HACARUS CHECK v1

Scroll down for code samples, example requests and responses. Select a language for code samples from the tabs above or the mobile navigation menu.

English | 日本語

ハカルスはメーカーとメディカル事業に向け、軽量で説明可能なAIソリューションズを提供しています。

弊社のスパースモデリングに基づいた技術は、人間のようにデータを理解できる独自の機械学習手法です。スパースモデリングは軽量に設計されているため、コンピューティングパワー、クラウド環境、および少量の学習データなどの資源の制限された環境下で特に有用です。

弊社のソリューションは組み込みシステムでオフライン環境かクラウドで実行できます。伝統的なディープラーニングベースの手法に比べ、資源を効率的に使用し、よりよい結果を提供します。

ハカルスの外観検査ソリューションの詳細を参照するには、こちらのリンク、https://hacarus.com/ja/visual-inspection/ をご覧ください。また、APIへのアクセスのリクエストはcontact us までご連絡ください。

Base URLs:

この外観検査用 API wrapper for C#は、あなたのC#のプロジェクトに、APIを通じて外観検査モジュールを組み込むことができます。このライブラリはC# .Net Framework 4.6.1 と .Net Core 2.0.をサポートしています。

このPython用外観検査SDKは、APIを用いてハカルス外観検査モジュールを使いたいソフトウェアエンジニアのために作成しています。このAPIによって、Pythonベースのアプリケーションに外観検査モジュールを簡単に統合することができます。

インストール

PM> Install-Package HacarusVisualInspectionApi -Version 1.1.2-beta

> pip install hacarus_visual_inspection

このパッケージをプロジェクトにインストールするため、こちらのコマンドをPackage Manager Consoleに入力してください。

このパッケージをインストールするために、以下のコマンドを実行してください。

NuGetパッケージマネージャを使用してプロジェクトに追加することもできます。Show prerelease packagesにチェックを入れ、 HacarusVisualInspectionApi を nuget.org のリポジトリで検索してください。

他のインストール方法はNuget Package Siteを参照ください。

用語説明

このドキュメントの全体で使用されている用語の簡単な説明：

モデル
モデルは一連のパラメータの設定とともに、データセットの学習データ（トレーニングデータ）をアルゴリズムに適用して、作成・学習できます。

アルゴリズム
アルゴリズムはモデルを構成するため使用する機械学習アルゴリズムのことです。外観検査の性質と予想される精度および性能によって選択します。

学習
新しいモデルを作成する操作です。

パラメータ
パラメータは、学習中にアルゴリズムを適用する方法を構成するために使用されます。例えば、最小または最大の画像解像度などがあります。

データ
データ（アイテムとも呼ばれる)は検査の対象となる単一の製品のデータを表します。 1つのデータに1つまたは複数の画像を関連付けることができます。例：倉庫内の梱包箱の場合、一つの箱に6つの面の画像があります。

使用方法

始めに、下記のものが必要です。

クライエントIDとクライエントシークレット
- これはSDKを認証するために使用されます。
- 認証されたユーザーはシステム機能にアクセスでき、データを追加したり、データ、アルゴリズム、モデルの一覧を取得ことができます。
- モデルの学習や推論を行うためにはライセンスをアクティベートする必要があります。
ライセンスファイル
- これはライセンスをアクティベートするため使用されます。

学習画像
- モデルを作成するためには、画像をアップロードして、学習データを提供する必要があります。
- モデルを作成するには学習データが使用されます。
- 画像をアップロードすることで推論のためのデータを追加します。
- 作成済みのモデルを用いてデータを推論します。
- サンプル画像はこちらからダウンロードすることができます。（Metal Plates、Wood Blocks）
  - モデルの作成に使用できる良品と不良品の画像とパラメータの一覧が含まれています。
  - 学習用の画像はtrainフォルダにあり、推論用の画像はpredictフォルダにあります。
  - NGとは不良品の画像、OKとは良品の画像という意味です。

Authentication

HTTP Authentication, scheme: bearer
oAuth2 authentication.
- Flow: clientCredentials
- Token URL = http://localhost:3000/api/auth/token

Scope	Scope Description
all	all scopes

targets

Inspection Target related operations

getInspectionTargets

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/targets/ \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/targets/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/targets/

Get List of Inpection Targets

Get all inspection targets created by user

Example responses

200 Response

{
  "data": [
    {
      "preprocessing": [
        {
          "active": true,
          "name": "string",
          "config": "string"
        }
      ],
      "name": "string",
      "target_id": 0
    }
  ]
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	InspectionTargetListWrapper
default	Default	Default error response	BaseError

getInspectionTargetById

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/targets/{target_id} \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/targets/{target_id} HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/targets/{target_id}

Get Inspection Target by ID

Get a single inspection target using the inspection target ID.

Parameters

Name	In	Type	Required	Description
target_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "preprocessing": [
      {
        "active": true,
        "name": "string",
        "config": "string"
      }
    ],
    "name": "string",
    "target_id": 0
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	InspectionTargetWrapper
default	Default	Default error response	BaseError

setInspectionTargetById

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/targets/{target_id} \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/targets/{target_id} HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

POST /api/v1/targets/{target_id}

Set Default Inspection Target by ID

Set the default inspection target using the inspection target ID.

Parameters

Name	In	Type	Required	Description
target_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "preprocessing": [
      {
        "active": true,
        "name": "string",
        "config": "string"
      }
    ],
    "name": "string",
    "target_id": 0
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	InspectionTargetWrapper
default	Default	Default error response	BaseError

getModelResults

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/targets/model/results \
  -H 'Accept: application/octet-stream' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/targets/model/results HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/octet-stream

POST /api/v1/targets/model/results

Parameters

Name	In	Type	Required	Description
model_ids	query	array[string]	false	none

Example responses

200 Response

422 Response

{
  "errors": {
    "message": [
      {
        "key": "string",
        "errors": [
          "string"
        ]
      }
    ],
    "status": 0,
    "detail": "string",
    "source": {
      "pointer": "string"
    },
    "title": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	CSV file containing the model/s results	string
422	Unprocessable Entity	Unprocessable Entity	BaseError

algorithms

Algorithm related operations

アルゴリズムを取得する

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/algorithms/ \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/algorithms/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/algorithms/

Get List of Algorithms

学習に使用できるアルゴリズムと使用できるパラメータの一覧をリストで返します。
アルゴリズムの説明について、こちらをご覧ください。

Example responses

200 Response

{
  "data": [
    {
      "type": "string",
      "algorithm_id": "string",
      "version": "string",
      "updated_at": "2022-03-03T06:48:30Z",
      "file_name": "string",
      "class_name": "string",
      "parameters": [
        {
          "value": "string",
          "data_type": "string",
          "algorithm_id": "string",
          "updated_at": "2022-03-03T06:48:30Z",
          "algorithm_parameter_id": 0,
          "display_name": "string",
          "range_lower": "string",
          "created_at": "2022-03-03T06:48:30Z",
          "model_parameter": true,
          "name": "string",
          "range_upper": "string"
        }
      ],
      "created_at": "2022-03-03T06:48:30Z",
      "name": "string",
      "attribute": "string"
    }
  ]
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	AlgorithmWithParameterListWrapper
default	Default	Default error response	BaseError

getAlgorithmById

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/algorithms/{algorithm_id} \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/algorithms/{algorithm_id} HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/algorithms/{algorithm_id}

