WO2017145600A1 - 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム - Google Patents
画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム Download PDFInfo
- Publication number
- WO2017145600A1 WO2017145600A1 PCT/JP2017/001914 JP2017001914W WO2017145600A1 WO 2017145600 A1 WO2017145600 A1 WO 2017145600A1 JP 2017001914 W JP2017001914 W JP 2017001914W WO 2017145600 A1 WO2017145600 A1 WO 2017145600A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- distance
- area
- image
- unit
- parallax
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 97
- 238000003384 imaging method Methods 0.000 title claims abstract description 75
- 238000003672 processing method Methods 0.000 title claims description 3
- 238000001514 detection method Methods 0.000 claims abstract description 89
- 230000009467 reduction Effects 0.000 claims abstract description 8
- 239000000284 extract Substances 0.000 claims description 2
- 238000000034 method Methods 0.000 description 61
- 230000000875 corresponding effect Effects 0.000 description 57
- 230000008569 process Effects 0.000 description 50
- 238000010586 diagram Methods 0.000 description 25
- 238000010191 image analysis Methods 0.000 description 17
- 238000004364 calculation method Methods 0.000 description 16
- 238000009826 distribution Methods 0.000 description 16
- 238000000926 separation method Methods 0.000 description 10
- 238000000605 extraction Methods 0.000 description 8
- 238000002372 labelling Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000002093 peripheral effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003706 image smoothing Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000002250 progressing effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/593—Depth or shape recovery from multiple images from stereo images
-
- 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
-
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- 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
- H04N13/239—Image signal generators using stereoscopic image cameras using two 2D image sensors having a relative position equal to or related to the interocular distance
-
- 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
- H04N13/271—Image signal generators wherein the generated image signals comprise depth maps or disparity maps
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2554/00—Input parameters relating to objects
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30261—Obstacle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
- H04N2013/0081—Depth or disparity estimation from stereoscopic image signals
Definitions
- the present invention relates to an image processing apparatus, an imaging apparatus, a mobile device control system, an image processing method, and a program.
- reduction is performed to reduce the number of pixels of a distance image having a pixel value according to the distance of an object located within a predetermined distance from the plurality of imaging units in accordance with frames captured by the plurality of imaging units.
- a rejection unit configured to reject data of the detected object based on the number of pixels included in the region of the object in the advancing region.
- FIG. 1 is a view showing the configuration of an in-vehicle device control system as a mobile device control system according to an embodiment of the present invention.
- the on-vehicle device control system 1 is mounted on a host vehicle 100 such as a car which is a moving body, and includes an imaging unit 101, an image analysis unit 102, a display monitor 103, and a vehicle travel control unit 104. Then, the imaging unit 101 detects an object in front of the host vehicle from a plurality of captured image data (frames) of the area ahead of the host vehicle in the traveling direction (imaging area). Control mobiles and various in-vehicle devices.
- the control of the moving body includes, for example, notification of a warning, control of the steering wheel of the own vehicle 100 (self-moving body), or a brake of the own vehicle 100 (self-moving body).
- the imaging unit 101 is installed, for example, in the vicinity of a rearview mirror (not shown) of the windshield 105 of the vehicle 100.
- Various data such as captured image data obtained by imaging of the imaging unit 101 is input to an image analysis unit 102 as an image processing unit.
- the image analysis unit 102 analyzes the data transmitted from the imaging unit 101, and on the traveling road surface ahead of the vehicle with respect to the road surface portion on which the vehicle 100 is traveling (a road surface portion located directly below the vehicle). The relative height (position information) at each point is detected, and the three-dimensional shape of the traveling road surface ahead of the host vehicle is grasped. In addition, other objects ahead of the host vehicle, pedestrians, and various obstacles and other objects to be recognized are recognized.
- the analysis result of the image analysis unit 102 is sent to the display monitor 103 and the vehicle travel control unit 104.
- the display monitor 103 displays the captured image data and the analysis result obtained by the imaging unit 101.
- the vehicle travel control unit 104 notifies, for example, a warning to the driver of the own vehicle 100 based on the recognition result of the recognition target such as another vehicle ahead of the own vehicle, a pedestrian, and various obstacles by the image analysis unit 102.
- driving support control such as controlling the steering wheel and brake of the host vehicle.
- FIG. 2 is a diagram showing the configuration of the imaging unit 101 and the image analysis unit 102.
- the imaging unit 101 is configured of a stereo camera including two imaging units 110a and 110b as imaging means, and the two imaging units 110a and 110b are the same.
- the respective imaging units 110a and 110b respectively include imaging lenses 111a and 111b, sensor substrates 114a and 114b including image sensors 113a and 113b on which light receiving elements are two-dimensionally arranged, and analog signals output from the sensor substrates 114a and 114b.
- Electric signal (electrical signal corresponding to the amount of light received by each light receiving element on the image sensors 113a, 113b) is converted into digital electric signal, and it is composed of signal processing units 115a, 115b for generating and outputting captured image data ing. For example, luminance image data is output from the imaging unit 101.
- the imaging unit 101 also includes a processing hardware unit 120 configured of an FPGA (Field-Programmable Gate Array) or the like.
- the processing hardware unit 120 calculates the parallax values of corresponding image portions between the imaged images captured by the imaging units 110a and 110b.
- a disparity calculating unit 121 is provided as a disparity image information generating unit to calculate.
- the parallax value referred to here is a comparison image with respect to an image portion on a reference image corresponding to the same point in the imaging region, with one of the images taken by each of the imaging units 110a and 110b as a reference image and the other as a comparison image.
- the positional displacement amount of the upper image portion is calculated as the parallax value of the image portion.
- FIG. 3 is a diagram for explaining the principle of calculating the distance from the parallax value by using the principle of triangulation.
- f is the focal length of each of the imaging lenses 111a and 111b
- D is the distance between the optical axes.
- Z is the distance from the imaging lenses 111a and 111b to the subject 301 (the distance in the direction parallel to the optical axis).
- the imaging positions in the left and right images with respect to the point O on the subject 301 are respectively ⁇ 1 and ⁇ 2 from the imaging center.
- the image analysis unit 102 is configured of an image processing board or the like, and includes a storage unit 122 configured of a RAM, a ROM, or the like that stores luminance image data and parallax image data and the like output from the imaging unit 101; And a central processing unit (CPU) 123 for executing a computer program for performing parallax calculation control and the like, a data I / F (interface) 124, and a serial I / F 125.
- CPU central processing unit
- the FPGA that configures the processing hardware unit 120 performs processing that requires real-time processing on image data, such as gamma correction, distortion correction (parallelization of right and left captured images), and parallax calculation by block matching to generate a parallax image. And the like, and performs processing such as writing out in the RAM of the image analysis unit 102.
- the CPU of the image analysis unit 102 is responsible for control of the image sensor controller of each imaging unit 110A, 110B and overall control of the image processing board, detection processing of the three-dimensional shape of the road surface, guardrails and other various objects (objects)
- a program for executing detection processing of the image is loaded from ROM, and various processing is executed with luminance image data and parallax image data stored in RAM as input, and the processing result is data I / F 124 and serial I / F 125 Output from the outside.
- vehicle operation information such as the vehicle speed, acceleration (acceleration mainly generated in the front and rear direction of the vehicle) of the vehicle 100, steering angle, yaw rate etc. It can also be used as a processing parameter.
- the data output to the outside is used as input data for performing control (brake control, vehicle speed control, warning control, etc.) of various devices of the host vehicle 100.
- the imaging unit 101 and the image analysis unit 102 may be configured as an imaging device 2 that is an integrated device.
- FIG. 4 is a functional block diagram of an imaging apparatus including an image processing apparatus that performs object detection processing implemented by the processing hardware unit 120 and the image analysis unit 102 in FIG.
- the imaging apparatus includes a stereo camera image acquisition unit 130, a short distance processing unit 140, a normal distance processing unit 150, and an object data list 170.
- the stereo camera image acquisition unit 130 acquires an image captured by a stereo camera.
- the short distance processing unit 140 detects and tracks an object located at a short distance (within a predetermined distance from the host vehicle).
- the normal distance processing unit 150 detects and tracks an object located at a distance greater than the short distance.
- the object data list 170 stores data of an object (object) detected by the short distance processing unit 140 and the normal distance processing unit 150, and outputs the data to the display monitor 103 and the vehicle travel control unit 104.
- processing may be executed at least partially in parallel, or processing is sequentially performed. It is also good.
- the processing by the disparity calculation unit 121 of the processing hardware unit 120 may be performed by performing near-field disparity image generation processing after performing disparity image generation processing and performing processing by the other image analysis units 102 in parallel. Good.
- the stereo camera image acquisition unit 130 acquires luminance image data from the two imaging units 110 a and 110 b constituting the stereo camera. At this time, when the imaging units 110a and 110b are in color, color luminance conversion to obtain a luminance signal (Y) from the RGB signals is performed using, for example, the following formula [1].
- FIG. 5 is a functional block diagram of the short distance processing unit 140.
- the short distance processing unit 140 detects and tracks an object in a short distance (within a predetermined distance from the host vehicle).
