WO2016208198A1 - 立体物検出装置、ステレオカメラ装置、車両及び立体物検出方法 - Google Patents
立体物検出装置、ステレオカメラ装置、車両及び立体物検出方法 Download PDFInfo
- Publication number
- WO2016208198A1 WO2016208198A1 PCT/JP2016/003043 JP2016003043W WO2016208198A1 WO 2016208198 A1 WO2016208198 A1 WO 2016208198A1 JP 2016003043 W JP2016003043 W JP 2016003043W WO 2016208198 A1 WO2016208198 A1 WO 2016208198A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- distance
- dimensional object
- parallax
- subject
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C3/00—Measuring distances in line of sight; Optical rangefinders
- G01C3/10—Measuring distances in line of sight; Optical rangefinders using a parallactic triangle with variable angles and a base of fixed length in the observation station, e.g. in the instrument
- G01C3/14—Measuring distances in line of sight; Optical rangefinders using a parallactic triangle with variable angles and a base of fixed length in the observation station, e.g. in the instrument with binocular observation at a single point, e.g. stereoscopic type
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/204—Image signal generators using stereoscopic image cameras
- H04N13/207—Image signal generators using stereoscopic image cameras using a single 2D image sensor
- H04N13/225—Image signal generators using stereoscopic image cameras using a single 2D image sensor using parallax barriers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
- G01C11/06—Interpretation of pictures by comparison of two or more pictures of the same area
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
- G01S11/12—Systems for determining distance or velocity not using reflection or reradiation using electromagnetic waves other than radio waves
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V8/00—Prospecting or detecting by optical means
- G01V8/10—Detecting, e.g. by using light barriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/275—Image signal generators from 3D object models, e.g. computer-generated stereoscopic image signals
Definitions
- the present invention relates to a three-dimensional object detection device, a stereo camera device, a vehicle, and a three-dimensional object detection method.
- the solid object detection device of the present disclosure includes an input unit and a control unit.
- the input unit can accept input of a first image and a second image having parallax in the first direction.
- the control unit generates a parallax image in which a difference between the position of the subject in the first image and the position of the subject in the second image in the first direction is represented by a pixel value.
- the control unit divides at least a part of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction.
- the control unit divides the distance in the direction toward the subject into a plurality of distance sections for each of the partial regions.
- the control unit calculates the distance section to which each pixel of the partial region belongs from the pixel value, and sets an evaluation value corresponding to the length in the second direction of the three-dimensional object belonging to the distance section as the distance section. It calculates based on the number of pixels of the pixel to which it belongs.
- the control unit determines the presence of a three-dimensional object for each distance section based on the evaluation value.
- the control unit uses a different threshold value for each distance section, and determines that a three-dimensional object exists when the evaluation value is larger than the threshold value.
- the stereo camera device of the present disclosure includes a stereo camera and a control unit.
- the stereo camera outputs a first image and a second image having parallax in the first direction.
- the control unit generates a parallax image in which a difference between the position of the subject in the first image and the position of the subject in the second image in the first direction is expressed by a pixel value.
- the control unit divides at least a part of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction.
- the control unit divides the distance in the direction toward the subject into a plurality of distance sections for each of the partial regions.
- the control unit calculates the distance section to which each pixel of the partial region belongs from the pixel value, and sets an evaluation value corresponding to the length in the second direction of the three-dimensional object belonging to the distance section as the distance section. It calculates based on the number of pixels of the pixel to which it belongs.
- the control unit determines the presence of a three-dimensional object for each distance section based on the evaluation value.
- the control unit uses a different threshold value for each distance section, and determines that a three-dimensional object exists when the evaluation value is larger than the threshold value.
- the vehicle of the present disclosure includes a stereo camera and a control unit.
- the stereo camera outputs a first image and a second image having parallax in the first direction.
- the control unit generates a parallax image in which a difference between the position of the subject in the first image and the position of the subject in the second image in the first direction is expressed by a pixel value.
- the control unit divides at least a part of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction.
- the control unit divides the distance in the direction toward the subject into a plurality of distance sections for each of the partial regions.
- the control unit calculates the distance section to which each pixel of the partial region belongs from the pixel value, and sets an evaluation value corresponding to the length in the second direction of the three-dimensional object belonging to the distance section as the distance section. It calculates based on the number of pixels of the pixel to which it belongs.
- the control unit determines the presence of a three-dimensional object for each distance section based on the evaluation value.
- the control unit uses a different threshold value for each distance section, and determines that a three-dimensional object exists when the evaluation value is larger than the threshold value.
- the stereo camera images the external environment of the vehicle.
- the solid object detection method of the present disclosure accepts input of a first image and a second image having parallax in the first direction by a stereo camera.
- the three-dimensional object detection method generates a parallax image in which a difference between the position of the subject in the first image and the position of the subject in the second image in the first direction is represented by a pixel value.
- at least a part of the parallax image is divided into a plurality of partial regions extending along a second direction intersecting the first direction.
- the three-dimensional object detection method divides the distance in the direction toward the subject for each of the partial areas into a plurality of distance sections, and calculates the distance section to which each pixel of the partial area belongs from the pixel value, An evaluation value corresponding to the length in the second direction of the three-dimensional object belonging to the distance section is calculated based on the number of pixels belonging to the distance section.
- the three-dimensional object detection method uses a different threshold value for each distance section, and a three-dimensional object exists when the evaluation value is larger than the threshold value.
