DeepRead - 生产级OCR API

DeepRead 是一个AI原生的OCR平台，可在数分钟内将文档转化为高精度数据。通过采用多模型共识机制，DeepRead实现了97%以上的准确率，并仅对不确定字段进行人机协同审核标记，从而将人工工作量从100%降低至5-10%。无需任何提示词工程。

功能说明

DeepRead 是生产级文档处理API，可在数分钟内提供高精度结构化数据输出，并通过人工审核标记功能，使人工复核仅需处理被标记的异常情况

核心特性：

• 文本提取：将PDF和图像转换为清晰标记文本
• 结构化数据：提取带置信度评分的JSON字段
• 人机协同界面：内置人机协同审核机制——不确定字段会被标记（hil_flag），确保仅需对例外情况进行人工复核
• 多轮次处理：通过多重验证流程实现最高精度
• 多模型共识：模型间交叉验证确保可靠性
• 免费套餐：每月 2,000 页（无需信用卡）

设置

1. 获取您的 API 密钥

注册并创建 API 密钥：

# Visit the dashboard
https://www.deepread.tech/dashboard

# Or use this direct link
https://www.deepread.tech/dashboard/?utm_source=clawdhub

保存您的 API 密钥：

export DEEPREAD_API_KEY="sk_live_your_key_here"

2. Clawdbot 配置（可选）

添加到您的clawdbot.config.json5：

{
  skills: {
    entries: {
      "deepread": {
        enabled: true
        // API key is read from DEEPREAD_API_KEY environment variable
        // Do NOT hardcode your API key here
      }
    }
  }
}

3. 处理您的第一个文档

选项 A：使用 Webhook（推荐）

# Upload PDF with webhook notification
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@document.pdf" \
  -F "webhook_url=https://your-app.com/webhooks/deepread"

# Returns immediately
{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "queued"
}

# Your webhook receives results when processing completes (2-5 minutes)

选项 B：轮询结果

# Upload PDF without webhook
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@document.pdf"

# Returns immediately
{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "queued"
}

# Poll until completed
curl https://api.deepread.tech/v1/jobs/550e8400-e29b-41d4-a716-446655440000 \
  -H "X-API-Key: $DEEPREAD_API_KEY"

使用示例

基础 OCR（仅文本）

提取为干净的 Markdown 文本：

# With webhook (recommended)
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf" \
  -F "webhook_url=https://your-app.com/webhook"

# OR poll for completion
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf"

# Then poll
curl https://api.deepread.tech/v1/jobs/JOB_ID \
  -H "X-API-Key: $DEEPREAD_API_KEY"

完成时的响应：

{
  "id": "550e8400-...",
  "status": "completed",
  "result": {
    "text": "# INVOICE\n\n**Vendor:** Acme Corp\n**Total:** $1,250.00..."
  }
}

结构化数据提取

提取特定字段并附带置信度评分：

curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf" \
  -F 'schema={
    "type": "object",
    "properties": {
      "vendor": {
        "type": "string",
        "description": "Vendor company name"
      },
      "total": {
        "type": "number",
        "description": "Total invoice amount"
      },
      "invoice_date": {
        "type": "string",
        "description": "Invoice date in MM/DD/YYYY format"
      }
    }
  }'

响应包含置信度标记：

{
  "status": "completed",
  "result": {
    "text": "# INVOICE\n\n**Vendor:** Acme Corp...",
    "data": {
      "vendor": {
        "value": "Acme Corp",
        "hil_flag": false,
        "found_on_page": 1
      },
      "total": {
        "value": 1250.00,
        "hil_flag": false,
        "found_on_page": 1
      },
      "invoice_date": {
        "value": "2024-10-??",
        "hil_flag": true,
        "reason": "Date partially obscured",
        "found_on_page": 1
      }
    },
    "metadata": {
      "fields_requiring_review": 1,
      "total_fields": 3,
      "review_percentage": 33.3
    }
  }
}

复杂模式（嵌套数据）

提取数组和嵌套对象：

curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf" \
  -F 'schema={
    "type": "object",
    "properties": {
      "vendor": {"type": "string"},
      "total": {"type": "number"},
      "line_items": {
        "type": "array",
        "items": {
          "type": "object",
          "properties": {
            "description": {"type": "string"},
            "quantity": {"type": "number"},
            "price": {"type": "number"}
          }
        }
      }
    }
  }'

逐页解析

获取带质量标记的每页OCR结果：

curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@contract.pdf" \
  -F "include_pages=true"

响应：

{
  "result": {
    "text": "Combined text from all pages...",
    "pages": [
      {
        "page_number": 1,
        "text": "# Contract Agreement\n\n...",
        "hil_flag": false
      },
      {
        "page_number": 2,
        "text": "Terms and C??diti??s...",
        "hil_flag": true,
        "reason": "Multiple unrecognized characters"
      }
    ],
    "metadata": {
      "pages_requiring_review": 1,
      "total_pages": 2
      }
  }
}

