Moving object detection apparatus and moving object detection method

Patent metadata

Field	Value
Publication number	US-11024042-B2
Application number	US-201916546081-A
Country	US
Kind code	B2
Filing date	Aug 20, 2019
Priority date	Aug 24, 2018
Publication date	Jun 1, 2021
Grant date	Jun 1, 2021

Abstract

Provided are an inexpensive and safe moving object detection apparatus and moving object detection method that enable accurate detection of a moving object from an image sequence captured by a monocular camera at a high speed. A representative configuration of the moving object detection apparatus according to the present invention is provided with a horizon line detection unit that detects a horizon line in a frame image, an edge image generation unit that generates an edge image from a frame image, and a moving object detection unit that sets a detection box on a moving object, and the edge image generation unit extracts an edge image below a horizon line detected by the horizon line detection unit, and the moving object detection unit generates a foreground by combining the difference between the edge image below the horizon line and a background image of the edge image with the difference between a gray scale image and a background image of the gray scale image.

First claim

The invention claimed is:

1. A moving object detection apparatus comprising:
an input unit for inputting an image sequence;
a frame acquisition unit that continuously extracts a plurality of frame images from an image sequence;
a horizon line detection unit that detects a horizon line in a frame image;
an edge image generation unit that generates an edge image from a frame image; and
a moving object detection unit that sets a detection box on a moving object,
wherein the edge image generation unit extracts an edge image below the horizon line detected by the horizon line detection unit, and
extracts a difference between the edge image below the horizon line and a background image of the edge image using a background subtraction method, and
the moving object detection unit converts a frame image into a gray scale image and extracts a difference between the gray scale image and a background image of the gray scale image using the background subtraction method, and
generates a foreground by combining the difference between the edge image below the horizon line and the background image of the edge image with the difference between the gray scale image and the background image of the gray scale image.

2. The moving object detection apparatus according to claim 1 ,
wherein the moving object detection unit combines the difference between the edge image below the horizon line and the background image of the edge image with the difference between the gray scale image and the background image of the gray scale image if the difference between the edge image below the horizon line and the background image of the edge image contains only little noise.

3. The moving object detection apparatus according to claim 2 ,
wherein the edge image generation unit extracts an edge image that is connected to the horizon line detected by the horizon line detection unit, and
the moving object detection unit filters the foreground using the edge image that is connected to the horizon line.

4. The moving object detection apparatus according to claim 3 ,
wherein, when the moving object detection apparatus is moving, the moving object detection unit filters the foreground using the edge image that is connected to the horizon line.

5. The moving object detection apparatus according to claim 1 ,
wherein the edge image generation unit extracts an edge image that is connected to the horizon line detected by the horizon line detection unit, and
the moving object detection unit filters the foreground using the edge image that is connected to the horizon line.

6. The moving object detection apparatus according to claim 5 ,
wherein, when the moving object detection apparatus is moving, the moving object detection unit filters the foreground using the edge image that is connected to the horizon line.

7. A moving object detection method comprising:
a step of inputting an image sequence;
a step of continuously extracting a plurality of frame images from an image sequence;
a step of extracting an edge image below a horizon line from a frame image;
a step of extracting a difference between the edge image below the horizon line and a background image of the edge image using a background subtraction method;
a step of converting a frame image into a gray scale image and extracting a difference between the gray scale image and a background image of the gray scale image using the background subtraction method; and
a step of generating a foreground by combining the difference between the edge image below the horizon line and the background image of the edge image with the difference between the gray scale image and the background image of the gray scale image.

Assignees

• Incorporated National Univ Iwate Univ
• Suzuki Motor Corp

Inventors

• Akashi Takuya
• Ganbold Uuganbayar
• Tomita Hiroyuki

Classifications

• 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

• G06T7/254 Primary

  involving subtraction of images · CPC title

• G06T7/215

  Motion-based segmentation · CPC title

• G06T2207/20224

  Image subtraction · CPC title

• G06T7/194

  involving foreground-background segmentation · CPC title