Get Algorithm by ID

Get a single algorithm by ID

Parameters

Name	In	Type	Required	Description
algorithm_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "type": "string",
    "algorithm_id": "string",
    "version": "string",
    "updated_at": "2022-03-03T06:48:30Z",
    "file_name": "string",
    "class_name": "string",
    "parameters": [
      {
        "value": "string",
        "data_type": "string",
        "algorithm_id": "string",
        "updated_at": "2022-03-03T06:48:30Z",
        "algorithm_parameter_id": 0,
        "display_name": "string",
        "range_lower": "string",
        "created_at": "2022-03-03T06:48:30Z",
        "model_parameter": true,
        "name": "string",
        "range_upper": "string"
      }
    ],
    "created_at": "2022-03-03T06:48:30Z",
    "name": "string",
    "attribute": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	AlgorithmWithParameterWrapper
default	Default	Default error response	BaseError

学習

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/algorithms/{algorithm_id}/train \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/algorithms/{algorithm_id}/train HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/json

POST /api/v1/algorithms/{algorithm_id}/train

Train

推論に使用するモデルを作成します。
モデルの学習に使用するデータIDの配列を含むオプションのパラメータを入力します。
アルゴリズムの設定を調整するため%{algorithmParameter}の配列を入力します。
新しく作成したモデルを確認するには、%{getModels}メソッドを使用します。

Body parameter

{
  "name": "string",
  "item_ids": [
    "string"
  ],
  "override_params": [
    {
      "algorithm_parameter_id": 0,
      "value": "string"
    }
  ]
}

Parameters

Name	In	Type	Required	Description
body	body	TrainRequest	true	none
algorithm_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "algorithm_type": "string",
    "image_url": "string",
    "algorithm_id": "string",
    "version": "string",
    "ok_img_count": 0,
    "precision": 0,
    "recall": 0,
    "status_text": "string",
    "active": true,
    "training_time": 0,
    "updated_at": "2022-03-03T06:48:30Z",
    "ng_img_count": 0,
    "context_id": 0,
    "model_id": "string",
    "stats": {
      "accuracy": 0,
      "precision": {
        "projected": 0,
        "computed": 0
      },
      "recall": {
        "projected": 0,
        "computed": 0
      }
    },
    "created_at": "2022-03-03T06:48:30Z",
    "status": "string",
    "name": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	HModelWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

models

Model related operations

モデルを取得する

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/models/ \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/models/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/models/

Get List of Models

データを推論するのに使用できる作成済みモデルのリストを取得します。
モデルIDが渡されなかった場合（%{serve}メソッドを使用して）、activeが%{true}のモデルをデフォルトとして使用します。
statusキーは、モデルがactiveかfailedかを示します。
- statusが activeのモデルは正常に作成され、予測に使用できます
- statusが failedのモデルは作成に失敗し、予測には使用できません

Parameters

Name	In	Type	Required	Description
algorithm_id	query	string	false	none

Example responses

200 Response

{
  "data": [
    {
      "algorithm_type": "string",
      "image_url": "string",
      "algorithm_id": "string",
      "version": "string",
      "ok_img_count": 0,
      "precision": 0,
      "recall": 0,
      "status_text": "string",
      "active": true,
      "training_time": 0,
      "updated_at": "2022-03-03T06:48:30Z",
      "ng_img_count": 0,
      "context_id": 0,
      "model_id": "string",
      "stats": {
        "accuracy": 0,
        "precision": {
          "projected": 0,
          "computed": 0
        },
        "recall": {
          "projected": 0,
          "computed": 0
        }
      },
      "created_at": "2022-03-03T06:48:30Z",
      "status": "string",
      "name": "string"
    }
  ]
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	HModelListWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

deleteModels

Code samples

# You can also use wget
curl -X DELETE https://sdd-api-staging.hacarus.com/api/v1/models/ \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

DELETE https://sdd-api-staging.hacarus.com/api/v1/models/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

DELETE /api/v1/models/

Delete Models

Delete models by id (list)

Parameters

Name	In	Type	Required	Description
model_ids	query	array[string]	false	none

Example responses

200 Response

{
  "data": {
    "model_ids": [
      "string"
    ]
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	ModelIdsQueryWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

getModelById

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/models/{model_id} \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/models/{model_id} HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/models/{model_id}

Get Model by ID

Get a single model by ID

Parameters

Name	In	Type	Required	Description
model_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "algorithm_type": "string",
    "image_url": "string",
    "algorithm_id": "string",
    "version": "string",
    "ok_img_count": 0,
    "precision": 0,
    "recall": 0,
    "status_text": "string",
    "active": true,
    "training_time": 0,
    "updated_at": "2022-03-03T06:48:30Z",
    "ng_img_count": 0,
    "context_id": 0,
    "model_id": "string",
    "stats": {
      "accuracy": 0,
      "precision": {
        "projected": 0,
        "computed": 0
      },
      "recall": {
        "projected": 0,
        "computed": 0
      }
    },
    "created_at": "2022-03-03T06:48:30Z",
    "status": "string",
    "name": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	HModelWrapper
default	Default	Default error response	BaseError

exportModelsWithId

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/models/export \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/zip' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/models/export HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/zip

POST /api/v1/models/export

Export models

Export models given the Id

Body parameter

{
  "model_ids": [
    "string"
  ]
}

Parameters

Name	In	Type	Required	Description
body	body	ModelIdsQuery	true	none

Example responses

200 Response

422 Response

{
  "errors": {
    "message": [
      {
        "key": "string",
        "errors": [
          "string"
        ]
      }
    ],
    "status": 0,
    "detail": "string",
    "source": {
      "pointer": "string"
    },
    "title": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	Zip file containing the model/s	string
422	Unprocessable Entity	Unprocessable Entity	BaseError

importModels

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/models/import \
  -H 'Content-Type: multipart/form-data' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/models/import HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: multipart/form-data
Accept: application/json

POST /api/v1/models/import

Import models

Import models from the zip file

Body parameter

models: string

Parameters

Name	In	Type	Required	Description
body	body	ImportModels	true	none

Example responses

200 Response

{
  "data": [
    {
      "algorithm_type": "string",
      "image_url": "string",
      "algorithm_id": "string",
      "version": "string",
      "ok_img_count": 0,
      "precision": 0,
      "recall": 0,
      "status_text": "string",
      "active": true,
      "training_time": 0,
      "updated_at": "2022-03-03T06:48:30Z",
      "ng_img_count": 0,
      "context_id": 0,
      "model_id": "string",
      "stats": {
        "accuracy": 0,
        "precision": {
          "projected": 0,
          "computed": 0
        },
        "recall": {
          "projected": 0,
          "computed": 0
        }
      },
      "created_at": "2022-03-03T06:48:30Z",
      "status": "string",
      "name": "string"
    }
  ]
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	HModelListWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

setModelPredictParameters

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/models/{model_id}/update_predict_parameters \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/models/{model_id}/update_predict_parameters HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/json

