WO2023175708A1 - 外界認識装置、および、外界認識方法 - Google Patents
外界認識装置、および、外界認識方法 Download PDFInfo
- Publication number
- WO2023175708A1 WO2023175708A1 PCT/JP2022/011524 JP2022011524W WO2023175708A1 WO 2023175708 A1 WO2023175708 A1 WO 2023175708A1 JP 2022011524 W JP2022011524 W JP 2022011524W WO 2023175708 A1 WO2023175708 A1 WO 2023175708A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- distance
- distance measurement
- external world
- recognition device
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/122—Improving the three-dimensional [3D] impression of stereoscopic images by modifying image signal contents, e.g. by filtering or adding monoscopic depth cues
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/02—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness
- G01B21/04—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness by measuring coordinates of points
- G01B21/045—Correction of measurements
-
- 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
- G01C3/02—Details
- G01C3/06—Use of electric means to obtain final indication
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/60—Extraction of image or video features relating to illumination properties, e.g. using a reflectance or lighting model
-
- 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
Definitions
- the present invention relates to an external world recognition device and an external world recognition method that improve the distance measurement accuracy of a stereo camera.
- An in-vehicle stereo camera device which is a type of in-vehicle camera device, can measure visual information from images and distance information to the object at the same time, so it can measure various objects around the car (people, cars, three-dimensional objects, etc.). It is a device that can provide a detailed understanding of white lines, road surfaces, road signs, etc., and contributes to improving safe driving performance.
- one of the characteristics of a stereo camera is that it has high spatial resolution and accuracy when measuring a distance to an object. This is because the distance to any object can be measured based on triangulation.
- stereo cameras have the problem that distance measurement performance is affected by the characteristics of the optical system and imaging system (lens, image sensor, image processing) because they measure distance with sub-pixel level resolution using triangulation. be.
- aiming In order to improve this problem, various techniques and devices have been proposed to correct distance measurement errors of stereo cameras (generally referred to as "aiming” or “aiming”), and one example is Patent Document 1. .
- Patent Document 1 As described above, the problem of Patent Document 1 is to reduce the influence of a transparent body (specifically, a windshield), which is one of the characteristics of the optical system, on parallax calculation. Therefore, this patent document describes a method in which images are captured twice, with and without glass, to obtain calibration parameters, and aiming processing is performed.
- a transparent body specifically, a windshield
- this patent document describes a method in which images are captured twice, with and without glass, to obtain calibration parameters, and aiming processing is performed.
- Patent Document 2 states, ⁇ Parallax detection device (3) that can suppress parallax detection errors even when a camera whose lens resolution is low relative to the pixel pitch of the image sensor is used.
- the PSF equalization unit (5) corrects at least one of the plurality of images obtained from each of the plurality of optical systems, and the image corrected by the PSF identification unit (5) is used to and a parallax calculation unit (9) that calculates the parallax occurring between the optical systems.”
- the light point spread function PSF which is the response characteristic of the left and right lenses
- PSF Point Spread Function
- Patent Documents 1 and 2 described above are uniformly applied to all captured images, uniformly improve parallax calculation, and uniformly reduce distance measurement errors.
- An object of the present invention is to provide a recognition device and an external world recognition method.
- a parallax calculation unit that calculates parallax from images captured by an in-vehicle camera device having a plurality of cameras; a distance measurement unit that calculates a distance to an object outside the vehicle from the parallax; and response characteristics of lenses of each of the plurality of cameras. and a storage unit in which brightness characteristics are recorded; a distance measurement correction necessity determining unit that uses the response characteristics of the lens and the brightness characteristics to determine whether or not it is necessary to correct the distance to the object; an external world recognition device, comprising: a distance measurement correction section that corrects a distance to the object based on the response characteristic of the lens and the brightness characteristic when it is determined that the above-mentioned brightness characteristic is necessary.
- the external world recognition device and the external world recognition method of the present invention by focusing on the characteristics (individual differences) of the optical system and imaging system of the in-vehicle stereo camera device, measurement due to temporal changes in the optical system and imaging system can be performed. The influence on distance error can be reduced. Note that problems, configurations, and effects other than those described above will be made clear by the description of the following examples.
- FIG. 1 is a schematic configuration diagram of an in-vehicle stereo camera device according to an embodiment.
- FIG. 1 is a functional block diagram of an in-vehicle stereo camera system according to an embodiment.