- FIG. 1 is a diagram schematically illustrating a vehicle equipped with a stereo camera device that travels on a road.
- FIG. 2 is a block diagram illustrating a schematic configuration of a stereo camera device according to one of a plurality of embodiments.
- FIG. 3 is a flowchart showing a three-dimensional object discrimination process in the control unit.
- FIG. 4 is a diagram illustrating an image obtained by photographing the front of the vehicle with one of the stereo cameras.
- FIG. 5 is a diagram for explaining a method of dividing a parallax image into partial regions.
- FIG. 6 is a diagram illustrating an example of a distance section divided in front of the vehicle.
- FIG. 7 is a diagram illustrating the height on the image corresponding to the road surface of one distance section.
- FIG. 8 is a diagram illustrating the height on the image of a three-dimensional object belonging to one distance section.
- FIG. 9 is a diagram illustrating a distance histogram and a threshold for three-dimensional discrimination
- the traveling direction of vehicle 1 (upward in FIG. 1) is the Z direction
- the vehicle width direction of vehicle 1 (rightward in FIG. 1) is the X direction
- a height direction that is a direction perpendicular to the paper surface is defined as a Y direction.
- the Z direction is also a direction toward the subject.
- vehicle in the present application includes, but is not limited to, an automobile, a railway vehicle, an industrial vehicle, and a living vehicle.
- the vehicle may include an airplane traveling on a runway.
- the automobile includes, but is not limited to, a passenger car, a truck, a bus, a two-wheeled vehicle, a trike, and a trolley bus, and may include other vehicles that travel on the road.
- Rail vehicles include, but are not limited to, locomotives, freight cars, passenger cars, trams, guided railroads, ropeways, cable cars, linear motor cars, and monorails, and may include other vehicles that travel along the track.
- Industrial vehicles include industrial vehicles for agriculture and construction. Industrial vehicles include but are not limited to forklifts and golf carts. Industrial vehicles for agriculture include, but are not limited to, tractors, tillers, transplanters, binders, combines, and lawn mowers.
- Industrial vehicles for construction include, but are not limited to, bulldozers, scrapers, excavators, cranes, dump trucks, and road rollers.
- Living vehicles include, but are not limited to, bicycles, wheelchairs, baby carriages, wheelbarrows, and electric two-wheelers.
- Vehicle power engines include, but are not limited to, internal combustion engines including diesel engines, gasoline engines, and hydrogen engines, and electrical engines including motors. Vehicles include those that travel by human power.
- the vehicle classification is not limited to the above. For example, an automobile may include an industrial vehicle capable of traveling on a road, and the same vehicle may be included in a plurality of classifications.
- the stereo camera device 10 includes a stereo camera 11 including two cameras, a left camera 11a and a right camera 11b, and a three-dimensional object detection device 12 electrically connected to the stereo camera 11.
- the “stereo camera” is a plurality of cameras having parallax and cooperating with each other.
- the stereo camera includes at least two cameras. In a stereo camera, it is possible to image a target from a plurality of directions by cooperating a plurality of cameras.
- Stereo cameras include those capable of simultaneously imaging a subject by cooperating a plurality of cameras. “Simultaneous” shooting is not limited to the exact same time.
- the imaging time reference includes an imaging start time, an imaging end time, a captured image data transmission time, and a time at which the counterpart device receives the image data.
- the stereo camera may be a device in which a plurality of cameras are included in one housing.
- the stereo camera may be a device including two or more cameras which are independent from each other and located apart from each other.
- a stereo camera 11 is a camera in which two cameras, a left camera 11a and a right camera 11b, are arranged in the X direction.
- the optical axis directions of the left camera 11a and the right camera 11b are parallel to each other and substantially in the Z direction.
- the optical axis direction may be inclined in the vertical direction from the Z direction.
- a camera having an optical mechanism that guides light incident on two distant locations to one light receiving element can be adopted as a stereo camera.
- two cameras, a left camera 11a and a right camera 11b, which are independent from each other are arranged.
- a plurality of images obtained by capturing the same subject from different viewpoints may be referred to as “stereo images”.
- the entire object imaged by the left camera 11a and the right camera 11b is collectively referred to as “subject”.
- the subject includes a road, a vehicle ahead, an obstacle on the road, a background, and the like.
- the “three-dimensional object” means an object in real space, and particularly an object having a size in the height direction.
- the left camera 11a and the right camera 11b include an image sensor.
- the image sensor includes a CCD image sensor (Charge-Coupled Device Device Image Sensor) and a CMOS image sensor (Complementary MOS Image Sensor).
- the left camera 11a and the right camera 11b may include a lens mechanism.
- the optical axes of the left camera 11a and the right camera 11b face the direction in which the same object can be imaged.
- the left camera 11a and the right camera 11b have different optical axes.
- the left camera 11a and the right camera 11b have their optical axes and positions determined so that at least the same object is included in the captured image.
- the optical axes of the left camera 11a and the right camera 11b are directed parallel to each other.
- the optical axes of the left camera 11a and the right camera 11b are not limited to being parallel but may be in different directions.
- the left camera 11a and the right camera 11b are fixed to the vehicle body of the vehicle 1 so that changes in the position and orientation with respect to the vehicle 1 are reduced.
- the optical axes of the left camera 11a and the right camera 11b may be directed in the Z direction, and may be inclined toward the sky side from the Z direction. The directions of the optical axes of the left camera 11a and the right camera 11b are appropriately changed according to the application.