- the short distance processing unit 140 includes a short distance parallax image generation unit 141, a reduced parallax image generation unit 142, a short distance object detection unit 143, a short distance object rejection unit 144, and a short distance object tracking unit 145.
- the short distance parallax image generation unit 141 may be realized by, for example, the parallax calculation unit 121 of the processing hardware unit 120.
- the short distance parallax image generation unit 141 performs short distance parallax image generation processing for generating parallax image data (parallax image information) for an object positioned within a predetermined distance from the host vehicle (stereo camera).
- the short distance parallax image generation process first, luminance image data of one of the two imaging units 110a and 110b is used as reference image data, and luminance image data of the other imaging unit 110b is used as comparative image data, Using these, the parallax for the object located within a predetermined distance from the host vehicle is calculated, and parallax image data is generated and output.
- the parallax image data represents a parallax image in which pixel values corresponding to the parallax value d calculated for each image portion on the reference image data are represented as pixel values of the respective image portions.
- the short distance parallax image generation unit 141 defines a block including a plurality of pixels (for example, 16 pixels ⁇ 1 pixel) centered on one target pixel for a row of reference image data. Then, in the same row in the comparison image data, a block having the same size as the block of the defined reference image data is shifted in the horizontal line direction (x direction) by one pixel. At this time, in the short distance parallax image generation processing, the range in the horizontal line direction in which the blocks are shifted is set as a range corresponding to within a predetermined distance (for example, within 1 m to 6 m) from the host vehicle. Thereby, the parallax with respect to the object located within the predetermined distance from the host vehicle is searched.
- a predetermined distance for example, within 1 m to 6 m
- the short distance parallax image generation unit 141 generates a correlation value indicating the correlation between the feature amount indicating the feature of the pixel value of the block defined in the reference image data and the feature amount indicating the feature of the pixel value of each block in the comparison image data. Are calculated respectively. Then, based on the calculated correlation value, a matching process is performed to select a block of comparison image data having the highest correlation with the block of reference image data among the blocks of comparison image data. Thereafter, the amount of positional deviation between the target pixel of the block of the reference image data and the corresponding pixel of the block of the comparison image data selected in the matching process is calculated as the parallax value d. It is possible to obtain parallax image data by performing such a process of calculating the parallax value d for the entire area of the reference image data or a specific area.
- the short-distance parallax image generation unit 141 brings the correlation value with the block of the reference image data closer to the highest value among the blocks in the comparison image data (the highest value Within a value) It is determined whether there are a plurality of blocks, and if there are a plurality of blocks, it is determined that a repeated pattern such as a fence has been detected, and a value indicating a repeated pattern is set as the parallax value d.
- the value (brightness value) of each pixel in the block can be used as the feature amount of the block used for the matching process, and the correlation value can be, for example, the value of each pixel in the block of the reference image data (brightness A sum of absolute values of differences between the values) and the values (brightness values) of the respective pixels in the block of the comparison image data respectively corresponding to these pixels can be used. In this case, it can be said that the block with the smallest total sum is most correlated.
- the matching processing in the short distance parallax image generation unit 141 is realized by hardware processing, for example, SSD (Sum of Squared Difference), ZSSD (Zero-mean Sum of Squared Difference), SAD (Sum of Absolute Difference), A method such as Zero-mean Sum of Absolute Difference (ZSAD) or Normalized cross correlation (NCC) can be used.
- ZSAD Zero-mean Sum of Absolute Difference
- NCC Normalized cross correlation
- the reduced parallax image generation unit 142 is reduced by extracting pixels at predetermined intervals from the parallax image for the object located within the predetermined distance from the own vehicle, which is generated by the short distance parallax image generation unit 141 ( Create a reduced parallax image with a small number of pixels. For example, when the parallax image of 1280 ⁇ 640 pixels is reduced to, for example, 1/10 vertically and horizontally, the reduced parallax image is 128 ⁇ 64 pixels.
- the reduced parallax image generation unit 142 extracts pixels from the parallax image at constant intervals, and the parallax value of the extracted pixels is, for example, a value indicating that the distance is infinity, or a repeated pattern In the case of an invalid value such as a value indicating that there is a value, the value of the extracted pixel is determined (complemented) based on the pixel value around the extracted pixel.
- the reduced parallax image generation unit 142 searches for a pixel having a valid value in the predetermined priority order around the extracted pixel, and if there is a pixel having a non-valid (valid) value, A non-ineffective value is taken as the value of the extracted pixel.
- the predetermined priority order may be, for example, an order of closest distance among a predetermined pixel group separated by a predetermined distance (pixel) from the extracted pixel 661.
- the reduced parallax image generation unit 142 may set, for example, the average value of the extracted pixels and the peripheral pixels thereof as the value of the extracted pixels.
- the parallax value of the extracted pixel when the reliability as parallax is low for example, when the repeated pattern of a fence or the like is detected by the short distance parallax image generation unit 141, for example. Is determined to be an invalid value, and the value of the extracted pixel is determined based on the surrounding pixel values.
- FIGS. 6A and 6B are diagrams for describing a reference image and a parallax image.
- a parallax image is generated for an image in which a preceding vehicle appears as shown in FIG. 6A, the difference in luminance is small in the same color part, so that the parallax value can not be calculated and becomes an invalid value (for example, infinity).
- an invalid value for example, infinity
- the short distance object detection unit 143 detects an object based on the reduced parallax image for the object located within a predetermined distance from the host vehicle, which is generated by the reduced parallax image generation unit 142.
- the short distance object detection unit 143 calculates the actual distance from the vehicle of the object shown in each pixel based on the parallax value of each pixel of the reduced parallax image, and calculates the actual distance from the vehicle of adjacent pixels If the difference is within a predetermined value (for example, 20 cm), the same label is added.
- a predetermined value for example, 20 cm
- “adjacent” may be four directions including upper and lower, right and left, or eight directions including upper right, upper left, lower right, and lower left.
- the short distance object detection unit 143 detects a unit labeled (with the same label added) as one object. Then, when the adjacent objects satisfy the predetermined coupling condition, the short distance object detection unit 143 detects the adjacent objects as one object.
- the predetermined combining condition is, for example, that “any one side of the vertical or horizontal sides of adjacent objects overlaps by half or more, and the difference in actual distance between the adjacent objects from the vehicle is less than 1 m”, etc. Do. Thus, it is possible to capture an object, such as a wall, located obliquely to the traveling direction of the host vehicle as a single object.
- the short distance object detection unit 143 detects a rectangular or square area including an object in the reduced parallax image as an area of the object.
- the short-distance object discarding unit 144 discards (discards) data of objects that are not targeted for tracking among the objects detected by the short-distance object detection unit 143.
- the short-distance object discarding unit 144 calculates the size of the object based on the number of vertical and horizontal pixels of the region of the object detected by the short-distance object detection unit 143 and the parallax value of the pixels included in the region. . Then, objects that are too large (greater than or equal to a predetermined threshold) and objects that are too small (equal to or less than a predetermined threshold) are rejected (data is discarded) as compared with the size of a person or vehicle set in advance.
- the short-distance object discarding unit 144 calculates the three-dimensional position of the object based on the position (coordinates) of the vertical and horizontal pixels of the area of the object and the parallax value of the pixels included in the area. Note that the calculation method of the three-dimensional position may be the same as the calculation method by the three-dimensional position determination unit 161 described later. Then, based on the previous frame (the image of the stereo camera), the height from the road surface is too high by referring to the height of each road surface portion tabulated by the road surface height table calculation unit 154 described later. Objects that are above the threshold) are rejected.
- the short-distance object discarding unit 144 calculates the parallax density of the area of the object, and discards the object whose parallax density is too low.
- the parallax density in this case is, for example, the number (A) of pixels having parallax values according to the object among the pixels contained in the area of the object, the total number of pixels (S) included in the area including the object Calculated by dividing (A / S) by.
- the short-distance object discarding unit 144 performs a progressing area object discarding process that discards the object based on the parallax density of the object in the area in which the host vehicle travels.
- the short distance object discarding unit 144 counts the number of pixels having a parallax value according to the object in the traveling area in which the host vehicle is traveling, and the number of pixels included in the area of the object in the traveling area. And discards (discards) the data of the object detected by the short distance object detection unit 143.
- the short distance object discarding unit 144 calculates the traveling area based on the height and width of the own vehicle and the distance (actual distance) from the own vehicle of the object detected by the short range object detection unit 143.
- FIG. 7 is a flow chart showing an example of the process of discarding the progress area object.
- the short-distance object discarding unit 144 initializes (for example, 0) the number of advancing area elements corresponding to the object (step S201).
- the short-distance object discarding unit 144 determines that the pixel is a component of the object by the short-distance object detection unit 143, and a label indicating the component of the object is added to the pixel. Is detected (step S202).
- the short-distance object discarding unit 144 determines whether the position (coordinates) of the pixel to which the label is added is within a region (traveling region, central region) where the host vehicle travels (step S203).
