4 |
5 |
6 |
--------------------------------------------------------------------------------
/.idea/misc.xml:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
--------------------------------------------------------------------------------
/.idea/modules.xml:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
--------------------------------------------------------------------------------
/.idea/paddleOCRPOC.iml:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
--------------------------------------------------------------------------------
/.idea/vcs.xml:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
1 | Apache License
2 | Version 2.0, January 2004
3 | http://www.apache.org/licenses/
4 |
5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6 |
7 | 1. Definitions.
8 |
9 | "License" shall mean the terms and conditions for use, reproduction,
10 | and distribution as defined by Sections 1 through 9 of this document.
11 |
12 | "Licensor" shall mean the copyright owner or entity authorized by
13 | the copyright owner that is granting the License.
14 |
15 | "Legal Entity" shall mean the union of the acting entity and all
16 | other entities that control, are controlled by, or are under common
17 | control with that entity. For the purposes of this definition,
18 | "control" means (i) the power, direct or indirect, to cause the
19 | direction or management of such entity, whether by contract or
20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the
21 | outstanding shares, or (iii) beneficial ownership of such entity.
22 |
23 | "You" (or "Your") shall mean an individual or Legal Entity
24 | exercising permissions granted by this License.
25 |
26 | "Source" form shall mean the preferred form for making modifications,
27 | including but not limited to software source code, documentation
28 | source, and configuration files.
29 |
30 | "Object" form shall mean any form resulting from mechanical
31 | transformation or translation of a Source form, including but
32 | not limited to compiled object code, generated documentation,
33 | and conversions to other media types.
34 |
35 | "Work" shall mean the work of authorship, whether in Source or
36 | Object form, made available under the License, as indicated by a
37 | copyright notice that is included in or attached to the work
38 | (an example is provided in the Appendix below).
39 |
40 | "Derivative Works" shall mean any work, whether in Source or Object
41 | form, that is based on (or derived from) the Work and for which the
42 | editorial revisions, annotations, elaborations, or other modifications
43 | represent, as a whole, an original work of authorship. For the purposes
44 | of this License, Derivative Works shall not include works that remain
45 | separable from, or merely link (or bind by name) to the interfaces of,
46 | the Work and Derivative Works thereof.
47 |
48 | "Contribution" shall mean any work of authorship, including
49 | the original version of the Work and any modifications or additions
50 | to that Work or Derivative Works thereof, that is intentionally
51 | submitted to Licensor for inclusion in the Work by the copyright owner
52 | or by an individual or Legal Entity authorized to submit on behalf of
53 | the copyright owner. For the purposes of this definition, "submitted"
54 | means any form of electronic, verbal, or written communication sent
55 | to the Licensor or its representatives, including but not limited to
56 | communication on electronic mailing lists, source code control systems,
57 | and issue tracking systems that are managed by, or on behalf of, the
58 | Licensor for the purpose of discussing and improving the Work, but
59 | excluding communication that is conspicuously marked or otherwise
60 | designated in writing by the copyright owner as "Not a Contribution."
61 |
62 | "Contributor" shall mean Licensor and any individual or Legal Entity
63 | on behalf of whom a Contribution has been received by Licensor and
64 | subsequently incorporated within the Work.
65 |
66 | 2. Grant of Copyright License. Subject to the terms and conditions of
67 | this License, each Contributor hereby grants to You a perpetual,
68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69 | copyright license to reproduce, prepare Derivative Works of,
70 | publicly display, publicly perform, sublicense, and distribute the
71 | Work and such Derivative Works in Source or Object form.
72 |
73 | 3. Grant of Patent License. Subject to the terms and conditions of
74 | this License, each Contributor hereby grants to You a perpetual,
75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76 | (except as stated in this section) patent license to make, have made,
77 | use, offer to sell, sell, import, and otherwise transfer the Work,
78 | where such license applies only to those patent claims licensable
79 | by such Contributor that are necessarily infringed by their
80 | Contribution(s) alone or by combination of their Contribution(s)
81 | with the Work to which such Contribution(s) was submitted. If You
82 | institute patent litigation against any entity (including a
83 | cross-claim or counterclaim in a lawsuit) alleging that the Work
84 | or a Contribution incorporated within the Work constitutes direct
85 | or contributory patent infringement, then any patent licenses
86 | granted to You under this License for that Work shall terminate
87 | as of the date such litigation is filed.
88 |
89 | 4. Redistribution. You may reproduce and distribute copies of the
90 | Work or Derivative Works thereof in any medium, with or without
91 | modifications, and in Source or Object form, provided that You
92 | meet the following conditions:
93 |
94 | (a) You must give any other recipients of the Work or
95 | Derivative Works a copy of this License; and
96 |
97 | (b) You must cause any modified files to carry prominent notices
98 | stating that You changed the files; and
99 |
100 | (c) You must retain, in the Source form of any Derivative Works
101 | that You distribute, all copyright, patent, trademark, and
102 | attribution notices from the Source form of the Work,
103 | excluding those notices that do not pertain to any part of
104 | the Derivative Works; and
105 |
106 | (d) If the Work includes a "NOTICE" text file as part of its
107 | distribution, then any Derivative Works that You distribute must
108 | include a readable copy of the attribution notices contained
109 | within such NOTICE file, excluding those notices that do not
110 | pertain to any part of the Derivative Works, in at least one
111 | of the following places: within a NOTICE text file distributed
112 | as part of the Derivative Works; within the Source form or
113 | documentation, if provided along with the Derivative Works; or,
114 | within a display generated by the Derivative Works, if and
115 | wherever such third-party notices normally appear. The contents
116 | of the NOTICE file are for informational purposes only and
117 | do not modify the License. You may add Your own attribution
118 | notices within Derivative Works that You distribute, alongside
119 | or as an addendum to the NOTICE text from the Work, provided
120 | that such additional attribution notices cannot be construed
121 | as modifying the License.