- Processing flowchart when manufacturing an in-vehicle stereo camera. Processing flowchart during driving. Flowchart of distance measurement arbitration.
- FIG. 1 is a block diagram showing a schematic hardware configuration of an in-vehicle stereo camera device 10 according to this embodiment.
- the in-vehicle stereo camera device 10 is a device that is installed in a vehicle as part of an in-vehicle system, and is a device that recognizes the environment outside the vehicle based on image information of a region to be photographed in front of the vehicle.
- the environment outside the vehicle that this device recognizes includes, for example, white lines on roads, pedestrians, vehicles, other three-dimensional objects, traffic lights, signs, illuminated lamps, and the distance to them.
- the drive system, braking system, steering system, etc. of the own vehicle are controlled according to the external environment recognized by the system 10, thereby realizing desired automatic driving control, driving support control, etc.
- the in-vehicle stereo camera device 10 includes a left camera 11, a right camera 12, an image input interface 13, an image processing section 14, an arithmetic processing section 15, a storage section 16, a control processing section 17, a CAN interface 18, It has an internal bus 19.
- the parts of the in-vehicle stereo camera device 10 excluding the left camera 11 and the right camera 12 are specifically a computer equipped with hardware such as an arithmetic unit such as a CPU, a storage device such as a semiconductor memory, and a communication device. It is.
- the arithmetic device executes a predetermined program to realize the functions of the image processing section 14, the arithmetic processing section 15, and the control processing section 17, but in the following, such well-known techniques will be omitted as appropriate. The details of each part will be explained.
- the left camera 11 and the right camera 12 are a pair of left and right cameras installed at the upper part of the inner surface of the windshield of the vehicle so as to capture a pair of left and right images P (left image PL , right image PR ) in front of the vehicle. camera.
- the image input interface 13 is an interface that controls imaging by both cameras and captures the captured image P.
- the image P captured through the image input interface 13 is transmitted to the image processing section 14 and the arithmetic processing section 15 via the internal bus 19.
- the image processing unit 14 compares the left image P L from the left camera 11 and the right image P R from the right camera 12, and corrects device-specific deviations caused by the image sensor for each image. , after performing image correction such as noise interpolation, the corrected left image P L and right image PR are stored in the storage unit 16 . Further, the image processing unit 14 identifies mutually corresponding locations between the corrected left and right images, calculates parallax information, and stores the calculated parallax information in the storage unit 16.
- the arithmetic processing unit 15 uses the corrected image P and parallax information (distance information for each point on the image) stored in the storage unit 16 to recognize various objects necessary for perceiving the environment around the vehicle. and stores it in the storage unit 16.
- the various objects recognized by the arithmetic processing unit 15 include people, cars, other obstacles, traffic lights, signs, tail lamps and headlights of cars, and the like.
- the control processing unit 17 uses the information on the various objects recognized by the arithmetic processing unit 15 to calculate a vehicle control policy necessary to realize desired automatic driving control, driving support control, etc. Further, the control processing unit 17 monitors whether each processing unit is operating abnormally or not, and whether an error has occurred during data transfer, and is designed to prevent abnormal operations.
- the CAN interface 18 is an interface that transmits the vehicle control policy obtained as the calculation result of the control processing section 17 and a part of the object recognition result of the calculation processing section 15 to the in-vehicle network CAN.
- the internal bus 19 is a bus that connects the image input interface 13, image processing section 14, arithmetic processing section 15, storage section 16, control processing section 17, and CAN interface 18 described above.
- FIG. 2 is a diagram representing the in-vehicle stereo camera device 10 of FIG. 1 as a functional block.
- the in-vehicle stereo camera device 10 includes a left camera 11, a right camera 12, an image correction section 1, a parallax calculation section 2, an object detection/distance measurement section 3, an object recognition section 4, a recognition dictionary 5, and a measurement dictionary. It includes a distance adjustment section 6, an optical characteristic database 7, and a vehicle control section 8.
- the image correction unit 1 and the parallax calculation unit 2 in FIG. 2 are functional units realized by the image processing unit 14 in FIG.
- the detection/distance measurement unit 3, the object recognition unit 4, and the distance measurement arbitration unit 6 are functional units realized by the arithmetic processing unit 15 in FIG. 1, and the vehicle control unit 8 in FIG. 2 is the control processing unit in FIG. 17, and the recognition dictionary 5 and optical property database 7 in FIG. 2 are data stored in advance in the storage unit 16 in FIG.
