Skip to content

Commit

Permalink
doc:add lambda layout analysis docs
Browse files Browse the repository at this point in the history
  • Loading branch information
@yanbasic
yanbasic committed Mar 24, 2024
1 parent 76a12ad commit a06c02b
Show file tree
Hide file tree
Showing 4 changed files with 163 additions and 28 deletions.
54 changes: 28 additions & 26 deletions docs/en/deploy-layout-analysis.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,70 +15,72 @@ include "include-deploy-description.md"

## API reference

### Text classification

- HTTP request method: `POST`

- Request body parameters

| **Name** | **Type** | **Required** | **Description** |
|----------|-----------|------------|------------|
| url | *String* |Choose url or img.| Image URL address, which supports HTTP/HTTPS and S3 protocols. Supported image formats are jpg/jpeg/png/bmp, with the longest side not exceeding 4096px.|
| img | *String* |Choose url or img.|Base64 encoded image data.|
| output_type | *String* | |`json` or `markdown`, Whether the result is returned in json or converted to markdown.|
| table_type | *String* | |`html` or `markdown`, Whether the table in the result is returned in html or converted to markdown.|
| **Name** | **Type** | **Required** | **Description** |
| ----------- | -------- | ------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------- |
| url | _String_ | Choose url or img. | Image URL address, which supports HTTP/HTTPS and S3 protocols. Supported image formats are jpg/jpeg/png/bmp, with the longest side not exceeding 4096px. |
| img | _String_ | Choose url or img. | Base64 encoded image data. |
| output_type | _String_ | | `json` or `markdown`, Whether the result is returned in json or converted to markdown. |
| table_type | _String_ | | `html` or `markdown`, Whether the table in the result is returned in html or converted to markdown. |

- Example Request

**Example 1**

``` json
```json
{
"url": "{{page.meta.sample_image}}"
"url": "{{page.meta.sample_image}}",
"output_type": "json"
}
```

``` json
```json
{
"img": "Base64-encoded image data"
"img": "Base64-encoded image data",
"output_type": "json"
}
```

**Example 2**

``` json
```json
{
"url": "{{page.meta.sample_image}}"
"url": "{{page.meta.sample_image}}",
"output_type": "markdown"
}
```

``` json
```json
{
"img": "Base64-encoded image data"
"img": "Base64-encoded image data",
"output_type": "markdown"
}
```

- Response parameters

When `output_type` is `json`, a list is returned, and an item in the list is a block in a document.

| **Name** | **Type** | **Description** |
|----------|-----------|------------|
|BlockType |*String* |These elements correspond to the different portions of the layout, and are: text, title, figure or table|
|location |*Dict* |location information about the location of detected items on a document page.|
|Text |*String* |The text content of the current block. When BlockType is `table`, the html or markdown of the table will be returned depending on the `table_type`.|
| **Name** | **Type** | **Description** |
| --------- | -------- | --------------------------------------------------------------------------------------------------------------------------------------------------- |
| BlockType | _String_ | These elements correspond to the different portions of the layout, and are: text, title, figure or table |
| Geometry | _Dict_ | location information about the location of detected items on a document page. |
| Text | _String_ | The text content of the current block. When BlockType is `table`, the html or markdown of the table will be returned depending on the `table_type`. |

When `output_type` is `markdown`, a dict is returned.

| **Name** | **Type** | **Description** |
|----------|-----------|------------|
|markdown |*String* |Converted images to markdown result.|
| **Name** | **Type** | **Description** |
| -------- | -------- | ------------------------------------ |
| markdown | _String_ | Converted images to markdown result. |

- Example JSON response

**Example 1 `output_type` is `json` response**

``` json
```json
[
{
"BlockType": "text",
Expand Down Expand Up @@ -110,7 +112,7 @@ When `output_type` is `markdown`, a dict is returned.

**Example 2 `output_type` is `markdown` response**

``` json
```json
{
"Markdown": "核准日期：xxx年xx月xx日 \n\n修改日期：xxx年xx月xx日...."
}
Expand Down
1 change: 1 addition & 0 deletions docs/en/deployment.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,7 @@ information, see [ICP Recordal](https://www.amazonaws.cn/en/support/icp/?nc1=h_l
| **LiteOCR** | no | Deploy [General OCR (Simplified Chinese)](deploy-general-ocr.md) |
| **GeneralOCRTraditional** | no | Deploy [General OCR (Traditional Chinese)](deploy-general-ocr-traditional.md) |
| **AdvancedOCR** | no | Deploy [Advanced OCR](deploy-general-ocr-traditional.md) |
| **DocumentLayoutAnalysis** | no | Deploy [Document Layout Analysis](deploy-layout-analysis.md) |
| **CustomOCR** | no | Deploy [Custom OCR](deploy-custom-ocr.md) |
| **CarLicensePlate** | no | Deploy [Car License Plate](deploy-car-license-plate.md) |
| **FaceComparison** | no | Deploy [Face Comparison](deploy-face-comparison.md) |
Expand Down
131 changes: 131 additions & 0 deletions docs/zh/deploy-layout-analysis.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,131 @@
---
feature_id: DocumentLayoutAnalysis
feature_name: 文档版面分析
feature_endpoint: layout_analysis
deployment_time: 20 Minutes
destroy_time: 15 Minutes
sample_image: 图像的URL地址
feature_description: 将文档图像转换为 Markdown/JSON 格式输出，并以 Markdown/HTML 格式生成表格。
feature_scenario: 适用于将纸质文件更改为电子格式、文件识别和内容审查，以提高信息处理效率。
---