122 |
123 | You may add Your own copyright statement to Your modifications and
124 | may provide additional or different license terms and conditions
125 | for use, reproduction, or distribution of Your modifications, or
126 | for any such Derivative Works as a whole, provided Your use,
127 | reproduction, and distribution of the Work otherwise complies with
128 | the conditions stated in this License.
129 |
130 | 5. Submission of Contributions. Unless You explicitly state otherwise,
131 | any Contribution intentionally submitted for inclusion in the Work
132 | by You to the Licensor shall be under the terms and conditions of
133 | this License, without any additional terms or conditions.
134 | Notwithstanding the above, nothing herein shall supersede or modify
135 | the terms of any separate license agreement you may have executed
136 | with Licensor regarding such Contributions.
137 |
138 | 6. Trademarks. This License does not grant permission to use the trade
139 | names, trademarks, service marks, or product names of the Licensor,
140 | except as required for reasonable and customary use in describing the
141 | origin of the Work and reproducing the content of the NOTICE file.
142 |
143 | 7. Disclaimer of Warranty. Unless required by applicable law or
144 | agreed to in writing, Licensor provides the Work (and each
145 | Contributor provides its Contributions) on an "AS IS" BASIS,
146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147 | implied, including, without limitation, any warranties or conditions
148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149 | PARTICULAR PURPOSE. You are solely responsible for determining the
150 | appropriateness of using or redistributing the Work and assume any
151 | risks associated with Your exercise of permissions under this License.
152 |
153 | 8. Limitation of Liability. In no event and under no legal theory,
154 | whether in tort (including negligence), contract, or otherwise,
155 | unless required by applicable law (such as deliberate and grossly
156 | negligent acts) or agreed to in writing, shall any Contributor be
157 | liable to You for damages, including any direct, indirect, special,
158 | incidental, or consequential damages of any character arising as a
159 | result of this License or out of the use or inability to use the
160 | Work (including but not limited to damages for loss of goodwill,
161 | work stoppage, computer failure or malfunction, or any and all
162 | other commercial damages or losses), even if such Contributor
163 | has been advised of the possibility of such damages.
164 |
165 | 9. Accepting Warranty or Additional Liability. While redistributing
166 | the Work or Derivative Works thereof, You may choose to offer,
167 | and charge a fee for, acceptance of support, warranty, indemnity,
168 | or other liability obligations and/or rights consistent with this
169 | License. However, in accepting such obligations, You may act only
170 | on Your own behalf and on Your sole responsibility, not on behalf
171 | of any other Contributor, and only if You agree to indemnify,
172 | defend, and hold each Contributor harmless for any liability
173 | incurred by, or claims asserted against, such Contributor by reason
174 | of your accepting any such warranty or additional liability.
175 |
176 | END OF TERMS AND CONDITIONS
177 |
178 | APPENDIX: How to apply the Apache License to your work.
179 |
180 | To apply the Apache License to your work, attach the following
181 | boilerplate notice, with the fields enclosed by brackets "[]"
182 | replaced with your own identifying information. (Don't include
183 | the brackets!) The text should be enclosed in the appropriate
184 | comment syntax for the file format. We also recommend that a
185 | file or class name and description of purpose be included on the
186 | same "printed page" as the copyright notice for easier
187 | identification within third-party archives.
188 |
189 | Copyright [yyyy] [name of copyright owner]
190 |
191 | Licensed under the Apache License, Version 2.0 (the "License");
192 | you may not use this file except in compliance with the License.
193 | You may obtain a copy of the License at
194 |
195 | http://www.apache.org/licenses/LICENSE-2.0
196 |
197 | Unless required by applicable law or agreed to in writing, software
198 | distributed under the License is distributed on an "AS IS" BASIS,
199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200 | See the License for the specific language governing permissions and
201 | limitations under the License.
202 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # paddleOCR for Det and Rec
2 | Optical Character Recognition (OCR) is a powerful technology that enables machines to recognize and extract text from images or scanned documents. OCR finds applications in various fields, including document digitization, text extraction from images, and text-based data analysis. In this article, we will explore how to use PaddleOCR, an advanced OCR toolkit based on deep learning, for text detection and recognition tasks. We will walk through a code snippet that demonstrates the process step-by-step.
