You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
42 lines
1.1 KiB
42 lines
1.1 KiB
import sys
|
|
import os
|
|
sys.path.append('/root/.agents/skills/ocr-document-processor/scripts')
|
|
|
|
from ocr_processor import OCRProcessor
|
|
|
|
# 测试图片路径 - 请替换为实际图片路径
|
|
image_path = "/path/to/your/image.png"
|
|
|
|
if not os.path.exists(image_path):
|
|
print(f"错误：图片文件 {image_path} 不存在")
|
|
print("请将图片上传到服务器，然后修改脚本中的 image_path 为实际路径")
|
|
sys.exit(1)
|
|
|
|
try:
|
|
# 创建处理器，指定中文
|
|
processor = OCRProcessor(image_path, lang='chi_sim')
|
|
|
|
# 启用预处理提高准确率
|
|
processor.preprocess(
|
|
deskew=True,
|
|
denoise=True,
|
|
threshold=True,
|
|
contrast=1.5
|
|
)
|
|
|
|
# 提取文本
|
|
text = processor.extract_text()
|
|
|
|
print("=" * 60)
|
|
print("OCR 识别结果：")
|
|
print("=" * 60)
|
|
print(text)
|
|
print("=" * 60)
|
|
|
|
# 获取结构化结果
|
|
result = processor.extract_structured()
|
|
print(f"\n识别置信度: {result['confidence']:.1f}%")
|
|
print(f"检测到的语言: {result['language']}")
|
|
|
|
except Exception as e:
|
|
print(f"识别失败: {str(e)}")
|
|
|