- the image correction unit 1 performs a non-linear process on the left image P L captured by the left camera 11 and the right image P R captured by the right camera 12 in order to absorb the inherent peculiarities of the image sensor and lens of each camera. Performs image correction such as image position mapping and brightness correction.
- the parallax calculation unit 2 performs stereo viewing, that is, parallax calculation processing for each pixel or each local area on the images, on the left image P L and the right image PR corrected by the image correction unit 1 . As a result, the parallax calculation unit 2 creates a stereo image. Details of this parallax calculation processing will be described later.
- the object detection/distance measurement unit 3 cuts out a predetermined portion of the stereo image generated by the parallax calculation unit 2, extracts clusters that are considered to be clusters of objects located at the same position from the distribution of parallax values, and extracts clusters that are considered to be clusters of objects located at the same position. Measure the distance to the block.
- the object recognition unit 4 uses image recognition to determine whether the object is a pedestrian, bicycle, motorcycle, or vehicle, and if it is nighttime, a light spot (for example, a tail lamp of a preceding vehicle). performs object recognition such as whether the vehicle is a vehicle or brake light). For object recognition in the object recognition unit 4, a well-known pattern matching determination is performed using a recognition dictionary 5 recorded in advance.
- the distance measurement arbitration unit 6 executes distance measurement arbitration processing depending on the target object. Note that the distance measurement arbitration is performed with reference to the optical characteristic database 7 that indicates the individual characteristics of the left camera 11 and the right camera 12 incorporated in the in-vehicle stereo camera device. Details of this distance measurement arbitration process will be described later.
- the vehicle control unit 8 determines the control content of the vehicle equipped with the vehicle-mounted stereo camera device 10. For example, a warning is issued to the occupants, and vehicle braking such as acceleration, deceleration, and steering of the own vehicle is determined.
- FIG. 3 is a physical model that simplifies the process of parallax calculation in the vehicle-mounted stereo camera device 10, and describes the mechanism of the present invention.
- objects in the external world can be considered to be composed of a group of point light sources, and when a group of point light sources is imaged on an image sensor, the There will be two processing steps.
- the first processing step is a processing step using a lens.
- Lens characteristics include response characteristics such as a light spot spread function PSF.
- the light spot spread function PSF is a response characteristic that can define how a point light source spreads on the image sensor surface (CMOS surface), as illustrated in FIG. 4, and is defined by this light spot spread function PSF.
- the shape determines the amount of light projected from the point light source onto the image sensor surface.
- the second processing step is a processing step in the image sensor.
- the light projected onto the image sensor surface is first converted into electronic intensity according to its intensity (number of photons).
- the information on the amount of light received electronically stored in each image sensor is converted into an analog/digital signal strength on an electrical signal, and finally the information is sent to the left and right cameras.
- This is a captured image, that is, a collection of two-dimensional brightness values.
- the processing process in this image sensor is a process in which light intensity information discretized in a lattice pattern is converted into brightness values on the image by focusing on a certain dynamic range, which represents the characteristics here.
- a brightness conversion table (converted from FIG. 5 into table-like data) that defines the relationship between the amount of received light and image brightness plays an important role.
- the parallax calculation unit 2 performs parallax calculation on the left image P L and right image PR obtained through these two processing steps.
- the lens characteristics that affect the first processing process and the brightness conversion characteristics that affect the second processing process change depending on factors such as the passage of time, environmental temperature, and the outside world, and are not always constant. Therefore, an error may occur in the parallax calculation by the parallax calculation unit 2 due to changes in lens characteristics or brightness conversion characteristics.
- the difference between distance measurement arbitration by the distance measurement arbitration unit 6 and conventional automatic adjustment and aiming is that it is not uniformly applied to all captured images, but is applied locally (specific objects, specific imaging environments, etc.). It is applied to specific driving scenes). For example, even if the in-vehicle stereo camera device 10 has the same specifications, distance measurement is good in daytime driving scenes due to individual differences in the optical system and imaging system, but at night the distance measurement is good due to the influence of individual differences in the optical system and imaging system. There are rare individuals with poor ranging performance for brake lights). For such an individual, the aim of the distance measurement mediation of the present invention is to perform distance measurement correction only when distance measurement is performed based on a light spot imaged at night.
