From eab0f0e0d4e309020606a281c8924f80b9150398 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=86=B0=E6=9C=94?= <565183519@qq.com>
Date: Wed, 24 Jun 2026 13:05:43 +0800
Subject: [PATCH] =?UTF-8?q?D144=20=C2=B7=20=E4=BF=AE=E5=A4=8D=20reference-?=
 =?UTF-8?q?subtitle-analyzer=20=E5=9D=90=E6=A0=87=E7=B3=BBbug?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 轮廓合并逻辑错误：max(all_x) + max(all_w) 改为 max(all_x[i]+all_w[i])
- 坐标系混用：y提前加offset但max_y没加，导致h=-1597
- 修复：先在bottom_region坐标系算完，最后加一次offset
- 验证：测试图检测y=1750/h=55px/描边2px全部正确

冰朔 TCS-0002∞ 见证
⊢ 铸渊 ICE-GL-ZY001 · D144 · 2026-06-24
---
 .../reference-subtitle-analyzer.py            | 25 +++++++++++++------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/video-ai-system/engines/subtitle-pipeline/reference-analysis/reference-subtitle-analyzer.py b/video-ai-system/engines/subtitle-pipeline/reference-analysis/reference-subtitle-analyzer.py
index 68052ec..721cea8 100644
--- a/video-ai-system/engines/subtitle-pipeline/reference-analysis/reference-subtitle-analyzer.py
+++ b/video-ai-system/engines/subtitle-pipeline/reference-analysis/reference-subtitle-analyzer.py
@@ -111,6 +111,7 @@ def analyze_subtitle_region(image_path: str, interactive: bool = False) -> dict:
             h = int(height * 0.12)
         else:
             # 合并所有轮廓的边界框
+            # 正确算式：找到所有轮廓的最小外接矩形
             all_x = []
             all_y = []
             all_w = []
@@ -123,20 +124,28 @@ def analyze_subtitle_region(image_path: str, interactive: bool = False) -> dict:
                 all_w.append(cw)
                 all_h.append(ch)
             
-            # 计算合并后的边界框
-            x = min(all_x)
-            y = min(all_y) + int(height * 0.85)  # 加上裁剪偏移
-            w = max(all_x) + max(all_w) - x
-            h = max(all_y) + max(all_h) - y
-            
+            # 正确合并：先在 bottom_region 坐标系里算完，最后再加偏移
+            x_bottom = min(all_x)
+            y_bottom = min(all_y)
+            max_x_bottom = max([all_x[i] + all_w[i] for i in range(len(all_x))])
+            max_y_bottom = max([all_y[i] + all_h[i] for i in range(len(all_y))])
+            w = max_x_bottom - x_bottom
+            h = max_y_bottom - y_bottom
+
             # 扩展边界框（包含描边）
             padding = 10
-            x = max(0, x - padding)
-            y = max(0, y - padding)
+            x = max(0, x_bottom - padding)
+            y = max(0, y_bottom - padding) + int(height * 0.85)  # 偏移只加一次，在最后
             w = min(width - x, w + 2 * padding)
             h = min(height - y, h + 2 * padding)
     
     # 计算字幕参数
+    # 先做边界校验，防止裁剪出空数组
+    x = max(0, min(x, width - 1))
+    y = max(0, min(y, height - 1))
+    w = max(1, min(w, width - x))
+    h = max(1, min(h, height - y))
+    
     subtitle_box = {"x": x, "y": y, "width": w, "height": h}
     subtitle_height_px = h
     bottom_distance_px = height - (y + h)