{%
include "include-deploy-description.md"
%}

## API 参数说明

- HTTP 方法: `POST`

- Body 请求参数

| **名称** | **类型** | **是否必选** | **说明** |
| --------------------------- | -------- | ----------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| url     | _String_ | 与 img 参数二选一 | 图像的 URL 地址。支持 HTTP/HTTPS 和 S3 协议。要求图像格式为 jpg/jpeg/png/bmp ，图像大小建议不超过 1920 _ 1080，在开启人像增强的情况下，图像大小建议不超过 1280 _ 720。AWS Lambda 版本方案由于性能限制，图像大小建议不超过 400 \* 400 |
| img | _String_ | 与 url 参数二选一 | 进行 Base64 编码的图像数据 |
| output_type | _String_ | | json`或`markdown`，返回结果是 json 格式还是转换为 markdown 格式。 |
| table_type | _String_ | | html "或 "markdown"，结果中的表格是以 html 格式返回还是转换为 markdown 格式。 |

- 请求 Body 示例

**示例 1**

```json
{
"url": "{{page.meta.sample_image}}",
"output_type": "json"
}
```

```json
{
"img": "Base64-encoded image data",
"output_type": "json"
}
```

**示例 2**

```json
{
"url": "{{page.meta.sample_image}}",
"output_type": "markdown"
}
```

```json
{
"img": "Base64-encoded image data",
"output_type": "markdown"
}
```

- 返回参数

- 返回参数

| **名称** | **类型** | **说明** |
| --------- | -------- | ---------------------------------------------------------- |
| BlockType | _String_ | 这些元素与版面的不同部分相对应，它们是：文本、标题、图或表 |
| Geometry | _Dict_ | 文档页面上检测到的项目的位置信息 |
| Text | _String_ | 当前块的文本内容。当 BlockType 为 "table "时，将根据 "table_type "返回表格的 html 或 markdown。 |

当 `output_type` 为 `markdown` 时，将返回一个 dict。

| **名称** | **类型** | **说明** |
| -------- | -------- | ------------------------------------ |
| markdown | _String_ | 将图像转换为 markdown 结果 |

- 返回示例

** 示例 1 `output_type` 是 `json` 格式**

```json
[
{
"BlockType": "text",
"Geometry": {
"BoundingBox": {
"Width": 972,
"Height": 79,
"Left": 561,
"Top": 564
}
},
"Text": "核准日期：xxx年xx月xx日"
},
{
"BlockType": "text",
"Geometry": {
"BoundingBox": {
"Width": 1543,
"Height": 80,
"Left": 569,
"Top": 678
}
},
"Text": "修改日期：xxx年xx月xx日"
}
...
]
```

**示例 2 `output_type` 是 `markdown` 格式**

```json
{
"Markdown": "核准日期：xxx年xx月xx日 \n\n修改日期：xxx年xx月xx日...."
}
```

{%
include-markdown "include-deploy-code.md"
%}

{%
include "include-deploy-cost-8GB.md"
%}

{%
include-markdown "include-deploy-uninstall.md"
%}
5 changes: 3 additions & 2 deletions docs/zh/deployment.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -37,10 +37,11 @@
来部署所需要的功能。所有功能的参数默认值均为 **no**。

| 参数名称 | 默认值 | 描述 |
|-------------------|-------| -------- |
|-------------------|-------| -------- |
| **Lite OCR - Simplified Chinese** | no | 部署[轻量级文字识别（简体中文）](deploy-general-ocr.md) |
| **General OCR - Traditional Chinese** | no | 部署[通用文字识别（繁体中文）](deploy-general-ocr-traditional.md) |
| **Advanced OCR** | no | 部署[高阶文字识别](deploy-advanced-ocr.md) |
| **Advanced OCR** | no | 部署[高阶文字识别](deploy-advanced-ocr.md) |
| **DocumentLayoutAnalysis** | no | 部署 [文档版面分析](deploy-layout-analysis.md) |
| **Custom Template OCR** | no | 部署[自定义模板文字识别](deploy-custom-ocr.md) |
| **Car License Plate** | no | 部署[车牌信息识别](deploy-car-license-plate.md) |
| **Face Comparison** | no | 部署[人脸相似度比对](deploy-face-comparison.md) |
Expand Down

0 comments on commit a06c02b

Please sign in to comment.