Title:
ＣＮＮ（Ｃｏｎｖｏｌｕｔｉｏｎａｌ Ｎｅｕｒａｌ Ｎｅｔｗｏｒｋ）を利用して車線を検出するための学習方法及び学習装置そしてこれを利用したテスト方法及びテスト装置｛ＬＥＡＲＮＩＮＧ ＭＥＴＨＯＤ， ＬＥＡＲＮＩＮＧ ＤＥＶＩＣＥ ＦＯＲ ＤＥＴＥＣＴＩＮＧ ＬＡＮＥ ＵＳＩＮＧ ＣＮＮ ＡＮＤ ＴＥＳＴ ＭＥＴＨＯＤ， ＴＥＳＴ ＤＥＶＩＣＥ ＵＳＩＮＧ ＴＨＥ ＳＡＭＥ｝
Document Type and Number:
Japanese Patent JP6847463
Kind Code:
B2
Abstract:
A learning method of a CNN for detecting lanes is provided. The method includes steps of: a learning device (a) instructing convolutional layers to generate feature maps by applying convolution operations to an input image from an image data set; (b) instructing an FC layer to generate an estimated result vector of cluster ID classifications of the lanes by feeding a specific feature map among the feature maps into the FC layer; and (c) instructing a loss layer to generate a classification loss by referring to the estimated result vector and a cluster ID GT vector, and backpropagate the classification loss, to optimize device parameters of the CNN; wherein the cluster ID GT vector is GT information on probabilities of being cluster IDs per each of cluster groups assigned to function parameters of a lane modeling function by clustering the function parameters based on information on the lanes.
Inventors:
Kim, Kehyun
Kim, Young Jung
Kim, ins
Kim, Ha Kyung
Nam, Eun Hyun
Bo, Shukhun
Sun, Munchul
Yeoh, Dong Hoon
Ryu, Uge
Jean, Taewon
John, Kyunjeong
Jae, Hongmo
Cho, Ho Jin
Kim, Young Jung
Kim, ins
Kim, Ha Kyung
Nam, Eun Hyun
Bo, Shukhun
Sun, Munchul
Yeoh, Dong Hoon
Ryu, Uge
Jean, Taewon
John, Kyunjeong
Jae, Hongmo
Cho, Ho Jin
Application Number:
JP2019160025A
Publication Date:
March 24, 2021
Filing Date:
September 03, 2019
Export Citation:
Assignee:
Stradvision,Inc.
International Classes:
G06T7/00; G06N3/04; G06N3/08; G06T7/60; G06V10/764
Domestic Patent References:
| JP9223218A | ||||
| JP2018088151A | ||||
| JP2016194925A |
Attorney, Agent or Firm:
Suzue International Patent Office