POST /api/v1/models/{model_id}/update_predict_parameters

Set Prediction Parameters

Updates predict parameters

Body parameter

{
  "threshold": 0,
  "ignore_outer": "string",
  "min_detected_area": 0
}

Parameters

Name	In	Type	Required	Description
body	body	UpdatePredictParameters	true	none
model_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "threshold": 0,
    "ignore_outer": "string",
    "min_detected_area": 0
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	UpdatePredictParametersResponseWrapper
400	Bad Request	Invalid parameters	BaseError
default	Default	Default error response	BaseError

setModelPredictCallback

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/models/{model_id}/update_predict_callback \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/models/{model_id}/update_predict_callback HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/json

POST /api/v1/models/{model_id}/update_predict_callback

Set Prediction Callback URL

Updates the callback URL for prediction. Once this endpoint has been set, a successful prediction will do a POST or GET operation to the callback URL with the data PredictionItemCallbackResponse

Body parameter

{
  "callback_url": "string",
  "callback_method": "string"
}

Parameters

Name	In	Type	Required	Description
body	body	UpdatePredictCallback	true	none
model_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "callback_url": "string",
    "callback_method": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	UpdatePredictCallbackResponseWrapper
400	Bad Request	Invalid parameters	BaseError
default	Default	Default error response	BaseError

predictItems

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/models/{model_id}/predict \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/models/{model_id}/predict HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/json

POST /api/v1/models/{model_id}/predict

Predict Items

Predict items in the background

Body parameter

{
  "item_ids": [
    "string"
  ]
}

Parameters

Name	In	Type	Required	Description
body	body	Predict	true	none
model_id	path	string	true	The model id, set to "active" to use current active model

Example responses

200 Response

{
  "data": {
    "model_version": "string",
    "item_ids": [
      "string"
    ]
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	PredictItemsResponseWrapper
400	Bad Request	model_id not found	BaseError
404	Not Found	item_id not found	BaseError
425	Unknown	The server is busy processing requests. Try again in a while.	BaseError
default	Default	Default error response	BaseError

items

Item related operations

getItemsByType

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/items/?item_type=string \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/items/?item_type=string HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/items/

Get List of Items

Returns items based on given item_type (prediction, training)

Parameters

Name	In	Type	Required	Description
count	query	integer(int32)	false	none
page	query	integer(int32)	false	none
model_id	query	string	false	none
item_type	query	string	true	none

Example responses

200 Response

{
  "data": {
    "end_page": 0,
    "current_page": 0,
    "count_per_page": 0,
    "items": [
      {
        "active": true,
        "is_training_data": true,
        "user_assessment": {
          "result": "string",
          "annotations": [
            {
              "y_min": 0,
              "x_max": 0,
              "y_max": 0,
              "label": 0,
              "defect_size": 0,
              "created_at": "2022-03-03T06:48:30Z",
              "x_min": 0,
              "annotation_id": 0,
              "notes": "string"
            }
          ],
          "label": "string",
          "images": [
            {
              "url": "string",
              "width": 0,
              "file_size": 0,
              "height": 0,
              "name": "string"
            }
          ]
        },
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ],
        "item_id": "string",
        "assessments": [
          {
            "annotations": [
              {
                "y_min": 0,
                "x_max": 0,
                "y_max": 0,
                "label": 0,
                "defect_size": 0,
                "created_at": "2022-03-03T06:48:30Z",
                "x_min": 0,
                "annotation_id": 0,
                "notes": "string"
              }
            ],
            "label": "string",
            "images": [
              {
                "url": "string",
                "width": 0,
                "file_size": 0,
                "height": 0,
                "name": "string"
              }
            ],
            "anomaly_score": "string",
            "model_id": "string",
            "detected_objects": 0,
            "detection_accuracy": 0,
            "result": "string"
          }
        ],
        "status": "string",
        "name": "string"
      }
    ]
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	GetItemsResponseWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

データを追加する

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/items/ \
  -H 'Content-Type: multipart/form-data' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/items/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: multipart/form-data
Accept: application/json

POST /api/v1/items/

Add Item

このメソッドを使用して、学習のためのデータをアップロードしてラベルを付けます。
データに良品または不良品としてラベルを付ける場合は、%{isGood}パラメータをブール値で%{true}（良品）または%{false}（不良品）に設定します。

Body parameter

files:
  - string
training: true
good: true
model_id: string

Parameters

Name	In	Type	Required	Description
body	body	AddItemRequest	true	none

Example responses

201 Response

{
  "data": {
    "item_ids": [
      "string"
    ]
  }
}

Responses

Status	Meaning	Description	Schema
201	Created	CREATED	AddItemResponseWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

delItem

Code samples

# You can also use wget
curl -X DELETE https://sdd-api-staging.hacarus.com/api/v1/items/ \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

DELETE https://sdd-api-staging.hacarus.com/api/v1/items/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/json

DELETE /api/v1/items/

Body parameter

{
  "item_ids": [
    "string"
  ]
}

Parameters

Name	In	Type	Required	Description
body	body	DelItemRequest	true	none

Example responses

422 Response

{
  "errors": {
    "message": [
      {
        "key": "string",
        "errors": [
          "string"
        ]
      }
    ],
    "status": 0,
    "detail": "string",
    "source": {
      "pointer": "string"
    },
    "title": "string"
  }
}

Responses

Status	Meaning	Description	Schema
422	Unprocessable Entity	Unprocessable Entity	BaseError

getItemById

Code samples

# You can also use wget
curl -X GET https://sdd-api-staging.hacarus.com/api/v1/items/{item_id} \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

GET https://sdd-api-staging.hacarus.com/api/v1/items/{item_id} HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

GET /api/v1/items/{item_id}

Get Item by Id

Get specific item and annotations

Parameters

Name	In	Type	Required	Description
show_assessments	query	boolean	false	none
show_annotations	query	boolean	false	none
model_id	query	string	false	none
item_id	path	string	true	none

Example responses

200 Response

{
  "data": {
    "active": true,
    "is_training_data": true,
    "user_assessment": {
      "result": "string",
      "annotations": [
        {
          "y_min": 0,
          "x_max": 0,
          "y_max": 0,
          "label": 0,
          "defect_size": 0,
          "created_at": "2022-03-03T06:48:30Z",
          "x_min": 0,
          "annotation_id": 0,
          "notes": "string"
        }
      ],
      "label": "string",
      "images": [
        {
          "url": "string",
          "width": 0,
          "file_size": 0,
          "height": 0,
          "name": "string"
        }
      ]
    },
    "images": [
      {
        "url": "string",
        "width": 0,
        "file_size": 0,
        "height": 0,
        "name": "string"
      }
    ],
    "item_id": "string",
    "assessments": [
      {
        "annotations": [
          {
            "y_min": 0,
            "x_max": 0,
            "y_max": 0,
            "label": 0,
            "defect_size": 0,
            "created_at": "2022-03-03T06:48:30Z",
            "x_min": 0,
            "annotation_id": 0,
            "notes": "string"
          }
        ],
        "label": "string",
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ],
        "anomaly_score": "string",
        "model_id": "string",
        "detected_objects": 0,
        "detection_accuracy": 0,
        "result": "string"
      }
    ],
    "status": "string",
    "name": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	ItemWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError

annotate

Code samples

# You can also use wget
curl -X PATCH https://sdd-api-staging.hacarus.com/api/v1/items/{item_id} \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

PATCH https://sdd-api-staging.hacarus.com/api/v1/items/{item_id} HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: application/json
Accept: application/json