- step S203 If the position (coordinates) of the pixel to which the label is added is not within the advancing area (NO in step S203), the process proceeds to step S205 described later.
- Step S204 If the position (coordinates) of the pixel to which the label is added is within the advancing area (YES in step S203), the number of advancing area elements stored in association with the object is incremented (increased by 1) Step S204).
- the short-distance object discarding unit 144 determines whether or not there is another pixel that is determined to be the object and is added with the label in the frame of one reduced parallax image by the short-distance object detection unit 143 (Step S205).
- step S205 If there is another pixel that is determined to be the object and to which the label is added in the frame of one reduced parallax image (YES in step S205), the process proceeds to step S202.
- step S205 if there is no other pixel determined to be the object and to which the label is attached (NO in step S205), the area of the object (area such as the smallest rectangle including the object) The total number of pixels in the area included in the advancing area is calculated (step S206).
- the short-distance object discarding unit 144 calculates the parallax density of the object in the area overlapping the moving area, based on the number of moving area elements described above and the total number of pixels included in the moving area calculated in step S206 (Ste S207). Note that the parallax density of an object in the area overlapping the advancing area is calculated, for example, by dividing the number of elements of the advancing area of the object by the total number of pixels included in the advancing area within the minimum rectangular area including the object. Do.
- the short-distance object discarding unit 144 determines whether the parallax density of the object in the area overlapping the advancing area is equal to or less than a predetermined threshold (for example, 20%) (step S208).
- a predetermined threshold for example, 20%
- step S208 If the parallax density of the object in the area overlapping the advancing area is not less than or equal to the predetermined threshold (NO in step S208), the process ends.
- step S208 If the parallax density of the object in the area overlapping the advancing area is less than or equal to the predetermined threshold (YES in step S208), it is determined that the object does not prevent the progress of the own vehicle, and the object is rejected (step S209) , End the process.
- the traveling area may be set in advance as, for example, a predetermined area in the reduced parallax image, or from the own vehicle of the pixel to which the height and width of the own vehicle and the label which have been preset are added. It may be calculated based on the actual distance of
- the traveling area When the travel area is calculated based on the height and width of the vehicle and the actual distance of the labeled pixel from the vehicle, the reduction corresponding to the height and width of the vehicle at the actual distance
- the area of the parallax image may be set as the advancing area.
- the actual distance from the vehicle of the pixel to which the label is added instead of using the value of the actual distance from the vehicle of the pixel to which the label is added, the actual distance from the vehicle of the object including the pixel to which the label is added (for example, each pixel included in the object
- the traveling region may be calculated using the average value of the actual distances of
- FIG. 8 is a diagram for explaining the traveling area object discarding process.
- the short-distance object discarding unit 144 calculates the parallax density of the object in the area overlapping the advancing area as described above in the area 254 where the area 253 such as the smallest rectangle including the object overlaps the advancing area 252.
- FIG. 9A to FIG. 9D are diagrams for explaining the traveling area object discarding process when the tunnel is detected.
- FIG. 9A is a view showing an example of a tunnel shown in a reference image.
- FIG. 9B is a diagram showing an example of a minimum rectangular area 253 a including a tunnel.
- FIG. 9C is a diagram showing an example in which the traveling area 252a is set in advance.
- the traveling area 252a of the host vehicle before the host vehicle enters the tunnel, the traveling area 252a of the host vehicle is included in the area 253a such as the smallest rectangle including the tunnel. Therefore, in the area 253a such as the smallest rectangle including the tunnel, the area 254a included in the advancing area 252a is the entire advancing area 252a.
- the parallax density of the object in the area overlapping the advancing area is, for example, the pixels of the area 255 having the parallax value for the tunnel included in the advancing area 252a (pixels labeled by near-field object detection processing, advancing area element Is calculated by dividing the number of pixels by the number of pixels included in the region 254a.
- FIG. 9D is a diagram showing an example in which the traveling area 252a is calculated based on the height and width of the vehicle and the actual distance from the vehicle of the pixel to which the label is added.
- the predetermined threshold used in step S208 of FIG. 7 may be, for example, 0%.
- the short-distance object tracking unit 145 tracks the objects detected by the short-distance object detection unit 143 and not rejected by the short-distance object discarding unit 144.
- the near distance object tracking unit 145 refers to the object data list 170, and determines that the object detected by the near distance object detection unit 143 is a new detection if it is not stored in the object data list 170.
- the objects are stored in the object data list 170. If the detected object is stored in the object data list 170, the data for the object stored in the object data list 170 is updated to data based on the detected object.
- the determination as to whether the detected object is stored in the object data list 170 may be made, for example, of the actual distance of the detected object from the vehicle and the actual distance of the object stored in the object data list 170 from the vehicle. It is judged according to whether the difference with the distance is equal to or less than a predetermined value.
- FIG. 10 is a functional block diagram of the normal distance processing unit 150. As shown in FIG. The normal distance processing unit 150 detects and tracks an object at a normal distance which is equal to or greater than a predetermined distance (for example, 2 m) from the host vehicle.
- a predetermined distance for example, 2 m
- the parallax image generation unit 151 may be realized by, for example, the parallax calculation unit 121 of the processing hardware unit 120.
- the parallax image generation unit 151 generates parallax image data (parallax image information) for an object located in a range (for example, about 2 m to 80 m from the host vehicle) farther than the short distance parallax image generation unit 141 described above. I do.
- the parallax image generation process is similar to the short distance parallax image generation process by the short distance parallax image generation unit 141 described above, but the parallax image generation process is different from the short distance parallax image generation process, for example, in the following.
- the parallax image generation unit 151 shifts blocks defined in the reference image data in the same row in the comparison image data in the horizontal line direction
- the range in the horizontal line direction in which the blocks are shifted is within a predetermined distance (for example, 2 m) from the host vehicle
- the range corresponds to ⁇ 80m).
- V map generation unit 152 executes V map generation processing for generating a V map.
- Each piece of parallax pixel data included in parallax image data is indicated by a set (x, y, d) of an x-direction position, a y-direction position, and a parallax value d.
- the parallax histogram information of the present embodiment includes three-dimensional coordinate information (d, y, f), and the three-dimensional histogram information distributed in an XY two-dimensional coordinate system is a V map (parallax histogram map , V-disparity map).
- the V map generation unit 152 calculates a disparity value frequency distribution for each row region of disparity image data obtained by dividing an image into a plurality of pieces in the vertical direction.
- Information indicating this disparity value frequency distribution is disparity histogram information.
- FIG. 11A and 11B are diagrams for explaining parallax image data and a V map generated from the parallax image data.
- FIG. 11A is a diagram showing an example of the parallax value distribution of the parallax image
- FIG. 11B is a diagram showing a V map showing the parallax value frequency distribution for each row of the parallax image of FIG. 11A.
- the V map generation unit 152 calculates a parallax value frequency distribution that is a distribution of the number of data of each parallax value for each row, This is output as disparity histogram information.
- a V map as shown in FIG. 11B can be obtained.
- This V map can also be expressed as an image in which pixels having pixel values according to the frequency f are distributed on the two-dimensional orthogonal coordinate system.
- FIG. 12A and 12B are diagrams showing an image example of a captured image as a reference image captured by one imaging unit and a V map corresponding to the captured image.
- FIG. 12A is a captured image
- FIG. 12B is a V map. That is, the V map shown in FIG. 12B is generated from the photographed image as shown in FIG. 12A.
- the V map since parallax is not detected in the area below the road surface, parallax is not counted in the hatched area A.
- FIG. 12A In the example of an image shown to FIG. 12A, the road surface 401 in which the own vehicle is drive
- This image example is a virtual reference obtained by extending the surface in front of the host vehicle relatively flat, ie, the surface in front of the host vehicle parallel to the road surface portion directly below the host vehicle to the front of the host vehicle It corresponds to the road surface (virtual reference movement surface).
- the high frequency points are distributed in a substantially straight line with an inclination such that the parallax value d becomes smaller toward the upper side of the image. Pixels showing such a distribution are present at approximately the same distance in each row on the parallax image and have the highest occupancy rate, and pixels that project an object of identification whose distance increases continuously as they move upward in the image It can be said that
- the parallax value d of the road surface decreases toward the upper side of the image as shown in FIG. 12A.
- the high-frequency points distributed in a substantially straight line on the V map correspond to the features possessed by the pixels projecting the road surface (moving surface). Therefore, it is possible to estimate that the pixels of the point distributed on the approximate straight line obtained by straight line approximation of the high frequency points on the V map or in the vicinity thereof are the pixels showing the road surface with high accuracy. Further, the distance to the road surface portion shown in each pixel can be determined with high accuracy from the parallax value d of the corresponding point on the approximate straight line.
- the road surface shape detection unit 153 detects the three-dimensional shape of the road surface ahead of the vehicle 100 from the information (parallax histogram information) of the V map generated by the V map generation unit 152. Processing is performed.
- working the preceding vehicle 402 which exists in front of the own vehicle 100, and the utility pole 403 which exists out of a road are shown.
