Unlock instant, AI-driven research and patent intelligence for your innovation.

Object detection device and object detection method

An object detection and object technology, which is used in measurement devices, optical observation devices, three-dimensional object recognition, etc., can solve problems such as poor accuracy and misjudgment of road markings as objects.

Inactive Publication Date: 2016-03-02
PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0017] However, existing object detection devices may misjudge road markings (numbers indicating maximum speed, etc.) drawn on distant roads as objects.
[0018] This is because, in a stereoscopic image, the accuracy of parallax calculation in the area where a distant object is captured is low, and it is difficult to calculate the inclination in the height direction of a distant object with high accuracy.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Object detection device and object detection method
  • Object detection device and object detection method
  • Object detection device and object detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0066] figure 2 It is a block diagram showing the configuration of the object detection device according to Embodiment 1 of the present invention.

[0067] This embodiment is an example of an object detection device applied to a vehicle-mounted stereo camera to detect objects existing in the road environment such as vehicles and pedestrians in front of the vehicle (for example, three-dimensional objects such as vehicles and pedestrians, except for road surfaces).

[0068] Such as figure 2 As shown, the object detection device 100 includes: a stereo image acquisition unit 110, a disparity map generation unit 120, a road surface estimation unit 130, an object candidate position extraction unit 140, an object judgment area extraction unit (three-dimensional object judgment area extraction unit) 150, a shape feature An extraction unit 160 , an object judging unit (three-dimensional object judging unit) 170 , and an object judging parameter memory (three-dimensional object judgi...

Embodiment approach 2

[0167] Figure 16 It is a diagram showing an extraction method based on correlation features in the shape feature extraction unit of the object detection device according to Embodiment 2 of the present invention.

[0168] The shape feature extraction unit of this embodiment replaces figure 2 The shape feature extraction unit 160 of the object detection device 100 is used. The basic structure and operation of the object detection device according to the second embodiment of the present invention are the same as those of the first embodiment.

[0169] Figure 16 (a) is a diagram showing the extraction method of relevant features, Figure 16 (b) is a diagram showing shape feature vectors based on correlation features. Such as Figure 16 As shown, the shape feature extraction unit of this embodiment divides the object determination area into grids, and generates shape feature vectors based on the correlation values ​​of the parallax data representative values ​​of each grid....

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Provided are a detection device, an object detection method, and a program capable of accurately determining whether an object captured in a stereoscopic image is an object or a road surface object. The object detection device (100) includes: a disparity map generation unit (120), which generates a disparity map based on the acquired stereo image; a road surface estimation unit (130), which estimates a road surface based on the generated disparity map; and an object candidate position extraction unit (140), which is based on The generated disparity map and the road surface, extract the position of the object candidate existing on the upper side of the road surface in the real space; and the object determination area extraction unit (150), from the generated disparity map, extract the object candidate surrounding area also included. Stereoscopic determination area. In addition, the object detection device (100) further includes: a shape feature extraction unit (160) for extracting shape features of object candidates based on the object determination area; and an object determination unit (170) for determining objects based on the extracted shape features and stored in advance in the object determination area. The object determination parameter of the parameter memory (180) determines whether the object candidate is an object or a road surface.

Description

technical field [0001] The present invention relates to an object detection device, an object detection method and a program for detecting an object captured in a stereoscopic image. Background technique [0002] Conventionally, there is known an object detection device that calculates a deviation (parallax) between a reference image and a reference image by performing stereo matching on stereo images (a reference image and a reference image), and detects an object based on the calculated parallax. The application of this object detection device to, for example, a device for detecting objects (such as three-dimensional objects such as vehicles and pedestrians, not including road surfaces) and road surfaces captured in a stereoscopic image taken in front of a vehicle by an on-vehicle camera has been studied. [0003] As a method adopted by a conventional object detection device, it is proposed to calculate the inclination in the height direction of an object captured in a ste...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T1/00G01B11/00G01B11/24G01C3/06G08G1/16B60R1/00
CPCG01C21/3602G08G1/166G01B11/00G01B11/24G06T2207/10012G06T2207/30261G06T7/11G06T7/174G06V20/58G06V20/64
Inventor 南里卓也丸谷健介西村洋文
Owner PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD