使用skimage库骨架化报错，求更正，谢谢。

Justheartyoung

代码已经完成了图像的二值化处理，但是要提取沙丘脊线，我们需要进行一些图像处理操作，例如形态学操作和骨架化。这里我将使用 skimage 库来进行这些操作。但是报错了，请大家帮忙解决并修改代码，谢谢。
代码：
from tifffile import imread
import numpy as np
from osgeo import gdal, osr
from skimage.morphology import opening, skeletonize, disk


# 读取TIFF格式无人机影像数据
image_path = 'F:\duneline\dune\dune.tif'
output_path = 'overlay_image1.tif'


image = imread(image_path)


# 将输入图像转换为灰度图像
gray_image = np.mean(image, axis=2)


# 自适应阈值二值化
binary = np.zeros_like(gray_image, dtype=np.uint8)
window_size = 131 # 窗口大小
k = 0.3 # 控制阈值的参数
for i in range(window_size//2, gray_image.shape[0]-window_size//2):
    for j in range(window_size//2, gray_image.shape[1]-window_size//2):
        window = gray_image[i-window_size//2:i+window_size//2+1, j-window_size//2:j+window_size//2+1]
        threshold = np.mean(window) - k * np.std(window)
        if gray_image[i, j] > threshold:
            binary[i, j] = 255


# 形态学开操作，去除噪声
selem = disk(6) # 创建一个半径为6的圆形结构元素
binary_opened = opening(binary, selem)


# 骨架化操作，提取沙丘脊线
skeleton = skeletonize(binary_opened)


# 配置输出的空间参考信息
in_ds = gdal.Open(image_path)
projection = in_ds.GetProjection()
geotransform = in_ds.GetGeoTransform()


# 创建新的输出图像
driver = gdal.GetDriverByName('GTiff')
out_ds = driver.Create(output_path, skeleton.shape[1], skeleton.shape[0], 1, gdal.GDT_Byte,
                      options=['COMPRESS=NONE'])  # 修改数据类型为gdal.GDT_Byte且关闭压缩


# 设置输出图像的空间参考信息
out_ds.SetProjection(projection)
out_ds.SetGeoTransform(geotransform)


# 将沙脊线部分写入输出图像
out_band = out_ds.GetRasterBand(1)
out_band.WriteArray(skeleton.astype(np.uint8) * 255)  # 直接写入skeleton数组


# 设置颜色表
color_table = gdal.ColorTable()
for i in range(256):  # 对所有可能的像素值进行映射
    if i == 0:
        color_table.SetColorEntry(i, (0, 0, 0, 255))  # 黑色，对应值为0
    elif i == 255:
        color_table.SetColorEntry(i, (255, 255, 255, 255))  # 白色，对应值为255
    else:
        color_table.SetColorEntry(i, (i, i, i, 255))  # 将其他的值映射为相应的灰度色
out_band.SetColorTable(color_table)


# 释放资源
out_band.FlushCache()
out_ds = None
in_ds = None


print("保存成功！")
报错：
Traceback (most recent call last):
  File "F:/duneline/duneline.py", line 264, in <module>
    skeleton = skeletonize(binary_opened)
  File "C:\安装软件\Python\lib\site-packages\skimage\morphology\_skeletonize.py", line 80, in skeletonize
    skeleton = skeletonize_2d(image)
  File "C:\安装软件\Python\lib\site-packages\skimage\morphology\_skeletonize.py", line 172, in skeletonize_2d
    raise ValueError('Image contains values other than 0 and 1')
ValueError: Image contains values other than 0 and 1

Process finished with exit code 1

Justheartyoung

Mike_python小 发表于 2023-8-6 12:28
该报错是因为骨架化（skeletonize）函数要求输入的二值图像只能包含0和1，而在你的代码中，二值化后的图像 ...

运行了你的代码，最终结果太粗了，提取的不够精细，应该怎么改