- the distance measurement error of the stereo image is comprehensively influenced by the lens structure (first processing step), image processing (second processing step), etc. of the in-vehicle stereo camera device 10. Therefore, it is essential to deal with the nighttime distance measurement error problem by embedding lens-related information in the manufacturing process of the in-vehicle stereo camera device 10 and using a distance measurement correction method that utilizes that information.
- optical characteristic data such as the response characteristics and temperature characteristics of the left and right lenses, and the brightness conversion curve (brightness conversion table) of the image are measured, and the storage unit 16 to be memorized.
- distance measurement correction is necessary depending on the optical characteristic data stored at the time of manufacture and the environment at the time of driving (exposure conditions, temperature, and brightness of the target object). The brightness is determined and correction is performed using brightness conversion.
- step S1 a pair of imaging modules that combine an imaging element, a lens, etc. are assembled.
- step S2 a stereo camera is assembled using the pair of imaging modules assembled in step S1.
- step S3 basic matters such as whether the stereo camera has been assembled correctly and whether the image sensor is functioning properly are inspected.
- step S4 the optical characteristics of the stereo camera are inspected. Specifically, lens characteristics such as the light spot spread function PSF are evaluated for each lens of the left camera 11 and the right camera 12 (see FIG. 4). Furthermore, the brightness characteristics indicating how brightness conversion is performed after imaging are evaluated for each of the image sensors of the left camera 11 and the right camera 12 (see FIG. 5). Furthermore, we will evaluate temperature dependence, which shows how lens characteristics and brightness characteristics are affected by temperature changes.
- lens characteristics such as the light spot spread function PSF are evaluated for each lens of the left camera 11 and the right camera 12 (see FIG. 4).
- the brightness characteristics indicating how brightness conversion is performed after imaging are evaluated for each of the image sensors of the left camera 11 and the right camera 12 (see FIG. 5). Furthermore, we will evaluate temperature dependence, which shows how lens characteristics and brightness characteristics are affected by temperature changes.
- step S5 the information obtained in step S4 is registered in the storage unit 16 as the optical characteristic database 7 for the individual stereo camera.
- the optical property database 7 of this embodiment includes a lens property 7a (see FIG. 4), a temperature property 7b regarding the temperature dependence of the lens property 7a, and a brightness property 7c (see FIG. 5). ) and a temperature characteristic 7d regarding the temperature dependence of the brightness characteristic 7c.
- the image input interface 13 causes the left camera 11 and the right camera 12 to synchronously capture a pair of images P (left image PL , right image PR ), and It is output to the image processing section 14.
- step S12 the image input interface 13 outputs the imaging environment (exposure time, environmental temperature, etc.) at the time of imaging in step S11 to the arithmetic processing unit 15 (distance measurement arbitration unit 6).
- step S13 the image processing unit 14 (image correction unit 1, parallax calculation unit 2) performs desired image correction on the left image PL and the right image PR , and then performs desired image correction on the left image PL and the right image after the correction.
- a parallax image is generated by calculating parallax based on PR .
- step S14 the image processing unit 14 (object detection/distance measurement unit 3) performs object detection and distance measurement to the object based on the corrected left image PL and right image PR .
- step S15 the calculation processing unit 15 (object recognition unit 4, distance measurement arbitration unit 6) performs distance measurement arbitration according to the optical property database 7 prepared in advance at the time of manufacturing and the imaging environment acquired in step S12. . Note that details of this step will be described later.
- step S16 the arithmetic processing unit 15 (object recognition unit 4, distance measurement arbitration unit 6) integrates the type information of the object recognized in step S15 and the distance information to the object.
- step S17 the arithmetic processing unit 15 (object recognition unit 4, distance measurement arbitration unit 6) outputs the information integrated in step S16 to the control processing unit 17 (vehicle control unit 8).
- the control processing unit 17 vehicle control unit 8
- the control processing unit 17 can transmit appropriate information according to the surrounding environment of the own vehicle to the vehicle system, regardless of individual differences between stereo cameras or the imaging environment.
- FIG. 8 is a detailed processing flowchart of step S15 (distance measurement arbitration processing) in FIG.
- step S15a the arithmetic processing unit 15 obtains the lens characteristic 7a (see FIG. 4), the temperature characteristic 7b related to the lens characteristic 7a, and the brightness characteristic 7c (see FIG. 5) from the optical characteristic database 7 prepared in advance at the time of manufacturing.
- a temperature characteristic 7d related to the brightness characteristic 7c is obtained.