- the optical axes of the left camera 11a and the right camera 11b are arranged forward, that is, slightly downward in the Z direction.
- the stereo camera device 10 can image various objects such as a preceding vehicle, an obstacle, and white lines 14a and 14b on the road surface 13 included in the external environment of the vehicle while traveling.
- the white line may be called a lane marking.
- the left camera 11a and the right camera 11b are located side by side along the vehicle width direction, that is, the X direction.
- the left camera 11a is located on the left side of the right camera 11b when facing forward, and the right camera 11b is located on the right side of the left camera 11a when facing forward.
- the left camera 11 a and the right camera 11b Due to the difference in position between the left camera 11a and the right camera 11b, the positions of the corresponding subjects in the two images, the first image captured by the left camera 11a and the second image captured by the right camera 11b, are different.
- the first image output from the left camera 11a and the second image output from the right camera 11b are stereo images captured from different viewpoints.
- the left camera 11 a and the right camera 11 b can be installed inside the windshield of the vehicle 1 with the optical axis directed in the Z direction of the vehicle 1.
- the left camera 11a and the right camera 11b may be fixed to any of the front bumper, fender grille, side fender, light module, and bonnet of the vehicle 1, respectively.
- the left camera 11a and the right camera 11b output captured images to the three-dimensional object detection device 12 as digital data.
- the three-dimensional object detection device 12 performs various processes such as image calibration and brightness adjustment on the first image output from the left camera 11a and the second image output from the right camera 11b. Can do. Furthermore, the three-dimensional object detection device 12 performs processing such as creation of a parallax image and determination of a three-dimensional object described below.
- the parallax image is a pixel value representing the difference in the first direction between the position of the subject in the first image and the position of the subject in the second image. The pixel value of each pixel of the parallax image corresponds to the distance of the subject.
- the “distance” is a distance in a direction from the left camera 11a and the right camera 11b toward the subject.
- “distance” means a distance in the Z direction from the left camera 11a and the right camera 11b.
- the “distance” may be a distance in the depth direction of the stereo image by the stereo camera 11.
- the stereo camera device 10 includes a stereo camera 11 and a three-dimensional object detection device 12 as shown in FIG.
- the three-dimensional object detection device 12 includes an input unit 15, a control unit 16, an output unit 17, and an image memory 18.
- the input unit 15 is an input interface for inputting an image to the three-dimensional object detection device 12.
- the input unit 15 can employ a physical connector and a wireless communication device.
- the physical connector includes an electrical connector that supports transmission using an electrical signal, an optical connector that supports transmission using an optical signal, and an electromagnetic connector that supports transmission using electromagnetic waves.
- Electrical connectors include connectors conforming to IEC 60603, connectors conforming to USB standards, connectors corresponding to RCA terminals, connectors corresponding to S terminals defined in EIAJ CP-1211A, D terminals prescribed in EIAJ RC-5237 The connector corresponding to is included.
- the electrical connector includes a connector conforming to the HDMI (registered trademark) standard and a connector corresponding to a coaxial cable including BNC.
- the optical connector includes various connectors conforming to IEC 61754.
- the wireless communication device includes a wireless communication device that complies with each standard including Bluetooth (registered trademark) and IEEE802.11.
- the wireless communication device includes at least one antenna.
- the input unit 15 can accept input of images taken from the left camera 11a and the right camera 11b.
- the input unit 15 delivers the received image to the control unit 16.
- the input to the input unit 15 includes a signal input via a wired cable and a signal input via a wireless connection.
- the input unit 15 may correspond to the imaging signal transmission method of the stereo camera 11.
- the control unit 16 includes one or a plurality of processors and a ROM (Read Only Memory) and a RAM (Random Access Memory) that store programs and data, and executes various processes.
- the processor includes a general-purpose processor that reads a specific program and executes a specific function, or a dedicated processor specialized for a specific process.
- the dedicated processor includes an application specific IC (ASIC: Application Specific Circuit).
- the processor includes a programmable logic device (PLD: Programmable Logic Device).
- PLD includes FPGA (Field-ProgrammablemGate Array).
- the control unit 16 may be one of SoC (System-on-a-Chip) and SiP (System-In-a-Package) in which one or a plurality of processors cooperate.
- the control unit 16 preprocesses the first image and the second image input from the input unit 15, and then generates an image processing unit 19 that generates a parallax image, and a three-dimensional object based on the parallax image generated by the image processing unit 19.
- the determination part 20 which performs these determinations is provided.
- Each of the image processing unit 19 and the determination unit 20 may be a hardware module or a software module.
- the control unit 16 can execute operations that each of the image processing unit 19 and the determination unit 20 can perform.
- the control unit 16 is not limited to the form including the image processing unit 19 and the determination unit 20, and one of the image processing unit 19 and the determination unit 20 may be omitted. In one of the embodiments, the control unit 16 may execute the operations of both the image processing unit 19 and the determination unit 20.
- the operations performed by the image processing unit 19 and the determination unit 20 may be rephrased as operations performed by the control unit 16.
- the processing performed by the control unit 16 using either the image processing unit 19 or the determination unit 20 may be
- the output unit 17 is an output interface that outputs data from the three-dimensional object detection device 12.
- the output unit 17 can employ a physical connector and a wireless communication device.
- the output unit 17 is connected to the network of the vehicle 1 such as CAN (Control Area Network).
