Title:
検出対象位置特定システム
Document Type and Number:
Japanese Patent JP7153306
Kind Code:
B2
Abstract:
To provide a system for specifying a detection object position, that identifies a detection object from image data photographed using a drone 11, and specifies a position of the identified detection object as a height of a probability value.SOLUTION: In a system 10 for specifying a position of a detection object, multiple pieces of image data photographed by a drone 11 using a camera 14 are transmitted to an arithmetic unit 12, and the arithmetic unit 12 generates a set of probability cells of a region in which the drone 11 is flying, and then, artificial intelligence is utilized to obtain a detection likelihood of the detection object and a position and size of the detection object on the image data contained in the image data. Further, in the system for specifying a position of the detection object, an update object range in the probability cell set is specified on the basis of position information of the drone 11 and a photographing direction of the image data, and a probability value of each cell contained in the update object range is updated on the basis of the detection likelihood of the detection object and the position and size of the detection object on the image data, by using an updating numerical expression based on the Bayes' theorem. These procedures are carried out once or more in order to specify the detection object as a height of the probability value for each cell in the probability cell set.SELECTED DRAWING: Figure 1
Inventors:
Daisuke Iwakura
Application Number:
JP2018079448A
Publication Date:
October 14, 2022
Filing Date:
April 17, 2018
Export Citation:
Assignee:
Rock Garage Co., Ltd.
International Classes:
G06T7/70; B64C39/02; B64D45/00; B64D47/08; G05D1/00
Domestic Patent References:
| JP2017022574A | ||||
| JP2010257341A |
Foreign References:
| WO2010101227A1 | ||||
| US20120020523 |
Other References:
宮里 太也 外2名,多目的自律型ドローンのための俯瞰視点人物検知システムに関する研究,電気学会研究会資料,日本,一般社団法人電気学会,2017年09月21日,pp. 11-16
園田 潤 外4名,屋内外シームレス電波環境解析のためのUAV-SfMによるFDTD実環境マルチスケール数値モデル構築,電子情報通信学会論文誌B,電子情報通信学会,2017年09月01日,第J100-B巻 第9号,pp. 773-781
園田 潤 外4名,屋内外シームレス電波環境解析のためのUAV-SfMによるFDTD実環境マルチスケール数値モデル構築,電子情報通信学会論文誌B,電子情報通信学会,2017年09月01日,第J100-B巻 第9号,pp. 773-781
Attorney, Agent or Firm:
Yoshitaka Kobayashi