3 | # Prerequisites
4 | Before we dive into the code, let's ensure we have everything set up to run the PaddleOCR library. Make sure you have the following prerequisites installed on your machine:
5 | Python (3.6 or higher)
6 | PaddleOCR library
7 | Other necessary dependencies (e.g., NumPy, pandas, etc)
8 |
9 | Before running the code snippet, make sure you have the necessary libraries installed. You can find all the required packages and their versions in the requirements.txt file
10 | ```python
11 | pip install -r requirements.txt
12 | ```
13 | # Getting started
14 | In the provided ```main.py```, we have an example usage of the RecMain class for performing text recognition (OCR) on a folder of images and generating an output Excel file with evaluation metrics:
15 | ```
16 | # Example usage: Replace with your image folder path, label file, and desired output file name
17 | image_folder = "path/to/image_folder"
18 | label_file = "path/to/txt_file"
19 | output_file = "raw_output.xlsx"
20 |
21 | if __name__ == "__main__":
22 | RecMain(image_folder=image_folder, rec_file=label_file, output_file=output_file).run_rec()
23 | ```
24 |
25 | same can be initiated for detection
26 | ```
27 | #Dec
28 | # Example usage: Replace with your image folder path, label file, and desired output file name
29 | image_folder = "path/to/image_folder"
30 | label_file = "path/to/txt_file"
31 | output_file = "raw_output.xlsx"
32 |
33 | if __name__ == "__main__":
34 | DecMain(image_folder_path=image_folder, label_file_path=label_file, output_file=output_file) \
35 | .run_dec()
36 | ```
37 |
38 | # 1. Text Detection
39 | The code provided is a part of a class named DecMain, which seems to be designed for Optical Character Recognition (OCR) evaluation using ground truth data. It appears to use PaddleOCR to extract text from images and then calculates metrics like precision, recall, and Character Error Rate (CER) to evaluate the performance of the OCR system.
40 | ```python
41 | class DecMain:
42 | def __init__(self, image_folder_path, label_file_path, output_file):
43 | self.image_folder_path = image_folder_path
44 | self.label_file_path = label_file_path
45 | self.output_file = output_file
46 |
47 | def run_dec(self):
48 | # Check and update the ground truth file
49 | CheckAndUpdateGroundTruth(self.label_file_path).check_and_update_ground_truth_file()
50 |
51 | df = OcrToDf(image_folder=self.image_folder_path, label_file=self.label_file_path, det=True, rec=True, cls=False).ocr_to_df()
52 |
53 | ground_truth_data = ReadGroundTruthFile(self.label_file_path).read_ground_truth_file()
54 |
55 | # Get the extracted text as a list of dictionaries (representing the OCR results)
56 | ocr_results = df.to_dict(orient="records")
57 |
58 | # Calculate precision, recall, and CER
59 | precision, recall, total_samples = CalculateMetrics(ground_truth_data, ocr_results).calculate_precision_recall()
60 |
61 | CreateSheet(dataframe=df, precision=precision, recall=recall, total_samples=total_samples,
62 | file_name=self.output_file).create_sheet()
63 | ```
64 | # Note: Format of the DET Ground Truth Label File
65 | ```
66 | To perform OCR evaluation using the DecMain class and the provided code, it's crucial to format the ground truth label file correctly.
67 | The label file should be in JSON format and follow the structure as shown below:
68 |
69 | image_name.jpg [{"transcription": "215mm 18", "points": [[199, 6], [357, 6], [357, 33], [199, 33]], "difficult": False, "key_cls": "digits"}, {"transcription": "XZE SA", "points": [[15, 6], [140, 6], [140, 36], [15, 36]], "difficult": False, "key_cls": "text"}]
70 |
71 | The label file should be in JSON format.
72 | Each line of the file represents an image's OCR ground truth.
73 | Each line contains the filename of the image, followed by the OCR results for that image in the form of a JSON object.
74 | The JSON object should have the following keys:
75 | "transcription": The ground truth text transcription of the image.
76 | "points": A list of four points representing the bounding box coordinates of the text region in the image.
77 | "difficult": A boolean value indicating whether the text region is difficult to recognize.
78 | "key_cls": The class label of the OCR result, e.g., "digits" or "text".
79 | Make sure to follow this format while creating the ground truth label file for accurate OCR evaluation.
80 | ```
81 | # 2. Text Recognition
82 | The code provided defines a class named RecMain, which is designed to run text recognition (OCR) using a pre-trained OCR model on a folder of images and generate an evaluation Excel sheet.
83 | ```python
84 | class RecMain:
85 | def __init__(self, image_folder, rec_file, output_file):
86 | self.image_folder = image_folder
87 | self.rec_file = rec_file
88 | self.output_file = output_file
89 |
90 | def run_rec(self):
91 | image_paths = GetImagePathsFromFolder(self.image_folder, self.rec_file). \
92 | get_image_paths_from_folder()
93 |
94 | ocr_model = LoadRecModel().load_model()
95 |
96 | results = ProcessImages(ocr=ocr_model, image_paths=image_paths).process_images()
97 |
98 | ground_truth_data = ConvertTextToDict(self.rec_file).convert_txt_to_dict()
99 |
100 | model_predictions, ground_truth_texts, image_names, precision, recall, \
101 | overall_model_precision, overall_model_recall, cer_data_list = EvaluateRecModel(results,
102 | ground_truth_data).evaluate_model()
103 |
104 | # Create Excel sheet
105 | CreateMetricExcel(image_names, model_predictions, ground_truth_texts,
106 | precision, recall, cer_data_list, overall_model_precision, overall_model_recall,
107 | self.output_file).create_excel_sheet()
108 | ```
109 | # Note: Format of the Ground Truth Text File
110 | ```
111 | To perform OCR evaluation using the RecMain class and the provided code, it's essential to format the ground truth (GT) text file correctly.
112 | The GT text file should be in the following format:
113 |
114 | image_name.jpg text
115 |
116 | Each line of the file represents an image's GT text.
117 | Each line contains the filename of the image, followed by a tab character (\t), and then the GT text for that image.
118 | Ensure that the GT text file contains GT text entries for all the images present in the image folder specified in the RecMain class. The GT text should match the actual text content present in the images. This format is necessary for accurate evaluation of the OCR model's performance.
