Serendipity/cDNA-image-processing
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: 后处理min_size改为格子面积百分比，自适应不同分辨率

Browse Source 
keep_largest_object: 格子面积的1%
remove_small_objects: 格子面积的2%
This commit is contained in:
刘航宇
2026-05-08 08:54:47 +08:00
parent 00836cd302
commit efc6704b14
1 changed files with 15 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/cDNA_segmentation.py
+15 -7
View File
@@ -203,8 +203,9 @@ def gridding(gray: np.ndarray) -> tuple:
# 第四部分：后处理（参考choice.m, choosemaxobj.m） # 第四部分：后处理（参考choice.m, choosemaxobj.m）
# ============================================================ # ============================================================
def remove_small_objects(binary: np.ndarray, min_size: int = 20) -> np.ndarray: def remove_small_objects(binary: np.ndarray, cell_area: int = 1225, pct: float = 0.02) -> np.ndarray:
"""去除面积小于min_size的连通域""" """去除面积小于 cell_area*pct 的连通域（pct默认2%）"""
min_size = int(cell_area * pct)
labeled, num = ndimage.label(binary) labeled, num = ndimage.label(binary)
result = binary.copy() result = binary.copy()
for i in range(1, num + 1): for i in range(1, num + 1):
@@ -213,8 +214,9 @@ def remove_small_objects(binary: np.ndarray, min_size: int = 20) -> np.ndarray:
return result return result
def keep_largest_object(binary: np.ndarray, min_size: int = 20) -> np.ndarray: def keep_largest_object(binary: np.ndarray, cell_area: int = 1225, pct: float = 0.01) -> np.ndarray:
"""只保留最大连通域""" """只保留最大连通域，若最大块面积 < cell_area*pct 则整块抹黑（pct默认1%）"""
min_size = int(cell_area * pct)
labeled, num = ndimage.label(binary) labeled, num = ndimage.label(binary)
if num == 0: if num == 0:
return binary return binary
@@ -373,6 +375,11 @@ def main() -> None:
# ---- 步骤3: 全图逐块分割（Otsu + TV去噪） ---- # ---- 步骤3: 全图逐块分割（Otsu + TV去噪） ----
print("\n[步骤3] 全图逐块分割...") print("\n[步骤3] 全图逐块分割...")
# 计算每个格子的面积，用于自适应后处理阈值
cell_h = int(np.median(np.diff(y_grid))) if len(y_grid) > 1 else 35
cell_w = int(np.median(np.diff(x_grid))) if len(x_grid) > 1 else 35
cell_area = cell_h * cell_w
bw_full = np.zeros_like(gray) bw_full = np.zeros_like(gray)
if len(x_grid) >= 2 and len(y_grid) >= 2: if len(x_grid) >= 2 and len(y_grid) >= 2:
for i in range(len(y_grid) - 1): for i in range(len(y_grid) - 1):
@@ -396,11 +403,12 @@ def main() -> None:
bw_blk = (blk_denoised > T).astype(np.uint8) bw_blk = (blk_denoised > T).astype(np.uint8)
except ValueError: except ValueError:
bw_blk = np.zeros(blk.shape, dtype=np.uint8) bw_blk = np.zeros(blk.shape, dtype=np.uint8)
# 后处理：保留最大连通域 # 后处理：保留最大连通域（最小面积 = 格子面积的1%）
bw_blk = keep_largest_object(bw_blk, min_size=8) bw_blk = keep_largest_object(bw_blk, cell_area=cell_area)
bw_full[r1:r2, c1:c2] = bw_blk bw_full[r1:r2, c1:c2] = bw_blk
bw_full = remove_small_objects(bw_full, min_size=20) # 全局去小连通域（最小面积 = 格子面积的2%）
bw_full = remove_small_objects(bw_full, cell_area=cell_area)
fig_full = plot_full_segmentation(gray, bw_full, "全图逐块Otsu分割结果") fig_full = plot_full_segmentation(gray, bw_full, "全图逐块Otsu分割结果")
fig_full.savefig(os.path.join(OUTPUT_DIR, 'result_full_segmentation.png'), dpi=150, bbox_inches='tight') fig_full.savefig(os.path.join(OUTPUT_DIR, 'result_full_segmentation.png'), dpi=150, bbox_inches='tight')