Driving assist apparatus, driving assist method, and non-transitory computer readable recording medium storing program

Patent metadata

Field	Value
Publication number	US-10192122-B2
Application number	US-201415323922-A
Country	US
Kind code	B2
Filing date	Aug 21, 2014
Priority date	Aug 21, 2014
Publication date	Jan 29, 2019
Grant date	Jan 29, 2019

Abstract

A driving assist apparatus includes an image acquisition part to acquire a captured image around a vehicle, a position information acquisition part to acquire position information of a first object existing around the vehicle and detected by a sensor, a detection range determination part to determine a detection range of the first object within the captured image based on the position information of the first object, and an object recognition part to perform image processing on, within the captured image, an image existing in a range other than the detection range of the first object, thereby to recognize a second object that is different from the first object. Hence, a processing time taken for recognizing the second object from the captured image can be shortened.

First claim

The invention claimed is:

1. A driving assist apparatus comprising:
a memory; and
a processor configured to
acquire, using a camera, a captured image around a vehicle;
acquire, from one or more sensors other than the camera, position information of one or more objects detected by the one or more sensors to exist around the vehicle;
determine a detection range of the one or more sensor detected objects within the captured image based on the acquired position information;
determine one or more regions of the captured image which correspond to the determined detection range of the one or more sensor detected objects; and
perform image processing on a region of the captured image other than a region corresponding to a detection range of an object detected by the one or more sensors to recognize a second object that is different from any sensor detected objects, wherein

the processor is further configured to
extract a pixel group having a color density equal to or higher than a threshold, from a region of the captured image other than the one or more regions corresponding to the detection range of the one or more sensor detected objects;
calculate a detecting region based on the extracted pixel group; and
performing image processing on the calculated detecting region of the captured image to recognize the second object, and

the processor extracts the pixel group having the color density equal to or higher than the threshold from a region of the capture image in a traffic lane included in the captured image.

2. The driving assist apparatus according to claim 1 , wherein the memory stores an edge image expressing a feature of an object to be detected, and
recognizing whether a second object is in the calculated detecting region of the captured image by matching the image within the calculated detecting region with the stored edge image.

3. The driving assist apparatus according to claim 1 , wherein the processor extracts the pixel group having the color density equal to or higher than the threshold from a region within the captured image existing below a horizontal-direction line at such a height where a vanishing point is included.

4. The driving assist apparatus according to claim 1 , wherein the processor is further configured to generate a model that retains distance information of a distance between the vehicle and the recognized second object.

5. The driving assist apparatus according to claim 4 , wherein when distances among a plurality of generated models are each equal to or lower than a threshold, the processor stores the plurality of models as a single model.

6. A driving assist apparatus comprising:
a memory; and
a processor configured to
acquire, using a camera, a captured image around a vehicle;
acquire, from one or more sensors other than the camera, position information of at least two objects detected by the one or more sensors to exist around the vehicle, each of the at least two objects being detected by radiation with one of the one or more sensors in a first-line direction;
determine a first detection range of a first object of the at least two radiation detected objects within the captured image based on the acquired position information of the at least two radiation detected objects,
determine a second detection range of a second object of the at least two radiation detected objects within the captured image based on the acquired position information of the at least two radiation detected objects;
determine a region within the captured image which corresponds to a composite of the determined first and second detection ranges; and
perform image processing on a region of the captured image other than a region determined to correspond to the composite range of the first and second detection ranges of the at least two radiation detected objects to recognize a third object that is different from a objects detected by the one or more sensors, wherein

the processor is further configured to
extract a pixel group having a color density equal to or higher than a threshold, from a region of the captured image other than the one or more regions corresponding to the detection range of the one or more sensor detected objects;
calculate a detecting region based on the extracted pixel group; and
performing image processing on the calculated detecting region of the captured image to recognize the second object, and

the processor extracts the pixel group having the color density equal to or higher than the threshold from a region of the capture image in a traffic lane included in the captured image.

7. A driving assist method comprising:
acquiring, using a camera, a captured image around a vehicle;
acquiring; from one or more sensors, position information of one or more objects detected by the one or more sensors to exist around the vehicle;
determining, using a processor, a detection range of the one or more sensor detected objects within the captured image based on the acquired position information;
determining one or more regions of the captured image which correspond to the determined detection range of the one or more sensor detected objects;
extracting a pixel group having a color density equal to or higher than a threshold, from a region of the captured image in a traffic lane included in the captured image and other than the one or more regions corresponding to the detection range of the one or more sensor detected objects;
calculating a detecting region based on the extracted pixel group;
performing image processing on the calculated detecting region to recognize a second object that is different from any sensor detected objects.

8. A non-transitory computer readable recording medium storing a program that when executed causes a computer to:
acquire, using a camera, a captured image around a vehicle;
acquire, from one or more sensors other than the camera, position information of one or more objects detected by the one or more sensors to exist around the vehicle;
determine a detection range of the one or more sensor detected objects within the captured image based on the acquired position information;
determine one or more regions of the captured image which correspond to the determined detection range of the one or more sensor detected objects;
extract a pixel group having a color density equal to or higher than a threshold, from a region of the captured image in a traffic lane included in the captured image and other than the one or more regions corresponding to the detection range of the one or more sensor detected objects;
calculate a detecting region based on the extracted pixel group; and
perform image processing on the calculated detecting region to recognize a second object that is different from any sensor detected objects.

Assignees

• Mitsubishi Electric Corp

Inventors

• Yoshida Michinori
• Tsushima Naoyuki

Classifications

• G08G1/165 Primary

  for passive traffic, e.g. including static obstacles, trees · CPC title

• G06V10/803

  of input or preprocessed data · CPC title

• G06V20/58

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G08G1/166

  for active traffic, e.g. moving vehicles, pedestrians, bikes · CPC title

• H04N7/183

  for receiving images from a single remote source · CPC title