PATCH /api/v1/items/{item_id}

Body parameter

{
  "annotations": [
    {
      "x_max": 0,
      "x_min": 0,
      "y_max": 0,
      "y_min": 0,
      "label": 0,
      "notes": "string"
    }
  ]
}

Parameters

Name	In	Type	Required	Description
body	body	AnnotateItemRequest	true	none
item_id	path	string	true	none

Example responses

422 Response

{
  "errors": {
    "message": [
      {
        "key": "string",
        "errors": [
          "string"
        ]
      }
    ],
    "status": 0,
    "detail": "string",
    "source": {
      "pointer": "string"
    },
    "title": "string"
  }
}

Responses

Status	Meaning	Description	Schema
202	Accepted	ACCEPTED	None
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

utils

Support and utility functions

setLicense

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/utils/ \
  -H 'Content-Type: multipart/form-data' \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/utils/ HTTP/1.1
Host: sdd-api-staging.hacarus.com
Content-Type: multipart/form-data
Accept: application/json

POST /api/v1/utils/

Body parameter

license_data: string

Parameters

Name	In	Type	Required	Description
body	body	License	true	none

Example responses

200 Response

{
  "data": {
    "license_expiry": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	LicenseResponseWrapper
422	Unprocessable Entity	Unprocessable Entity	BaseError
default	Default	Default error response	BaseError

getLogs

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/utils/debug \
  -H 'Accept: application/octet-stream' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/utils/debug HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/octet-stream

POST /api/v1/utils/debug

Example responses

200 Response

Responses

Status	Meaning	Description	Schema
200	OK	ZIP file containing the debug logs	string

getHashId

Code samples

# You can also use wget
curl -X POST https://sdd-api-staging.hacarus.com/api/v1/utils/node_hash \
  -H 'Accept: application/json' \
  -H 'Authorization: Bearer {access-token}'

POST https://sdd-api-staging.hacarus.com/api/v1/utils/node_hash HTTP/1.1
Host: sdd-api-staging.hacarus.com
Accept: application/json

POST /api/v1/utils/node_hash

Example responses

200 Response

{
  "data": {
    "hash_id": "string"
  }
}

Responses

Status	Meaning	Description	Schema
200	OK	OK	LicenseHashResponseWrapper
default	Default	Default error response	BaseError

token

API authentication operations

Code samples

LoginParams loginParams = new LoginParams(
  clientId: "client id",
  clientSecret: "client secret",
  grantType: "client_credentials"
);
var resp = instance.Login(loginParams);

POST /api/auth/token

Authorization

Returns the access token

Body parameter

{
  "grant_type": "client_credentials",
  "client_secret": "client_secret",
  "client_id": "client_id"
}

Name	In	Type	Required	Description
body	body	LoginParams	true	none

Example responses

200 Response

{
  "access_token": "access_token"
}

Status	Meaning	Description	Schema
200	OK	Successful login	LoginResponse
401	Unauthorized	Invalid credentials	BaseError
default	Default	default error handler	BaseError

Schemas

KeyValueError

{
  "key": "string",
  "errors": [
    "string"
  ]
}

Properties

Name	Type	Required	Restrictions	Description
key	string	false	none	Error key
errors	[string]	false	none	none

ErrorSource

{
  "pointer": "string"
}

Properties

Name	Type	Required	Restrictions	Description
pointer	string	false	none	Endpoint errored

Error

{
  "message": [
    {
      "key": "string",
      "errors": [
        "string"
      ]
    }
  ],
  "status": 0,
  "detail": "string",
  "source": {
    "pointer": "string"
  },
  "title": "string"
}

Properties

Name	Type	Required	Restrictions	Description
message	[KeyValueError]	false	none	none
status	integer(int32)	false	none	Error status code
detail	string	false	none	Error Details
source	ErrorSource	false	none	Error Source
title	string	false	none	Error Title

BaseError

{
  "errors": {
    "message": [
      {
        "key": "string",
        "errors": [
          "string"
        ]
      }
    ],
    "status": 0,
    "detail": "string",
    "source": {
      "pointer": "string"
    },
    "title": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
errors	Error	false	none	none

LoginResponse

{
  "access_token": "access_token"
}

response object of login

Properties

Name	Type	Required	Restrictions	Description
access_token	string	true	none	none

LoginParams

{
  "grant_type": "client_credentials",
  "client_secret": "client_secret",
  "client_id": "client_id"
}

Properties

Name	Type	Required	Restrictions	Description
client_id	string	true	none	none
client_secret	string	true	none	none
grant_type	string	true	none	none

Preprocessing

{
  "active": true,
  "name": "string",
  "config": "string"
}

Properties

Name	Type	Required	Restrictions	Description
active	boolean¦null	false	none	none
name	string	true	none	none
config	string¦null	false	none	none

InspectionTarget

{
  "preprocessing": [
    {
      "active": true,
      "name": "string",
      "config": "string"
    }
  ],
  "name": "string",
  "target_id": 0
}

Properties

Name	Type	Required	Restrictions	Description
preprocessing	[Preprocessing]	false	none	none
name	string	false	none	none
target_id	integer(int32)	false	none	none

InspectionTargetListWrapper

{
  "data": [
    {
      "preprocessing": [
        {
          "active": true,
          "name": "string",
          "config": "string"
        }
      ],
      "name": "string",
      "target_id": 0
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
data	[InspectionTarget]	false	none	none

InspectionTargetWrapper

{
  "data": {
    "preprocessing": [
      {
        "active": true,
        "name": "string",
        "config": "string"
      }
    ],
    "name": "string",
    "target_id": 0
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	InspectionTarget	false	none	none

AlgorithmParameter

{
  "value": "string",
  "data_type": "string",
  "algorithm_id": "string",
  "updated_at": "2022-03-03T06:48:30Z",
  "algorithm_parameter_id": 0,
  "display_name": "string",
  "range_lower": "string",
  "created_at": "2022-03-03T06:48:30Z",
  "model_parameter": true,
  "name": "string",
  "range_upper": "string"
}

Properties

Name	Type	Required	Restrictions	Description
value	string¦null	false	none	none
data_type	string	false	none	none
algorithm_id	string	true	none	none
updated_at	string(date-time)	true	none	none
algorithm_parameter_id	integer(int32)	false	none	none
display_name	string¦null	false	none	none
range_lower	string¦null	false	none	none
created_at	string(date-time)	true	none	none
model_parameter	boolean¦null	false	none	none
name	string	true	none	none
range_upper	string¦null	false	none	none

AlgorithmWithParameter

{
  "type": "string",
  "algorithm_id": "string",
  "version": "string",
  "updated_at": "2022-03-03T06:48:30Z",
  "file_name": "string",
  "class_name": "string",
  "parameters": [
    {
      "value": "string",
      "data_type": "string",
      "algorithm_id": "string",
      "updated_at": "2022-03-03T06:48:30Z",
      "algorithm_parameter_id": 0,
      "display_name": "string",
      "range_lower": "string",
      "created_at": "2022-03-03T06:48:30Z",
      "model_parameter": true,
      "name": "string",
      "range_upper": "string"
    }
  ],
  "created_at": "2022-03-03T06:48:30Z",
  "name": "string",
  "attribute": "string"
}

