Vehicle surroundings monitoring apparatus

• US 20030099377A1
• Filed: 12/20/2001
• Published: 05/29/2003
• Est. Priority Date: 01/30/1998
• Status: Active Grant

• Current Assignee: Subaru Corp.
• Original Assignee: Fuji Heavy Industries Limited (Subaru Corp.)
• Inventors: Hanawa, Keiji
• Granted Patent: US 6,888,953 B2
• Time in Patent Office: Days
• US Class Current: 382/104
• CPC Class Codes: G01S 11/12 using electromagnetic waves other than radio waves
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 2013/932 using own vehicle data, e.g. ground speed, steering wheel direction
• CPC Class Codes: G06T 17/10 Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 7/593 from stereo images
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: 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
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: H04N 13/239 using two 2D image sensors having a relative position equal to or related to the interocular distance H04N13/243 takes precedence
• CPC Class Codes: H04N 2013/0081 Depth or disparity estimation from stereoscopic image signals

Abstract

Distance image outputted from an image processor is read by a recognition/judgment computer, in which the distance image is divided into strips having a specified interval and a histogram is prepared for each strip. Based on the histogram, the position of solid objects and the distance thereto are obtained and these positional data of the solid objects are classified into groups of solid object and groups of side wall. Further, the position of a wall surface forming a boundary of a road is detected from the data of the groups of side wall. On the other hand, assuming a wall surface model composed of a series of nodes, respective nodes are determined and corrected by the data of side wall. Thus obtained nodes form a successive wall surface along a curved road.

Claims

1. A vehicle surroundings monitoring apparatus having a stereoscopic image detecting unit for detecting a stereoscopic image of solid objects around a self vehicle, an image processor for processing said image into a distance image and a plurality of micro-processors based on said distance image for recognizing said solid objects, comprising:

a wall surface detecting means for dividing positional data of said solid objects into groups and for detecting a wall surface formed along a boundary of a road based on said grouped positional data of said solid objects;

a wall surface model forming means for interconnecting a plurality of nodes and for forming a wall surface model based on said interconnected nodes to express an outline of said wall surface; and

a wall surface model correcting means for correcting said wall surface model based on said grouped positional data of said solid objects.

2. The apparatus according to claim 1, wherein

said wall surface model correcting means comprises a means for applying a pattern matching to said grouped positional data to search a position of said wall surface corresponding to said respective nodes and a means for correcting coordinates of said nodes based on said position of said wall surface.

3. The apparatus according to claim 2, wherein

said pattern matching uses a wall surface pattern having such weight coefficient as becoming large in the outside direction thereof.

4. The apparatus according to claim 2, wherein

said wall surface model correcting means comprises a means for correcting said coordinates of said corresponding nodes in the direction where said positional data of said solid objects partially exist, when said pattern matching detects no wall surface.

5. The apparatus according to claim 1, wherein

said wall surface model correcting means comprises a means for correcting said coordinates of said respective nodes in the direction of bringing them close to a straight line connecting one adjacent node and the other adjacent node after said wall surface model is corrected based on said grouped positional data of said solid objects.