JP2020204812A5 - - Google Patents
Download PDFInfo
- Publication number
- JP2020204812A5 JP2020204812A5 JP2019110963A JP2019110963A JP2020204812A5 JP 2020204812 A5 JP2020204812 A5 JP 2020204812A5 JP 2019110963 A JP2019110963 A JP 2019110963A JP 2019110963 A JP2019110963 A JP 2019110963A JP 2020204812 A5 JP2020204812 A5 JP 2020204812A5
- Authority
- JP
- Japan
- Prior art keywords
- recognition
- model
- information processing
- evaluation
- information
- 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
Links
- 230000010365 information processing Effects 0.000 claims description 19
- 238000011156 evaluation Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims 5
- 230000005856 abnormality Effects 0.000 claims 2
- 238000003672 processing method Methods 0.000 claims 2
- 238000012854 evaluation process Methods 0.000 claims 1
- 239000000284 extract Substances 0.000 claims 1
- 230000006870 function Effects 0.000 claims 1
Description
本発明の情報処理装置は、モデルに基づいて、画像から特徴を認識する認識手段と、前記モデルの追加学習を行う学習手段と、前記モデルの追加学習の履歴の情報を管理する管理手段と、前記モデルの前記認識の精度を評価する評価手段と、前記履歴の情報が管理されているモデルのなかから、前記認識の精度に基づいてモデルを選択する選択手段と、を有し、前記認識手段は、前記選択手段が選択したモデルを前記認識に用いることを特徴とする。 The information processing apparatus of the present invention includes a recognition means for recognizing features from an image based on a model, a learning means for performing additional learning of the model, and a management means for managing information on the history of additional learning of the model. The recognition means has an evaluation means for evaluating the recognition accuracy of the model and a selection means for selecting a model based on the recognition accuracy from the models in which the history information is managed. Is characterized in that the model selected by the selection means is used for the recognition.
Claims (16)
前記モデルの追加学習を行う学習手段と、
前記モデルの追加学習の履歴の情報を管理する管理手段と、
前記モデルによる前記認識の精度を評価する評価手段と、
前記履歴の情報が管理されているモデルのなかから、前記認識の精度に基づいてモデルを選択する選択手段と、を有し、
前記認識手段は、前記選択手段が選択したモデルを前記認識に用いることを特徴とする情報処理装置。 A recognition means that recognizes features from images based on a model,
A learning means for performing additional learning of the model and
A management means for managing information on the history of additional learning of the model, and
An evaluation means for evaluating the accuracy of the recognition by the model and
It has a selection means for selecting a model based on the accuracy of the recognition from the models in which the history information is managed.
The recognition means is an information processing apparatus characterized in that the model selected by the selection means is used for the recognition.
前記選択手段は、前記履歴の情報が管理されているそれぞれのモデルを用いて、前記評価セットを前記認識手段で認識した精度を前記評価手段で評価した結果を基に前記モデルを選択することを特徴とする請求項1に記載の情報処理装置。 The evaluation means holds a set of recognition targets recognized by the recognition means as an evaluation set.
The selection means selects the model based on the result of evaluating the accuracy of the evaluation set recognized by the recognition means by the evaluation means by using each model in which the history information is managed. The information processing apparatus according to claim 1.
前記評価手段は、前記選択されたクラスタに含まれる結果に基づいて前記認識の精度を評価することを特徴とする請求項1乃至6のいずれか1項に記載の情報処理装置。 The selection means clusters the recognition targets recognized by the recognition means, and at least one or more clusters among the results of the clustering are selected for each model.
The information processing apparatus according to any one of claims 1 to 6, wherein the evaluation means evaluates the accuracy of the recognition based on the result included in the selected cluster.
前記管理手段は、前記認識対象に前記追加学習が行われた世代を表す世代番号を付与することによって、前記モデルの履歴を管理し、
前記認識手段は、前記モデルが含むデータの部分集合を前記世代番号に基づいて定め、前記部分集合を前記履歴に含まれるモデルとすることを特徴とする請求項1乃至8のいずれか1項に記載の情報処理装置。 The learning means performs additional learning by adding or removing recognition targets to the model.
The management means manages the history of the model by assigning a generation number representing the generation in which the additional learning was performed to the recognition target.
The recognition means according to any one of claims 1 to 8, wherein a subset of data included in the model is determined based on the generation number, and the subset is a model included in the history. The information processing device described.
前記認識の精度は、前記世代番号のモデルを用いた際の所定の認識対象のデータに最も近い点までの距離を示すスコアであって、The recognition accuracy is a score indicating the distance to the point closest to the predetermined recognition target data when the model of the generation number is used.
前記選択手段は、最新のモデルの前記世代番号から初期のモデルまでの各モデルに含まれる前記認識対象のデータと、前記所定の認識対象に基づいて、前記スコアが最小となる前記認識対象のデータを含むモデルを選択し、The selection means includes the data of the recognition target included in each model from the generation number of the latest model to the initial model, and the data of the recognition target having the minimum score based on the predetermined recognition target. Select a model that contains
前記評価手段は、前記選択されたモデルの前記世代番号が前記初期のモデルの前記世代番号と異なる場合は、前記選択されたモデルの前記世代番号以降のモデルに含まれる前記認識対象のデータを除いた前記認識対象のデータと、前記所定の認識対象に基づいて前記スコアを評価し、前記選択されたモデルの前記世代番号が前記初期のモデルの前記世代番号である場合は、評価を終了することを特徴とする請求項9に記載の情報処理装置。When the generation number of the selected model is different from the generation number of the initial model, the evaluation means excludes the data to be recognized included in the models after the generation number of the selected model. The score is evaluated based on the data of the recognition target and the predetermined recognition target, and if the generation number of the selected model is the generation number of the initial model, the evaluation is terminated. 9. The information processing apparatus according to claim 9.