- This image example can be obtained by a road surface in which the front surface of the vehicle 100 is relatively flat, that is, a surface in front of the vehicle 100 extends a plane parallel to the road surface portion directly below the vehicle 100 to the front of the vehicle. It is a case where it corresponds to a virtual reference road surface (virtual reference movement surface).
- the high frequency points (road surface 501) are distributed in a substantially straight line with an inclination such that the parallax value d becomes smaller toward the upper side of the image. Pixels showing such a distribution are present at approximately the same distance in each row on the parallax image and have the highest occupancy ratio, and pixels that project a detection target whose distance is continuously farther away toward the upper side of the image It can be said that
- the parallax value d of the road surface decreases toward the upper side of the image as shown in FIG. 12B. Further, in the same line (horizontal line), pixels projecting the road surface have substantially the same parallax value d. Therefore, the high-frequency points (road surface 501) distributed in a substantially straight line on the V map correspond to the features possessed by the pixels projecting the road surface (moving surface). Therefore, it is possible to estimate that the pixels of the point distributed on the approximate straight line obtained by straight line approximation of the high frequency points on the V map or in the vicinity thereof are the pixels showing the road surface with high accuracy. Further, the distance to the road surface portion shown in each pixel can be determined with high accuracy from the parallax value d of the corresponding point on the approximate straight line.
- the road surface height table calculation unit 154 performs road surface height table calculation processing of calculating the road surface height (relative height with respect to the road surface portion immediately below the vehicle) and converting it into a table. From the information on the approximate straight line on the V map generated by the road surface shape detection unit 153, it is possible to calculate the distance to each road surface portion shown in each row region (each position in the vertical direction of the image) on the captured image. On the other hand, each surface portion in the vehicle's traveling direction of a virtual plane extending forward to the vehicle's traveling direction so that the road surface portion located directly below the vehicle is parallel to that surface is shown in any row region in the captured image.
- the virtual plane (reference road surface) is represented by a straight line (reference straight line) on the V map.
- a straight line reference straight line
- the height of each road surface portion ahead of the host vehicle can be obtained.
- the road surface height table calculation unit 154 the heights of the road surface portions obtained from the approximate straight line are tabulated for the necessary parallax range.
- the height from the road surface of the object shown in the captured image portion corresponding to the point where the Y-axis position is y ′ at a certain parallax value d is equal to the Y-axis position on the approximate straight line in the parallax value d And can be calculated from (y'-y0).
- the height H from the road surface for the object corresponding to the coordinates (d, y ′) on the V map can be calculated by the following equation (2).
- “f” is the camera focal length (y′ ⁇ It is a value converted into the same unit as y0).
- BF is a value obtained by multiplying the base length of the stereo camera and the focal length
- “offset” is a parallax value when an object at infinity is photographed.
- the U map generation unit 155 executes frequency U map generation processing and height U map generation processing as U map generation processing for generating a U map (U-disparity map).
- a pair (x, y, d) of an x-direction position, a y-direction position, and a parallax value d in each parallax pixel data included in parallax image data is Set the frequency on the Z-axis, and create XY 2D histogram information. This is called a frequency U map.
- the height H from the road surface is in a predetermined height range (for example, 20 cm to 3 m) based on the height of each road surface portion tabulated by the road surface height table calculation unit 154.
- the frequency U map is created only for the point (x, y, d) of the parallax image in. In this case, an object present in the predetermined height range can be appropriately extracted from the road surface.
- a pair (x, y, d) of the x-direction position, the y-direction position, and the parallax value d in each parallax pixel data included in parallax image data Set the d axis on the axis and the height from the road surface on the Z axis to create XY two-dimensional histogram information. This is called a height U map.
- the value of the height at this time is the highest from the road surface.
- FIG. 13 is an example of an image schematically representing an example of a reference image captured by the imaging unit 110a
- FIGS. 14A and 14B are U maps corresponding to the example of the image in FIG.
- FIG. 14A is a frequency U map
- FIG. 14B is a height U map.
- guard rails 413 and 414 exist on the left and right sides of the road surface, and as the other vehicle, one leading vehicle 411 and one oncoming vehicle 412 exist.
- high frequency points corresponding to the left and right guard rails 413 and 414 are substantially straight lines 603 and 604 extending upward from the left and right ends toward the center.
- high frequency points corresponding to the leading vehicle 411 and the oncoming vehicle 412 are distributed between the left and right guardrails in the states 601 and 602 of line segments extending substantially parallel to the X-axis direction.
- a high frequency point corresponding to another vehicle is a state in which a line segment extending parallel to the substantially X axis direction and a line segment inclined to the substantially X axis direction are connected.
- the height U map points having the highest height from the road surface in the left and right guard rails 413 and 414, the leading vehicle 411, and the oncoming vehicle 412 are distributed similarly to the frequency U map.
- the heights of the distribution 701 of points corresponding to the leading vehicle and the distribution 702 of points corresponding to the oncoming vehicle are higher than the distributions 703 and 704 of points corresponding to the guardrail.
- the real U map generation unit 156 generates a real frequency U map generation process and a real height U map generation process as a U map generation process of generating a real U map (an example of “distribution data”). Run.
- the real U map is obtained by converting the horizontal axis in the U map from the pixel unit of the image to an actual distance, and converting the parallax value on the vertical axis into a thinned parallax having a thinning ratio according to the distance.
- the real U map generation unit 156 In the real frequency U map generation process, the real U map generation unit 156 generates a pair (x, y, d) of the x direction position, the y direction position, and the parallax value d in each parallax pixel data included in parallax image data.
- the two-dimensional histogram information of XY is created by setting the actual distance in the horizontal direction to the X axis, the thinning parallax to the Y axis, and the frequency to the Z axis.
- the real U map generation unit 156 of the present embodiment is based on the heights of the road surface portions tabulated by the road surface height table calculation unit 154, and the height from the road surface.
- a real frequency U map is created only for the point (x, y, d) of the parallax image in which H is in a predetermined height range.
- the real U map generation unit 156 may be configured to generate the real U map based on the U map generated by the U map generation unit 155.
- FIG. 15 is a diagram showing a real U map (hereinafter, real frequency U map) corresponding to the frequency U map shown in FIG. 14A.
- the left and right guardrails are represented by vertical linear patterns 803 and 804, and the preceding vehicle and the oncoming vehicle are also represented by patterns 801 and 802 close to the actual shape.
- the thinning parallax on the vertical axis is not thinned for a long distance (here, 50 m or more), 1/2 for an intermediate distance (20 m or more and less than 50 m), and 1 / l for a short distance (10 m or more and less than 20 m) It is thinned out to 3 and thinned out to 1/8 for a short distance (10 m or more and less than 20 m).
- the amount of thinning is reduced as the distance increases.
- the reason is that since the object appears small at a distance, there is little parallax data and the distance resolution is small, so thinning is reduced, and conversely, an object is large at a short distance, so there is much parallax data and the distance resolution is large.
- a width of 10 m on each side is set as the object detection range. Assuming that the width of one pixel in the horizontal direction of the real U map is 10 cm, the horizontal size of the real U map is 200 pixels.
- the focal length of the camera is f
- the lateral position of the sensor from the camera center is p
- the distance from the camera to the subject is Z
- the lateral position from the camera center to the subject is X.
- Z Bf / d”.
- a real U map (hereinafter, real height U map) corresponding to the height U map shown in FIG. 14B can also be created in the same procedure.
- the real U map has an advantage that the processing becomes faster because the vertical and horizontal lengths can be made smaller than the U map.
- the horizontal direction becomes independent of distance, the same object can be detected with the same width in both far and near areas, and it is possible to judge the processing branch to the peripheral area removal, horizontal separation, and vertical separation in the latter stage.
- FIG. 17 is a flowchart showing the flow of the isolated area detection process performed by the isolated area detection unit 157.
- the isolated area detection unit 157 first smoothes the information of the real frequency U map generated by the real U map generation unit 156 (step S111).
- the real U map differs from the schematic diagram shown in FIG. Contains. Therefore, the real U map is smoothed to remove noise and to facilitate separation of the object to be detected. This is considered to be noise by applying a smoothing filter (for example, a simple average of 3 ⁇ 3 pixels) to a frequency value (real frequency U map) of the real U map as in the case of image smoothing. The frequency is reduced, and in the part of the object, the group is grouped more frequently than the surroundings, which has the effect of facilitating the subsequent isolated area detection processing.
- a smoothing filter for example, a simple average of 3 ⁇ 3 pixels
- a real frequency U map is used to detect isolated regions (referred to as islands) whose frequency is higher than that of the surroundings.
- the real frequency U map is first binarized (step S113).
- binarization is performed with a threshold value of 0. This is a measure against the fact that some islands are isolated and some are connected to other islands because of the height of the object, its shape, and the separation from the road surface parallax. That is, by binarizing the real frequency U map from a small threshold, an island of an appropriately isolated size is detected at first, and then the connected islands are separated by increasing the threshold, and the island is isolated. It is possible to detect as an island of appropriate size.
- the method of detecting the island after binarization uses labeling (step S114). Coordinates that are black after binarization (coordinates whose frequency value is higher than the binarization threshold) are labeled based on their connectivity, and areas with the same label are regarded as islands.
