python读取tif文件，tif文件包括多页时的处理方式

python利用plt读取tiff图像(多个页面)_iamge.open python tif_Li_Yurun的博客-CSDN博客

学习以上blog

 首先执行seek 再读取数据

from paddleocr import PaddleOCR, draw_ocr
import paddle
import paddleocr
import time
from PIL import Image
import numpy as np
import os
# paddleocr.DEFAULT_OCR_MODEL_VERSION = 'PP-OCRv4'
if __name__ == '__main__':
    # paddle.device.set_device("gpu:1")
    # print(paddle.device.get_device())
    #os.environ['CUDA_VISIBLE_DEVICES']='1'
    ocr = PaddleOCR(use_angle_cls=True, lang="ch",ocr_version="PP-OCRv4",gpu_id=1)
    img = Image.open('00000001.tif').convert('RGB')
    img2 = Image.open('00000001.tif')
    nframesNum = img2.n_frames
    print(nframesNum)
    for i in range(nframesNum):
        print(i)
        start = time.time()
        # result = ocr.ocr(np.array(img), cls=True)
        img2.seek(i)
        result = ocr.ocr(np.array(img2.convert('RGB')))
        end = time.time()
        print('time:', (end-start))

        for line in result:
            print(line)

保存时处理方式：通过save函数中的save_all =1 保存所有的页

	img = Image.open(img_file)
	print(img.n_frames)
	img.save("test.tif",save_all=1)

多页数据中某一页清除成白色，需要通过imglist先中转数据，再保存，通过append_images 字段保存数据

from paddleocr import PaddleOCR, draw_ocr
import paddleocr
import time
import img2pdf
from PIL import Image
import numpy as np

save_pdf_file = 'output.pdf'

paddleocr.DEFAULT_OCR_MODEL_VERSION = 'PP-OCR'
if __name__ == '__main__':
    # ocr = PaddleOCR(use_angle_cls=True, lang="ch",ocr_version="PP-OCRv3",device='cuda:0')
    img = Image.open('00000001.tif') #.convert('RGB')
    print(img.n_frames)
    nFramesNum = img.n_frames
    imglist=[]
    for i in range(nFramesNum):
        start = time.time()
        # result = ocr.ocr(np.array(img), cls=True)
        img.seek(i)
        tmpimg = np.array(img.convert('RGB'))
        if i == 1:
            tmpimg[:,:,0:3]=255
        img2 = Image.fromarray(tmpimg)
        imglist.append(img2)
        # result = ocr.ocr(np.array(img.convert('RGB')))
        print(img.info)
        end = time.time()
        print('time:', (end-start))


    with open(save_pdf_file, "wb") as f:
        # convert函数 用来转PDF
        img3 = imglist.pop(0)
        img3.save("test.tif",save_all=True,append_images=imglist)
        write_content = img2pdf.convert("test.tif")
        f.write(write_content)  # 写入文件

    print("转换成功！")  # 提示语
    # for line in result:
    #     print(line)