DocumentAnalyzerSchema

Type: object
No Additional Properties

paragraphs Required

root paragraphs

Paragraphs

Type: array

List of detected paragraphs

No Additional Items

Each item of this array must be:

root paragraphs ParagraphSchema

ParagraphSchema

Type: object
No Additional Properties

box Required

root paragraphs ParagraphSchema box

Box

Type: array of integer

Bounding box of the paragraph in the format [x1, y1, x2, y2]

Must contain a minimum of 4 items

Must contain a maximum of 4 items

No Additional Items

Each item of this array must be:

root paragraphs ParagraphSchema box box items

Type: integer

contents Required

root paragraphs ParagraphSchema contents

Contents

Text content of the paragraph

Any of

• Option 1
• Option 2

root paragraphs ParagraphSchema contents anyOf item 0

Type: string

root paragraphs ParagraphSchema contents anyOf item 1

Type: null

direction Required

root paragraphs ParagraphSchema direction

Direction

Text direction, e.g., ['horizontal' or 'vertical']

Any of

• Option 1
• Option 2

root paragraphs ParagraphSchema direction anyOf item 0

Type: string

root paragraphs ParagraphSchema direction anyOf item 1

Type: null

order Required

root paragraphs ParagraphSchema order

Order

Order of the paragraph in the document

Any of

• Option 1
• Option 2

root paragraphs ParagraphSchema order anyOf item 0

Type: integer

root paragraphs ParagraphSchema order anyOf item 1

Type: null

role Required

root paragraphs ParagraphSchema role

Role

Role of the paragraph, e.g., ['sectionheadings', 'pageheader', 'page_footer'])

Any of

• Option 1
• Option 2

root paragraphs ParagraphSchema role anyOf item 0

Type: string

root paragraphs ParagraphSchema role anyOf item 1

Type: null

tables Required

root tables

Tables

Type: array

List of detected tables

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema

TableStructureRecognizerSchema

Type: object
No Additional Properties

box Required

root tables TableStructureRecognizerSchema box

Box

Type: array of integer

Bounding box of the table in the format [x1, y1, x2, y2]

Must contain a minimum of 4 items

Must contain a maximum of 4 items

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema box box items

Type: integer

n_row Required

root tables TableStructureRecognizerSchema n_row

N Row

Type: integer

Number of rows in the table

n_col Required

root tables TableStructureRecognizerSchema n_col

N Col

Type: integer

Number of columns in the table

rows Required

root tables TableStructureRecognizerSchema rows

Rows

Type: array

List of table lines representing rows

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema rows TableLineSchema

TableLineSchema

Type: object
No Additional Properties

box Required

root tables TableStructureRecognizerSchema rows TableLineSchema box

Box

Type: array of integer

Bounding box of the table line in the format [x1, y1, x2, y2]

Must contain a minimum of 4 items

Must contain a maximum of 4 items

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema rows TableLineSchema box box items

Type: integer

score Required

root tables TableStructureRecognizerSchema rows TableLineSchema score

Score

Type: number

Confidence score of the table line detection

cols Required

root tables TableStructureRecognizerSchema cols

Cols

Type: array

List of table lines representing columns

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema cols TableLineSchema

TableLineSchema

Type: object
Same definition as TableLineSchema

spans Required

root tables TableStructureRecognizerSchema spans

Spans

Type: array

List of table lines representing spans

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema spans TableLineSchema

TableLineSchema

Type: object
Same definition as TableLineSchema

cells Required

root tables TableStructureRecognizerSchema cells

Cells

Type: array

List of table cells

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema cells TableCellSchema

TableCellSchema

Type: object
No Additional Properties

col Required

root tables TableStructureRecognizerSchema cells TableCellSchema col

Col

Type: integer

Column index of the cell

row Required

root tables TableStructureRecognizerSchema cells TableCellSchema row

Row

Type: integer

Row index of the cell

col_span Required

root tables TableStructureRecognizerSchema cells TableCellSchema col_span

Col Span

Type: integer

Number of columns spanned by the cell

row_span Required

root tables TableStructureRecognizerSchema cells TableCellSchema row_span

Row Span

Type: integer

Number of rows spanned by the cell

box Required

root tables TableStructureRecognizerSchema cells TableCellSchema box

Box

Type: array of integer

Bounding box of the cell in the format [x1, y1, x2, y2]

Must contain a minimum of 4 items

Must contain a maximum of 4 items

No Additional Items

Each item of this array must be:

root tables TableStructureRecognizerSchema cells TableCellSchema box box items

Type: integer

contents Required

root tables TableStructureRecognizerSchema cells TableCellSchema contents

Contents

Text content of the cell

Any of

• Option 1
• Option 2

root tables TableStructureRecognizerSchema cells TableCellSchema contents anyOf item 0

Type: string

root tables TableStructureRecognizerSchema cells TableCellSchema contents anyOf item 1

Type: null

order Required

root tables TableStructureRecognizerSchema order

Order

Type: integer

Order of the table in the document

words Required

root words

Words

Type: array

List of recognized words

No Additional Items

Each item of this array must be:

root words WordPrediction

WordPrediction

Type: object
No Additional Properties

points Required

root words WordPrediction points

Points

Type: array of array

Bounding box of the word in the format [[x1, y1], [x2, y2], [x3, y3], [x4, y4]]

Must contain a minimum of 4 items

Must contain a maximum of 4 items

No Additional Items

Each item of this array must be:

root words WordPrediction points points items

Type: array of integer

Must contain a minimum of 2 items

Must contain a maximum of 2 items

No Additional Items

Each item of this array must be:

root words WordPrediction points points items points items items

Type: integer

content Required

root words WordPrediction content

Content

Type: string

Text content of the word

direction Required

root words WordPrediction direction

Direction

Type: string

Text direction, e.g., 'horizontal' or 'vertical'

rec_score Required

root words WordPrediction rec_score

Rec Score

Type: number

Confidence score of the word recognition

det_score Required

root words WordPrediction det_score

Det Score

Type: number

Confidence score of the word detection

figures Required

root figures

Figures

Type: array

List of detected figures

No Additional Items

Each item of this array must be:

root figures FigureSchema

FigureSchema

Type: object
No Additional Properties

box Required

root figures FigureSchema box

Box

Type: array of integer

Bounding box of the figure in the format [x1, y1, x2, y2]

Must contain a minimum of 4 items

Must contain a maximum of 4 items

No Additional Items

Each item of this array must be:

root figures FigureSchema box box items

Type: integer

order Required

root figures FigureSchema order

Order

Order of the figure in the document

Any of

• Option 1
• Option 2

root figures FigureSchema order anyOf item 0

Type: integer

root figures FigureSchema order anyOf item 1

Type: null

paragraphs Required

root figures FigureSchema paragraphs

Paragraphs

Type: array

List of paragraphs associated with the figure

No Additional Items

Each item of this array must be:

root figures FigureSchema paragraphs ParagraphSchema

ParagraphSchema

Type: object
Same definition as ParagraphSchema

direction Required

root figures FigureSchema direction

Direction

Text direction, e.g., ['horizontal' or 'vertical']

Any of

• Option 1
• Option 2

root figures FigureSchema direction anyOf item 0

Type: string

root figures FigureSchema direction anyOf item 1

Type: null