- Size determination is performed for each of the plurality of isolated areas detected (step S115). This is because the object to be detected is a pedestrian to a large car, so it is determined whether the width of the isolated area is within the range of its size. If the size is large (step S115: YES), the binarization threshold is incremented by 1 (step S112), and binarization is performed only in the isolated region of the real frequency U map (step S113). Then, labeling is performed, a smaller isolated area is detected (step S114), and the size is determined (step S115).
- step S115 peripheral area removal is performed next (step S116). This is an object at a distance, the road surface detection accuracy is poor, the parallax of the road surface is introduced into the real U map, and the parallax between the object and the road surface is the height when it is detected as one mass Is a process of deleting the area near the road surface (the peripheral area in the isolated area). If there is a removal area (step S117: YES), labeling is performed again to reset the isolated area (step S114).
- step S117 If there is no more removal area (step S117: NO), the size (width, height, distance) of the isolated area subjected to the removal of the surrounding area is determined (step S118), and the separation in the lateral direction is performed S119) or vertical separation (step S120) or nothing is registered as an object candidate. If horizontal separation or vertical separation has been performed (step S121: YES, step S122: YES), labeling is performed again to reset the isolated area (step S114).
- FIG. 18 is a view showing a real frequency U map in which a rectangular area inscribed in the isolated area detected by the isolated area detection unit is set.
- FIG. 19 is a diagram in which a scanning range corresponding to the rectangular area in FIG.
- FIG. 20 is a diagram showing an image
- FIG. 20 is a diagram showing a parallax image in which an object region is set by searching for the scanning range in FIG.
- the width of the rectangular area corresponds to the width of the identification target (object) corresponding to the isolated area.
- the height of the set rectangular area corresponds to the depth (the length in the traveling direction of the host vehicle) of the identification target (object) corresponding to the isolated area.
- the height of the identification object (object) corresponding to each isolated area is unknown at this stage.
- the corresponding area detection unit 158 of the parallax image detects the corresponding area on the parallax image corresponding to the isolated area in order to obtain the height of the object corresponding to the isolated area related to the object candidate area.
- the corresponding area detection unit 158 of the parallax image detects the positions, widths, and minimums of the first detection island 811 and the second detection island 812 detected from the real U map based on the information of the isolated area output from the isolated area detection unit 157. From the parallax, the x-direction range (xmin, xmax) of the first detected island corresponding area scanning range 481 and the second detected island corresponding area scanning range 482 which are ranges to be detected in the parallax image shown in FIG. 19 can be determined.
- the set scanning range is scanned, and a pixel having a value of a range of rectangular depth (minimum parallax dmin, maximum parallax dmax) detected by the isolated area detection unit 157 as parallax Are extracted as candidate pixels. Then, among the extracted candidate pixel groups, a line having a predetermined ratio or more in the horizontal direction with respect to the detection width is set as a candidate line.
- scanning is performed in the vertical direction, and when there is a predetermined density or more of another object candidate line around a certain focused object candidate line, the focused object candidate line is determined as an object line .
- the object area extraction unit 159 searches for an object line in the search area of the parallax image to determine the lowermost end and the uppermost end of the object line, and as shown in FIG.
- the area 462 is determined as the areas 451 and 452 of the objects (first and second vehicles) in the parallax image.
- FIG. 21 is a flowchart showing the flow of processing performed by the corresponding area detection unit 158 and the object area extraction unit 159 of the parallax image.
- the search range in the x-axis direction with respect to the parallax image is set based on the position, width and minimum parallax of the island in the real U map (step S161).
- the maximum search value ymax in the y-axis direction with respect to the parallax image is set (step S162).
- the minimum search value ymin in the y-axis direction for the parallax image is obtained and set from the maximum height of the island in the real height U map and the ymax and dmax set in step S172, thereby the y-axis for the parallax image
- the search range of the direction is set (step S163).
- the parallax image is searched in the set search range, and pixels in the range of the minimum parallax dmin of the island and the maximum parallax dmax are extracted, and are set as object candidate pixels (step S164).
- the object candidate pixel is in the horizontal direction at a certain ratio or more, the line is extracted as an object candidate line (step S165).
- the density of the object candidate line is calculated, and if the density is larger than a predetermined value, the line is determined as an object line (step S166). Finally, a circumscribed rectangle of the object line group is detected as an object area in the parallax image (step S167).
- an identification object object, object
- object object
- the identification object (object) displayed in the image area corresponding to the object area The actual height Ho can be calculated.
- “zo” is the distance between the object corresponding to the object area and the vehicle calculated from the minimum parallax value d in the object area
- “f” is the focal length of the camera (yomax-yomin) It is a value converted to the same unit as the unit of.
- the object type classification unit 160 classifies the object type based on the information on the height, width, and depth of the object corresponding to the object area that can be calculated in this manner.
- the table shown in FIG. 22 shows an example of table data for classifying object types. According to this, it is possible to distinguish and recognize whether an identification target (object) existing ahead of the host vehicle is a pedestrian, a bicycle, a small car, a truck or the like.
- the three-dimensional position determination unit 161 Since the distance to the object corresponding to the detected object region and the distance on the image between the image center of the parallax image and the center of the object region on the parallax image are also grasped, the three-dimensional position of the object is determined. Can.
- the relative lateral to the imaging units 110a and 110b of the identification target (object) can be calculated from the following equation [8] and equation [9].
- the object tracking unit 162 executes processing of tracking (tracking) an object (object) detected from a frame of a previous (past) parallax image, using the object data list 170.
- the object selection unit 163 selects whether or not to track the object based on whether the object stored in the object data list 170 is at a position suitable for tracking, or the like. Specifically, after a prediction range in parallax image data in which the object is predicted to be located is set based on the object prediction data of the object data list 170 and the height of the object within the prediction range is specified, The width of the object is identified from the identified height with reference to the object feature amount of the object data list 170, and the lateral position of the object on the parallax image data is estimated from the identified width.
- the lateral position of the object estimated in this way is a predetermined tracking target condition (when the object is in a position where the probability of being present in the image is high, the position is appropriate for the object to be the tracking target, etc.) If the above is satisfied, the object is selected as a tracking target.
- the object selection unit 163 may delete data of an object not selected as a tracking target from the object data list 170.
- the object data list 170 is used, for example, to track data of each object detected from one frame over a plurality of subsequent frames.
- the object data list 170 includes data of “object data”, “object prediction data”, and “object feature amount”, as shown in FIG.
- Object data includes data of "position”, “size”, “distance”, “relative velocity”, “actual position”, and “actual size”.
- “Position” is the upper left coordinates of the object in the parallax image.
- the “size” is the size (the number of pixels of the width and the height) of the object in the parallax image.
- “Distance” is an actual distance from the host vehicle to the object.
- the “relative velocity” is the relative velocity between the vehicle and the object, which is calculated based on the previous and current frames.
- the “relative speed” includes data of the amount of movement (m: meters) per frame in the traveling direction Z (depth direction) of the host vehicle and in the horizontal direction X (lateral direction).
- the "actual position” is the actual relative position of the object relative to the host vehicle calculated based on the current frame.
- the “actual position” includes, for example, data of the position in the lateral direction Z (m: meter) and the position in the depth Z direction (m: meter) of the center of the left and right ends of the object.
- the "actual size” is the actual size of the object relative to the host vehicle calculated based on the current frame.
- the “actual size” includes data of the width (m: meter) and the height (m: meter) of the object.
- the “object prediction data” includes data such as “position”, “size”, “distance”, “relative velocity” and the like.
- the “object prediction data” is data which is calculated based on the previous frame (for example, the immediately preceding frame) and which predicts the position of the object in the next frame.
- the “object feature amount” includes data of “histogram peak position” and “distance between peaks”.
- the “histogram peak position” is coordinates of a position where the value of disparity frequency in the real U map is relatively large.
- the “peak-to-peak distance” is an actual distance between “histogram peak positions” in the real U map.
- the data of the object data list 170 is sent to the display monitor 103 and the vehicle travel control unit 104 as an analysis result of the image analysis unit 102.
- the display monitor 103 displays, for example, a green frame of the area of the object in the current frame based on, for example, the "position” and the "size” of the "object data” among the data of the object data list 170 which is the analysis result. Display with.
- the "position” and the "size” of the "object data” are data indicating an area including the object.
- the display monitor 103 corresponds to the predicted position of the object in the current frame based on, for example, the “position” and the “size” of the “object predicted data” among the data of the object data list 170 which is the analysis result.
- An area to be displayed (hereinafter referred to as "tracking frame") is displayed, for example, with a red frame.
- a parallax image for an object located at a distance of about 2 m to 80 m (normal distance) from the host vehicle is generated at high speed to detect and track the object. It is conceivable.
- an object in a short distance is detected by an area such as the smallest rectangle including the object in the parallax image, for example, an area such as a tunnel is different from a bar at an entrance of a parking lot etc. Since the area is high, the area in which the host vehicle travels may be detected as an obstacle although it is vacant.