Properties

Name	Type	Required	Restrictions	Description
type	string	false	none	none
algorithm_id	string	true	none	none
version	string¦null	false	none	none
updated_at	string(date-time)	true	none	none
file_name	string	true	none	none
class_name	string	true	none	none
parameters	[AlgorithmParameter]	false	none	none
created_at	string(date-time)	true	none	none
name	string	true	none	none
attribute	string	false	none	none

AlgorithmWithParameterListWrapper

{
  "data": [
    {
      "type": "string",
      "algorithm_id": "string",
      "version": "string",
      "updated_at": "2022-03-03T06:48:30Z",
      "file_name": "string",
      "class_name": "string",
      "parameters": [
        {
          "value": "string",
          "data_type": "string",
          "algorithm_id": "string",
          "updated_at": "2022-03-03T06:48:30Z",
          "algorithm_parameter_id": 0,
          "display_name": "string",
          "range_lower": "string",
          "created_at": "2022-03-03T06:48:30Z",
          "model_parameter": true,
          "name": "string",
          "range_upper": "string"
        }
      ],
      "created_at": "2022-03-03T06:48:30Z",
      "name": "string",
      "attribute": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
data	[AlgorithmWithParameter]	false	none	none

AlgorithmWithParameterWrapper

{
  "data": {
    "type": "string",
    "algorithm_id": "string",
    "version": "string",
    "updated_at": "2022-03-03T06:48:30Z",
    "file_name": "string",
    "class_name": "string",
    "parameters": [
      {
        "value": "string",
        "data_type": "string",
        "algorithm_id": "string",
        "updated_at": "2022-03-03T06:48:30Z",
        "algorithm_parameter_id": 0,
        "display_name": "string",
        "range_lower": "string",
        "created_at": "2022-03-03T06:48:30Z",
        "model_parameter": true,
        "name": "string",
        "range_upper": "string"
      }
    ],
    "created_at": "2022-03-03T06:48:30Z",
    "name": "string",
    "attribute": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	AlgorithmWithParameter	false	none	none

OverrideParam

{
  "algorithm_parameter_id": 0,
  "value": "string"
}

Properties

Name	Type	Required	Restrictions	Description
algorithm_parameter_id	integer(int32)	false	none	none
value	string	false	none	none

TrainRequest

{
  "name": "string",
  "item_ids": [
    "string"
  ],
  "override_params": [
    {
      "algorithm_parameter_id": 0,
      "value": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
name	string	false	none	none
item_ids	[string]	false	none	none
override_params	[OverrideParam]	false	none	none

HModelPrecision

{
  "projected": 0,
  "computed": 0
}

Properties

Name	Type	Required	Restrictions	Description
projected	number¦null	false	none	none
computed	number¦null	false	none	none

HModelStats

{
  "accuracy": 0,
  "precision": {
    "projected": 0,
    "computed": 0
  },
  "recall": {
    "projected": 0,
    "computed": 0
  }
}

Properties

Name	Type	Required	Restrictions	Description
accuracy	number¦null	false	none	none
precision	HModelPrecision	false	none	none
recall	HModelPrecision	false	none	none

HModel

{
  "algorithm_type": "string",
  "image_url": "string",
  "algorithm_id": "string",
  "version": "string",
  "ok_img_count": 0,
  "precision": 0,
  "recall": 0,
  "status_text": "string",
  "active": true,
  "training_time": 0,
  "updated_at": "2022-03-03T06:48:30Z",
  "ng_img_count": 0,
  "context_id": 0,
  "model_id": "string",
  "stats": {
    "accuracy": 0,
    "precision": {
      "projected": 0,
      "computed": 0
    },
    "recall": {
      "projected": 0,
      "computed": 0
    }
  },
  "created_at": "2022-03-03T06:48:30Z",
  "status": "string",
  "name": "string"
}

Properties

Name	Type	Required	Restrictions	Description
algorithm_type	string	true	none	none
image_url	string¦null	false	none	none
algorithm_id	string	true	none	none
version	string	true	none	none
ok_img_count	integer(int32)	false	none	none
precision	number¦null	false	none	none
recall	number¦null	false	none	none
status_text	string¦null	false	none	none
active	boolean¦null	false	none	none
training_time	integer(int32)¦null	false	none	none
updated_at	string(date-time)	true	none	none
ng_img_count	integer(int32)	false	none	none
context_id	integer(int32)	true	none	none
model_id	string	true	none	none
stats	HModelStats	false	none	none
created_at	string(date-time)	true	none	none
status	string	false	none	none
name	string	true	none	none

HModelWrapper

{
  "data": {
    "algorithm_type": "string",
    "image_url": "string",
    "algorithm_id": "string",
    "version": "string",
    "ok_img_count": 0,
    "precision": 0,
    "recall": 0,
    "status_text": "string",
    "active": true,
    "training_time": 0,
    "updated_at": "2022-03-03T06:48:30Z",
    "ng_img_count": 0,
    "context_id": 0,
    "model_id": "string",
    "stats": {
      "accuracy": 0,
      "precision": {
        "projected": 0,
        "computed": 0
      },
      "recall": {
        "projected": 0,
        "computed": 0
      }
    },
    "created_at": "2022-03-03T06:48:30Z",
    "status": "string",
    "name": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	HModel	false	none	none

HModelListWrapper

{
  "data": [
    {
      "algorithm_type": "string",
      "image_url": "string",
      "algorithm_id": "string",
      "version": "string",
      "ok_img_count": 0,
      "precision": 0,
      "recall": 0,
      "status_text": "string",
      "active": true,
      "training_time": 0,
      "updated_at": "2022-03-03T06:48:30Z",
      "ng_img_count": 0,
      "context_id": 0,
      "model_id": "string",
      "stats": {
        "accuracy": 0,
        "precision": {
          "projected": 0,
          "computed": 0
        },
        "recall": {
          "projected": 0,
          "computed": 0
        }
      },
      "created_at": "2022-03-03T06:48:30Z",
      "status": "string",
      "name": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
data	[HModel]	false	none	none

ModelIdsQuery

{
  "model_ids": [
    "string"
  ]
}

Properties

Name	Type	Required	Restrictions	Description
model_ids	[string]	false	none	none

ModelIdsQueryWrapper

{
  "data": {
    "model_ids": [
      "string"
    ]
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	ModelIdsQuery	false	none	none

ImportModels

{
  "models": "string"
}

Properties

Name	Type	Required	Restrictions	Description
models	string(binary)	false	none	none

UpdatePredictParameters

{
  "threshold": 0,
  "ignore_outer": "string",
  "min_detected_area": 0
}

Properties

Name	Type	Required	Restrictions	Description
threshold	number(float)	false	none	none
ignore_outer	string	false	none	none
min_detected_area	number(float)	false	none	none

UpdatePredictParametersResponse

{
  "threshold": 0,
  "ignore_outer": "string",
  "min_detected_area": 0
}

Properties

Name	Type	Required	Restrictions	Description
threshold	number(float)	false	none	none
ignore_outer	string	false	none	none
min_detected_area	number(float)	false	none	none

UpdatePredictParametersResponseWrapper

{
  "data": {
    "threshold": 0,
    "ignore_outer": "string",
    "min_detected_area": 0
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	UpdatePredictParametersResponse	false	none	none

