Python3+gdal如何读取tiff格式数据

def readTif(fileName):
	
	merge_img = 0
	driver = gdal.GetDriverByName('GTiff')
	driver.Register()
 
	dataset = gdal.Open(fileName)
	if dataset == None:
		print(fileName+ "掩膜失败，文件无法打开")
		return
	im_width = dataset.RasterXSize #栅格矩阵的列数
	print('im_width:', im_width) 
 
	im_height = dataset.RasterYSize #栅格矩阵的行数
	print('im_height:', im_height) 
	im_bands = dataset.RasterCount #波段数
	im_geotrans = dataset.GetGeoTransform()#获取仿射矩阵信息
	im_proj = dataset.GetProjection()#获取投影信息
	
 
	if im_bands == 1:
		band = dataset.GetRasterBand(1)
		im_data = dataset.ReadAsArray(0,0,im_width,im_height) #获取数据
		cdata = im_data.astype(np.uint8)
		merge_img = cv2.merge([cdata,cdata,cdata])
 
		cv2.imwrite('C:/Users/summer/Desktop/a.jpg', merge_img)
# 
	elif im_bands == 4:
	# 	# im_data = dataset.ReadAsArray(0,0,im_width,im_height)#获取数据
	# 	# im_blueBand = im_data[0,0:im_width,0:im_height] #获取蓝波段
	# 	# im_greenBand = im_data[1,0:im_width,0:im_height] #获取绿波段
	# 	# im_redBand = im_data[2,0:im_width,0:im_height] #获取红波段
	# 	# # im_nirBand = im_data[3,0:im_width,0:im_height] #获取近红外波段
	# 	# merge_img=cv2.merge([im_redBand,im_greenBand,im_blueBand])
 
	# 	# zeros = np.zeros([im_height,im_width],dtype = "uint8")
 
	# 	# data1 = im_redBand.ReadAsArray
 
	# 	band1=dataset.GetRasterBand(1)
	# 	band2=dataset.GetRasterBand(2)
	# 	band3=dataset.GetRasterBand(3)
	# 	band4=dataset.GetRasterBand(4)
	
		data1=band1.ReadAsArray(0,0,im_width,im_height).astype(np.uint16) #r #获取数据
		data2=band2.ReadAsArray(0,0,im_width,im_height).astype(np.uint16) #g #获取数据
		data3=band3.ReadAsArray(0,0,im_width,im_height).astype(np.uint16) #b #获取数据
		data4=band4.ReadAsArray(0,0,im_width,im_height).astype(np.uint16) #R #获取数据
	# 	print(data1[1][45])
	# 	output1= cv2.convertScaleAbs(data1, alpha=(255.0/65535.0))
	# 	print(output1[1][45])
	# 	output2= cv2.convertScaleAbs(data2, alpha=(255.0/65535.0))
	# 	output3= cv2.convertScaleAbs(data3, alpha=(255.0/65535.0))
 
		merge_img1 = cv2.merge([output3,output2,output1]) #B G R
		
		cv2.imwrite('C:/Users/summer/Desktop/merge_img1.jpg', merge_img1)

import cv2
import numpy as np
import os
 
tiff_file = './try_img/2.tiff'
save_folder = './try_img_re/'
if not os.path.exists(save_folder):
	os.makedirs(save_folder)
 
tif_img = cv2.imread(tiff_file)
width, height, channel = tif_img.shape
# print height, width, channel : 6908 7300 3
threshold = 1000
overlap = 100
 
step = threshold - overlap
x_num = width/step + 1
y_num = height/step + 1
print x_num, y_num
 
N = 0
yj = 0 
 
for xi in range(x_num):
	for yj in range(y_num):
	# print xi
		if yj <= y_num:
			print yj
			x = step*xi
	  y = step*yj
 
	  wi = min(width,x+threshold)
	  hi = min(height,y+threshold)
	  # print wi , hi
 
	  if wi-x < 1000 and hi-y < 1000:
	  	im_block = tif_img[wi-1000:wi, hi-1000:hi]
 
	  elif wi-x > 1000 and hi-y < 1000:
	  	im_block = tif_img[x:wi, hi-1000:hi]
 
	  elif wi-x < 1000 and hi-y > 1000:
	  	im_block = tif_img[wi-1000:wi, y:hi]
 
	 	else:
	  	im_block = tif_img[x:wi,y:hi]
	  	
	  cv2.imwrite(save_folder + 'try' + str(N) + '.jpg', im_block)
	  N += 1