119 | ```
120 |
121 | **If you find this library useful, please consider starring this repository from the top of this page.**
122 | [](#)
123 |
124 | # Support my work
125 |
126 |
127 | # License
128 | ```
129 | Copyright [2023] [Vinod Baste]
130 |
131 | Licensed under the Apache License, Version 2.0 (the "License");
132 | you may not use this file except in compliance with the License.
133 | You may obtain a copy of the License at
134 |
135 | http://www.apache.org/licenses/LICENSE-2.0
136 |
137 | Unless required by applicable law or agreed to in writing, software
138 | distributed under the License is distributed on an "AS IS" BASIS,
139 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
140 | See the License for the specific language governing permissions and
141 | limitations under the License.
142 | ```
143 |
--------------------------------------------------------------------------------
/detection/CalculateMetrics.py:
--------------------------------------------------------------------------------
1 | class CalculateMetrics:
2 | def __init__(self, ground_truth_data_value, ocr_results_value):
3 | self.ground_truth_data_value = ground_truth_data_value
4 | self.ocr_results_value = ocr_results_value
5 |
6 | def calculate_precision_recall(self):
7 | total_image_samples = len(self.ground_truth_data_value)
8 | true_positives = 0
9 | false_positives = 0
10 | false_negatives = 0
11 |
12 | for gt_entry, ocr_entry in zip(self.ground_truth_data_value, self.ocr_results_value):
13 | gt_regions = gt_entry["Regions"]
14 | ocr_text = ocr_entry.get("Extracted_Text", "") # Replace "Extracted_Text" with the correct key name
15 |
16 | if not ocr_text: # Handle empty or None "Extracted_Text" as a special case
17 | ocr_text = ""
18 |
19 | # Combine all ground truth transcriptions into one string for comparison
20 | gt_text = " ".join(region["transcription"] for region in gt_regions)
21 |
22 | # Print the ground truth and detected text for each image
23 | print(f"Image: {gt_entry['Image_Path']}")
24 | print(f"Ground Truth: {gt_text}")
25 | print(f"Detected Text: {ocr_text}\n")
26 |
27 | # Calculate precision and recall
28 | true_positives_img = sum(1 for ocr_char, gt_char in zip(ocr_text, gt_text) if ocr_char == gt_char)
29 | false_positives_img = len(ocr_text) - true_positives_img
30 | false_negatives_img = len(gt_text) - true_positives_img
31 |
32 | # Update overall metrics
33 | true_positives += true_positives_img
34 | false_positives += false_positives_img
35 | false_negatives += false_negatives_img
36 |
37 | # Check for zero true positives to avoid division by zero
38 | if true_positives == 0:
39 | precision_value = 0.0
40 | recall_value = 0.0
41 | else:
42 | precision_value = true_positives / (true_positives + false_positives)
43 | recall_value = true_positives / (true_positives + false_negatives)
44 |
45 | return precision_value, recall_value, total_image_samples
46 |
--------------------------------------------------------------------------------
/detection/GetImagePath.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 |
4 | class GetImagePath:
5 |
6 | def __init__(self, image_name, image_folder_path):
7 | self.image_name = image_name
8 | self.image_folder_path = image_folder_path
9 |
10 | def get_image_path_by_name(self):
11 | image_path = os.path.join(self.image_folder_path, self.image_name)
12 | if os.path.exists(image_path) and os.path.isfile(image_path):
13 | return image_path
14 | else:
15 | print(f"Image with name '{self.image_name}' not found in the folder '{self.image_folder_path}'.")
16 | return None
17 |
--------------------------------------------------------------------------------
/detection/OcrToDf.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 | import pandas as pd
4 | from paddleocr import PaddleOCR
5 |
6 | from detection.GetImagePath import GetImagePath
7 |
8 |
9 | class OcrToDf:
10 | def __init__(self,
11 | image_folder,
12 | label_file,
13 | lang='en',
14 | use_gpu=False,
15 | rec_path=None,
16 | det_db_thresh=0.5,
17 | rec_thresh=0.5,
18 | image_shape=(640, 640),
19 | cls=False,
20 | det=True,
21 | rec=True
22 | ):
23 | self.image_folder_path = image_folder
24 | self.label_file_path = label_file
25 | self.lang = lang
26 | self.use_gpu = use_gpu
27 | self.rec_path = rec_path
28 | self.det_db_thresh = det_db_thresh
29 | self.rec_thresh = rec_thresh
30 | self.image_shape = image_shape
31 | self.cls = cls
32 | self.det = det,
33 | self.rec = rec
34 |
35 | def ocr_to_df(self):
36 | # Initialize PaddleOCR with the text detection_poc model
37 | ocr = PaddleOCR(
38 | lang=self.lang,
39 | use_gpu=self.use_gpu,
40 | rec_path=self.rec_path,
41 | det_db_thresh=self.det_db_thresh,
42 | rec_thresh=self.rec_thresh,
43 | image_shape=self.image_shape,
44 | )
45 |
46 | results = []
47 |
48 | with open(self.label_file_path, 'r') as f:
49 | ground_truth_labels = f.readlines()
50 |
51 | for ground_truth in ground_truth_labels:
52 |
53 | image_path = GetImagePath(image_name=ground_truth.strip().split('\t')[0],
54 | image_folder_path=self.image_folder_path).get_image_path_by_name()
55 |
56 | if not image_path or image_path is None:
57 | continue
58 |
59 | # Perform text detection_poc and recognition (OCR) on the image
60 | result = ocr.ocr(image_path, cls=self.cls, det=self.det, rec=self.rec)
61 | if result[0]:
62 | print(result[0])
63 | extracted_text = []
64 |
65 | # Extract the text from OCR results
66 | for line in result[0]:
67 | for word_info in line[1]:
68 | extracted_text.append(str(word_info))
69 | break
70 |
71 | regions = eval(ground_truth.strip().split('\t')[1])
72 |
73 | # Combine all ground truth transcriptions into one string for comparison
74 | gt_text = " ".join(region["transcription"] for region in regions)
75 |
76 | # Calculate precision and recall
77 | true_positives = sum(1 for ocr_char, gt_char in zip(extracted_text, gt_text) if ocr_char == gt_char)
78 | false_positives = len(extracted_text) - true_positives
79 | false_negatives = len(gt_text) - true_positives
80 |
81 | # Check for division by zero
82 | if (true_positives + false_positives) == 0:
83 | precision_value = 0.0
84 | else:
85 | precision_value = true_positives / (true_positives + false_positives)
86 |
87 | if (true_positives + false_negatives) == 0:
88 | recall_value = 0.0
89 | else:
90 | recall_value = true_positives / (true_positives + false_negatives)
91 |
92 | # Append the extracted text and ground truth label to the results list
93 | results.append({"Image_Path": str(os.path.basename(image_path)),
94 | "Extracted_Text": " ".join(extracted_text),
95 | "Ground_Truth": gt_text,
96 | "Precision": precision_value,
97 | "Recall": recall_value
98 | })
99 |
100 | # Create a DataFrame from the results list
101 | dataframe = pd.DataFrame(results)
102 |
103 | return dataframe
104 |
--------------------------------------------------------------------------------
/detection/ReadGroundTruthFile.py:
--------------------------------------------------------------------------------
1 | class ReadGroundTruthFile:
2 | def __init__(self, ground_truth_file):
3 | self.ground_truth_file = ground_truth_file
4 |
5 | def read_ground_truth_file(self):
6 | with open(self.ground_truth_file, 'r') as f:
7 | lines = f.readlines()
8 |
9 | ground_truth_data_value = []
10 | for line in lines:
11 | image_path, regions = line.strip().split('\t')
12 | regions = eval(regions) # Convert the string representation of the list to a Python list
13 | ground_truth_data_value.append({"Image_Path": image_path, "Regions": regions})
14 |
15 | return ground_truth_data_value
16 |
--------------------------------------------------------------------------------
/detection/__pycache__/CalculateMetrics.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/CalculateMetrics.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/__pycache__/GetImagePath.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/GetImagePath.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/__pycache__/OcrToDf.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/OcrToDf.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/__pycache__/ReadGroundTruthFile.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/ReadGroundTruthFile.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/__pycache__/check_update_ground_truth.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/check_update_ground_truth.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/__pycache__/create_sheet.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/create_sheet.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/__pycache__/decmain.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/detection/__pycache__/decmain.cpython-310.pyc
--------------------------------------------------------------------------------
/detection/check_update_ground_truth.py:
--------------------------------------------------------------------------------
1 | class CheckAndUpdateGroundTruth:
2 | def __init__(self, ground_truth_file):
3 | self.ground_truth_file = ground_truth_file
4 |
5 | def check_and_update_ground_truth_file(self):
6 | # Read the contents of the ground truth file
7 | with open(self.ground_truth_file, 'r') as f:
8 | content = f.read()
9 |
10 | # Replace all occurrences of 'false' with 'False'
11 | content = content.replace('false', 'False')
12 |
13 | # Write the modified content back to the ground truth file
14 | with open(self.ground_truth_file, 'w') as f:
15 | f.write(content)
16 |
--------------------------------------------------------------------------------
/detection/create_sheet.py:
--------------------------------------------------------------------------------
1 | import pandas as pd
2 |
3 |
4 | class CreateSheet:
5 | def __init__(self, dataframe, precision, recall, total_samples, file_name="output_file.xlsx",
6 | extracted_text_sheet="Extracted_Texts_Metrics", model_metrics_sheet="Model_Metrics"):
7 | self.precision = precision
8 | self.recall = recall
9 | self.total_samples = total_samples
10 | self.df = dataframe
11 | self.file_name = file_name
12 | self.extracted_text_sheet = extracted_text_sheet
13 | self.model_metrics_sheet = model_metrics_sheet
14 |
15 | def create_sheet(self):
16 | # Create a DataFrame for the entire model's metrics
17 | df_model_metrics = pd.DataFrame(
18 | {"Model": "en_PP-OCRv3_det", "Precision": [self.precision], "Recall": [self.recall],
19 | "Total Samples": [self.total_samples]})
20 |
21 | # Create a Pandas Excel writer object
22 | with pd.ExcelWriter(self.file_name, engine="xlsxwriter") as writer:
23 | # Save the DataFrame with extracted texts to the first sheet
24 | self.df.to_excel(writer, sheet_name=self.extracted_text_sheet, index=False)
25 | worksheet = writer.sheets[self.extracted_text_sheet]
26 | # Set the column width
27 | worksheet.set_column("A:C", 50) # Adjust the width as needed
28 | # Freeze the first row
29 | worksheet.freeze_panes(1, 0)
30 |
31 | # Save the DataFrame with model metrics to the second sheet
32 | df_model_metrics.to_excel(writer, sheet_name=self.model_metrics_sheet, index=False)
33 | worksheet = writer.sheets[self.model_metrics_sheet]
34 | # Set the column width
35 | worksheet.set_column("A:D", 20) # Adjust the width as needed
36 | # Freeze the first row
37 | worksheet.freeze_panes(1, 0)
38 |
--------------------------------------------------------------------------------
/detection/decmain.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 | from detection.CalculateMetrics import CalculateMetrics
4 | from detection.OcrToDf import OcrToDf
5 | from detection.ReadGroundTruthFile import ReadGroundTruthFile
6 | from detection.check_update_ground_truth import CheckAndUpdateGroundTruth
7 | from detection.create_sheet import CreateSheet
8 |
9 | os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"
10 |
11 |
12 | class DecMain:
13 | def __init__(self, image_folder_path, label_file_path, output_file):
14 | self.image_folder_path = image_folder_path
15 | self.label_file_path = label_file_path
16 | self.output_file = output_file
17 |
18 | def run_dec(self):
19 | # Check and update the ground truth file
20 | CheckAndUpdateGroundTruth(self.label_file_path).check_and_update_ground_truth_file()
21 |