UpdatePredictCallback

{
  "callback_url": "string",
  "callback_method": "string"
}

Properties

Name	Type	Required	Restrictions	Description
callback_url	string	false	none	The callback URL that HACARUS CHECK will call to send prediction details after a prediction call has been completed.
callback_method	string	false	none	The method in which the callback will be called. Valid values are 'post' and 'get'

UpdatePredictCallbackResponse

{
  "callback_url": "string",
  "callback_method": "string"
}

Properties

Name	Type	Required	Restrictions	Description
callback_url	string	false	none	none
callback_method	string	false	none	none

UpdatePredictCallbackResponseWrapper

{
  "data": {
    "callback_url": "string",
    "callback_method": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	UpdatePredictCallbackResponse	false	none	none

Predict

{
  "item_ids": [
    "string"
  ]
}

Properties

Name	Type	Required	Restrictions	Description
item_ids	[string]	false	none	none

PredictItemsResponse

{
  "model_version": "string",
  "item_ids": [
    "string"
  ]
}

Properties

Name	Type	Required	Restrictions	Description
model_version	string	false	none	none
item_ids	[string]	false	none	none

PredictItemsResponseWrapper

{
  "data": {
    "model_version": "string",
    "item_ids": [
      "string"
    ]
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	PredictItemsResponse	false	none	none

Annotation

{
  "y_min": 0,
  "x_max": 0,
  "y_max": 0,
  "label": 0,
  "defect_size": 0,
  "created_at": "2022-03-03T06:48:30Z",
  "x_min": 0,
  "annotation_id": 0,
  "notes": "string"
}

Properties

Name	Type	Required	Restrictions	Description
y_min	number¦null	false	none	none
x_max	number¦null	false	none	none
y_max	number¦null	false	none	none
label	integer(int32)¦null	false	none	none
defect_size	number¦null	false	none	none
created_at	string(date-time)	true	none	none
x_min	number¦null	false	none	none
annotation_id	integer(int32)	false	none	none
notes	string¦null	false	none	none

ItemImageMeta

{
  "url": "string",
  "width": 0,
  "file_size": 0,
  "height": 0,
  "name": "string"
}

Properties

Name	Type	Required	Restrictions	Description
url	string	false	none	none
width	integer(int32)	false	none	none
file_size	integer(int32)	false	none	none
height	integer(int32)	false	none	none
name	string	false	none	none

AssessmentItem

{
  "result": "string",
  "annotations": [
    {
      "y_min": 0,
      "x_max": 0,
      "y_max": 0,
      "label": 0,
      "defect_size": 0,
      "created_at": "2022-03-03T06:48:30Z",
      "x_min": 0,
      "annotation_id": 0,
      "notes": "string"
    }
  ],
  "label": "string",
  "images": [
    {
      "url": "string",
      "width": 0,
      "file_size": 0,
      "height": 0,
      "name": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
result	string	false	none	none
annotations	[Annotation]	false	none	none
label	string	false	none	none
images	[ItemImageMeta]	false	none	none

AssessmentItem1

{
  "annotations": [
    {
      "y_min": 0,
      "x_max": 0,
      "y_max": 0,
      "label": 0,
      "defect_size": 0,
      "created_at": "2022-03-03T06:48:30Z",
      "x_min": 0,
      "annotation_id": 0,
      "notes": "string"
    }
  ],
  "label": "string",
  "images": [
    {
      "url": "string",
      "width": 0,
      "file_size": 0,
      "height": 0,
      "name": "string"
    }
  ],
  "anomaly_score": "string",
  "model_id": "string",
  "detected_objects": 0,
  "detection_accuracy": 0,
  "result": "string"
}

Properties

Name	Type	Required	Restrictions	Description
annotations	[Annotation]	false	none	none
label	string	false	none	none
images	[ItemImageMeta]	false	none	none
anomaly_score	string	false	none	none
model_id	string	false	none	none
detected_objects	integer(int32)¦null	false	none	none
detection_accuracy	integer(int32)¦null	false	none	none
result	string	false	none	none

Item

{
  "active": true,
  "is_training_data": true,
  "user_assessment": {
    "result": "string",
    "annotations": [
      {
        "y_min": 0,
        "x_max": 0,
        "y_max": 0,
        "label": 0,
        "defect_size": 0,
        "created_at": "2022-03-03T06:48:30Z",
        "x_min": 0,
        "annotation_id": 0,
        "notes": "string"
      }
    ],
    "label": "string",
    "images": [
      {
        "url": "string",
        "width": 0,
        "file_size": 0,
        "height": 0,
        "name": "string"
      }
    ]
  },
  "images": [
    {
      "url": "string",
      "width": 0,
      "file_size": 0,
      "height": 0,
      "name": "string"
    }
  ],
  "item_id": "string",
  "assessments": [
    {
      "annotations": [
        {
          "y_min": 0,
          "x_max": 0,
          "y_max": 0,
          "label": 0,
          "defect_size": 0,
          "created_at": "2022-03-03T06:48:30Z",
          "x_min": 0,
          "annotation_id": 0,
          "notes": "string"
        }
      ],
      "label": "string",
      "images": [
        {
          "url": "string",
          "width": 0,
          "file_size": 0,
          "height": 0,
          "name": "string"
        }
      ],
      "anomaly_score": "string",
      "model_id": "string",
      "detected_objects": 0,
      "detection_accuracy": 0,
      "result": "string"
    }
  ],
  "status": "string",
  "name": "string"
}

Properties

Name	Type	Required	Restrictions	Description
active	boolean	false	none	none
is_training_data	boolean	false	none	none
user_assessment	AssessmentItem	false	none	none
images	[ItemImageMeta]	false	none	none
item_id	string	false	none	none
assessments	[AssessmentItem1]	false	none	none
status	string	false	none	none
name	string	false	none	none

GetItemsResponse

{
  "end_page": 0,
  "current_page": 0,
  "count_per_page": 0,
  "items": [
    {
      "active": true,
      "is_training_data": true,
      "user_assessment": {
        "result": "string",
        "annotations": [
          {
            "y_min": 0,
            "x_max": 0,
            "y_max": 0,
            "label": 0,
            "defect_size": 0,
            "created_at": "2022-03-03T06:48:30Z",
            "x_min": 0,
            "annotation_id": 0,
            "notes": "string"
          }
        ],
        "label": "string",
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ]
      },
      "images": [
        {
          "url": "string",
          "width": 0,
          "file_size": 0,
          "height": 0,
          "name": "string"
        }
      ],
      "item_id": "string",
      "assessments": [
        {
          "annotations": [
            {
              "y_min": 0,
              "x_max": 0,
              "y_max": 0,
              "label": 0,
              "defect_size": 0,
              "created_at": "2022-03-03T06:48:30Z",
              "x_min": 0,
              "annotation_id": 0,
              "notes": "string"
            }
          ],
          "label": "string",
          "images": [
            {
              "url": "string",
              "width": 0,
              "file_size": 0,
              "height": 0,
              "name": "string"
            }
          ],
          "anomaly_score": "string",
          "model_id": "string",
          "detected_objects": 0,
          "detection_accuracy": 0,
          "result": "string"
        }
      ],
      "status": "string",
      "name": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
end_page	integer(int32)	false	none	none
current_page	integer(int32)	false	none	none
count_per_page	integer(int32)	false	none	none
items	[Item]	false	none	none

