opensource/WeKnora
3
0
Fork 0
mirror of https://github.com/Tencent/WeKnora.git synced 2026-09-24 16:29:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update ocr_engine.py

Browse Source
This commit is contained in:
Liwx
2025-08-29 12:24:58 +08:00
committed by lyingbug
parent 11910048c0
commit 7d0037fc2d
1 changed files with 1 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
services/docreader/src/parser/ocr_engine.py
+1 -8
View File
@@ -70,12 +70,6 @@ class PaddleOCRBackend(OCRBackend):
Extracted text string
"""
try:
# # Ensure image is in RGB format
# if hasattr(image, "convert") and image.mode != "RGBA":
# img_for_ocr = image.convert("RGBA")
# logger.info(f"Converted image from {image.mode} to RGB format")
# else:
# img_for_ocr = image
logger.info(f"Initial image mode: {image.mode if hasattr(image, 'mode') else 'N/A'}")
# Ensure image is in RGB format
if hasattr(image, "convert"):
@@ -98,9 +92,8 @@ class PaddleOCRBackend(OCRBackend):
else:
image_array = img_for_ocr
logger.info(f"Input array shape for OCR: {image_array.shape}, dtype: {image_array.dtype}")
ocr_result = self.ocr.predict(image_array)
logger.info(f"OCR result type: {type(ocr_result)}")
# Extract text
if ocr_result and any(ocr_result):
ocr_text = ""