JP2021051542A5 - - Google Patents

Download PDF

Info

Publication number
JP2021051542A5
JP2021051542A5 JP2019174028A JP2019174028A JP2021051542A5 JP 2021051542 A5 JP2021051542 A5 JP 2021051542A5 JP 2019174028 A JP2019174028 A JP 2019174028A JP 2019174028 A JP2019174028 A JP 2019174028A JP 2021051542 A5 JP2021051542 A5 JP 2021051542A5
Authority
JP
Japan
Prior art keywords
images
detected
obstacle
image
road
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP2019174028A
Other languages
Japanese (ja)
Other versions
JP7383954B2 (en
JP2021051542A (en
Filing date
Publication date
Application filed filed Critical
Priority to JP2019174028A priority Critical patent/JP7383954B2/en
Priority claimed from JP2019174028A external-priority patent/JP7383954B2/en
Publication of JP2021051542A publication Critical patent/JP2021051542A/en
Publication of JP2021051542A5 publication Critical patent/JP2021051542A5/ja
Application granted granted Critical
Publication of JP7383954B2 publication Critical patent/JP7383954B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Claims (8)

車載カメラにより撮影された複数フレームの画像について、予め学習された学習済みモデルを用いた前記画像中の路上障害物の検出及び前記画像中の物体の検出において、前記路上障害物が検出された２つの画像の間の画像のうち、前記路上障害物が存在すると推定されるエリア内に前記路上障害物が検出されず、かつ前記物体が検出された画像を、前記学習済みモデルの学習用データとして用いて更新された、学習済みモデル。Regarding a plurality of frames of an image taken by an in-vehicle camera, the road obstacle was detected in the detection of a road obstacle in the image and the detection of an object in the image using a pre-learned model. 2 Among the images between the two images, the image in which the road obstacle is not detected in the area where the road obstacle is presumed to exist and the object is detected is used as training data of the trained model. Trained model updated with. 請求項１に記載の更新された学習済みモデルを用いて、路上障害物を検出する障害物検出部を含む制御装置。A control device including an obstacle detection unit that detects an obstacle on the road using the updated trained model according to claim 1. 予め学習された学習済みモデルを用いて画像中の路上障害物を検出すると共に、Detects road obstacles in images using pre-trained trained models and
前記画像中の物体を検出し、The object in the image is detected
車載カメラにより撮影された複数フレームの画像について、前記路上障害物が検出された２つの画像の間の画像のうち、前記路上障害物が存在すると推定されるエリア内に前記路上障害物が検出されず、かつ前記物体が検出された画像を、前記学習済みモデルの学習用データとして、前記学習済みモデルを更新する、Regarding the images of a plurality of frames taken by the in-vehicle camera, the road obstacle is detected in the area where the road obstacle is presumed to exist among the images between the two images in which the road obstacle is detected. The trained model is updated by using the image in which the object is detected as training data of the trained model.
処理をコンピュータが実行する学習済みモデルの更新方法。 How to update a trained model whose processing is performed by a computer. 予め学習された学習済みモデルを用いて画像中の路上障害物を検出する障害物検出部と、
前記画像中の物体を検出する物体検出部と、
車載カメラにより撮影された複数フレームの画像について、前記路上障害物が検出された２つの画像の間の画像のうち、前記路上障害物が存在すると推定されるエリア内に前記路上障害物が検出されず、かつ前記物体検出部により前記物体が検出された画像を、前記学習済みモデルの学習用データとして出力する出力部と、
を備えた学習用データ収集装置。 An obstacle detection unit that detects road obstacles in an image using a pre-learned model, and an obstacle detection unit.
An object detection unit that detects an object in the image,
Regarding the images of a plurality of frames taken by the in-vehicle camera, the road obstacle is detected in the area where the road obstacle is presumed to exist among the images between the two images in which the road obstacle is detected. An output unit that outputs an image in which the object is detected by the object detection unit as training data of the trained model, and an output unit.
Data acquisition device for learning equipped with. 予め学習された学習済みモデルを用いて画像中の路上障害物を検出する障害物検出部と、An obstacle detection unit that detects road obstacles in an image using a pre-learned model, and an obstacle detection unit.
前記画像中の物体を検出する物体検出部と、An object detection unit that detects an object in the image,
車載カメラにより撮影された複数フレームの画像について、前記路上障害物が検出された２つの画像の間の画像のうち、前記物体が存在すると推定されるエリア内に前記物体が検出されず、かつ前記路上障害物が検出された画像を、前記学習済みモデルの学習用データとして出力する出力部と、Regarding the images of a plurality of frames taken by the in-vehicle camera, among the images between the two images in which the road obstacle is detected, the object is not detected in the area where the object is presumed to exist, and the object is said to be present. An output unit that outputs an image in which a road obstacle is detected as training data of the trained model, and an output unit.
を備えた学習用データ収集装置。Data acquisition device for learning equipped with. 前記障害物検出部は、前記画像を分割して得られた局所領域ＳThe obstacle detection unit is a local region S obtained by dividing the image. ｎn と、前記学習済みモデルから出力された確率ＰAnd the probability P output from the trained model ｍm とに基づいて、画像のｉ番目の局所領域ＳBased on, the i-th local region S of the image ｉi における路上障害物らしさＬRoad obstacle-likeness L ｉi を算出することにより、前記路上障害物を検出する請求項４又は請求項５に記載の学習用データ収集装置。The learning data collecting device according to claim 4 or 5, wherein the road obstacle is detected by calculating. 前記車載カメラを搭載した車両の車速が遅いほど、連続する複数フレームの画像を多い枚数を用いて、前記２つの画像の間の画像のうちから、前記学習用データとする画像を得る請求項４～請求項６の何れか１項に記載の学習用データ収集装置。4. The slower the vehicle speed of the vehicle equipped with the in-vehicle camera, the larger the number of images of a plurality of consecutive frames are used to obtain an image to be the training data from the images between the two images. The learning data collecting device according to any one of claims 6. 車載カメラにより撮影された複数フレームの画像について、路上障害物が検出された２つの画像の間の画像のうち、前記路上障害物が存在すると推定されるエリア内に前記路上障害物が検出されず、かつ物体が検出された画像を、前記路上障害物のデータとして出力する出力部、Of the images between the two images in which road obstacles are detected in the images of multiple frames taken by the in-vehicle camera, the road obstacles are not detected in the area where the road obstacles are presumed to exist. And an output unit that outputs an image in which an object is detected as data of the road obstacle,
を備えた障害物データ収集装置。Obstacle data collection device equipped with.
JP2019174028A 2019-09-25 2019-09-25 Learning data collection device Active JP7383954B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2019174028A JP7383954B2 (en) 2019-09-25 2019-09-25 Learning data collection device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2019174028A JP7383954B2 (en) 2019-09-25 2019-09-25 Learning data collection device

