used lane boundary line selection processing. As shown in FIG. 10A, the use lane boundary selection unit 530 receives the lane boundary information 720 from the lane boundary detection unit 520 as the detected lane boundary information ("left Ms"). Then, the use lane boundary selection unit 530 calculates “3” as a selection range (“left NumUse”) by executing the use lane boundary selection process. Therefore, the use lane boundary selection unit 530 selects three information elements ("leftMs.LaneMark [0]" to "leftMs.LaneMark [2]") from the detected lane boundary information 720.

  As a result, the use lane boundary selection unit 530 notifies the traveling lane judgment unit 540 of the lane boundary information 1010 including the selected information element ("sel_leftMs.LaneMarks [0]" to "sel_leftMs.LaneMarks [2]"). Do.

  Further, as shown in FIG. 10B, the use lane boundary selection unit 530 receives the lane boundary information 730 as the detected lane boundary information ("right Ms") from the lane boundary detection unit 520. . Then, the use lane boundary selection unit 530 calculates “2” as a selection range (“right NumUse”) by executing the use lane boundary selection process. Therefore, the use lane boundary selection unit 530 selects two information elements ("right Ms. Lane Mark [0]" to "right Ms. Lane Mark [1]") from the lane boundary information 730.

  As a result, the use lane boundary selection unit 530 notifies the traveling lane judgment unit 540 of the lane boundary information 1020 including the selected information element ("sel_rightMs.LaneMarks [0]" to "sel_rightMs.LaneMarks [1]"). Do.

<Processing result of traffic lane information output process>
Next, the processing result of the travel lane information output processing will be described. FIG. 11 is a diagram showing the processing result of the travel lane information output processing. Among these, FIG. 11 (a) shows the processing result of the traffic lane information output process before the predetermined condition is established as a comparison target, and FIG. 11 (b) shows the result after the predetermined condition is established, The processing result of a driving lane information output process is shown.

  As shown in FIG. 11A, before the predetermined condition is established, the traveling lane is recognized by using all the information elements included in the lane boundary information detected by the lane boundary detection unit 520, and the lane number Is judged. For this reason, as shown to Fig.11 (a), the lane number ("LanePosition") of the present traveling lane of the vehicle 300 might be misjudged with "3".

  On the other hand, as shown in FIG. 11B, after the predetermined condition is satisfied, the use lane boundary selection unit 530 selects an information element to be used for recognizing the traveling lane from the detected lane boundary information. Then, the traveling lane determination unit 540 recognizes the traveling lane using the lane boundary information including the selected information element. Therefore, as shown in FIG. 11B, the lane number ("LanePosition") of the current travel lane of the vehicle 300 can be correctly determined as "2".

<Effect of lane boundary selection process>
Next, the effect of the use lane boundary selection process will be described. FIG. 12 is a diagram showing the effect of the use lane boundary line selection process. In FIG. 12, the horizontal axis represents the elapsed time since the start of the travel lane information output processing, and the vertical axis is travel lane information (the processing result of the travel lane information output processing at each elapsed time) It represents the correct answer rate for the lane number ("LanePosition") of the travel lane.

  As shown in FIG. 12, when the in-use lane boundary line selection process is started at timing 1200 when a predetermined condition is satisfied, the correct answer rate greatly increases after timing 1200.

<Summary>
As apparent from the above description, the image processing apparatus 120 according to the present embodiment recognizes the traveling lane based on the lane boundary line information detected from the photographed image,
・ Acquire information on the number of lanes of the road on which the vehicle travels.
-The number of information elements of the detected lane boundary information is compared with the number of lane boundaries specified according to the acquired information about the number of lanes, and the information used for recognizing the traveling lane based on the comparison result Select an element

  Thereby, according to the image processing apparatus 120 in the present embodiment, it is possible to recognize the traveling lane using the lane boundary line information from which the objects other than the detection object are excluded. As a result, according to the image processing apparatus 120 in the present embodiment, it is possible to improve the recognition accuracy of the traveling lane.