GetItemsResponseWrapper

{
  "data": {
    "end_page": 0,
    "current_page": 0,
    "count_per_page": 0,
    "items": [
      {
        "active": true,
        "is_training_data": true,
        "user_assessment": {
          "result": "string",
          "annotations": [
            {
              "y_min": 0,
              "x_max": 0,
              "y_max": 0,
              "label": 0,
              "defect_size": 0,
              "created_at": "2022-03-03T06:48:30Z",
              "x_min": 0,
              "annotation_id": 0,
              "notes": "string"
            }
          ],
          "label": "string",
          "images": [
            {
              "url": "string",
              "width": 0,
              "file_size": 0,
              "height": 0,
              "name": "string"
            }
          ]
        },
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ],
        "item_id": "string",
        "assessments": [
          {
            "annotations": [
              {
                "y_min": 0,
                "x_max": 0,
                "y_max": 0,
                "label": 0,
                "defect_size": 0,
                "created_at": "2022-03-03T06:48:30Z",
                "x_min": 0,
                "annotation_id": 0,
                "notes": "string"
              }
            ],
            "label": "string",
            "images": [
              {
                "url": "string",
                "width": 0,
                "file_size": 0,
                "height": 0,
                "name": "string"
              }
            ],
            "anomaly_score": "string",
            "model_id": "string",
            "detected_objects": 0,
            "detection_accuracy": 0,
            "result": "string"
          }
        ],
        "status": "string",
        "name": "string"
      }
    ]
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	GetItemsResponse	false	none	none

PredictionItemCallbackResponse

{
  "items": [
    {
      "active": true,
      "is_training_data": true,
      "user_assessment": {
        "result": "string",
        "annotations": [
          {
            "y_min": 0,
            "x_max": 0,
            "y_max": 0,
            "label": 0,
            "defect_size": 0,
            "created_at": "2022-03-03T06:48:30Z",
            "x_min": 0,
            "annotation_id": 0,
            "notes": "string"
          }
        ],
        "label": "string",
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ]
      },
      "images": [
        {
          "url": "string",
          "width": 0,
          "file_size": 0,
          "height": 0,
          "name": "string"
        }
      ],
      "item_id": "string",
      "assessments": [
        {
          "annotations": [
            {
              "y_min": 0,
              "x_max": 0,
              "y_max": 0,
              "label": 0,
              "defect_size": 0,
              "created_at": "2022-03-03T06:48:30Z",
              "x_min": 0,
              "annotation_id": 0,
              "notes": "string"
            }
          ],
          "label": "string",
          "images": [
            {
              "url": "string",
              "width": 0,
              "file_size": 0,
              "height": 0,
              "name": "string"
            }
          ],
          "anomaly_score": "string",
          "model_id": "string",
          "detected_objects": 0,
          "detection_accuracy": 0,
          "result": "string"
        }
      ],
      "status": "string",
      "name": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
items	[Item]	false	none	none

PredictionItemCallbackWrapper

{
  "data": {
    "items": [
      {
        "active": true,
        "is_training_data": true,
        "user_assessment": {
          "result": "string",
          "annotations": [
            {
              "y_min": 0,
              "x_max": 0,
              "y_max": 0,
              "label": 0,
              "defect_size": 0,
              "created_at": "2022-03-03T06:48:30Z",
              "x_min": 0,
              "annotation_id": 0,
              "notes": "string"
            }
          ],
          "label": "string",
          "images": [
            {
              "url": "string",
              "width": 0,
              "file_size": 0,
              "height": 0,
              "name": "string"
            }
          ]
        },
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ],
        "item_id": "string",
        "assessments": [
          {
            "annotations": [
              {
                "y_min": 0,
                "x_max": 0,
                "y_max": 0,
                "label": 0,
                "defect_size": 0,
                "created_at": "2022-03-03T06:48:30Z",
                "x_min": 0,
                "annotation_id": 0,
                "notes": "string"
              }
            ],
            "label": "string",
            "images": [
              {
                "url": "string",
                "width": 0,
                "file_size": 0,
                "height": 0,
                "name": "string"
              }
            ],
            "anomaly_score": "string",
            "model_id": "string",
            "detected_objects": 0,
            "detection_accuracy": 0,
            "result": "string"
          }
        ],
        "status": "string",
        "name": "string"
      }
    ]
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	PredictionItemCallbackResponse	false	none	none

AddItemRequest

{
  "files": [
    "string"
  ],
  "training": true,
  "good": true,
  "model_id": "string"
}

Properties

Name	Type	Required	Restrictions	Description
files	[string]	false	none	none
training	boolean	true	none	none
good	boolean	false	none	none
model_id	string	false	none	none

AddItemResponse

{
  "item_ids": [
    "string"
  ]
}

Properties

Name	Type	Required	Restrictions	Description
item_ids	[string]	false	none	none

AddItemResponseWrapper

{
  "data": {
    "item_ids": [
      "string"
    ]
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	AddItemResponse	false	none	none

DelItemRequest

{
  "item_ids": [
    "string"
  ]
}

Properties

Name	Type	Required	Restrictions	Description
item_ids	[string]	false	none	none

ItemWrapper

{
  "data": {
    "active": true,
    "is_training_data": true,
    "user_assessment": {
      "result": "string",
      "annotations": [
        {
          "y_min": 0,
          "x_max": 0,
          "y_max": 0,
          "label": 0,
          "defect_size": 0,
          "created_at": "2022-03-03T06:48:30Z",
          "x_min": 0,
          "annotation_id": 0,
          "notes": "string"
        }
      ],
      "label": "string",
      "images": [
        {
          "url": "string",
          "width": 0,
          "file_size": 0,
          "height": 0,
          "name": "string"
        }
      ]
    },
    "images": [
      {
        "url": "string",
        "width": 0,
        "file_size": 0,
        "height": 0,
        "name": "string"
      }
    ],
    "item_id": "string",
    "assessments": [
      {
        "annotations": [
          {
            "y_min": 0,
            "x_max": 0,
            "y_max": 0,
            "label": 0,
            "defect_size": 0,
            "created_at": "2022-03-03T06:48:30Z",
            "x_min": 0,
            "annotation_id": 0,
            "notes": "string"
          }
        ],
        "label": "string",
        "images": [
          {
            "url": "string",
            "width": 0,
            "file_size": 0,
            "height": 0,
            "name": "string"
          }
        ],
        "anomaly_score": "string",
        "model_id": "string",
        "detected_objects": 0,
        "detection_accuracy": 0,
        "result": "string"
      }
    ],
    "status": "string",
    "name": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	Item	false	none	none

RequestAnnotation

{
  "x_max": 0,
  "x_min": 0,
  "y_max": 0,
  "y_min": 0,
  "label": 0,
  "notes": "string"
}

Properties

Name	Type	Required	Restrictions	Description
x_max	integer(int32)¦null	false	none	none
x_min	integer(int32)¦null	false	none	none
y_max	integer(int32)¦null	false	none	none
y_min	integer(int32)¦null	false	none	none
label	integer(int32)¦null	false	none	none
notes	string	false	none	none

AnnotateItemRequest

{
  "annotations": [
    {
      "x_max": 0,
      "x_min": 0,
      "y_max": 0,
      "y_min": 0,
      "label": 0,
      "notes": "string"
    }
  ]
}