- the three-dimensional object detection device 12 is connected to a control device, an alarm device, and the like of the vehicle 1 through the CAN.
- the three-dimensional object detection device 12 outputs three-dimensional object information detected by image processing, distance information to the three-dimensional object, and the like from the output unit 17 to the control device, the alarm device, and the like. Such information is appropriately used in each of the control device and the alarm device.
- output part 17 is separated from input part 15 in one of a plurality of embodiments, it is not restricted to this.
- the input unit 15 and the output unit 17 may be one communication unit.
- This communication unit serves as a communication interface of the stereo camera device 10.
- the communication unit can employ a physical connector and a wireless communication device.
- the image memory 18 is a memory for temporarily storing images captured by the left camera 11a and the right camera 11b input from the input unit 15.
- the image memory 18 includes a memory independent of the processor and a cache memory built in the processor.
- the memory may include DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), and the like.
- the image processing unit 19 receives input of the first image and the second image from the input unit 15 (step S01).
- the image processing unit 19 adjusts the image for each of the input first image from the left camera 11a and the second image from the right camera 11b.
- This adjustment of the image may include correction of misorientation of the left camera 11a and the right camera 11b.
- the image processing unit 19 corrects the image based on the orientation deviation measured in advance.
- This image adjustment may include adjusting the brightness and contrast of the image according to the external environment.
- the stereo camera 11 may perform part or all of the adjustment of the image.
- the image processing unit 19 stores the first image and the second image in which these images have been adjusted in the image memory 18.
- the first image and the second image have parallax in the x direction corresponding to the X direction in real space.
- the x direction is the first direction.
- the direction intersecting the x direction is defined as the y direction.
- the y direction may be orthogonal to the x direction.
- the y direction is the second direction.
- the image processing unit 19 divides the second image into small areas (step S02).
- the small area is a small rectangular area having about 2 to 10 pixels in the x and y directions.
- the small area is an area having 4 pixels each in the x direction and the y direction.
- the image processing unit 19 shifts one pixel at a time in the x direction from the position corresponding to the small area of the second image on the first image, and changes the luminance or color of the second image.
- a small area matching process for searching for a small area that matches the pattern is performed (step S03).
- the image processing unit 19 compares the first image and the second image for the corresponding 4 ⁇ 4 pixels, and when the sum of the absolute values of the luminance differences of the pixels becomes the minimum value, It is determined that the small areas of the first image and the second image match.
- the image processing unit 19 can calculate the distance using the principle of triangulation for each small area, using the shift amount between the first image and the second image between the small areas at this time.
- the image processing unit 19 can obtain the parallax for the entire screen by dividing all regions of the second image into small regions and performing matching with the first image. When the parallax is obtained for the entire screen, the image processing unit 19 generates a parallax image representing the distance with the pixel value of each pixel (step S04).
- the pixel value of each pixel of the parallax image represents the difference in the first direction between the position of the subject in the first image and the position of the subject in the second image.
- the pixel value of each pixel of the parallax image represents the distance to the subject corresponding to the pixel.
- the parallax image may be called a distance image.
- the second image is divided into small areas and the matching process is performed on the first image.
- the matching process on the image may be performed by dividing the first image into small regions and performing the matching process on the second image.
- a small region including the pixel and its surrounding pixels may be extracted from the second image for each pixel, and the matching process may be performed on the first image.
- the number of pixels of the parallax image and the first image and the second image can be substantially the same.
- a numerical value representing parallax is associated with each pixel.
- the parallax image can be displayed as a two-dimensional image in which luminance or color is assigned to each pixel based on the numerical value representing the parallax.
- the parallax image drawn in this way is significantly different from the first image and the second image.
- the three-dimensional object included in the first image and the second image can be recognized as a region having substantially the same luminance or color on the parallax image. Therefore, it is possible to grasp the distance, position, and size of the three-dimensional object by viewing the parallax image.
- generating a parallax image does not require that the parallax image be displayed as a two-dimensional image.
- the parallax image may be generated and used only for arithmetic processing in the control unit 16 as numerical data representing the parallax associated with the pixel.
- the determination unit 20 determines the presence / absence of a three-dimensional object and the distance to the three-dimensional object based on the parallax image created by the image processing unit 19 (steps S05 to S07).
- the determination unit 20 divides the parallax image into a plurality of long regions along the y direction (step S05).
- each region of the divided parallax image may be referred to as a partial region.
- each region of the divided parallax image may be referred to as a strip-shaped region.
- the strip shape means a rectangular shape whose long side is significantly longer than the short side.
- the strip-shaped long side is, for example, 5 times or more of the short side, and may be 10 times or more, or 20 times or more.
- each partial region is used for determining the presence of a three-dimensional object in the region and its distance. Therefore, if the width in the x direction of each partial region is made fine, the resolution for detecting the three-dimensional object in the x direction increases, and a plurality of three-dimensional objects located at different x-direction positions can be simultaneously determined.
- the width in the x direction of each partial region can be several pixels to several tens of pixels.
- the determination unit 20 divides a part of the parallax image into a plurality of partial areas. In other words, the determination unit 20 may divide at least a part of the parallax image into a plurality of partial areas.
- the partial region may cover from the top to the bottom of the parallax image.
- the determination unit 20 creates a distance histogram for each partial region R (step S06).
- the distance histogram shows the distribution of the number of pixels for each distance section by calculating the distance section to which each pixel included in the partial region R belongs from the pixel value.
