From 1a5608d0f80c3625edc44535dde66bcc580a9842 Mon Sep 17 00:00:00 2001
From: Stephen Hu <stephenhu@seismic.com>
Date: Mon, 28 Apr 2025 13:35:34 +0800
Subject: [PATCH] Fix: Add title_tks for Pictures (#7365)

### What problem does this PR solve?
https://github.com/infiniflow/ragflow/issues/7362

append title_tks
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---
 rag/app/picture.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/rag/app/picture.py b/rag/app/picture.py
index 97a954c9c..f5fb3ae72 100644
--- a/rag/app/picture.py
+++ b/rag/app/picture.py
@@ -15,6 +15,7 @@
 #
 
 import io
+import re
 
 import numpy as np
 from PIL import Image
@@ -24,6 +25,8 @@ from api.db.services.llm_service import LLMBundle
 from deepdoc.vision import OCR
 from rag.nlp import tokenize
 from rag.utils import clean_markdown_block
+from rag.nlp import rag_tokenizer
+
 
 ocr = OCR()
 
@@ -32,6 +35,7 @@ def chunk(filename, binary, tenant_id, lang, callback=None, **kwargs):
     img = Image.open(io.BytesIO(binary)).convert('RGB')
     doc = {
         "docnm_kwd": filename,
+        "title_tks": rag_tokenizer.tokenize(re.sub(r"\.[a-zA-Z]+$", "", filename)),
         "image": img
     }
     bxs = ocr(np.array(img))