何时使用此功能

✅ 适用场景：

• 发票处理：提取供应商、总额、明细项目
• 收据OCR识别：解析商户、商品、总额
• 合同分析：提取签约方、日期、条款
• 表格数字化：将纸质表格转为结构化数据
• 文档工作流：任何需要OCR+数据提取的流程
• 质量关键型应用：需要识别不确定提取内容时

❌ 不适用场景：

• 实时处理：处理需2-5分钟（异步工作流）
• 月度批量>2000页：请升级至PRO或SCALE套餐

工作原理

多轮次处理流程

PDF → Convert → Rotate Correction → OCR → Multi-Model Validation → Extract → Done

该流程自动处理：

• 文档旋转与方向校正
• 多轮次验证以确保准确性
• 跨模型共识以确保可靠性
• 字段级置信度评分

人机交互（HIL）界面

DeepRead内置人机交互（HIL）审核系统。人工智能将提取的文本与原始图像进行比对，并为每个字段设置hil_flag：

• hil_flag: false= 清晰、置信度高的提取 → 自动处理
• hil_flag: true= 不确定的提取 → 转交人工审核

HIL 工作原理：

1. 高置信度提取的字段将自动核准
2. 不确定的字段会标记为hil_flag: true并附上原因
3. 仅需对标注字段进行人工审核（通常占总字段的5-10%）
4. 在DeepRead预览界面(preview.deepread.tech)中审核标注字段——这是一个专门的人机协同审核界面，审核员可并排查看原始文档与提取数据，修正标注字段并批准结果
5. 或通过API响应中的hil_flag数据集成至自有审核队列

AI在以下情况会对提取内容进行标注：

• 文本为手写体、模糊或质量低下
• 存在多种可能解读
• 字符部分可见或模糊不清
• 文档中未找到对应字段

此为多模态AI判定，非基于规则的系统

高级功能

1.蓝图（优化架构）

为特定文档类型创建可复用的优化架构：

# List your blueprints
curl https://api.deepread.tech/v1/blueprints \
  -H "X-API-Key: $DEEPREAD_API_KEY"

# Use blueprint instead of inline schema
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf" \
  -F "blueprint_id=660e8400-e29b-41d4-a716-446655440001"

优势：

• 相比基准方案，准确率提升20-30%
• 可在同类文档间复用
• 支持版本管理与回滚

如何创建蓝图：

# Create a blueprint from training data
curl -X POST https://api.deepread.tech/v1/optimize \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "utility_invoice",
    "description": "Optimized for utility invoices",
    "document_type": "invoice",
    "initial_schema": {
      "type": "object",
      "properties": {
        "vendor": {"type": "string", "description": "Vendor name"},
        "total": {"type": "number", "description": "Total amount"}
      }
    },
    "training_documents": ["doc1.pdf", "doc2.pdf", "doc3.pdf"],
    "ground_truth_data": [
      {"vendor": "Acme Power", "total": 125.50},
      {"vendor": "City Electric", "total": 89.25}
    ],
    "target_accuracy": 95.0,
    "max_iterations": 5
  }'

# Returns: {"job_id": "...", "blueprint_id": "...", "status": "pending"}

# Check optimization status
curl https://api.deepread.tech/v1/blueprints/jobs/JOB_ID \
  -H "X-API-Key: $DEEPREAD_API_KEY"

# Use blueprint (once completed)
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf" \
  -F "blueprint_id=BLUEPRINT_ID"

2. Webhooks（生产环境推荐）

处理完成时接收通知，无需轮询：

curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@invoice.pdf" \
  -F "webhook_url=https://your-app.com/webhooks/deepread"

处理完成后您的webhook将收到此数据载荷：

{
  "job_id": "550e8400-...",
  "status": "completed",
  "created_at": "2025-01-27T10:00:00Z",
  "completed_at": "2025-01-27T10:02:30Z",
  "result": {
    "text": "...",
    "data": {...}
  },
  "preview_url": "https://preview.deepread.tech/abc1234"
}

优势：

• 无需轮询
• 任务完成即时通知
• 更低延迟
• 更适合生产工作流

3. 预览功能（人机协同审核界面）

DeepRead预览平台（preview.deepread.tech）是内置的人机协同审核界面。审核人员可并排查看原始文档与提取数据，修正标记字段并批准结果。预览链接支持免认证分享：

# Request preview URL
curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@document.pdf" \
  -F "include_images=true"

# Get preview URL in response
{
  "result": {
    "text": "...",
    "data": {...}
  },
  "preview_url": "https://preview.deepread.tech/Xy9aB12"
}