- the arithmetic processing unit 15 acquires the exposure time and the environmental temperature from the imaging environment acquired in step S12. Note that since the brighter the object to be imaged, the shorter the exposure time, the brightness of the object to be imaged can be estimated from the exposure time.
- step S15b the arithmetic processing unit 15 determines whether the distance measurement result in step S14 needs to be corrected based on the various information acquired in step s15a. If correction is necessary, the process proceeds to step S15c, and if correction is not necessary, the process of step S15 is carried out.
- the situation in which it is determined that correction of the distance measurement result is necessary is, for example, the following. (1) When the brightness of the object to be imaged is greater than or equal to a predetermined value (when the exposure time during photographing is less than a predetermined value). (2) When the environmental temperature at the time of imaging is equal to or higher than a predetermined value.
- step S15c the arithmetic processing unit 15 corrects the distance measurement result in step S14, taking into account individual differences among the stereo cameras.
- the distance measurement error that may occur is estimated by comparing the characteristics (of the individual stereo camera) with the environment (when driving), and based on this, the distance measurement correction is determined, and the distance measurement correction for the object is performed. It includes processing to perform.
- the measurement in step S14 may be performed in consideration of the lens characteristics 7a and brightness characteristics 7c of the individual stereo camera. Correct the distance result. Moreover, if the case corresponds to the above (2), the distance measurement result in step S14 is corrected in consideration of the temperature characteristics 7b and 7d.
- processing such as performing image filter calculations to equalize the light spot spread function PSF of the left and right lenses may have a low contribution to improving the accuracy of sub-pixel calculations, so the brightness on the image It is effective to perform a parallax calculation after performing a narrowing down calculation (limiting down the exposure in a general camera), or to apply a distance measurement correction table according to the brightness of the object and the temperature characteristics at the time of imaging.
- the present invention records information regarding the characteristics of the optical system and imaging system when manufacturing the stereo camera, and installs the stereo camera in the vehicle.
- distance measurement is performed by comparing the imaging environment such as exposure and temperature when images are taken by the left and right cameras for stereo viewing with the characteristic information recorded at the time of manufacturing the stereo camera. This method is characterized in that distance measurement performance is improved by performing arbitration processing on the results.
