Obstacle detecting apparatus and method

• US 7,091,837 B2
• Filed: 03/14/2003
• Issued: 08/15/2006
• Est. Priority Date: 03/29/2002
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Nakai, Hiroaki, Hattori, Hiroshi, Onoguchi, Kazunori, Takeda, Nobuyuki
• Primary Examiner(s): Mullen, Jr., Thomas J.
• Assistant Examiner(s): Previl, Daniel
• Application Number:
  US10/387,465
• Publication Number:
  US 20030227378A1
• Time in Patent Office: 1,250 Days
• Field of Search: 340/435, 340/425.5, 340/436, 340/901, 340/902, 340/903, 340/933, 340/937, 382/190, 382/195, 701/301
• US Class Current: 340/435
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06T 7/77 using statistical methods
• CPC Class Codes: G06T 7/97 Determining parameters from multiple pictures depth or shape recovery from multiple images G06T7/55; stereo camera calibration G06T7/85
• 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

Abstract

An obstacle detecting apparatus configured to detect an obstacle present on the road, such as a vehicle running ahead, a parked vehicle, and a pedestrian, by distinguishing the obstacle from a mere object causing no obstruction to driving of the vehicle, such as the texture like the white line or signs on the road surface, and the guardrail along the roadside.

Claims

1. An obstacle detecting apparatus for detecting whether there is an obstacle within a common range shot through the use of a plurality of image pick-up devices, said obstacle detecting apparatus comprising:

image input means for receiving a plurality of images inputted respectively from said plurality of image pick-up devices;

feature calculating means for calculating a predetermined, specific feature quantity in each of said plurality of input images;

feature comparing means for finding a difference in said calculated feature quantity between said plurality of input images; and

result judging means for comparing said found differences in said calculated feature quantity with a threshold to judge a presence or absence of an obstacle when the difference in said feature quantity is smaller than the threshold.

2. The obstacle detecting apparatus according claim 1, wherein said feature calculating means calculates a luminance gradient at each pixel in each image and uses the luminance gradient as said feature quantity.

3. A method of detecting an obstacle for detecting whether there is an obstacle within a common range shot through the use of a plurality of image pick-up devices, said method comprising:

an image inputting step of receiving a plurality of images inputted respectively from said plurality of image pick-up devices;

a feature calculating step of calculating a predetermined, specific feature quantity in each of said plurality of input images;

a feature comparing step of finding a difference in said calculated feature quantity between said plurality of input images; and

a result judging step for comparing said found differences in said calculated feature quantity with a threshold to judge a presence or absence of an obstacle when the difference in said feature quantity is smaller than the threshold.

4. A computer-executable program in a computer readable medium for a method of detecting an obstacle for detecting whether there is an obstacle within a common range shot through the use of a plurality of image pick-up devices, said program causing a computer to perform:

an image inputting function of receiving a plurality of images inputted respectively from said plurality of image pick-up devices;

a feature calculating function of calculating a predetermined, specific feature quantity in each of said plurality of input images;

a feature comparing function of finding a difference in said calculated feature quantity between said plurality of input images; and

result judging function of comparing said found differences in said calculated feature quantity with a threshold to judge a presence or absence of an obstacle when the difference in said feature quantity is smaller than the threshold.

5. An obstacle detecting apparatus for detecting whether there is an obstacle within a common range shot through the use of a plurality of image pick-up devices, said obstacle detecting apparatus performing of:

receiving a plurality of images inputted respectively from said plurality of image pick-up devices;

calculating a predetermined, specific feature quantity in each of said plurality of input images;

finding a difference in said calculated feature quantity between said plurality of input images; and

comparing said found differences in said calculated feature quantity with a threshold to judge a presence or absence of an obstacle when the difference in said feature quantity is smaller than the threshold.

6. An obstacle detecting apparatus for detecting an obstacle within a common viewing range through the use of a plurality of image pick-up devices, comprising:

input unit configured to receive a plurality of images respectively from said plurality of image pick-up devices;

a first calculating unit configured to calculate a predetermined feature data in each of said plurality of input images;

second calculating unit configured to calculate differences between a pair of the feature data among said plurality of input images; and

comparing said found differences in said calculated feature quantity with a threshold to judge a presence or absence of an obstacle when the difference in said feature quantity is smaller than the threshold.

7. A method of detecting an obstacle for detecting an obstacle within a common viewing range through the use of a plurality of image pick-up devices, said method comprising:

receiving a plurality of images respectively from said plurality of image pick-up devices;

calculating a predetermined feature data in each of said plurality of input images;

calculating differences between a pair of the feature data among said plurality of input images; and

comparing said found differences in said calculated feature quantity with a threshold to judge a presence or absence of an obstacle when the difference in said feature quantity is smaller than the threshold.