22 | df = OcrToDf(image_folder=self.image_folder_path, label_file=self.label_file_path, det=True, rec=True, cls=False).ocr_to_df()
23 |
24 | ground_truth_data = ReadGroundTruthFile(self.label_file_path).read_ground_truth_file()
25 |
26 | # Get the extracted text as a list of dictionaries (representing the OCR results)
27 | ocr_results = df.to_dict(orient="records")
28 |
29 | # Calculate precision, recall, and CER
30 | precision, recall, total_samples = CalculateMetrics(ground_truth_data, ocr_results).calculate_precision_recall()
31 |
32 | CreateSheet(dataframe=df, precision=precision, recall=recall, total_samples=total_samples,
33 | file_name=self.output_file).create_sheet()
34 |
--------------------------------------------------------------------------------
/main.py:
--------------------------------------------------------------------------------
1 | from detection.decmain import DecMain
2 | from recognition.rec_main import RecMain
3 |
4 | """
5 | Note: Format of the REC Ground Truth Text File
6 |
7 | To perform OCR evaluation using the RecMain class and the provided code, it's essential to format the ground truth (GT) text file correctly. The GT text file should be in the following format:
8 | T25166601GTD_202301181427_1 - Copy (2)_crop_0.jpg MICHELIN
9 | Each line of the file represents an image's GT text.
10 | Each line contains the filename of the image, followed by a tab character (\t), and then the GT text for that image.
11 | Ensure that the GT text file contains GT text entries for all the images present in the image folder specified in the RecMain class. The GT text should match the actual text content present in the images. This format is necessary for accurate evaluation of the OCR model's performance.
12 | """
13 |
14 | """
15 | Note: Format of the DET Ground Truth Label File
16 |
17 | To perform OCR evaluation using the DecMain class and the provided code, it's crucial to format the ground truth label file correctly. The label file should be in JSON format and follow the structure as shown below:
18 |
19 | identity_T40829322TB_202212131924_1 - Copy (2).PNG [{"transcription": "215mm 18", "points": [[199, 6], [357, 6], [357, 33], [199, 33]], "difficult": False, "key_cls": "digits"}, {"transcription": "XZE SA", "points": [[15, 6], [140, 6], [140, 36], [15, 36]], "difficult": False, "key_cls": "text"}]
20 | The label file should be in JSON format.
21 | Each line of the file represents an image's OCR ground truth.
22 | Each line contains the filename of the image, followed by the OCR results for that image in the form of a JSON object.
23 | The JSON object should have the following keys:
24 |
25 | "transcription": The ground truth text transcription of the image.
26 | "points": A list of four points representing the bounding box coordinates of the text region in the image.
27 | "difficult": A boolean value indicating whether the text region is difficult to recognize.
28 | "key_cls": The class label of the OCR result, e.g., "digits" or "text".
29 | Make sure to follow this format while creating the ground truth label file for accurate OCR evaluation."""
30 |
31 | # Example usage: Replace with your image folder path, label file, and desired output file name
32 | image_folder = "path/to/image_folder"
33 | label_file = "path/to/txt_file"
34 | output_file = "raw_output.xlsx"
35 |
36 | # if __name__ == "__main__":
37 | # DecMain(image_folder_path=image_folder, label_file_path=label_file, output_file=output_file) \
38 | # .run_dec()
39 |
40 | if __name__ == "__main__":
41 | RecMain(image_folder=image_folder, rec_file=label_file, output_file=output_file).run_rec()
42 |
--------------------------------------------------------------------------------
/recognition/__pycache__/calculate_cer.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/calculate_cer.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/convert_txt_to_dict.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/convert_txt_to_dict.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/create_excel_sheet.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/create_excel_sheet.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/evaluate_model.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/evaluate_model.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/get_image_paths.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/get_image_paths.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/load_model.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/load_model.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/process_images.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/process_images.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/__pycache__/rec_main.cpython-310.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/vinodbaste/paddleOCR_rec_dec/bff1af448e4d9629d5eb723f13dc802c56fef394/recognition/__pycache__/rec_main.cpython-310.pyc
--------------------------------------------------------------------------------
/recognition/calculate_cer.py:
--------------------------------------------------------------------------------
1 |
2 | class CalculateCER:
3 | def __init__(self, model_texts, ground_truth_texts_data):
4 | self.model_texts = model_texts
5 | self.ground_truth_texts_data = ground_truth_texts_data
6 |
7 | def calculate_cer(self):
8 | cer_scores = []
9 |
10 | for model_text, ground_truth_text in zip(self.model_texts, self.ground_truth_texts_data):
11 | # Obtain Sentence-Level Character Error Rate (CER)
12 | cer_score = fastwer.score_sent(model_text, ground_truth_text, char_level=True)
13 | cer_scores.append(cer_score)
14 |
15 | return cer_scores
16 |
--------------------------------------------------------------------------------
/recognition/convert_txt_to_dict.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 |
4 | class ConvertTextToDict:
5 | def __init__(self, txt_file):
6 | self.txt_file = txt_file
7 |
8 | def convert_txt_to_dict(self):
9 | if not os.path.exists(self.txt_file):
10 | raise FileNotFoundError(f"The TXT file '{self.txt_file}' does not exist.")