- the present invention also provides an in-vehicle stereo camera device that processes images captured by a pair of imaging units, in which the imaging unit of the optical system/imaging system performs parallax calculation from images captured by the two left and right cameras. , determine whether temporal changes in the optical system and imaging system (changes in imaging characteristics due to temperature and imaging conditions) satisfy the conditions for parallax calculation, or if the conditions are not met, consider environmental factors at the time of imaging (changes in imaging characteristics due to temperature and imaging conditions).
- the present invention by focusing on the characteristics (individual differences) of the optical system and imaging system of the in-vehicle stereo camera device, it is possible to reduce the influence on distance measurement errors caused by temporal changes in the optical system and imaging system. can be reduced.
- SYMBOLS 10 Vehicle-mounted stereo camera device, 1... Image correction unit, 2... Parallax calculation unit, 3... Object detection/distance measurement unit, 4... Object recognition unit, 5... Recognition dictionary, 6... Distance measurement arbitration unit, 7... Optical characteristics Database, 7a... Lens characteristics, 7c... Brightness characteristics, 7b, 7d... Temperature characteristics, 8... Vehicle control unit, 11... Left camera, 12... Right camera, 13... Image input interface, 14... Image processing unit, 15... Arithmetic processing unit, 16...Storage unit, 17...Control processing unit, 18...CAN interface, 19...Internal bus
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Electromagnetism (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Measurement Of Optical Distance (AREA)
- Studio Devices (AREA)
Abstract
Description
図1は、本実施例に係る車載ステレオカメラ装置10のハードウェアの概略構成を示すブロック図である。この車載ステレオカメラ装置10は、車載システムの一部として車両に搭載される装置であり、自車両前方の撮影対象領域の画像情報に基づいて車外環境を認識する装置である。この装置が認識する車外環境は、例えば、道路の白線、歩行者、車両、その他の立体物、信号、標識、点灯ランプ、および、それらまでの距離などであり、車載システムは、車載ステレオカメラ装置10が認識した車外環境に応じて、自車両の駆動系、制動系、操舵系などを制御し、所望の自動運転制御、運転支援制御等を実現する。
図2は、図1の車載ステレオカメラ装置10を機能ブロックとして表現した図である。機能ブロックとして表現した場合、車載ステレオカメラ装置10は、左カメラ11、右カメラ12、画像補正部1、視差演算部2、物体検知・測距部3、物体認識部4、認識辞書5、測距調停部6、光学特性データベース7、車両制御部8を備えている。
図3は、車載ステレオカメラ装置10における視差演算の過程を簡易化した物理モデルであり、本発明の機序を語るものである。点光源を使ったインパルス応答的な過程を考えると、外界の物体は点光源群で構成されていると見做すことができ、点光源群が撮像素子上で画像化される場合は、大きく2つの処理過程を経ることになる。
測距調停部6による測距調停が、従来の自動調整やエイミングと違う点は、全撮像画像に対して一律に適用されるのではなく、局所的(特定の対象物、特定の撮像環境、特定の走行シーン)に適用されることにある。例えば、同じ仕様の車載ステレオカメラ装置10であっても、光学系・撮像系の個体差の影響で、昼間走行シーンでの測距は良いが、夜間の光点(例えば、先行車のテールランプやブレーキランプ)に対する測距性能が劣る個体が稀にある。そのような個体であれば、夜間撮像した光点に基づいて測距する場合に限り、測距補正することが本発明の測距調停の狙いである。
ここで、図6のフローチャートに基づいて、車載ステレオカメラ装置10の製造時の処理の詳細を説明する。
次に、図7、図8のフローチャートに基づいて、車載ステレオカメラ装置10を装着した車両が実路を走行する場合に、ステレオカメラ装置10内で実施される処理の詳細を説明する。
(1)撮像対象物の明るさが所定値以上の場合(撮影時の露光時間が所定値未満の場合)。
(2)撮像時の環境温度が所定値以上である場合。
Claims (8)
- 複数のカメラを有する車載カメラ装置で撮像した画像から視差を演算する視差演算部と、
前記視差から車外の対象物までの距離を求める測距部と、
前記複数のカメラの各々のレンズの応答特性および輝度特性が記録された記憶部と、
前記レンズの応答特性および前記輝度特性を用いて、前記対象物までの距離を補正する必要の有無を判定する測距補正要否判定部と、
前記補正が必要と判定された場合に、前記レンズの応答特性および前記輝度特性に基づいて前記対象物までの距離を補正する測距補正部と、
を有することを特徴とする外界認識装置。 - 請求項1に記載の外界認識装置において、
前記レンズの応答特性は温度に依存するデータであって、
前記測距補正要否判定部または前記測距補正部は、前記レンズの応答特性に加えて温度依存性を用いることを特徴とする外界認識装置。 - 請求項1に記載の外界認識装置において、
前記レンズの輝度特性は温度に依存するデータであって、
前記測距補正要否判定部または前記測距補正部は、前記レンズの輝度特性に加えて温度依存性を用いることを特徴とする外界認識装置。 - 請求項1から請求項3の何れか一項に記載の外界認識装置において、
前記レンズの応答特性および輝度特性は、前記車載カメラ装置の製造時に検査され、前記記憶部に記録されたものであることを特徴とする外界認識装置。 - 請求項1に記載の外界認識装置において、
前記測距補正要否判定部は、対象物の明るさが所定値以上の場合に、前記対象物までの距離を補正する必要があると判定することを特徴とする外界認識装置。 - 請求項1に記載の外界認識装置において、
