# US 2018/0324406 A1 · Three-dimensional object detection apparatus, stereo camera apparatus, vehicle, and three-dimensional object detection method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0324406 A1 - Document ID: US20180324406A1 - Status: Published application - Kind: A1 - Application number: 15/739,481 - Filed: 2016-06-23 - Published: 2018-11-08 - Earliest claimed: 2015-06-23 - Assignees: KYOCERA Corporation - Applicants: KYOCERA Corporation - Inventors: Tomofumi KOISHI - Web page: https://patentgrep.com/patent/US20180324406A1 ## CPC - H04N 13/225: using parallax barriers - G01C 3/14: with binocular observation at a single point, e.g. stereoscopic type (G01C3/20 takes precedence) - G01C 11/06: by comparison of two or more pictures of the same area - G01S 11/12: using electromagnetic waves other than radio waves - G01V 8/10: Detecting, e.g. by using light barriers (by reflection from the object G01S17/00) - G06T 1/00: General purpose image data processing - G06T 7/00: Image analysis - G06V 20/58: Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads - G06V 20/584: {of vehicle lights or traffic lights} - H04N 13/275: from three-dimensional [3D] object models, e.g. computer-generated stereoscopic image signals - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) - G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - G01V: GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS (means for indicating the location of accidentally buried, e.g. snow-buried, persons A63B29/02) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A three-dimensional object detection apparatus includes an input interface and a controller. The input interface accepts input of first and second images having mutual parallax in a first direction. The controller generates a parallax image that uses a pixel value to represent a difference in the first direction between positions of a subject in the first image and in the second image and divides the parallax image into a plurality of partial regions extending along a second direction. For each partial region, the controller segments a distance in the direction towards the subject into a plurality of distance intervals; and calculates, on the basis of a pixel count of pixels belonging to the distance interval, an evaluation value of a three-dimensional object belonging to the distance interval. The controller uses a different threshold for each distance interval to determine whether or not a three-dimensional object is present. ## Claims ### Claim 1 A three-dimensional object detection apparatus comprising: an input interface; and a controller, wherein the input interface is configured to accept input of a first image and a second image having mutual parallax in a first direction, the controller is configured to generate a parallax image that uses a pixel value to represent a difference in the first direction between a position of a subject in the first image and a position of the subject in the second image, and divide at least a portion of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction, the controller is configured, for each partial region, to segment a distance in a direction towards the subject into a plurality of distance intervals, calculate, from the pixel value, the distance interval to which each pixel of the partial region belongs, calculate, on the basis of a pixel count of pixels belonging to the distance interval, an evaluation value corresponding to a length in the second direction of a three-dimensional object belonging to the distance interval, and determine whether a three-dimensional object is present in each distance interval on the basis of the evaluation value, and the controller is configured to use a different threshold for each distance interval to determine that a three-dimensional object is present when the evaluation value is greater than the threshold. ### Claim 2 (depends on 1) The three-dimensional object detection apparatus of claim 1, wherein the threshold becomes a smaller value with increasing distance in the direction towards the subject. ### Claim 3 (depends on 1) The three-dimensional object detection apparatus of claim 1, wherein the evaluation value is calculated by dividing the pixel count of each distance interval by a pixel count of the partial region in the first direction, and the threshold satisfies a conditional expression, f · ( tan [ tan - 1 ( h 0 Z 1 ) - θ 0 ] - tan [ tan - 1 ( h 0 Z 2 ) - θ 0 ] )  /  P < Th < f · Y min Z 3  /  P ( 1 ) where Th is the threshold, f is a focal length of an optical system that captured the first image and the second image, θ0 is the angle of depression of the optical system that captured the first image and the second image, h0 is the attachment height of the optical system that captured the first image and the second image, Z1 is a near-side boundary and Z2 is a far-side boundary of a distance interval of a determination target, Z3 is a deemed distance to a predetermined three-dimensional object for the distance interval of the determination target, Ymin is a minimum height of a three-dimensional object to be detected, and P is a pixel pitch of an image. ### Claim 4 (depends on 3) The three-dimensional object detection apparatus of claim 3, wherein the conditional expression is applied at a distance equal to or greater than a distance at which a three-dimensional object with a minimum height targeted for determination is entirely included in the first image and the second image. ### Claim 5 A stereo camera apparatus comprising: a stereo camera; and a controller, wherein the stereo camera is configured to output a first image and a second image having mutual parallax in a first direction, the controller is configured to generate a parallax image that uses a pixel value to represent a difference in the first direction between a position of a subject in the first image and a position of the subject in the second image, and divide at least a portion of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction, the controller is configured, for each partial region, to segment a distance in a direction towards the subject into a plurality of distance intervals, calculate, from the pixel value, the distance interval to which each pixel of the partial region belongs, calculate, on the basis of a pixel count of pixels belonging to the distance interval, an evaluation value corresponding to a length in the second direction of a three-dimensional object belonging to the distance interval, and determine whether a three-dimensional object is present in each distance interval on the basis of the evaluation value, and the controller is configured to use a different threshold for each distance interval to determine that a three-dimensional object is present when the evaluation value is greater than the threshold. ### Claim 6 A vehicle comprising: a stereo camera; and a controller, wherein the stereo camera is configured to output a first image and a second image having mutual parallax in a first direction, the controller is configured to generate a parallax image that uses a pixel value to represent a difference in the first direction between a position of a subject in the first image and a position of the subject in the second image, and divide at least a portion of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction, the controller is configured, for each partial region, to segment a distance in a direction towards the subject into a plurality of distance intervals, calculate, from the pixel value, the distance interval to which each pixel of the partial region belongs, calculate, on the basis of a pixel count of pixels belonging to the distance interval, an evaluation value corresponding to a length in the second direction of a three-dimensional object belonging to the distance interval, and determine whether a three-dimensional object is present in each distance interval on the basis of the evaluation value, the controller is configured to use a different threshold for each distance interval to determine that a three-dimensional object is present when the evaluation value is greater than the threshold, and the stereo camera is configured to image an external environment of the vehicle. ### Claim 7 A three-dimensional object detection method comprising: accepting, from a stereo camera, input of a first image and a second image having mutual parallax in a first direction; generating a parallax image that uses a pixel value to represent a difference in the first direction between a position of a subject in the first image and a position of the subject in the second image; and dividing at least a portion of the parallax image into a plurality of partial regions extending along a second direction intersecting the first direction; for each partial region, segmenting a distance in a direction towards the subject into a plurality of distance intervals; calculating, from the pixel value, the distance interval to which each pixel of the partial region belongs; and calculating, on the basis of a pixel count of pixels belonging to the distance interval, an evaluation value corresponding to a length in the second direction of a three-dimensional object belonging to the distance interval; and using a different threshold for each distance interval, and determining that a three-dimensional object is present when the evaluation value is greater than the threshold.