- the distance section is obtained by dividing the Z direction into a plurality of regions according to the distance in the Z direction.
- the distance axis is taken on the horizontal axis, and the number of pixels belonging to each distance section can be expressed by the height of the column in the distance section.
- a distance section corresponding to the distance in the Z direction (forward) starting from the stereo camera 11 is defined in advance.
- the plurality of distance sections may have different widths between the long distance side boundary and the short distance side boundary.
- the plurality of distance sections may determine the width of the next distance section based on the width of the previous distance section.
- the plurality of distance sections may start from the far side boundary of the measurement target region.
- the distance section may start from a distance of 5 m.
- the first distance section may be, for example, 5m as a short distance side boundary and 5m50cm as a long distance side boundary. In the first distance section, the width of the section between the long distance side boundary and the short distance side boundary is 50 cm.
- a plurality of distance sections may be set so that the width of the second and subsequent distance sections is increased by 10%.
- the width between the long-distance side boundary and the short-distance side boundary of each distance section may be the same length.
- FIG. 6 shows an example of a distance section divided in front of the vehicle 1. According to FIG. 6, the non-measurement area where the pixel count data is not counted is from the position of the stereo camera 11 of the vehicle 1 up to 5 m ahead.
- distance sections are set starting from the distance section S 1 closest to the vehicle 1 while sequentially increasing the width in the Z direction.
- Z 1 the distance from the vehicle 1 in the nth distance section Sn to the short distance side boundary
- Z 2 the distance from the long distance side boundary
- the non-measurement area is 5 m in advance.
- the non-measurement area is not limited to 5 m.
- the top of the three-dimensional object having the minimum height that the stereo camera device 10 must detect is higher than the position where the stereo camera 11 is installed.
- the closest distance on the road surface imaged by the stereo camera 11 and the closest distance at which the top of the three-dimensional object having the minimum height that the stereo camera device 10 must detect are calculated by the stereo camera 11 are calculated.
- the longer distance is the non-measurement distance.
- This non-measuring distance is a distance within which the height direction of the entire three-dimensional object having the minimum height to be measured fits in the stereo image if it is longer than this.
- the determination unit 20 determines the presence of a three-dimensional object from the distance histogram (step S07).
- the threshold value used for determining the presence of a three-dimensional object is set to a large value to avoid misrecognition.
- a certain threshold is provided in the distance histogram, a solid object at a long distance may not be detected.
- a different threshold is used for each distance section according to the distance, as will be described below.
- the focal length of the imaging lens of the stereo camera 11 is f
- the depression angle in the optical axis direction of the imaging lens of the stereo camera 11 is ⁇ 0
- the mounting height of the stereo camera 11 is h 0 . That is, f, ⁇ 0 , and h 0 are the focal length, the depression angle, and the mounting height of the optical system that captured the first image and the second image, respectively.
- the height y road image corresponding to the road surface between the far side of the border and the short distance side of the boundary of the distance interval S n can be expressed by the following equation (1) .
- the group of pixels cannot be distinguished from a flat surface having no thickness on the road surface or the road surface.
- the pixels of the parallax image belonging to the distance section where the three-dimensional object exists are arranged in the width in the x direction of the partial region. In this case, if the size obtained by multiplying the number of pixels arranged in the y direction by the pixel pitch is not larger than the height y load of the road surface image in the distance section, the three-dimensional object belonging to the distance section cannot be determined.
- the height of the image of the three-dimensional object is assumed to be located at the distance Z 3 (Z 1 ⁇ Z 3 ⁇ Z 2 ), and the height of the three-dimensional object is Y.
- the height y object of the three-dimensional object image can be expressed by the following equation (2).
- the three-dimensional object having the height Y is displayed as a group of pixels belonging to the same distance section having the height y object in the parallax image.
- the minimum height of the three-dimensional object to be detected is Y min
- the length in the y direction of the group of pixels belonging to the distance section where the three-dimensional object exists is obtained by substituting Y min for Y in Equation (2).
- it is larger than the obtained y object it is always determined as a three-dimensional object.
- a three-dimensional object higher than the height Y min that must be detected is determined as a three-dimensional object.
- the length of the group of pixels in the y direction may be a size obtained by multiplying the number of pixels arranged in the y direction when the pixels belonging to the distance section of the parallax image are arranged in the x direction width of the partial region by the pixel pitch.
- the determination unit 20 counts the number of pixels belonging to the distance section divided in advance for each partial area, and divides the value obtained by dividing the number of pixels in the x direction of the partial area by the distance section.
- the evaluation value corresponds to the number of pixels arranged in the y direction when the pixels included in the distance section are arranged in the width of the partial region in the x direction. Since the length is obtained by multiplying the number of pixels by the pixel pitch, the evaluation value corresponds to the length in the y direction of the object belonging to the distance section.
- the determination unit 20 compares the evaluation value with a threshold value Th that is different for each distance section. The determination unit 20 determines that a three-dimensional object exists when the evaluation value is larger than a threshold value.
- the threshold Th of each distance section is determined by f indicating the focal length of the optical system that has captured the stereo image, ⁇ 0 of the depression angle of the optical system that has captured the stereo image, and h 0 the mounting height of the optical system that has captured the stereo image.