Properties

Name	Type	Required	Restrictions	Description
annotations	[RequestAnnotation]	false	none	none

License

{
  "license_data": "string"
}

Properties

Name	Type	Required	Restrictions	Description
license_data	string(binary)	false	none	none

LicenseResponse

{
  "license_expiry": "string"
}

Properties

Name	Type	Required	Restrictions	Description
license_expiry	string	false	none	none

LicenseResponseWrapper

{
  "data": {
    "license_expiry": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	LicenseResponse	false	none	none

LicenseHashResponse

{
  "hash_id": "string"
}

Properties

Name	Type	Required	Restrictions	Description
hash_id	string	false	none	none

LicenseHashResponseWrapper

{
  "data": {
    "hash_id": "string"
  }
}

Properties

Name	Type	Required	Restrictions	Description
data	LicenseHashResponse	false	none	none

アルゴリズム V0.5.1

アルゴリズム概要

Reconstructor Inspection
- アノテーション不要の教師なし学習アルゴリズム
- 画像の全体的なテクスチャに注目し、正常画像から逸脱しているかどうか判定する。
Local Pattern Inspection
- アノテーション不要の教師なし学習アルゴリズム。
- 画像の局所的な部分に注目し、正常画像から逸脱しているかどうか判定する。
Pattern Matching
- Classical pattern matching algorithm which compares an image against a golden image.
Patch Sample Inspection (Beta)
- A new, generic inspection algorithm that can handle inspection objects with different placement or rotation. We have seen good inspection results and very high inspection accuracy even with difficult inspection targets across customer projects.

アルゴリズム詳細

Reconstructor Inspection

正常画像のみから学習するアノテーション不要の教師なし学習モデル。正常な入力画像に対して辞書学習を行う。推論時には新たな画像に対して、再構成誤差を計算し、その大小に基づいて異常検知を行う。これによって、画像の全体的なテクスチャに注目し、正常画像から逸脱しているかどうかを判定することができる。画像の位置合わせや、画像のサイズなどは特に気にしなくてもよい。

パラメータ	変数名	型 : 範囲	既定値	計算時間への影響	精度への影響	備考
学習サイズ	train_size	int : [1, inf]	10000	大 -> 大	大 -> 大
最小の検査サイズ	min_inspection_size	(int, int) : 0~height, 0~width	(8, 8)	大 -> 大	-	(4, 4) or (8,8) or (16,16) が経験的によい。
最小の検査サイズ幅	min_inspection_step	int : [1, inf]	4	大 -> 小	大 -> 小	"min_inspection_sizeの約数が望ましい。1のとき一番計算負荷がかかるが、もっとも性能がよい。"
特徴量の次元	n_components	int : [1, inf]	10	大 -> 大	"大 -> ほとんど正常	特徴量の次元が大きいと複雑な対象でも再構成できるようになる。
アルゴリズム反復回数	max_iter	Int : [1, inf]	10	大 -> 大	-	少なすぎると性能が悪くなるが、ある程度大きければ、性能はそれほど変わらない。
再構成誤差の閾値	reconstruct_thresh	float : [0, inf]	2	-	"大 -> ほとんど正常小 -> ほとんど異常"	再構成誤差に対する閾値、1.5 ~ 3程度の値が経験的によい
検知領域の異常度の閾値	threshold	float : [0, inf]	0.1	-	-	大きすぎるとほとんど正常、小さすぎるとほとんど異常
無視される外周からの幅	ignore_outer	(int, int) : 0~height, 0~width	(0,0)	-	-	"外周からの無視する領域の幅画像の縦, 横のピクセル数までの値にする。"
最小の検知サイズ	min_detected_area	Float : [1, inf]	0	-	-	検出される矩形の面積の下限

Local Pattern Inspection

正常画像のみから学習するアノテーション不要の教師なし学習モデル。画像の局所的な部分の特徴に注目し、正常画像から逸脱しているかどうかを判定する。使用する際は多少の位置ズレは許容するが、基本的には全て位置合わせをする必要があり、画像サイズも統一しておかなければいけない。大幅な位置ズレや、回転に対しては誤分類の原因になる。

パラメータ	変数名	型 : 範囲	既定値	計算時間への影響	精度への影響	備考
最小の検査サイズ	min_inspection_size	(int, int) : 0~height, 0~width	(32, 32)	大 -> 大	-	(16, 16) or (32,32) or (64,64) が経験的によい。
最小の検査サイズ幅	min_inspection_step	int : [1, inf]	16	大 -> 小	大 -> 小	min_inspection_sizeの1/2か1/4が望ましい
検知領域の異常度の閾値	threshold	float : [0, inf]	0.001	-	-	大きすぎるとほとんど正常、小さすぎるとほとんど異常
無視される外周からの幅	ignore_outer	(int, int) : 0~height, 0~width	(0,0)	-	-	"外周からの無視する領域の幅画像の縦, 横のピクセル数までの値にする。"
最小の検知サイズ	min_detected_area	Float : [1, inf]	0	-	-	検出される矩形の面積の下限

Pattern Matching

We calculate the golden image from “OK” images (good samples, without defects, no annotation required) and compare it to reference. Hence, the position alignment is required (position variation is not allowed), and the size of all images must be the same. Misalignment and rotation might be a cause of misclassification.

Parameter	Variable name	Type: Range	Default	Impact on processing time	Impact on accuracy	Remarks
Minimum inspection size	min_inspection_size	(int, int) : 0~height, 0~width	(16, 16)	large -> large	-	(16, 16) or (32, 32) or (64, 64) are empirically good.
Minimum inspection step size	min_inspection_step	int : [1, inf]	4	large -> small		Half or quarter of min_inspection_size is desirable.
Threshold of anomaly	threshold	float : [0, inf] 0.1	-	large->more tolerant
Width from the perimeter to be ignored	ignore_outer	(int, int) : 0~height, 0~width	(0,0)	-	-	The width of the ignored area from the perimeter. Use values up to the number of vertical and horizontal pixels in the image.
Minimum detection size	min_detected_area	float : [1, inf]	0	-	-	Threshold of area of detected rectangle
Anomaly range	anomaly_range	float : [0, 1]	0	-	-	Used to normalize heatmap colors across results.

Pattern Sample Inspection (Beta)

A new, generic inspection algorithm that can handle inspection objects with different placement or rotation. We have seen good inspection results and very high inspection accuracy even with difficult inspection targets across customer projects.

Parameter	Variable name	Type: Range	Default	Impact on processing time	Impact on accuracy	Remarks
Model type	model_type	int : [0, 1]	0	small -> large	-	0 will use a small model. 1 will use a larger model.
Number of Features	num_features	int : n	100	large -> large	-	100-n are empirically good.
Threshold of anomaly	threshold	float : [0, inf] 0.1	-	large->more tolerant
Width from the perimeter to be ignored	ignore_outer	(int, int) : 0~height, 0~width	(0,0)	-	-	The width of the ignored area from the perimeter. Use values up to the number of vertical and horizontal pixels in the image.
Minimum detection size	min_detected_area	float : [1, inf]	0	-	-	Threshold area of detected rectangle
Anomaly range	anomaly_range	float : [0, 1]	0	-	-	Used to normalize heatmap colors across results.

C# SDK

Open

Python SDK

Open