Second Embodiment
In the first embodiment, it is described that the success or failure of the predetermined condition is determined based on the elapsed time, the number of frames, and the like. On the other hand, in the second embodiment, the determination is made based on the reliability of the traveling lane information (the lane number of the traveling lane) output from the traveling lane determination unit 540. Hereinafter, the second embodiment will be described focusing on differences from the first embodiment.

<Functional Configuration of Driving Lane Information Output Unit>
FIG. 13 is a second diagram illustrating an example of a functional configuration of the traffic lane information output unit. The difference from the functional configuration shown in FIG. 5 is that in the case of FIG. 13, a reliability calculation unit 1310 is provided.

  The reliability calculation unit 1310 calculates the reliability of the lane number of the traveling lane of the vehicle 300, which is determined by the traveling lane determination unit 540. Specifically, the reliability calculation unit 1310 first acquires the selection range ("leftNumUse", "rightNumUse") from the in-use lane boundary selection unit 530. In the present embodiment, the use lane boundary selection unit 530 detects lane boundary information ("left Ms", "") detected by the lane boundary detection unit 520 regardless of whether a predetermined condition is satisfied. It is assumed that you have acquired rightMs "). Further, it is assumed that the use lane boundary selection unit 530 calculates the selection range ("leftNumUse", "rightNumUse") based on the detected lane boundary information.

  Subsequently, the reliability calculation unit 1310 acquires the information on the number of lanes (“numLanes”) acquired by the lane number information acquisition unit 550 via the use lane boundary selection unit 530.

  Then, the reliability calculation unit 1310 calculates the reliability ("reliability") using the acquired selection range ("leftNumUse", "rightNumUse") and the information on the acquired number of lanes ("numLanes"). . Further, the reliability calculation unit 1310 notifies the lane boundary detection unit 520 of the calculated reliability ("reliability").

  When the lane boundary detection unit 520 acquires the reliability ("reliability") from the reliability calculation unit 1310, the lane boundary detection unit 520 determines whether the acquired reliability is equal to or higher than a predetermined threshold. When it is less than the predetermined threshold value, it is determined that the predetermined condition is not satisfied. In this case, the lane boundary detection unit 520 directly notifies the traveling lane determination unit 540 of the detected lane boundary information (“left Ms”, “right Ms”).

  On the other hand, when it is determined that the received reliability is equal to or higher than a predetermined threshold value, it is determined that the predetermined condition is satisfied. In this case, the lane boundary detection unit 520 instructs the use lane boundary selection unit 530 to execute the use lane boundary selection process.

<Flow of driving lane information output process>
Next, the flow of the traveling lane information output processing by the traveling lane information output unit 130 will be described. FIG. 14 is a second flowchart showing the flow of the travel lane information output process. Steps different from the first flowchart described with reference to FIG. 6 are steps S1401 and S1402.

  In step S1401, the reliability calculation unit 1310 executes the “reliability calculation process” to calculate the reliability (“reliability”) of the lane number (“LanePosition”) of the traveling lane determined by the traveling lane determination unit 540. . In step S1402, the lane boundary detection unit 520 determines whether the reliability ("reliability") calculated by the reliability calculation unit 1310 is equal to or higher than a predetermined threshold.

  If it is determined in step S1402 that the threshold value is equal to or higher than the predetermined threshold (in the case of Yes in step S1402), the process proceeds to step S604. On the other hand, if it is determined in step S1402 that the value is smaller than the predetermined threshold (in the case of No in step S1402), the process proceeds to step S607.

<Details of reliability calculation process>
Next, details of the reliability calculation process (step S1401 in FIG. 14) will be described. FIG. 15 is a flowchart showing details of the reliability calculation process. In step S604, the number-of-lanes information acquisition unit 550 executes a "number-of-lanes information acquisition process" to acquire information on the number of lanes. In step S 1501, the reliability calculation unit 1310 acquires information on the number of lanes (“numLanes”) from the lane number information acquisition unit 550 via the used lane boundary selection unit 530.

  In step S605, the use lane boundary selection unit 530 executes the “use lane boundary selection process” to calculate a selection range. In step S 1502, the reliability calculation unit 1310 acquires the calculated selection range (“leftNumUse”, “rightNumUse”) from the in-use lane boundary selection unit 530.