- the value of the invalid pixel is complemented based on the pixel value of the periphery. Density is higher.
- an object detected based on the number of pixels having a parallax value according to the object in the advancing area of the vehicle and the number of pixels included in the area of the object in the advancing area Reject the data of Thereby, highly accurate detection can be performed at high speed.
- a distance image may be generated by integrating distance information generated using a detection device such as a millimeter wave radar or laser radar with respect to parallax images generated using a stereo camera.
- detection and tracking of an object located at a distance of about 0.5 m to 60 m may be performed.
- the detection accuracy may be further enhanced by using a stereo camera and a detection device such as a millimeter wave radar or a laser radar in combination and combining it with the above-described detection result of an object by the stereo camera.
- a detection device such as a millimeter wave radar or a laser radar in combination and combining it with the above-described detection result of an object by the stereo camera.
- each functional unit of the processing hardware unit 120 and the image analysis unit 102 may be configured to be realized by hardware, or may be configured to be realized by execution of a program stored in a storage device by a CPU. Good.
- This program may be recorded in a computer readable recording medium and distributed by a file in an installable format or an executable format.
- a CD-R Compact Disc Recordable
- DVD Digital Versatile Disk
- Blu-ray Disc etc.
- recording media such as CD-ROM in which each program is stored, and HD 504 in which these programs are stored can be provided domestically or abroad as a program product (Program Product).
- In-vehicle device control system (an example of "mobile device control system) 100 Self-vehicle 101 Imaging unit 102 Image analysis unit (an example of “image processing apparatus”) 103 Display monitor 104 Vehicle traveling control unit (an example of “control unit”) 110a, 110b imaging unit 120 processing hardware unit 140 short-distance processing unit 141 short-distance parallax image generating unit (an example of “generating unit”) 142 Reduced parallax image generation unit (an example of “reduction unit”) 143 Near object detection unit (an example of “detection unit”) 144 Short range object rejection part (an example of "rejection part”) 145 Short-distance object tracking unit 151 Parallax image generation unit 152 V map generation unit 153 Road surface shape detection unit 154 Road surface height table calculation unit 155 U map generation unit 156 Real U map generation unit 157 Isolated area detection unit 158 Correspondence area detection unit 159 Object region extraction unit 160 Object type classification unit 161 Three-dimensional position determination unit 162 Object tracking unit 163
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
- Image Processing (AREA)
Abstract
Description
図1は、本発明の実施形態に係る移動体機器制御システムとしての車載機器制御システムの構成を示す図である。
図2は、撮像ユニット101及び画像解析ユニット102の構成を示す図である。
ステレオカメラ画像取得部130は、ステレオカメラを構成する2つの撮像部110a,110bから、輝度画像データを取得する。このとき、撮像部110a,110bがカラーの場合には、そのRGB信号から輝度信号(Y)を得るカラー輝度変換を、例えば下記の式〔1〕を用いて行う。
〈近距離処理〉
次に、図5を参照し、近距離処理部140の機能について説明する。図5は、近距離処理部140の機能ブロック図である。近距離処理部140は、近距離(自車両から所定距離内)の物体の検出や追跡を行う。近距離処理部140は、近距離視差画像生成部141、縮小視差画像生成部142、近距離物体検出部143、近距離物体棄却部144、近距離物体追跡部145を有する。
近距離視差画像生成部141は、例えば、処理ハードウェア部120の視差演算部121によって実現してもよい。
次に、縮小視差画像生成部142について説明する。縮小視差画像生成部142は、近距離視差画像生成部141により生成された、自車両から所定距離内に位置する物体に対する視差画像から、所定の間隔で画素を抽出することにより、縮小された(画素数が少ない)縮小視差画像を作成する。例えば1280×640画素の視差画像を、縦及び横に例えば1/10に縮小する場合、縮小視差画像は128×64画素となる。
近距離物体検出部143は、縮小視差画像生成部142により生成された、自車両から所定距離内に位置する物体に対する縮小視差画像に基づいて、物体を検出する。
近距離物体棄却部144は、近距離物体検出部143により検出されたオブジェクトのうち、トラッキングの対象としないオブジェクトのデータを棄却(破棄)する。
近距離物体追跡部145は、近距離物体検出部143により検出され、近距離物体棄却部144に棄却されなかったオブジェクトの追跡を行う。
次に、図10を参照し、図4の通常距離処理部150の機能について説明する。図10は、通常距離処理部150の機能ブロック図である。通常距離処理部150は、自車両から所定距離(例えば2m)以上である通常距離の物体の検出や追跡を行う。
視差画像生成部151は、例えば、処理ハードウェア部120の視差演算部121によって実現してもよい。
視差画像補間処理を行ったら、次に、Vマップ生成部152において、Vマップを生成するVマップ生成処理を実行する。視差画像データに含まれる各視差画素データは、x方向位置とy方向位置と視差値dとの組(x,y,d)で示される。これを、X軸にd、Y軸にy、Z軸に頻度fを設定した三次元座標情報(d,y,f)に変換したもの、又はこの三次元座標情報(d,y,f)から所定の頻度閾値を超える情報に限定した三次元座標情報(d,y,f)を、視差ヒストグラム情報として生成する。本実施形態の視差ヒストグラム情報は、三次元座標情報(d,y,f)からなり、この三次元ヒストグラム情報をX-Yの2次元座標系に分布させたものを、Vマップ(視差ヒストグラムマップ、V-disparity map)と呼ぶ。
次に、本実施形態では、Vマップ生成部152が生成したVマップの情報(視差ヒストグラム情報)から、路面形状検出部153において、自車両100の前方路面の3次元形状を検出する路面形状検出処理が実行される。
次に、路面高さテーブル算出部154において、路面高さ(自車の真下の路面部分に対する相対的な高さ)を算出してテーブル化する路面高さテーブル算出処理を行う。路面形状検出部153により生成されたVマップ上の近似直線の情報から、撮像画像上の各行領域(画像上下方向の各位置)に映し出されている各路面部分までの距離を算出できる。一方、自車の真下に位置する路面部分をその面に平行となるように自車進行方向前方へ延長した仮想平面の自車進行方向における各面部分が、撮像画像中のどの各行領域に映し出されるかは予め決まっており、この仮想平面(基準路面)はVマップ上で直線(基準直線)により表される。路面形状検出部153から出力される近似直線を基準直線と比較することで、自車前方の各路面部分の高さを得ることができる。簡易的には、路面形状検出部153から出力される近似直線上のY軸位置から、これに対応する視差値から求められる距離だけ自車前方に存在する路面部分の高さを算出できる。路面高さテーブル算出部154では、近似直線から得られる各路面部分の高さを、必要な視差範囲についてテーブル化する。
《Uマップ生成処理》
次に、Uマップ生成部155は、Uマップ(U-disparity map)を生成するUマップ生成処理として、頻度Uマップ生成処理及び高さUマップ生成処理を実行する。
次に、リアルUマップ生成部156について説明する。リアルUマップ生成部156では、リアルUマップ(Real U-disparity map)(「分布データ」の一例)を生成するUマップ生成処理として、リアル頻度Uマップ生成処理及びリアル高さUマップ生成処理を実行する。
次に、孤立領域検出部157について説明する。図17は、孤立領域検出部157で行う孤立領域検出処理の流れを示すフローチャートである。孤立領域検出部157では、まずリアルUマップ生成部156で生成されたリアル頻度Uマップの情報の平滑化を行う(ステップS111)。
次に、視差画像の対応領域検出部158及びオブジェクト領域抽出部159について説明する。図18は、孤立領域検出部で検出された孤立領域が内接する矩形領域を設定したリアル頻度Uマップを示す図であり、図19は、図18における矩形領域に対応する走査範囲を設定した視差画像を示す図であり、図20は、図19における走査範囲を探索してオブジェクト領域を設定した視差画像を示す図である。
次に、オブジェクトタイプ分類部160について説明する。
同様に、オブジェクト領域抽出部159で抽出されるオブジェクト領域の幅(xomax-xomin)から、下記の式〔6〕より、そのオブジェクト領域に対応する画像領域に映し出されている識別対象物(オブジェクト)の実際の幅Woを計算できる。
また、当該オブジェクト領域に対応する画像領域に映し出されている識別対象物(オブジェクト)の奥行きDoは、当該オブジェクト領域に対応した孤立領域内の最大視差dmaxと最小視差dminから、下記の式〔7〕より計算することができる。
オブジェクトタイプ分類部160は、このようにして計算できるオブジェクト領域に対応するオブジェクトの高さ、幅、奥行きの情報から、そのオブジェクトタイプの分類を行う。図22に示す表は、オブジェクトタイプの分類を行うためのテーブルデータの一例を示すものである。これによれば、自車両前方に存在する識別対象物(オブジェクト)が、歩行者なのか、自転車なのか、小型車なのか、トラックなどか等を区別して認識することが可能となる。
次に、3次元位置決定部161について説明する。検出されたオブジェクト領域に対応するオブジェクトまでの距離や、視差画像の画像中心と視差画像上のオブジェクト領域の中心との画像上の距離も把握されることから、オブジェクトの3次元位置を決定することができる。
Yo=Z×(region_centerY-image_centerY)/f …式〔9〕