- Z 1 short-range side boundary of the object distance interval, the long distance side boundary Z 2, the distance deemed distance to predetermined three-dimensional object for each section Z 3, the minimum height of the three-dimensional object to be detected
- Y min and the pixel pitch of the image is P, selection is made so as to satisfy the following conditional expression (3).
- the threshold value Th of the conditional expression (3) is made smaller than the left side of the threshold value Th, there is a possibility that a road surface or a flat object having no thickness on the road surface may be detected. If the threshold value is larger than the right side of Th, a part of the three-dimensional object larger than Ymin may not be detected in the vertical direction that must be detected. By setting the threshold value within the range of conditional expression (3), it is possible to detect only a three-dimensional object and to prevent detection of a three-dimensional object larger than a desired size. Conditional expression (3) may be applied beyond the non-measurement region in the Z direction.
- Table 1 shows the specifications and settings of the stereo camera 11.
- the distance to the boundary of each distance section, the calculated height of the road surface image, and the height of the image of the three-dimensional object having the minimum height to be detected are as shown in Table 2.
- the distance of the boundary of the distance interval z (mm) is a diagram illustrating the distance Z 1 until the short distance side boundary from the first segment in order.
- the distance z is any distance Z 2 until the far side boundary of the previous distance segment.
- the distance z 5.5 m is the far distance side boundary of the first distance section and the near distance side boundary of the second distance section.
- Image height (mm) of the road surface represents the height of the image of the road surface in the distance section with the distance z as the short distance side boundary.
- the “minimum three-dimensional object image height (mm)” is the height of the minimum three-dimensional object image to be detected located in each distance section.
- Z 2 5.5 m
- the image height of the road surface is 0.091 mm
- the minimum image height is 0.400 mm. It is.
- Th the threshold value between these 0.091 mm and 0.400 mm
- the threshold value Th may be set to a smaller value as the distance z is larger.
- the horizontal axis direction represents a distance section.
- the height of the column in each distance section indicates an evaluation value.
- the evaluation value is calculated based on the number of pixels belonging to each distance section.
- an example of the threshold value of the evaluation value for three-dimensional discrimination is shown by a curve.
- the evaluation value as a whole has a high short distance and a low long distance.
- the forward vehicle 21 shown in FIG. 4 exists, and the evaluation value corresponding to the number of pixels belonging to the distance section Sn corresponding to the distance to the back surface of the forward vehicle 21 is high.
- the determination unit 20 uses a different threshold according to the distance for each distance section, and determines that a three-dimensional object exists when the evaluation value is larger than the threshold. Determination unit 20 reduces the to erroneous detection of the road surface or the like of a position close to the vehicle 1, a three-dimensional object located at a distance interval S n can be correctly detected as compared with the prior art. Since the threshold value Th is further lowered depending on the distance, the determination unit 20 can perform correct detection compared to the related art even when there is a three-dimensional object at a far distance.
- the determination unit 20 uses a different threshold for each partial region R according to the distance for each distance section, and a solid object exists when the evaluation value is larger than the threshold. judge. Thereby, the determination part 20 can discriminate
- a threshold value according to conditional expression (3) it is possible to reduce misrecognition of a road surface or a flat object having no thickness on the road surface as a three-dimensional object, and to detect a three-dimensional object having a height higher than that to be detected. It can be detected as a three-dimensional object.
- the determination unit 20 may determine at which position in the x direction the three-dimensional object is present depending on which of the partial areas divided in the x direction is determined to be present.
- the present invention is not limited to the above embodiment.
- the embodiments of the present disclosure may have many variations or modifications within the scope of the present invention.
- the number of cameras mounted on a stereo camera device is not limited to two, and three or more cameras can be used.
- the evaluation value definition and the evaluation value threshold setting method are not limited to those described above, and it is possible to set threshold values according to various evaluation values and evaluation value definitions.
- the control unit 16 can use the number of pixels belonging to each distance section of the distance histogram as an evaluation value, and set a threshold according to the evaluation value.
- the control unit 16 calculates a value obtained by dividing the number of pixels belonging to each distance section in the partial area by the number of pixels in the x direction of the partial area as an evaluation value of the distance section.
- the control unit 16 may calculate another value as the evaluation value.
- the control unit 16 may calculate the evaluation value by dividing the number of pixels described above by the number of pixels in the x direction of the partial region and then multiplying by the correction coefficient.
- the correction coefficient is a value obtained by dividing the number of pixels obtained in the partial area by the number of effective pixels obtained in the partial area.
- the control unit 16 may calculate the correction coefficient before calculating the evaluation value.
