MCPcopy
hub / github.com/sml2h3/ddddocr / classification

Method classification

ddddocr/compat/v1.py:95–127  ·  view source on GitHub ↗

OCR识别方法 Args: img: 图片数据(bytes、str、pathlib.PurePath或PIL.Image) png_fix: 是否修复PNG透明背景问题 probability: 是否返回概率信息 color_filter_colors: 颜色过滤预设颜色列表,如 ['red', 'blue'] color_filter_custom_ranges: 自定义HSV颜色范围列表,如 [((0,50,50), (

(self, img: Union[bytes, str, pathlib.PurePath, Image.Image], 
                      png_fix: bool = False, probability: bool = False,
                      color_filter_colors: Optional[List[str]] = None,
                      color_filter_custom_ranges: Optional[List[Tuple[Tuple[int, int, int], Tuple[int, int, int]]]] = None)

Source from the content-addressed store, hash-verified

93 self.slide_engine = SlideEngine()
94
95 def classification(self, img: Union[bytes, str, pathlib.PurePath, Image.Image],
96 png_fix: bool = False, probability: bool = False,
97 color_filter_colors: Optional[List[str]] = None,
98 color_filter_custom_ranges: Optional[List[Tuple[Tuple[int, int, int], Tuple[int, int, int]]]] = None) -> Union[str, Dict[str, Any]]:
99 """
100 OCR识别方法
101
102 Args:
103 img: 图片数据(bytes、str、pathlib.PurePath或PIL.Image)
104 png_fix: 是否修复PNG透明背景问题
105 probability: 是否返回概率信息
106 color_filter_colors: 颜色过滤预设颜色列表,如 ['red', 'blue']
107 color_filter_custom_ranges: 自定义HSV颜色范围列表,如 [((0,50,50), (10,255,255))]
108
109 Returns:
110 识别结果文本或包含概率信息的字典
111
112 Raises:
113 DDDDOCRError: 当功能未启用或识别失败时
114 """
115 if self.det:
116 raise DDDDOCRError("当前识别类型为目标检测")
117
118 if not self.ocr_engine:
119 raise DDDDOCRError("OCR功能未初始化")
120
121 return self.ocr_engine.predict(
122 image=img,
123 png_fix=png_fix,
124 probability=probability,
125 color_filter_colors=color_filter_colors,
126 color_filter_custom_ranges=color_filter_custom_ranges
127 )
128
129 def detection(self, img: Union[bytes, str, pathlib.PurePath, Image.Image]) -> List[List[int]]:
130 """

Callers 5

mainFunction · 0.95
call_toolMethod · 0.80
ocr_recognitionFunction · 0.80
ocr_recognitionFunction · 0.80
ocr_recognition_fileFunction · 0.80

Calls 2

DDDDOCRErrorClass · 0.85
predictMethod · 0.45

Tested by

no test coverage detected