《オブジェクトトラッキング》
次に、オブジェクトトラッキング部162について説明する。オブジェクトトラッキング部162は、オブジェクトデータリスト170を用いて、以前(過去)の視差画像のフレームから検出されたオブジェクト(物体)をトラッキング(追跡)する処理を実行する。
次に、オブジェクト選択部163について説明する。オブジェクト選択部163は、オブジェクトデータリスト170に保存されているオブジェクトについて、当該オブジェクトが追跡対象とするのにふさわしい位置にいるか等を判断基準にして、当該オブジェクトを追跡するか否かを選択する。具体的には、オブジェクトデータリスト170のオブジェクト予測データに基づいて当該オブジェクトが位置すると予測される視差画像データ中の予測範囲を設定し、その予測範囲内における当該オブジェクトの高さを特定した後、オブジェクトデータリスト170のオブジェクト特徴量を参照して、特定した高さから当該オブジェクトの幅を特定し、特定した幅から視差画像データ上におけるオブジェクトの横方向位置を推定する。このようにして推定したオブジェクトの横方向位置が所定の追跡対象条件(当該オブジェクトが画像内に存在する確度が高い位置である場合、当該オブジェクトが追跡対象とするのにふさわしい位置である場合等)を満たせば、当該オブジェクトを追跡対象として選択する。なお、オブジェクト選択部163は、追跡対象として選択しなかったオブジェクトのデータを、オブジェクトデータリスト170から削除してもよい。
次に、オブジェクトデータリスト170について説明する。オブジェクトデータリスト170は、例えば、一のフレームから検出された各オブジェクトのデータを、それ以降の複数のフレームに渡って追跡するために使用される。
ステレオカメラを用いて生成した視差画像において、自車両(ステレオカメラ)に近い距離に位置する物体に対する視差は大きいため、一方のカメラで撮像した画像における当該物体の画素の位置と、他方のカメラで撮像した画像における当該物体の画素の位置との距離は大きくなる。そのため、一方の画像の一の画素に対応する画素を他方の画像から探索する距離(当該一の画素の位置からの距離)が長くなり、視差画像を作る処理に時間がかかる。
100 自車両
101 撮像ユニット
102 画像解析ユニット(「画像処理装置」の一例)
103 表示モニタ
104 車両走行制御ユニット(「制御部」の一例)
110a,110b 撮像部
120 処理ハードウェア部
140 近距離処理部
141 近距離視差画像生成部(「生成部」の一例)
142 縮小視差画像生成部(「縮小部」の一例)
143 近距離物体検出部(「検出部」の一例)
144 近距離物体棄却部(「棄却部」の一例)
145 近距離物体追跡部
151 視差画像生成部
152 Vマップ生成部
153 路面形状検出部
154 路面高さテーブル算出部
155 Uマップ生成部
156 リアルUマップ生成部
157 孤立領域検出部
158 対応領域検出部
159 オブジェクト領域抽出部
160 オブジェクトタイプ分類部
161 3次元位置決定部
162 オブジェクトトラッキング部
163 オブジェクト選択部
170 オブジェクトデータリスト
2 撮像装置
Claims (8)
- 複数の撮像部で各々撮影されたフレームに応じた、前記複数の撮像部から所定距離内に位置する物体の距離に応じた画素値を有する距離画像の画素数を縮小する縮小部と、
前記縮小された距離画像に基づいて、前記物体を検出する検出部と、
前記複数の撮像部を備える移動体の進行する進行領域における前記物体に応じた距離の値を有する画素の数と、前記進行領域における前記物体の領域に含まれる画素の数に基づいて、前記検出された前記物体のデータを棄却する棄却部と、
を備えることを特徴とする画像処理装置。 - 前記棄却部は、前記移動体の高さ及び幅と、前記物体の前記移動体からの距離に基づいて、前記進行領域を算出することを特徴とする、請求項1記載の画像処理装置。
- 前記検出部は、前記縮小された距離画像における前記物体を含む矩形または正方形の領域を、前記物体の領域として検出することを特徴とする、請求項1または2記載の画像処理装置。
- 前記縮小部は、前記距離画像から所定の間隔で画素を抽出し、当該抽出した画素の値が無効である場合、当該抽出した画素の周辺の画素の値に基づいて、当該抽出した画素の値を決定することを特徴とする、請求項1乃至3のいずれか一項に記載の画像処理装置。
- 複数の撮像部と、
前記複数の撮像部で各々撮影されたフレームに応じた、前記複数の撮像部から所定距離内に位置する物体の視差に応じた画素値を有する距離画像を生成する生成部と、
前記距離画像の画素数を縮小する縮小部と、
前記縮小された距離画像に基づいて、前記物体を検出する検出部と、
前記複数の撮像部を備える移動体の進行する進行領域における前記物体に応じた距離の値を有する画素の数と、前記進行領域における前記物体の領域に含まれる画素の数に基づいて、前記検出された前記物体のデータを棄却する棄却部と、
を備えることを特徴とする撮像装置。 - 移動体に搭載され、前記移動体の前方を撮像する複数の撮像部と、
前記複数の撮像部で各々撮影されたフレームに応じた、前記複数の撮像部から所定距離内に位置する物体の視差に応じた画素値を有する距離画像を生成する生成部と、
前記距離画像の画素数を縮小する縮小部と、
前記縮小された距離画像に基づいて、前記物体を検出する検出部と、
前記複数の撮像部を備える移動体の進行する進行領域における前記物体に応じた距離の値を有する画素の数と、前記進行領域における前記物体の領域に含まれる画素の数に基づいて、前記検出された前記物体のデータを棄却する棄却部と、
前記検出部及び前記棄却部による検出及び棄却結果に基づいて、前記移動体の制御を行う制御部と、
を備える移動体機器制御システム。 - 画像処理装置が、
複数の撮像部で各々撮影されたフレームに応じた、前記複数の撮像部から所定距離内に位置する物体の距離に応じた画素値を有する距離画像の画素数を縮小するステップと、
前記縮小された距離画像に基づいて、前記物体を検出するステップと、
前記複数の撮像部を備える移動体の進行する進行領域における前記物体に応じた距離の値を有する画素の数と、前記進行領域における前記物体の領域に含まれる画素の数に基づいて、前記検出された前記物体のデータを棄却するステップと、
を実行する、画像処理方法。 - 画像処理装置に、
複数の撮像部で各々撮影されたフレームに応じた、前記複数の撮像部から所定距離内に位置する物体の距離に応じた画素値を有する距離画像の画素数を縮小するステップと、
前記縮小された距離画像に基づいて、前記物体を検出するステップと、
前記複数の撮像部を備える移動体の進行する進行領域における前記物体に応じた距離の値を有する画素の数と、前記進行領域における前記物体の領域に含まれる画素の数に基づいて、前記検出された前記物体のデータを棄却するステップと、
を実行させるプログラム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2018501055A JP6705496B2 (ja) | 2016-02-23 | 2017-01-20 | 画像処理装置、撮像装置、移動体機器制御システム、移動体、画像処理方法、及びプログラム |
EP17756042.2A EP3422289A4 (en) | 2016-02-23 | 2017-01-20 | IMAGE PROCESSING DEVICE, IMAGING DEVICE, MOBILE ENTITY DEVICE CONTROL SYSTEM, IMAGE PROCESSING METHOD, AND PROGRAM |
US16/105,529 US10860867B2 (en) | 2016-02-23 | 2018-08-20 | Image processing apparatus, imaging apparatus, mobile device control system, and recording medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016032118 | 2016-02-23 | ||
JP2016-032118 | 2016-02-23 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US16/105,529 Continuation US10860867B2 (en) | 2016-02-23 | 2018-08-20 | Image processing apparatus, imaging apparatus, mobile device control system, and recording medium |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017145600A1 true WO2017145600A1 (ja) | 2017-08-31 |
Family
ID=59686251
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2017/001914 WO2017145600A1 (ja) | 2016-02-23 | 2017-01-20 | 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム |
Country Status (4)
Country | Link |
---|---|
US (1) | US10860867B2 (ja) |
EP (1) | EP3422289A4 (ja) |
JP (1) | JP6705496B2 (ja) |
WO (1) | WO2017145600A1 (ja) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021054341A1 (ja) * | 2019-09-19 | 2021-03-25 | 京セラ株式会社 | 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 |
JP2021051476A (ja) * | 2019-09-24 | 2021-04-01 | 京セラ株式会社 | 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 |
JP2021086265A (ja) * | 2019-11-26 | 2021-06-03 | 株式会社Subaru | 画像処理装置 |
JP7138265B1 (ja) * | 2021-06-17 | 2022-09-15 | ヌヴォトンテクノロジージャパン株式会社 | 物体検知装置、物体検知方法およびプログラム |
WO2022264894A1 (ja) * | 2021-06-17 | 2022-12-22 | ヌヴォトンテクノロジージャパン株式会社 | 物体検知装置、物体検知方法およびプログラム |
WO2024121911A1 (ja) * | 2022-12-05 | 2024-06-13 | 日立Astemo株式会社 | 画像処理装置 |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6743882B2 (ja) * | 2016-03-14 | 2020-08-19 | 株式会社リコー | 画像処理装置、機器制御システム、撮像装置、画像処理方法及びプログラム |
JP6795027B2 (ja) * | 2016-03-15 | 2020-12-02 | 株式会社リコー | 情報処理装置、物体認識装置、機器制御システム、移動体、画像処理方法およびプログラム |
JP6904346B2 (ja) * | 2016-06-22 | 2021-07-14 | ソニーグループ株式会社 | 画像処理装置、画像処理システム、および画像処理方法、並びにプログラム |
US10347030B2 (en) | 2017-05-15 | 2019-07-09 | Envisics Ltd | Adjusting depth of augmented reality content on a heads up display |
US10929986B2 (en) * | 2018-12-19 | 2021-02-23 | Fca Us Llc | Techniques for using a simple neural network model and standard camera for image detection in autonomous driving |
JP2020160914A (ja) * | 2019-03-27 | 2020-10-01 | 株式会社豊田自動織機 | 物体検出装置 |
CN111192283A (zh) * | 2019-12-26 | 2020-05-22 | 昆山伟宇慧创智能科技有限公司 | 一种限高杆检测和高度计算方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001319224A (ja) * | 2000-05-08 | 2001-11-16 | Fujitsu Ltd | 3次元物体検出装置と3次元物体検出方法及び記録媒体 |
US20070071311A1 (en) * | 2005-09-28 | 2007-03-29 | Deere & Company, A Delaware Corporation | Method for processing stereo vision data using image density |
JP2011185664A (ja) * | 2010-03-05 | 2011-09-22 | Panasonic Electric Works Co Ltd | 対象物検出装置 |
JP2013250907A (ja) * | 2012-06-04 | 2013-12-12 | Ricoh Co Ltd | 視差算出装置、視差算出方法及び視差算出用プログラム |