前記測距補正要否判定部は、露光時間が所定値未満の場合に、前記対象物までの距離を補正する必要があると判定することを特徴とする外界認識装置。 - 請求項2または請求項3に記載の外界認識装置において、
前記測距補正要否判定部は、撮像環境の温度が所定値以上の場合に、前記対象物までの距離を補正する必要があると判定することを特徴とする外界認識装置。 - 複数のカメラを有する車載カメラ装置で撮像した画像から視差を演算する視差演算ステップと、
前記視差から車外の対象物までの距離を求める測距ステップと、
記憶部に記憶された前記複数のカメラの各々のレンズの応答特性および輝度特性を用いて、前記対象物までの距離を補正する必要の有無を判定する測距補正要否判定ステップと、
前記補正が必要と判定された場合に、前記レンズの応答特性および前記輝度特性に基づいて前記対象物までの距離を補正する測距補正ステップと、
を有することを特徴とする外界認識方法。
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2024507242A JP7769779B2 (ja) | 2022-03-15 | 2022-03-15 | 外界認識装置、および、外界認識方法 |
| US18/836,304 US12707033B2 (en) | 2022-03-15 | 2022-03-15 | External environment recognition device and external environment recognition method |
| PCT/JP2022/011524 WO2023175708A1 (ja) | 2022-03-15 | 2022-03-15 | 外界認識装置、および、外界認識方法 |
| DE112022006340.8T DE112022006340T5 (de) | 2022-03-15 | 2022-03-15 | Aussenumgebungs-erkennungsvorrichtung und aussenumgebungs-erkennungsverfahren |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/JP2022/011524 WO2023175708A1 (ja) | 2022-03-15 | 2022-03-15 | 外界認識装置、および、外界認識方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023175708A1 true WO2023175708A1 (ja) | 2023-09-21 |
Family
ID=88022918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2022/011524 Ceased WO2023175708A1 (ja) | 2022-03-15 | 2022-03-15 | 外界認識装置、および、外界認識方法 |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US12707033B2 (ja) |
| JP (1) | JP7769779B2 (ja) |
| DE (1) | DE112022006340T5 (ja) |
| WO (1) | WO2023175708A1 (ja) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009017020A (ja) * | 2007-07-02 | 2009-01-22 | Nissan Motor Co Ltd | 画像処理装置及び表示画像生成方法 |
| JP2012044317A (ja) * | 2010-08-16 | 2012-03-01 | Fujitsu Semiconductor Ltd | 画像処理装置、撮像装置、及び画像処理方法 |
| WO2018101055A1 (ja) * | 2016-11-30 | 2018-06-07 | 日立オートモティブシステムズ株式会社 | 画像補正装置および画像補正方法 |
| JP2019070610A (ja) * | 2017-10-11 | 2019-05-09 | キヤノン株式会社 | 測距装置および測距方法 |
Family Cites Families (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007024590A (ja) | 2005-07-13 | 2007-02-01 | Toyota Motor Corp | 物体検出装置 |
| JP2007081806A (ja) | 2005-09-14 | 2007-03-29 | Konica Minolta Holdings Inc | 撮像システム |
| JP4835332B2 (ja) | 2006-09-04 | 2011-12-14 | 住友電気工業株式会社 | 遠赤外線撮像システム及び遠赤外線撮像方法 |
| JP4858263B2 (ja) | 2007-03-28 | 2012-01-18 | 株式会社日立製作所 | 3次元計測装置 |
| JP5172314B2 (ja) * | 2007-12-14 | 2013-03-27 | 日立オートモティブシステムズ株式会社 | ステレオカメラ装置 |
| US8723926B2 (en) | 2009-07-22 | 2014-05-13 | Panasonic Corporation | Parallax detecting apparatus, distance measuring apparatus, and parallax detecting method |
| US8400511B2 (en) | 2009-12-04 | 2013-03-19 | Lockheed Martin Corporation | Optical detection and ranging sensor system for sense and avoid, and related methods |
| JP5545527B2 (ja) | 2010-02-22 | 2014-07-09 | 株式会社リコー | 距離画像取得装置 |
| JP5440364B2 (ja) | 2010-04-28 | 2014-03-12 | コニカミノルタ株式会社 | 画像処理装置、周辺監視システム、および、プログラム |
| JP2012147281A (ja) | 2011-01-13 | 2012-08-02 | Ricoh Co Ltd | 画像処理装置 |
| JP6197291B2 (ja) * | 2012-03-21 | 2017-09-20 | 株式会社リコー | 複眼カメラ装置、及びそれを備えた車両 |
| JP2015148887A (ja) * | 2014-02-05 | 2015-08-20 | 株式会社リコー | 画像処理装置、物体認識装置、移動体機器制御システム及び物体認識用プログラム |
| JP6427900B2 (ja) | 2014-03-07 | 2018-11-28 | 株式会社リコー | 校正方法、校正システム、プログラム及び移動体 |
| JP6453193B2 (ja) | 2015-09-25 | 2019-01-16 | 日立オートモティブシステムズ株式会社 | ステレオカメラ装置 |
| JP6997551B2 (ja) | 2017-07-28 | 2022-01-17 | 日立Astemo株式会社 | 車載環境認識装置 |