11 |
12 | data_list = []
13 | with open(self.txt_file, 'r') as file:
14 | for line_num, line in enumerate(file, 1):
15 | line = line.strip()
16 | if not line:
17 | continue # Skip empty lines
18 | parts = line.split('\t')
19 | if len(parts) != 2:
20 | raise ValueError(f"Invalid format at line {line_num} in '{self.txt_file}'. "
21 | f"Expected 'image_name\\tground_truth_text' format.")
22 | file_name, ground_truth_text = parts
23 | data_list.append({
24 | 'file_name': file_name,
25 | 'ground_truth_text': ground_truth_text
26 | })
27 |
28 | print(data_list)
29 | return data_list
30 |
--------------------------------------------------------------------------------
/recognition/create_excel_sheet.py:
--------------------------------------------------------------------------------
1 | import pandas as pd
2 |
3 |
4 | class CreateMetricExcel:
5 | def __init__(self, image_names, model_predictions, ground_truth_texts, precision_list,
6 | recall_list, cer_list, overall_model_precision, overall_model_recall,
7 | file_name="output_file.xlsx",
8 | extracted_text_sheet="Extracted_Texts_Metrics", model_metrics_sheet="Model_Metrics"):
9 | self.image_names = image_names
10 | self.model_predictions = model_predictions
11 | self.ground_truth_texts = ground_truth_texts
12 | self.precision_list = precision_list
13 | self.recall_list = recall_list
14 | self.cer_list = cer_list
15 | self.overall_model_precision = overall_model_precision
16 | self.overall_model_recall = overall_model_recall
17 | self.file_name = file_name
18 | self.extracted_text_sheet = extracted_text_sheet
19 | self.model_metrics_sheet = model_metrics_sheet
20 |
21 | def create_excel_sheet(self):
22 | df_metrics = pd.DataFrame({"Image": str(self.image_names),
23 | "Extracted_Text": self.model_predictions,
24 | "Ground_Truth": self.ground_truth_texts,
25 | "Precision": self.precision_list,
26 | "Recall": self.recall_list,
27 | "CER": self.cer_list
28 | })
29 |
30 | df_model_metrics = pd.DataFrame(
31 | {"Model": "en_PP-OCRv3_rec", "Precision": [self.overall_model_precision],
32 | "Recall": [self.overall_model_recall], "Total Samples": [len(self.image_names)]})
33 |
34 | # Create a Pandas Excel writer object
35 | with pd.ExcelWriter(self.file_name, engine="xlsxwriter") as writer:
36 | # Save the DataFrame with extracted texts to the first sheet
37 | df_metrics.to_excel(writer, sheet_name=self.extracted_text_sheet, index=False)
38 | worksheet = writer.sheets[self.extracted_text_sheet]
39 | # Set the column width
40 | worksheet.set_column("A:F", 50) # Adjust the width as needed
41 | # Freeze the first row
42 | worksheet.freeze_panes(1, 0)
43 |
44 | # Save the DataFrame with model metrics to the second sheet
45 | df_model_metrics.to_excel(writer, sheet_name=self.model_metrics_sheet, index=False)
46 | worksheet = writer.sheets[self.model_metrics_sheet]
47 | # Set the column width
48 | worksheet.set_column("A:D", 20) # Adjust the width as needed
49 | # Freeze the first row
50 | worksheet.freeze_panes(1, 0)
51 |
--------------------------------------------------------------------------------
/recognition/evaluate_model.py:
--------------------------------------------------------------------------------
1 | from sklearn.metrics import precision_recall_fscore_support, precision_score, recall_score, accuracy_score
2 |
3 | from recognition.calculate_cer import CalculateCER
4 |
5 |
6 | class EvaluateRecModel:
7 | def __init__(self, ocr_results, rec_ground_truth_data):
8 | self.ocr_results = ocr_results
9 | self.rec_ground_truth_data = rec_ground_truth_data
10 |
11 | def evaluate_model(self):
12 | model_predictions_data = []
13 | ground_truth_texts_data = []
14 | image_names_data = []
15 | precision_list = []
16 | recall_list = []
17 |
18 | for result, annotation in zip(self.ocr_results, self.rec_ground_truth_data):
19 | model_text = [word_info[0] for line in result for word_info in line]
20 | model_text = ' '.join(model_text).strip()
21 | model_predictions_data.append(model_text)
22 | # Handle missing ground truth text
23 | ground_truth_text = annotation.get('ground_truth_text', 'NILL').replace(' ', '')
24 | ground_truth_texts_data.append(ground_truth_text)
25 | image_names_data.append(annotation.get('file_name', 'NILL')) # Handle missing image names
26 |
27 | # Calculate precision and recall for the current image
28 | precision_value, recall_value, f_score, true_sum = \
29 | precision_recall_fscore_support([ground_truth_text], [model_text], average='weighted')
30 | precision_list.append(precision_value)
31 | recall_list.append(recall_value)
32 |
33 | overall_model_precision = precision_score(ground_truth_texts_data, model_predictions_data, average='weighted')
34 | overall_model_recall = recall_score(ground_truth_texts_data, model_predictions_data, average='weighted')
35 |
36 | cer_data_list = CalculateCER(model_predictions_data, ground_truth_texts_data).calculate_cer()
37 |
38 | return model_predictions_data, ground_truth_texts_data, \
39 | image_names_data, precision_list, recall_list, overall_model_precision, overall_model_recall, cer_data_list
40 |
--------------------------------------------------------------------------------
/recognition/get_image_paths.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 | from detection.GetImagePath import GetImagePath
4 |
5 |
6 | class GetImagePathsFromFolder:
7 | def __init__(self, image_folder, txt_file):
8 | self.image_folder = image_folder
9 | self.txt_file = txt_file
10 |
11 | def get_image_paths_from_folder(self):
12 | if not os.path.exists(self.image_folder):
13 | raise FileNotFoundError(f"The image folder '{self.image_folder}' does not exist.")