JP2015170306A (ja) * | 2014-03-10 | 2015-09-28 | サクサ株式会社 | 画像処理装置 |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1139596A (ja) | 1997-07-17 | 1999-02-12 | Fuji Heavy Ind Ltd | 車外監視装置 |
JP4624594B2 (ja) * | 2000-06-28 | 2011-02-02 | パナソニック株式会社 | 物体認識方法および物体認識装置 |
US7068815B2 (en) * | 2003-06-13 | 2006-06-27 | Sarnoff Corporation | Method and apparatus for ground detection and removal in vision systems |
EP1779295A4 (en) * | 2004-07-26 | 2012-07-04 | Automotive Systems Lab | SYSTEM FOR PROTECTING USERS OF THE ROAD IN A DANGER SITUATION |
JP4900103B2 (ja) * | 2007-07-18 | 2012-03-21 | マツダ株式会社 | 歩行者検出装置 |
US8774512B2 (en) * | 2009-02-11 | 2014-07-08 | Thomson Licensing | Filling holes in depth maps |
US8861842B2 (en) * | 2010-02-05 | 2014-10-14 | Sri International | Method and apparatus for real-time pedestrian detection for urban driving |
US9053562B1 (en) * | 2010-06-24 | 2015-06-09 | Gregory S. Rabin | Two dimensional to three dimensional moving image converter |
US8824733B2 (en) * | 2012-03-26 | 2014-09-02 | Tk Holdings Inc. | Range-cued object segmentation system and method |
US8768007B2 (en) * | 2012-03-26 | 2014-07-01 | Tk Holdings Inc. | Method of filtering an image |
EP2669846B1 (en) | 2012-06-01 | 2017-11-01 | Ricoh Company, Ltd. | Target recognition system and target recognition method executed by the target recognition system, target recognition program executed on the target recognition system, and recording medium storing the target recognition program |
EP2946336B1 (en) * | 2013-01-15 | 2023-06-21 | Mobileye Vision Technologies Ltd. | Stereo assist with rolling shutters |
JP6519262B2 (ja) * | 2014-04-10 | 2019-05-29 | 株式会社リコー | 立体物検出装置、立体物検出方法、立体物検出プログラム、及び移動体機器制御システム |
US10079827B2 (en) | 2015-03-16 | 2018-09-18 | Ricoh Company, Ltd. | Information processing apparatus, information processing method, and information processing system |
US9727793B2 (en) * | 2015-12-15 | 2017-08-08 | Honda Motor Co., Ltd. | System and method for image based vehicle localization |
JP6583527B2 (ja) | 2016-02-22 | 2019-10-02 | 株式会社リコー | 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム |
US10638109B2 (en) * | 2017-09-15 | 2020-04-28 | Elphel, Inc. | Method for the FPGA-based long range multi-view stereo with differential image rectification |
-
2017
- 2017-01-20 JP JP2018501055A patent/JP6705496B2/ja active Active
- 2017-01-20 WO PCT/JP2017/001914 patent/WO2017145600A1/ja active Application Filing
- 2017-01-20 EP EP17756042.2A patent/EP3422289A4/en active Pending
-
2018
- 2018-08-20 US US16/105,529 patent/US10860867B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001319224A (ja) * | 2000-05-08 | 2001-11-16 | Fujitsu Ltd | 3次元物体検出装置と3次元物体検出方法及び記録媒体 |
US20070071311A1 (en) * | 2005-09-28 | 2007-03-29 | Deere & Company, A Delaware Corporation | Method for processing stereo vision data using image density |
JP2011185664A (ja) * | 2010-03-05 | 2011-09-22 | Panasonic Electric Works Co Ltd | 対象物検出装置 |
JP2013250907A (ja) * | 2012-06-04 | 2013-12-12 | Ricoh Co Ltd | 視差算出装置、視差算出方法及び視差算出用プログラム |
JP2015170306A (ja) * | 2014-03-10 | 2015-09-28 | サクサ株式会社 | 画像処理装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3422289A4 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021054341A1 (ja) * | 2019-09-19 | 2021-03-25 | 京セラ株式会社 | 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 |
JP2021047741A (ja) * | 2019-09-19 | 2021-03-25 | 京セラ株式会社 | 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 |
JP7308112B2 (ja) | 2019-09-19 | 2023-07-13 | 京セラ株式会社 | 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 |
JP2021051476A (ja) * | 2019-09-24 | 2021-04-01 | 京セラ株式会社 | 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 |
WO2021060117A1 (ja) * | 2019-09-24 | 2021-04-01 | 京セラ株式会社 | 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 |
JP7308115B2 (ja) | 2019-09-24 | 2023-07-13 | 京セラ株式会社 | 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 |
JP2021086265A (ja) * | 2019-11-26 | 2021-06-03 | 株式会社Subaru | 画像処理装置 |
JP7406962B2 (ja) | 2019-11-26 | 2023-12-28 | 株式会社Subaru | 画像処理装置 |
JP7138265B1 (ja) * | 2021-06-17 | 2022-09-15 | ヌヴォトンテクノロジージャパン株式会社 | 物体検知装置、物体検知方法およびプログラム |
WO2022264894A1 (ja) * | 2021-06-17 | 2022-12-22 | ヌヴォトンテクノロジージャパン株式会社 | 物体検知装置、物体検知方法およびプログラム |
WO2024121911A1 (ja) * | 2022-12-05 | 2024-06-13 | 日立Astemo株式会社 | 画像処理装置 |
Also Published As
Publication number | Publication date |
---|---|
US20180357495A1 (en) | 2018-12-13 |
JPWO2017145600A1 (ja) | 2018-12-06 |
EP3422289A1 (en) | 2019-01-02 |
EP3422289A4 (en) | 2019-02-27 |
JP6705496B2 (ja) | 2020-06-03 |
US10860867B2 (en) | 2020-12-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017145600A1 (ja) | 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム | |
US10984509B2 (en) | Image processing apparatus, imaging device, moving body device control system, image information processing method, and program product | |
JP6657789B2 (ja) | 画像処理装置、撮像装置、機器制御システム、頻度分布画像生成方法、及びプログラム | |
JP6662388B2 (ja) | 画像処理装置、撮像装置、機器制御システム、分布データ生成方法、及びプログラム | |
JP6583527B2 (ja) | 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム | |
US11004215B2 (en) | Image processing apparatus, imaging device, moving body device control system, image information processing method, and program product | |
WO2017145634A1 (ja) | 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム | |
JP6769477B2 (ja) | 画像処理装置、撮像装置、移動体機器制御システム、画像処理方法、及びプログラム | |
US10628960B2 (en) | Information processing apparatus, imaging apparatus, device control system, moving object, information processing method, and recording medium | |
JP2018088233A (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、及びプログラム | |
JP6812701B2 (ja) | 画像処理装置、移動体機器制御システム、画像処理方法、及びプログラム | |
JP2018088234A (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、及びプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2018501055 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2017756042 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2017756042 Country of ref document: EP Effective date: 20180924 |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17756042 Country of ref document: EP Kind code of ref document: A1 |