Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
画像データ生成装置、画像認識装置、画像データ生成プログラム、及び画像認識プログラム
Document Type and Number:
Japanese Patent JP7002729
Kind Code:
B2
Abstract:
The purpose of the present invention is to recognize from images a dynamic object to be recognized. By means of a spatio-temporal image data generation unit 2, an image recognition device 1 executes a Hilbert scan of frame image data 6 constituting moving-image data 4 to generate one-dimensional spatial image data 7, and further arrays the one-dimensional spatial image data 7 in a time direction to generate two-dimensional spatio-temporal image data 8 that holds spatial information and temporal information. Thus, the image recognition device 1 converts the moving-image data 4 into the two-dimensional spatio-temporal image data 8 while holding the spatial information and temporal information. Next, by means of a CNN unit 3, the image recognition device 1 executes a convolution process in which a two-dimensional filter is used on the spatio-temporal image data 8 to image-recognize a behavior of a pedestrian who is a recognition object. Thus, the image recognition device 1 can execute behavioral recognition of the pedestrian that requires a three-dimensional analysis consisting of two dimensions for space and one dimension for time by means of the image recognition process of the two-dimensional image by the CNN, and thereby can estimate a state of the pedestrian.

Inventors:
Hideo Yamada
Tatsuya Muramatsu
Masatoshi Shibata
Enokida Shuichi
Application Number:
JP2018035744A
Publication Date:
January 20, 2022
Filing Date:
February 28, 2018
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
Aisin Co., Ltd.
Kyushu Institute of Technology
International Classes:
G06T3/00; G06T7/00; G06V10/764; G06V10/82
Domestic Patent References:
JP2002123834A
JP2017187954A
Other References:
橋村 佳祐,距離画像のフレーム連結画像を用いたConvolutional Neural Networkによる手話単語認識,電子情報通信学会技術研究報告 Vol.116 No.248,日本,一般社団法人電子情報通信学会,2016年10月09日,WIT2016-36 (2016-10),P.17-22
Attorney, Agent or Firm:
Takashi Kawai
Nakano Hitoshi