前記表示制御手段は、前記認識手段によって正常と認識された物体と、異常が発生したと認識された物体とをそれぞれ示す情報を表示させることを特徴とする請求項1乃至11のいずれか1項に記載の情報処理装置。The display control means is any one of claims 1 to 11, wherein the display control means displays information indicating an object recognized as normal by the recognition means and an object recognized as having an abnormality. The information processing device described in.
モデルに基づいて、画像から特徴を認識する認識工程と、A recognition process that recognizes features from images based on a model,
前記モデルの追加学習を行う学習工程と、The learning process for performing additional learning of the model and
前記モデルの追加学習の履歴の情報を管理する管理工程と、A management process that manages information on the history of additional learning of the model, and
前記モデルの前記認識の精度を評価する評価工程と、An evaluation process for evaluating the recognition accuracy of the model and
前記履歴の情報が管理されているモデルのなかから、前記認識の精度に基づいてモデルを選択する選択工程と、を有し、It has a selection step of selecting a model based on the accuracy of the recognition from the models in which the history information is managed.
前記認識工程は、前記選択工程で選択されたモデルを前記認識に用いることを特徴とする情報処理方法。The recognition step is an information processing method characterized in that the model selected in the selection step is used for the recognition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2019110963A JP7433782B2 (en) | 2019-06-14 | 2019-06-14 | Information processing device, information processing method, and program |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2019110963A JP7433782B2 (en) | 2019-06-14 | 2019-06-14 | Information processing device, information processing method, and program |
Publications (3)
Publication Number | Publication Date |
---|---|
JP2020204812A JP2020204812A (en) | 2020-12-24 |
JP2020204812A5 true JP2020204812A5 (en) | 2022-06-17 |
JP7433782B2 JP7433782B2 (en) | 2024-02-20 |
Family
ID=73838407
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2019110963A Active JP7433782B2 (en) | 2019-06-14 | 2019-06-14 | Information processing device, information processing method, and program |
Country Status (1)
Country | Link |
---|---|
JP (1) | JP7433782B2 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2023149113A (en) * | 2022-03-30 | 2023-10-13 | 株式会社神戸製鋼所 | Abnormality determination method, method for processing at the time of abnormality, information processing apparatus, welding system, and program |
CN117315565A (en) | 2023-08-31 | 2023-12-29 | 云南电网有限责任公司德宏供电局 | Abnormal behavior identification monitoring method based on increment space-time learning |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010050334A1 (en) | 2008-10-30 | 2010-05-06 | コニカミノルタエムジー株式会社 | Information processing device |
JP5335536B2 (en) | 2009-04-23 | 2013-11-06 | キヤノン株式会社 | Information processing apparatus and information processing method |
JP6061713B2 (en) | 2013-02-08 | 2017-01-18 | 本田技研工業株式会社 | Inspection apparatus, inspection method and program |
JP6149710B2 (en) | 2013-11-27 | 2017-06-21 | 富士ゼロックス株式会社 | Image processing apparatus and program |
JP2015116319A (en) | 2013-12-18 | 2015-06-25 | パナソニックIpマネジメント株式会社 | Diagnosis support device, diagnosis support method, and diagnosis support program |
WO2019087791A1 (en) | 2017-10-31 | 2019-05-09 | 富士フイルム株式会社 | Diagnosis support device, endoscope device, management device, diagnosis support method, management method, diagnosis support program, and management program |
-
2019
- 2019-06-14 JP JP2019110963A patent/JP7433782B2/en active Active
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9886669B2 (en) | Interactive visualization of machine-learning performance | |
US20180365557A1 (en) | Information processing method and information processing apparatus | |
JP2019016209A (en) | Diagnosis device, diagnosis method, and computer program | |
US20180246846A1 (en) | Information processing apparatus, information processing method, and storage medium | |
US20200116522A1 (en) | Anomaly detection apparatus and anomaly detection method | |
JP2016133895A (en) | Information processing device, information processing method, and program | |
WO2019026134A1 (en) | Information processing device and information processing method | |
JP2020204812A5 (en) | ||
JP2021099582A (en) | Information processing apparatus, information processing method, and program | |
JP7298825B2 (en) | Learning support device, learning device, learning support method, and learning support program | |
WO2018056973A1 (en) | Font detector | |
JP2020024665A (en) | Information processing method and information processing system | |
CN112001212B (en) | Waveform segmentation device and waveform segmentation method | |
JP7120914B2 (en) | Production performance data analyzer | |
JP6405851B2 (en) | Predictive detection support program, method, apparatus, and predictive detection program, | |
US20220405299A1 (en) | Visualizing feature variation effects on computer model prediction | |
JP6930195B2 (en) | Model identification device, prediction device, monitoring system, model identification method and prediction method | |
JP7339063B2 (en) | Machine learning program and machine learning device for learning about work processes | |
CN115349129A (en) | Generating performance predictions with uncertainty intervals | |
CN110706772A (en) | Sorting method and device, electronic equipment and storage medium | |
CN115461686A (en) | Method for generating a process model | |
CN112579587A (en) | Data cleaning method and device, equipment and storage medium | |
JP2016110448A (en) | Diagnosis and repair case example search device | |
JPWO2020085374A1 (en) | Proficiency index providing device, proficiency index providing method, and program | |
CN110796156A (en) | Information processing method and information processing system |