训练Unet模型时访问掩码触发ValueError错误求助

问题场景 #

参考DigitalSreeni的YouTube第208期多类别语义分割Unet教程复现代码，仅替换自有同尺寸图像后，在Spyder中运行时执行n, h, w = train_masks.shape触发错误：

ValueError: not enough values to unpack (expected 3, got 1)

相关代码 #

from simple_multi_unet_model import multi_unet_model #Uses softmax 

from keras.utils import normalize
import os
import glob
import cv2
import numpy as np
from matplotlib import pyplot as plt


#Resizing images, if needed
SIZE_X = 128 
SIZE_Y = 128
n_classes=5 #Number of classes for segmentation

#Capture training image info as a list
train_images = []

for directory_path in glob.glob("/Hydro/128bit/images/"):
    for img_path in glob.glob(os.path.join(directory_path, "*.tif")):
        img = cv2.imread(img_path, 0)       
        #img = cv2.resize(img, (SIZE_Y, SIZE_X))
        train_images.append(img)
       
#Convert list to array for machine learning processing        
train_images = np.array(train_images)

#Capture mask/label info as a list
train_masks = [] 
for directory_path in glob.glob("/Hydro/128bit/masks/"):
    for mask_path in glob.glob(os.path.join(directory_path, "*.tif")):
        mask = cv2.imread(mask_path, 0)       
        #mask = cv2.resize(mask, (SIZE_Y, SIZE_X), interpolation = cv2.INTER_NEAREST)  #Otherwise ground truth changes due to interpolation
        train_masks.append(mask)
        
#Convert list to array for machine learning processing           
train_masks = np.array(train_masks)

###############################################
#Encode labels... but multi dim array so need to flatten, encode and reshape
from sklearn.preprocessing import LabelEncoder
labelencoder = LabelEncoder()
n, h, w = train_masks.shape
train_masks_reshaped = train_masks.reshape(-1,1)
train_masks_reshaped_encoded = labelencoder.fit_transform(train_masks_reshaped)
train_masks_encoded_original_shape = train_masks_reshaped_encoded.reshape(n, h, w)

错误信息 #

File "C:\Users\anish\208_multiclass_Unet_sandstone.py", line 63, in <module>
n, h, w = train_masks.shape
ValueError: not enough values to unpack (expected 3, got 1)

问题原因与解决方法 #

核心原因 #

train_masks的维度为1维，而非预期的3维（样本数、高度、宽度），本质是未成功加载到掩码图像，导致数组维度异常。

具体排查方向 #

1. 路径匹配失败

   • 若使用Windows系统，路径格式错误：原路径/Hydro/128bit/masks/是Linux风格，Windows需改为C:/Hydro/128bit/masks/或\\Hydro\\128bit\\masks\\，同时确认目录真实存在。
   • 检查glob.glob("/Hydro/128bit/masks/")是否返回空列表，若为空则说明路径未匹配到目标目录。

2. 文件匹配规则问题

   • 掩码文件后缀可能是大写.TIF，原规则*.tif无法匹配，可修改为兼容大小写的匹配：glob.glob(os.path.join(directory_path, "*.[tT][iI][fF]"))。
   • 确认掩码目录下确实存在.tif格式文件，无拼写错误。

3. 图像加载失败

   • 部分掩码文件损坏或无读取权限，cv2.imread返回None，导致列表中混入无效值，转数组后维度异常。可在循环中添加验证：

     mask = cv2.imread(mask_path, 0)
     if mask is not None:
         train_masks.append(mask)
     else:
         print(f"无法加载文件: {mask_path}")
     

验证步骤 #

在train_masks = np.array(train_masks)后添加打印语句，确认数据加载状态：

print("掩码列表长度:", len(train_masks))
print("掩码数组形状:", train_masks.shape)

若长度为0，优先排查路径和文件匹配问题；若形状不是(n, 128, 128)，则检查图像加载和尺寸是否符合预期。

内容的提问来源于stack exchange，提问作者Anish Nayak