我正在尝试将 DBnet 模型与 EasyOCR 结合使用，通过使用：
reader = Reader(['ar'], gpu = False,detect_network = 'dbnet18')
EasyOCR 已下载模型，但是，在检测文本时出现以下错误：
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
Cell In[116], line 1
----> 1 easy_ocr(res,r"C:\Users\PC\OCR-App 1\data\interim\iscore\file.txt",r"C:\Users\PC\OCR-App 1\data\interim\iscore\out.jpg")
Cell In[115], line 15, in easy_ocr(image, out_txt, out_image)
12 ts = time.time()
13 #min_size=10,width_ths=0.5,add_margin=0.1,paragraph = False,text_threshold =0.15,low_text=0.2
14 #text_threshold =0.15,low_text=0.2,add_margin=0.09
---> 15 results = reader.readtext(image,paragraph = True)
16 te = time.time()
17 td = te - ts
File c:\Users\PC\anaconda3\envs\easyocr\Lib\site-packages\easyocr\easyocr.py:452, in Reader.readtext(self, image, decoder, beamWidth, batch_size, workers, allowlist, blocklist, detail, rotation_info, paragraph, min_size, contrast_ths, adjust_contrast, filter_ths, text_threshold, low_text, link_threshold, canvas_size, mag_ratio, slope_ths, ycenter_ths, height_ths, width_ths, y_ths, x_ths, add_margin, threshold, bbox_min_score, bbox_min_size, max_candidates, output_format)
446 '''
447 Parameters:
448 image: file path or numpy-array or a byte stream object
449 '''
450 img, img_cv_grey = reformat_input(image)
--> 452 horizontal_list, free_list = self.detect(img,
453 min_size = min_size, text_threshold = text_threshold,\
454 low_text = low_text, link_threshold = link_threshold,\
455 canvas_size = canvas_size, mag_ratio = mag_ratio,\
456 slope_ths = slope_ths, ycenter_ths = ycenter_ths,\
457 height_ths = height_ths, width_ths= width_ths,\
...
231 "Input type is {}, but 'deform_conv_{}.*.so' is not imported successfully.".format(device_, device_),
232 )
233 return output
RuntimeError: Input type is cpu, but 'deform_conv_cpu.*.so' is not imported successfully.
同样的问题你解决了吗？