- the control unit 16 can detect the three-dimensional object in consideration of missing pixels in the parallax image by multiplying the correction coefficient.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Electromagnetism (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Image Analysis (AREA)
- Measurement Of Optical Distance (AREA)
- Traffic Control Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
Description
10 ステレオカメラ装置
11 ステレオカメラ
11a 左側カメラ
11b 右側カメラ
12 立体物検出装置
13 路面
14a,14b 白線
15 入力部
16 制御部
17 出力部
18 画像メモリ
19 画像処理部
20 判定部
21 前方車両
R 部分領域
S 距離区間
Claims (7)
- 入力部と、制御部とを備え、
前記入力部は、互いに第1方向に視差を有する第1画像及び第2画像の入力を受け付け可能であり、
前記制御部は、
前記第1画像における被写体の位置、及び、前記第2画像における前記被写体の位置の前記第1方向における差を画素値で表現した視差画像を生成し、
前記視差画像の少なくとも一部を前記第1方向に交差する第2方向に沿って広がる複数の部分領域に分け、
前記制御部は、それぞれの前記部分領域について、
前記被写体に向かう方向における距離を複数の距離区間に区分し、
前記部分領域のそれぞれの画素の属する前記距離区間を前記画素値から算出し、
前記距離区間に属する立体物の前記第2方向の長さに対応する評価値を当該距離区間に属する画素の画素数に基づいて算出し、
該評価値に基づいて前記距離区間ごとの立体物の存在を判定し、
前記制御部は、前記距離区間ごとに異なる閾値を用い、前記評価値が前記閾値より大きい場合に立体物が存在すると判定する、立体物検出装置。 - 前記閾値は、前記被写体に向かう方向における距離が遠くなるほど小さい値となる請求項1に記載の立体物検出装置。
- 前記条件式は、判定対象となる最小の高さの立体物全体が、前記第1画像および前記第2画像に収まる距離から遠い距離において適用される請求項3に記載の立体物検出装置。
- ステレオカメラと、制御部と、を備え、
前記ステレオカメラは、互いに第1方向に視差を有する第1画像及び第2画像を出力し、
前記制御部は、
前記第1画像における被写体の位置、及び、前記第2画像における前記被写体の位置の前記第1方向における差を画素値で表現した視差画像を生成し、
前記視差画像の少なくとも一部を前記第1方向に交差する第2方向に沿って広がる複数の部分領域に分け、
前記制御部は、それぞれの前記部分領域について、
前記被写体に向かう方向における距離を複数の距離区間に区分し、
前記部分領域のそれぞれの画素の属する前記距離区間を前記画素値から算出し、
前記距離区間に属する立体物の前記第2方向の長さに対応する評価値を当該距離区間に属する画素の画素数に基づいて算出し、
該評価値に基づいて前記距離区間ごとの立体物の存在を判定し、
前記制御部は、前記距離区間ごとに異なる閾値を用い、前記評価値が前記閾値より大きい場合に立体物が存在すると判定する、ステレオカメラ装置。 - ステレオカメラと、制御部と、を備え、
前記ステレオカメラは、互いに第1方向に視差を有する第1画像及び第2画像を出力し、
前記制御部は、
前記第1画像における被写体の位置、及び、前記第2画像における前記被写体の位置の前記第1方向における差を画素値で表現した視差画像を生成し、
前記視差画像の少なくとも一部を前記第1方向に交差する第2方向に沿って広がる複数の部分領域に分け、
前記制御部は、それぞれの前記部分領域について、
前記被写体に向かう方向における距離を複数の距離区間に区分し、
前記部分領域のそれぞれの画素の属する前記距離区間を前記画素値から算出し、
前記距離区間に属する立体物の前記第2方向の長さに対応する評価値を当該距離区間に属する画素の画素数に基づいて算出し、
該評価値に基づいて前記距離区間ごとの立体物の存在を判定し、
前記制御部は、前記距離区間ごとに異なる閾値を用い、前記評価値が前記閾値より大きい場合に立体物が存在すると判定し、
前記ステレオカメラは、前記車両の外部環境を撮像する、車両。 - ステレオカメラによる、互いに第1方向に視差を有する第1画像及び第2画像の入力を受け付け、
前記第1画像における被写体の位置、及び、前記第2画像における前記被写体の位置の前記第1方向における差を画素値で表現した視差画像を生成し、
前記視差画像の少なくとも一部を前記第1方向に交差する第2方向に沿って広がる複数の部分領域に分け、
それぞれの前記部分領域について、
前記被写体に向かう方向における距離を複数の距離区間に区分し、
前記部分領域のそれぞれの画素の属する前記距離区間を前記画素値から算出し、
前記距離区間に属する立体物の前記第2方向の長さに対応する評価値を当該距離区間に属する画素の画素数に基づいて算出し、
前記距離区間ごとに異なる閾値を用い、前記評価値が前記閾値より大きい場合に立体物が存在する、立体物検出方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP16813970.7A EP3316006B1 (en) | 2015-06-23 | 2016-06-23 | Three-dimensional-object detection device, stereo camera device, vehicle, and three-dimensional-object detection method |
JP2017524646A JP6456499B2 (ja) | 2015-06-23 | 2016-06-23 | 立体物検出装置、ステレオカメラ装置、車両及び立体物検出方法 |
US15/739,481 US10567736B2 (en) | 2015-06-23 | 2016-06-23 | Three-dimensional object detection apparatus, stereo camera apparatus, vehicle, and three-dimensional object detection method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2015-125875 | 2015-06-23 | ||
JP2015125875 | 2015-06-23 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016208198A1 true WO2016208198A1 (ja) | 2016-12-29 |
Family
ID=57584769
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2016/003043 WO2016208198A1 (ja) | 2015-06-23 | 2016-06-23 | 立体物検出装置、ステレオカメラ装置、車両及び立体物検出方法 |
Country Status (4)
Country | Link |
---|---|
US (1) | US10567736B2 (ja) |
EP (1) | EP3316006B1 (ja) |
JP (1) | JP6456499B2 (ja) |
WO (1) | WO2016208198A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2021193026A (ja) * | 2017-01-20 | 2021-12-23 | 株式会社小糸製作所 | 車両用灯具の制御装置 |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6970568B2 (ja) * | 2017-09-22 | 2021-11-24 | 株式会社デンソー | 車両の周辺監視装置と周辺監視方法 |
US20210289113A1 (en) * | 2018-09-18 | 2021-09-16 | AI Gaspar Limited | System and process for identification and illumination of anatomical sites of a person and articles at such sites |