公开预览端点：

# No authentication required
curl https://api.deepread.tech/v1/preview/Xy9aB12

速率限制与定价

免费层级（无需信用卡）

• 每月2,000页
• 每分钟10次请求
• 完整功能访问（OCR + 结构化提取 + 蓝图）

付费方案

• 专业版： 每月50,000页，每分钟100次请求 @ 每月99美元
• 企业版： 定制化批量定价（联系销售）

升级： https://www.deepread.tech/dashboard/billing?utm_source=clawdhub

速率限制响应头

每个响应都包含配额信息：

X-RateLimit-Limit: 2000
X-RateLimit-Remaining: 1847
X-RateLimit-Used: 153
X-RateLimit-Reset: 1730419200

最佳实践

1. 生产环境使用Webhook

✅ 推荐：Webhook通知

curl -X POST https://api.deepread.tech/v1/process \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@document.pdf" \
  -F "webhook_url=https://your-app.com/webhook"

仅在以下情况使用轮询：

• 测试/开发
• 无法暴露Webhook端点
• 需要同步响应

2. 模式设计

✅ 良好：描述性的字段说明

{
  "vendor": {
    "type": "string",
    "description": "Vendor company name. Usually in header or top-left of invoice."
  }
}

❌ 不佳：无说明

{
  "vendor": {"type": "string"}
}

3. 轮询策略（如需要）

仅在无法使用webhook时，每5-10秒轮询一次：

import time
import requests

def wait_for_result(job_id, api_key):
    while True:
        response = requests.get(
            f"https://api.deepread.tech/v1/jobs/{job_id}",
            headers={"X-API-Key": api_key}
        )
        result = response.json()

        if result["status"] == "completed":
            return result["result"]
        elif result["status"] == "failed":
            raise Exception(f"Job failed: {result.get('error')}")

        time.sleep(5)

4. 处理质量标志

将置信度高的字段与不确定的字段分开：

def process_extraction(data):
    confident = {}
    needs_review = []

    for field, field_data in data.items():
        if field_data["hil_flag"]:
            needs_review.append({
                "field": field,
                "value": field_data["value"],
                "reason": field_data.get("reason")
            })
        else:
            confident[field] = field_data["value"]

    # Auto-process confident fields
    save_to_database(confident)

    # Send uncertain fields to review queue
    if needs_review:
        send_to_review_queue(needs_review)

故障排除

错误：配额超限

{"detail": "Monthly page quota exceeded"}

解决方案：升级到PRO版本或等待下一个结算周期。

错误：无效架构

{"detail": "Schema must be valid JSON Schema"}

解决方案：确保架构是有效的JSON并且包含类型和属性。

错误：文件过大

{"detail": "File size exceeds 50MB limit"}

解决方案：压缩PDF或分割成更小的文件。

任务状态：失败

{"status": "failed", "error": "PDF could not be processed"}

常见原因：

• 已损坏的PDF文件
• 受密码保护的PDF
• 不支持的PDF版本
• 图像质量过低，无法进行OCR识别

示例模式模板

发票模式

{
  "type": "object",
  "properties": {
    "invoice_number": {
      "type": "string",
      "description": "Unique invoice ID"
    },
    "invoice_date": {
      "type": "string",
      "description": "Invoice date in MM/DD/YYYY format"
    },
    "vendor": {
      "type": "string",
      "description": "Vendor company name"
    },
    "total": {
      "type": "number",
      "description": "Total amount due including tax"
    },
    "line_items": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "description": {"type": "string"},
          "quantity": {"type": "number"},
          "price": {"type": "number"}
        }
      }
    }
  }
}

收据模式

{
  "type": "object",
  "properties": {
    "merchant": {
      "type": "string",
      "description": "Store or merchant name"
    },
    "date": {
      "type": "string",
      "description": "Transaction date"
    },
    "total": {
      "type": "number",
      "description": "Total amount paid"
    },
    "items": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": {"type": "string"},
          "price": {"type": "number"}
        }
      }
    }
  }
}

合同模式

{
  "type": "object",
  "properties": {
    "parties": {
      "type": "array",
      "items": {"type": "string"},
      "description": "Names of all parties in the contract"
    },
    "effective_date": {
      "type": "string",
      "description": "Contract start date"
    },
    "term_length": {
      "type": "string",
      "description": "Duration of contract"
    },
    "termination_clause": {
      "type": "string",
      "description": "Conditions for termination"
    }
  }
}

支持与资源

• GitHub:https://github.com/deepread-tech
• 问题反馈:https://github.com/deepread-tech/deep-read-service/issues
• 电子邮件:hello@deepread.tech

重要提示

• 处理时间: 2-5分钟（异步，非实时）
• 异步工作流程: 使用webhooks（推荐）或轮询
• 速率限制免费套餐每分钟10次请求
• 文件大小限制: 每文件50MB
• 支持格式: PDF, JPG, JPEG, PNG

准备开始了吗？获取您的免费API密钥，请访问https://www.deepread.tech/dashboard/?utm_source=clawdhub