VEHICLE DETECTION DEVICE AND VEHICLE DETECTION METHOD

• US 20120147187A1
• Filed: 12/02/2011
• Published: 06/14/2012
• Est. Priority Date: 12/08/2010
• Status: Active Grant

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: HIRAI, Hideaki, LI, Xue
• Granted Patent: US 8,908,038 B2
• Time in Patent Office: Days
• US Class Current: 348/148
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle
• 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

A vehicle detection device having an imaging device to capture two respective polarized images from two polarized light beams having different polarization directions contained in light received from an imaging area including a road surface on which one'"'"'s own vehicle is traveling and a vehicle traveling on the road surface and a polarization difference calculation device to calculate a polarization difference indicating the ratio of a luminance difference between the two respective polarized images to the luminance total thereof for respective identification processing areas formed by dividing the two respective polarized images taken by the imaging device, and a vehicle area detection device to conduct a vehicle area detection process of detecting a vehicle area displaying the vehicle in the imaging area based on the polarization difference of the respective identification processing areas calculated by the polarization difference calculation device.

Claims

1. A vehicle detection device comprising:

an imaging device to capture two different polarized images from two differently polarized light beams reflected within a vehicle detection device imaging area, the imaging area containing a road surface;

a polarization difference calculation device that divides each of the two polarized images captured by the imaging device into predetermined identification processing areas and calculates a degree of polarization difference indicating a ratio of a luminance difference between the two polarized images to a total luminance thereof for each predetermined identification processing area; and

a vehicle area detection device to conduct a vehicle area detection process of detecting a vehicle area displaying a vehicle in the imaging area based on the polarization difference of the identification processing areas calculated by the polarization difference calculation device.

2. The vehicle detection device according to claim 1, wherein the vehicle area detection device determines whether the polarization differences for the identification processing areas calculated by the polarization difference calculation device are within a predetermined range, identifies an identification processing area having a value within the predetermined range as a window of the vehicle, and conducts the vehicle area detection process using the identification.

3. The vehicle detection device according to claim 2, further comprising a vehicle characteristic data memory device that stores vehicle characteristic data indicating characteristics of the vehicle,

wherein the vehicle area detection device determines whether the identification processing areas identified as the window are the vehicle areas that display the vehicle using the luminance total of the two polarized images of the identification processing areas and the vehicle characteristic data stored in the vehicle characteristic data memory device.

4. The vehicle detection device according to claim 1, further comprising a three-dimensional object area identifying device to identify a three-dimensional object area that displays a three-dimensional object existing in the imaging area based on the polarization difference of the identification processing areas calculated by the polarization difference calculation device,

wherein the vehicle area detection device conducts the vehicle area detection process for the polarization differences of the three-dimensional object area identified by the three-dimensional object area identifying device.

5. The vehicle detection device according to claim 4, further comprising a road surface characteristic data memory device that stores road surface characteristic data indicating characteristics of the road surface,

wherein the three-dimensional object area identifying device conducts a digitization process of the identification processing areas by comparing a polarization difference of the identification processing areas calculated by the difference polarization calculation device with a road surface identification threshold, identifies an identification processing area that displays the road surface as the road surface area using the road surface characteristic data stored in the road surface characteristic data memory device, and identifies remaining identification processing areas as the three-dimensional object areas.

6. The vehicle detection device according to claim 1, further comprising a particular three-dimensional object area identification device that identifies a particular three-dimensional object area that displays a predetermined particular three-dimensional object other than the vehicle in as portion of the imaging area that excludes the vehicle area identified by the vehicle area detection device.

7. A vehicle detection method comprising:

capturing two different polarized images from two differently polarized light beams reflected within an imaging area of an imaging device, the imaging area containing road surface;

dividing the two polarized images taken by the imaging device into multiple identification processing areas;

calculating a polarization difference indicating a ratio of a luminance difference between the two polarized images to a total luminance for each of the identification processing areas; and

detecting a vehicle area displaying the vehicle in the imaging area based on the calculated polarization difference of the identification processing areas.