Publications (3)

Publication Number Publication Date
JP2021051542A JP2021051542A (en) 2021-04-01
JP2021051542A5 true JP2021051542A5 (en) 2022-07-21
JP7383954B2 JP7383954B2 (en) 2023-11-21

Family

ID=75158245

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2019174028A Active JP7383954B2 (en) 2019-09-25 2019-09-25 Learning data collection device

Country Status (1)

Country Link
JP (1) JP7383954B2 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114343641A (en) * 2022-01-24 2022-04-15 广州熠华教育咨询服务有限公司 Learning difficulty intervention training guidance method and system thereof

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6682833B2 (en) * 2015-12-04 2020-04-15 トヨタ自動車株式会社 Database construction system for machine learning of object recognition algorithm
JP6565967B2 (en) * 2017-05-12 2019-08-28 トヨタ自動車株式会社 Road obstacle detection device, method, and program
JP6986685B2 (en) * 2018-03-12 2021-12-22 パナソニックＩｐマネジメント株式会社 Information processing equipment

Similar Documents

Publication Publication Date Title
US10460470B2 (en) Recognition and reconstruction of objects with partial appearance
US20210096241A1 (en) Perception system
JP7147420B2 (en) OBJECT DETECTION DEVICE, OBJECT DETECTION METHOD AND COMPUTER PROGRAM FOR OBJECT DETECTION
JP2020504360A5 (en)
WO2018102697A1 (en) System and method for estimating vehicular motion based on monocular video data
US11200468B2 (en) Fast CNN classification of multi-frame semantic signals
JPWO2020171983A5 (en)
WO2017177367A1 (en) Method and system for object tracking
TW202022797A (en) Object detection method using cnn model and object detection apparatus using the same
JP2016062610A (en) Feature model creation method and feature model creation device
CN106919895B (en) Tracking method and system for moving object
JPWO2020159960A5 (en)
CN104616324B (en) Method for tracking target based on adaptive apparent model and point set learning distance metric
Rajeshwari et al. Object detection: an overview
KR20190119510A (en) Vision-based sample-efficient reinforcement learning framework for autonomous driving
JP2021051542A5 (en)
KR102246418B1 (en) Lane keeping method and apparatus thereof
JP2023020992A (en) Object detection method and apparatus
JP6241172B2 (en) Vehicle position estimation device and vehicle position estimation method
JPWO2021067056A5 (en)
Chen et al. Real-time lane detection model based on non bottleneck skip residual connections and attention pyramids
JP7383954B2 (en) Learning data collection device
CN103903267B (en) Image partition method based on average template and student's t mixed models
Chen et al. Imitating driver behavior for fast overtaking through bagging Gaussian process regression
US11893087B2 (en) Defending multimodal fusion models against single-source adversaries