DE102019212021B4 (de) * | 2019-08-09 | 2024-02-08 | Volkswagen Aktiengesellschaft | Verfahren und Vorrichtung zum Feststellen eines Parallaxenproblems in Sensordaten zweier Sensoren |
CN110567408A (zh) * | 2019-09-02 | 2019-12-13 | 安徽中电兴发与鑫龙科技股份有限公司 | 智能型直流手车检验工装 |
CN114677318A (zh) * | 2020-12-24 | 2022-06-28 | 苏州科瓴精密机械科技有限公司 | 一种障碍物识别方法、装置、设备、介质及除草机器人 |
US12082531B2 (en) | 2022-01-26 | 2024-09-10 | Deere & Company | Systems and methods for predicting material dynamics |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH05265547A (ja) * | 1992-03-23 | 1993-10-15 | Fuji Heavy Ind Ltd | 車輌用車外監視装置 |
JP2011185664A (ja) * | 2010-03-05 | 2011-09-22 | Panasonic Electric Works Co Ltd | 対象物検出装置 |
US20140125966A1 (en) * | 2010-04-14 | 2014-05-08 | Digital Ally, Inc. | Traffic scanning lidar |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6055330A (en) * | 1996-10-09 | 2000-04-25 | The Trustees Of Columbia University In The City Of New York | Methods and apparatus for performing digital image and video segmentation and compression using 3-D depth information |
JP5580233B2 (ja) | 2011-03-22 | 2014-08-27 | 富士重工業株式会社 | 車外監視装置および車外監視方法 |
JP5612621B2 (ja) * | 2012-02-28 | 2014-10-22 | 富士重工業株式会社 | 車外環境認識装置 |
-
2016
- 2016-06-23 JP JP2017524646A patent/JP6456499B2/ja active Active
- 2016-06-23 WO PCT/JP2016/003043 patent/WO2016208198A1/ja active Application Filing
- 2016-06-23 EP EP16813970.7A patent/EP3316006B1/en active Active
- 2016-06-23 US US15/739,481 patent/US10567736B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH05265547A (ja) * | 1992-03-23 | 1993-10-15 | Fuji Heavy Ind Ltd | 車輌用車外監視装置 |
JP2011185664A (ja) * | 2010-03-05 | 2011-09-22 | Panasonic Electric Works Co Ltd | 対象物検出装置 |
US20140125966A1 (en) * | 2010-04-14 | 2014-05-08 | Digital Ally, Inc. | Traffic scanning lidar |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2021193026A (ja) * | 2017-01-20 | 2021-12-23 | 株式会社小糸製作所 | 車両用灯具の制御装置 |
JP7133073B2 (ja) | 2017-01-20 | 2022-09-07 | 株式会社小糸製作所 | 車両用灯具の制御装置 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2016208198A1 (ja) | 2018-02-08 |
JP6456499B2 (ja) | 2019-01-23 |
EP3316006A4 (en) | 2019-03-06 |
EP3316006A1 (en) | 2018-05-02 |
US10567736B2 (en) | 2020-02-18 |
US20180324406A1 (en) | 2018-11-08 |
EP3316006B1 (en) | 2020-12-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6456499B2 (ja) | 立体物検出装置、ステレオカメラ装置、車両及び立体物検出方法 | |
EP3358295B1 (en) | Image processing device, stereo camera device, vehicle, and image processing method | |
US10373338B2 (en) | Calculation device, camera device, vehicle, and calibration method | |
US10679388B2 (en) | Image processing apparatus, device control system, imaging apparatus, and recording medium | |
JP7206583B2 (ja) | 情報処理装置、撮像装置、機器制御システム、移動体、情報処理方法およびプログラム | |
JP6141562B1 (ja) | 視差算出装置、ステレオカメラ装置、車両及び視差算出方法 | |
WO2017115732A1 (ja) | 画像処理装置、物体認識装置、機器制御システム、画像処理方法及び画像処理プログラム | |
WO2021045092A1 (ja) | 物体検出装置、物体検出システム、移動体及び物体検出方法 | |
JP6121641B1 (ja) | 画像処理装置、ステレオカメラ装置、車両、及び画像処理方法 | |
JP5027710B2 (ja) | 車両用環境認識装置および先行車追従制御システム | |
JP6674959B2 (ja) | 視差算出装置、ステレオカメラ装置、車両および視差算出方法 | |
JP6855325B2 (ja) | 画像処理装置、ステレオカメラシステム、移動体、路面形状検出方法およびプログラム | |
WO2021060117A1 (ja) | 物体検出装置、物体検出システム、移動体及び物体検出方法 | |
WO2018097269A1 (en) | Information processing device, imaging device, equipment control system, mobile object, information processing method, and computer-readable recording medium | |
WO2021049649A1 (ja) | 路面検出装置、物体検出装置、物体検出システム、移動体及び物体検出方法 | |
US11430147B2 (en) | Image processing apparatus, imaging apparatus, mobile body, and image processing method | |
JP2018088237A (ja) | 情報処理装置、撮像装置、機器制御システム、移動体、情報処理方法、及び、情報処理プログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16813970 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017524646 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15739481 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2016813970 Country of ref document: EP |