14 |
15 | image_paths = []
16 |
17 | with open(self.txt_file, 'r') as file:
18 | for line_num, line in enumerate(file, 1):
19 | line = line.strip()
20 | if not line:
21 | continue # Skip empty lines
22 |
23 | image_path = GetImagePath(image_name=line.split('\t')[0],
24 | image_folder_path=self.image_folder).get_image_path_by_name()
25 | image_paths.append(os.path.join(self.image_folder, image_path))
26 |
27 | return image_paths
28 |
--------------------------------------------------------------------------------
/recognition/load_model.py:
--------------------------------------------------------------------------------
1 | from paddleocr import PaddleOCR
2 |
3 |
4 | class LoadRecModel:
5 | def __init__(self, use_gpu=False, rec_path=None, lang='en', rec_thresh=0.5, image_shape=(640, 640)):
6 | self.rec_thresh = rec_thresh
7 | self.image_shape = image_shape
8 | self.lang = lang
9 | self.use_gpu = use_gpu
10 | self.rec_path = rec_path
11 |
12 | def load_model(self):
13 | ocr = PaddleOCR(lang=self.lang,
14 | use_gpu=self.use_gpu,
15 | rec_path=self.rec_path,
16 | image_shape=self.image_shape,
17 | rec_thresh=self.rec_thresh)
18 |
19 | return ocr
20 |
--------------------------------------------------------------------------------
/recognition/process_images.py:
--------------------------------------------------------------------------------
1 | from paddleocr import PaddleOCR
2 |
3 |
4 | class ProcessImages:
5 |
6 | def __init__(self, ocr, image_paths, cls=False, det=False, rec=True):
7 | self.ocr = ocr
8 | self.image_paths = image_paths
9 | self.cls = cls
10 | self.det = det
11 | self.rec = rec
12 |
13 | def process_images(self):
14 | results = []
15 |
16 | for image_path in self.image_paths:
17 | result = self.ocr.ocr(image_path, cls=self.cls, det=True, rec=self.rec) # No need for the detection model
18 | results.append(result)
19 |
20 | return results
21 |
--------------------------------------------------------------------------------
/recognition/rec_main.py:
--------------------------------------------------------------------------------
1 | from recognition.convert_txt_to_dict import ConvertTextToDict
2 | from recognition.create_excel_sheet import CreateMetricExcel
3 | from recognition.evaluate_model import EvaluateRecModel
4 | from recognition.get_image_paths import GetImagePathsFromFolder
5 | from recognition.load_model import LoadRecModel
6 | from recognition.process_images import ProcessImages
7 |
8 |
9 | class RecMain:
10 | def __init__(self, image_folder, rec_file, output_file):
11 | self.image_folder = image_folder
12 | self.rec_file = rec_file
13 | self.output_file = output_file
14 |
15 | def run_rec(self):
16 | image_paths = GetImagePathsFromFolder(self.image_folder, self.rec_file). \
17 | get_image_paths_from_folder()
18 |
19 | ocr_model = LoadRecModel().load_model()
20 |
21 | results = ProcessImages(ocr=ocr_model, image_paths=image_paths).process_images()
22 |
23 | ground_truth_data = ConvertTextToDict(self.rec_file).convert_txt_to_dict()
24 |
25 | model_predictions, ground_truth_texts, image_names, precision, recall, \
26 | overall_model_precision, overall_model_recall, cer_data_list = EvaluateRecModel(results,
27 | ground_truth_data).evaluate_model()
28 |
29 | # Create Excel sheet
30 | CreateMetricExcel(image_names, model_predictions, ground_truth_texts,
31 | precision, recall, cer_data_list, overall_model_precision, overall_model_recall,
32 | self.output_file).create_excel_sheet()
33 |
--------------------------------------------------------------------------------
/requirements.txt:
--------------------------------------------------------------------------------
1 | paddleocr==2.6.1.3
2 | paddlepaddle-gpu==2.5.0
3 | paddlepaddle==2.5.0
4 | pandas==2.0.3
5 | xlsxwriter==3.1.1
6 | scikit-learn==1.2.2
7 | fastwer
8 |
--------------------------------------------------------------------------------