| WO2020140049A1 (en) * | 2018-12-28 | 2020-07-02 | Nvidia Corporation | Distance to obstacle detection in autonomous machine applications |
| JP7297606B2 (ja) | 2019-09-04 | 2023-06-26 | 日立Astemo株式会社 | 画像処理装置 |
| JP7247063B2 (ja) | 2019-09-13 | 2023-03-28 | 日立Astemo株式会社 | 画像処理装置、及びそれを用いたステレオカメラ装置 |
| JP7402660B2 (ja) * | 2019-11-01 | 2023-12-21 | キヤノン株式会社 | 頭部装着型表示装置、頭部装着型表示装置の制御方法 |
| JP7371053B2 (ja) * | 2021-03-29 | 2023-10-30 | キヤノン株式会社 | 電子機器、移動体、撮像装置、および電子機器の制御方法、プログラム、記憶媒体 |
-
2022
- 2022-03-15 US US18/836,304 patent/US12707033B2/en active Active
- 2022-03-15 DE DE112022006340.8T patent/DE112022006340T5/de active Pending
- 2022-03-15 JP JP2024507242A patent/JP7769779B2/ja active Active
- 2022-03-15 WO PCT/JP2022/011524 patent/WO2023175708A1/ja not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009017020A (ja) * | 2007-07-02 | 2009-01-22 | Nissan Motor Co Ltd | 画像処理装置及び表示画像生成方法 |
| JP2012044317A (ja) * | 2010-08-16 | 2012-03-01 | Fujitsu Semiconductor Ltd | 画像処理装置、撮像装置、及び画像処理方法 |
| WO2018101055A1 (ja) * | 2016-11-30 | 2018-06-07 | 日立オートモティブシステムズ株式会社 | 画像補正装置および画像補正方法 |
| JP2019070610A (ja) * | 2017-10-11 | 2019-05-09 | キヤノン株式会社 | 測距装置および測距方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| JP7769779B2 (ja) | 2025-11-13 |
| DE112022006340T5 (de) | 2024-10-31 |
| US12707033B2 (en) | 2026-08-11 |
| JPWO2023175708A1 (ja) | 2023-09-21 |
| US20250159122A1 (en) | 2025-05-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10701341B2 (en) | Calibration method, calibration device, and computer program product | |
| CN107122770B (zh) | 多目相机系统、智能驾驶系统、汽车、方法和存储介质 | |
| US20180292201A1 (en) | Calibration apparatus, calibration method, and calibration program | |
| JP2009171122A (ja) | 光源色温度推定方法、ホワイトバランス調整装置、及び画像認識装置 | |
| JP6602982B2 (ja) | 車載カメラ、車載カメラの調整方法、車載カメラシステム | |
| US20160379066A1 (en) | Method and Camera System for Distance Determination of Objects from a Vehicle | |
| EP1069536A2 (en) | Stereo type vehicle monitoring apparatus with fail-safe function | |
| JP2016207030A (ja) | 画像処理装置 | |
| JP2021025868A (ja) | ステレオカメラ | |
| CN114252887A (zh) | 求取环境感测系统运行参数的方法、环境感测系统和控制器 | |
| WO2023067867A1 (ja) | 車載制御装置、および、3次元情報取得方法 | |
| JP2004257837A (ja) | ステレオアダプタ撮像システム | |
| JP6808753B2 (ja) | 画像補正装置および画像補正方法 | |
| US20140055641A1 (en) | System for recognizing surroundings of vehicle | |
| CN100554878C (zh) | 立体光学模块和立体摄像机 | |
| JP2011033594A (ja) | 車両用距離算出装置 | |
| JP7769779B2 (ja) | 外界認識装置、および、外界認識方法 | |
| JP2007293672A (ja) | 車両用撮影装置、車両用撮影装置の汚れ検出方法 | |
| JP7492599B2 (ja) | 車載カメラ装置 | |
| JP2017116321A (ja) | 車載ステレオカメラ用のキャリブレーション装置 | |
| JP7652897B2 (ja) | 画像処理装置、および、画像処理方法 | |
| JP4539400B2 (ja) | ステレオカメラの補正方法、ステレオカメラ補正装置 | |
| JP7818478B2 (ja) | 画像処理装置、および、画像処理方法 | |
| WO2025099908A1 (ja) | 画像処理装置 | |
| WO2026105391A1 (ja) | 画像処理装置及び画像校正方法 |
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: 22931993 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2024507242 Country of ref document: JP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 18836304 Country of ref document: US |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 112022006340 Country of ref document: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 22931993 Country of ref document: EP Kind code of ref document: A1 |
|
| WWP | Wipo information: published in national office |
Ref document number: 18836304 Country of ref document: US |