  In step S1503, the reliability calculation unit 1310 calculates the reliability based on the information on the number of lanes ("numLanes") and the selection range ("leftNumUse", "rightNumUse"). Specifically, the reliability calculation unit 1310 calculates the reliability ("reliability") based on the following equation (1).

(Formula 1)
reliability = (leftNumUse + rightNumUse) / (numLanes + 1) × 100
In other words, the reliability ("reliability") matches the number of lane boundaries calculated from the information on the number of lanes ("numLanes" + 1) and the sum value of the selection range ("leftNumUse", "rightNumUse") Represents the degree.

  When the "number of lanes information acquisition process" and the "use lane boundary line selection process" are performed in the reliability calculation process (FIG. 15), the "number of lanes information acquisition" is performed in the traveling lane information output process of FIG. The execution of the process "and the use lane boundary line selection process" may be omitted.

<Summary>
As apparent from the above description, the image processing apparatus 120 according to the present embodiment calculates the reliability of the traveling lane information output from the traveling lane determination unit 540, and uses the lane boundary based on the calculated reliability. Determine the start timing of the line selection process.

  Thereby, according to the image processing apparatus 120 in the present embodiment, the same effect as that of the first embodiment can be obtained, and it is possible to start the use lane boundary line selection processing at an appropriate timing.

Other Embodiments
In the first and second embodiments, it has been described that the lane number is defined from the left side with respect to the traveling direction of the vehicle 300. Therefore, in the use lane boundary selection process, the use lane boundary selection unit 530 adds 1 to the lane number ("LanePosition") of the travel lane of the vehicle 300 to calculate the selection range (leftNumUse).

  However, this is not the case when the lane number is specified from the right side with respect to the traveling direction of the vehicle 300. In this case, the use lane boundary selection unit 530 selects a selection range based on the lane number ("LanePosition") of the travel lane of the vehicle 300 acquired from the travel lane determination unit 540 and the information on the number of lanes ("numLanes"). Calculate (leftNumUse).

  That is, in the use lane boundary line selection process in the first and second embodiments, either the left side or the right side with respect to the traveling direction of the vehicle 300 is replaced with the first side, and the opposite is to the first side. By replacing the side with the second side, execution can be performed regardless of the method of defining the lane number.

  In the first and second embodiments described above, the imaging device 110 is disposed at the rear center position of the vehicle 300 and the rear of the vehicle 300 is photographed. However, the arrangement position and the imaging direction of the imaging device 110 are not limited to this, and for example, the imaging device 110 may be disposed at the front center position and configured to image the front of the vehicle 300.

  In the first and second embodiments, the image processing apparatus 120 functioning as a traveling lane recognition apparatus has been described as being configured separately from other vehicle-mounted devices, but it may be integrated with other vehicle-mounted devices You may configure it.

  Note that the present invention is not limited to the configurations shown here, such as combinations with other elements in the configurations and the like described in the above embodiments. These points can be modified without departing from the spirit of the present invention, and can be appropriately determined according to the application form.

100: image processing system 110: imaging device 120: image processing device 130: traveling lane information output unit 510: captured image acquisition unit 520: lane boundary detection unit 530: use lane boundary selection unit 540: traveling lane determination unit 550: Lane number information acquisition unit 1310: reliability calculation unit

Claims (11)

