revoke: revoke persepctive correction code and fix yellow car plate issue

mv-and-ip/.gitignore

@@ -4,6 +4,7 @@
 
 # All image files
 *.jpg
+*.jpeg
 *.png
 *.webp
 

mv-and-ip/car_plate.py

@@ -49,21 +49,29 @@ def _uniform_car_plate(img: cv.typing.MatLike) -> cv.typing.MatLike:
 class CarPlateHsvBoundary:
     lower_bound: cv.typing.MatLike
     upper_bound: cv.typing.MatLike
+    need_revert: bool
+    """是否取反黑白颜色，因为蓝牌和黄牌的操作正好是反的"""
 
 
 CAR_PLATE_HSV_BOUNDARIES: tuple[CarPlateHsvBoundary, ...] = (
     # 中国蓝牌 HSV 范围
-    CarPlateHsvBoundary(np.array([100, 80, 60]), np.array([130, 255, 255])),
+    CarPlateHsvBoundary(np.array([100, 80, 60]), np.array([130, 255, 255]), True),
     # 中国绿牌 HSV 范围
-    CarPlateHsvBoundary(np.array([35, 43, 46]), np.array([99, 255, 255])),
+    CarPlateHsvBoundary(np.array([35, 43, 46]), np.array([99, 255, 255]), False),
     # 中国黄牌 HSV 范围
-    CarPlateHsvBoundary(np.array([32, 43, 46]), np.array([68, 255, 255])),
+    CarPlateHsvBoundary(np.array([16, 43, 46]), np.array([34, 255, 255]), False),
 )
 
 
+@dataclass
+class CarPlateMask:
+    mask: cv.typing.MatLike
+    need_revert: bool
+    """是否对颜色取反，与CarPlateHsvBoundary中的同名字段含义一致"""
+
 def _batchly_mask_car_plate(
     hsv: cv.typing.MatLike,
-) -> typing.Iterator[cv.typing.MatLike]:
+) -> typing.Iterator[CarPlateMask]:
     """ """
     for boundary in CAR_PLATE_HSV_BOUNDARIES:
         # 以给定HSV范围检测符合该颜色的位置
@@ -76,7 +84,7 @@ def _batchly_mask_car_plate(
         mask = cv.morphologyEx(mask, cv.MORPH_OPEN, kernel_open)
 
         # Return value
-        yield mask
+        yield CarPlateMask(mask, boundary.need_revert)
 
 
 @dataclass
@@ -85,6 +93,8 @@ class CarPlateRegion:
     y: int
     w: int
     h: int
+    need_revert: bool
+    """是否对颜色取反，与CarPlateHsvBoundary中的同名字段含义一致"""
 
 
 MIN_AREA: float = 3000
@@ -94,10 +104,10 @@ BEST_ASPECT_RATIO: float = 3.5
 
 
 def _analyse_car_plate_connection(
-    mask: cv.typing.MatLike,
+    mask: CarPlateMask,
 ) -> typing.Optional[CarPlateRegion]:
     # 连通域分析，筛选最符合车牌长宽比的区域
-    num_labels, labels, stats, _ = cv.connectedComponentsWithStats(mask, connectivity=8)
+    num_labels, labels, stats, _ = cv.connectedComponentsWithStats(mask.mask, connectivity=8)
 
     best: typing.Optional[CarPlateRegion] = None
     best_score = 0
@@ -113,7 +123,7 @@ def _analyse_car_plate_connection(
             score = area * (1 - abs(ratio - BEST_ASPECT_RATIO) / BEST_ASPECT_RATIO)
             if score > best_score:
                 best_score = score
-                best = CarPlateRegion(x, y, w, h)
+                best = CarPlateRegion(x, y, w, h, mask.need_revert)
 
     return best
 
@@ -136,10 +146,11 @@ def extract_car_plate(img: cv.typing.MatLike) -> typing.Optional[cv.typing.MatLi
         # 连通域分析，筛选最符合车牌长宽比的区域作为车牌
         candidate = _analyse_car_plate_connection(mask)
         # 找到任意一个就退出
-        if candidate is not None: break
+        if candidate is not None:
+            break
 
     if candidate is None:
-        logging.error('Can not find any car plate.')
+        logging.error("Can not find any car plate.")
         return None
 
     # 稍微扩边获取最终车牌区域
@@ -149,7 +160,7 @@ def extract_car_plate(img: cv.typing.MatLike) -> typing.Optional[cv.typing.MatLi
     y1 = max(candidate.y - pad, 0)
     x2 = min(candidate.x + candidate.w + pad, w_img)
     y2 = min(candidate.y + candidate.h + pad, h_img)
-    logging.info(f'车牌区域: x={x1}, y={y1}, w={x2 - x1}, h={y2 - y1}')
+    logging.info(f"车牌区域: x={x1}, y={y1}, w={x2 - x1}, h={y2 - y1}")
 
     # # 在原图上标记（仅供调试）
     # debug = img.copy()
@@ -165,17 +176,20 @@ def extract_car_plate(img: cv.typing.MatLike) -> typing.Optional[cv.typing.MatLi
     # Otsu 自动阈值，得到白字黑底，再取反 → 黑字白底
     _, binary_otsu = cv.threshold(blurred, 0, 255, cv.THRESH_BINARY + cv.THRESH_OTSU)
     # 反转：字符变黑，背景变白
+    if candidate.need_revert:
         binary = cv.bitwise_not(binary_otsu)
+    else:
+        binary = binary_otsu
 
     # 去除小噪点（开运算）
     kernel_denoise = cv.getStructuringElement(cv.MORPH_RECT, (2, 2))
     binary = cv.morphologyEx(binary, cv.MORPH_OPEN, kernel_denoise)
 
-    #return binary
+    return binary
     # cv.imwrite('./plate_binary.png', binary)
     # print("二值化结果已保存: plate_binary.png")
 
-    # ── 4. 叠加边框轮廓（细化文字边缘，参考效果图）─────────────────────
+    # 叠加边框轮廓（细化文字边缘，参考效果图）
     # Canny 边缘叠加让效果更接近参考图
     edges = cv.Canny(blurred, 40, 120)
     edges_inv = cv.bitwise_not(edges)  # 边缘→黑色