An acquisition unit that acquires information on the number of lanes of a road on which the vehicle travels;
A detection unit that detects lane boundary line information from a captured image captured by an imaging device mounted on the vehicle;
And a selection unit that selects lane boundary information to be used for recognizing a traveling lane from the detected lane boundary information.
The traveling unit may drive based on a comparison result between the number of lane boundaries indicated by the detected lane boundary information and the number of lane boundaries identified according to the acquired information on the number of lanes. A traffic lane recognition device characterized by selecting lane boundary information used for lane recognition. The detection unit detects first lane boundary information positioned on a first side with respect to the traveling direction of the vehicle,
The selection unit is
A first side existing in the first direction with respect to the traveling direction of the vehicle based on information indicating which travel lane the vehicle is traveling from the first side with respect to the traveling direction Identify the number of lane boundaries in
A first lane boundary line number based on a comparison result between the number of first lane boundary lines indicated by the detected first lane boundary line information and the number of first lane boundary lines identified; 2. The traffic lane recognition device according to claim 1, wherein lane boundary line information of 1 is selected. The selection unit is
If the number of first lane boundaries specified is greater than the number of first lane boundaries indicated by the first lane boundary information detected, the first lane boundary detected 3. The travel lane recognition device according to claim 2, wherein the first lane boundary information used for recognizing the traveling lane is selected based on the number of first lane boundaries indicated by the information. The selection unit is
If the number of first lane boundaries specified is less than or equal to the number of first lane boundaries indicated by the detected first lane boundary information, the number of first lane boundaries specified is 4. The driving lane recognition device according to claim 2, wherein the first lane boundary information used to recognize the driving lane is selected based on the selected driving lane. The detection unit detects second lane boundary information located on a second side opposite to the first side with respect to the traveling direction of the vehicle.
The selection unit is
In the traveling direction of the vehicle, based on the information indicating which travel lane the vehicle is traveling from the first side with respect to the traveling direction and the acquired information on the number of lanes Identifying the number of second lane boundaries present on the second side,
A second lane boundary line number indicated by the detected second lane boundary line information, and a second lane boundary line number used to identify the traveling lane based on a comparison result between the identified second lane boundary line number; The traffic lane recognition device according to any one of claims 2 to 4, wherein the lane boundary line information of 2 is selected. The selection unit is
If the number of the second lane boundary identified is larger than the number of the second lane boundary indicated by the detected second lane boundary information, the detected second lane boundary The travel lane recognition device according to claim 5, wherein the second lane boundary information used for recognizing the traveling lane is selected based on the number of second lane boundaries indicated by the information. The selection unit is
If the number of second lane boundaries identified is less than or equal to the number of second lane boundaries indicated by the second lane boundary information detected, the number of second lane boundaries identified is 7. The travel lane recognition device according to claim 5, wherein second lane boundary line information used for recognizing the travel lane is selected based on the selected travel lane. The vehicle further includes a determination unit that determines a lane number indicating which travel lane the vehicle is traveling from the first side with respect to the traveling direction.
The judgment unit
The lane number is determined by recognizing a traveling lane using the detected lane boundary line information until a predetermined condition is established,
The travel according to claim 7, wherein the lane number is determined by recognizing a traveling lane using the lane boundary line information selected by the selection unit after the predetermined condition is satisfied. Lane recognition device. The number of lane boundaries calculated based on the acquired information on the number of lanes, and the number of first lane boundaries indicated by the first lane boundary information used to recognize the selected traveling lane, and A calculation unit for calculating the degree of coincidence between the sum of the number of second lane boundaries indicated by the lane boundary information of 2 and the sum of the second lane boundaries;
9. The driving lane recognition device according to claim 8, wherein the success or failure of the predetermined condition is determined based on the calculated degree of coincidence. An acquisition step of acquiring information on the number of lanes of a road on which the vehicle travels;
A detection step of detecting lane boundary line information from a captured image captured by an imaging device mounted on the vehicle;
And a selection step of selecting lane boundary information to be used for recognizing a traveling lane from the detected lane boundary information.
The traveling may be performed based on a comparison result between the number of lane boundaries indicated by the detected lane boundary information and the number of lane boundaries identified according to the acquired information on the number of lanes. A traffic lane recognition method comprising: selecting lane boundary information used for lane recognition. An acquisition step of acquiring information on the number of lanes of a road on which the vehicle travels;
A detection step of detecting lane boundary line information from a captured image captured by an imaging device mounted on the vehicle;
A program for causing a computer to execute a selection step of selecting lane boundary line information to be used for recognizing a traveling lane from the detected lane line boundary information;
The traveling may be performed based on a comparison result between the number of lane boundaries indicated by the detected lane boundary information and the number of lane boundaries identified according to the acquired information on the number of lanes. A program characterized by selecting lane boundary information used for lane recognition.

Images