OSCR

Dual-color augmented reality waveguide display for color vision assistance using color tracking.

Code ↔ Paper

1 match between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 1 match
  1. [1] § STAR★Methods › Method details › ANN for HSV-based red color object extraction ↔ AI_HSV_color extraction.py, lines 87–98 · score 0.58 · activation function, OpenCV, sigmoid, layer, Model, ANN

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 297 lines · 12 KB · no license · 1 match

  1. import cv2
  2. import os
  3. import numpy as np
  4. import time
  5. from sklearn.model_selection import train_test_split
  6. from sklearn.metrics import accuracy_score, precision_score, recall_score
  7. import xml.etree.ElementTree as ET
  8. import matplotlib.pyplot as plt
  9. import matplotlib.image as mpimg
  12. # 단일 이미지에서 훈련 데이터 수집
  13. def collect_training_data(image_path, lower_red1, upper_red1, lower_red2, upper_red2, save_mask=False, output_dir='masks'):
  14. image = cv2.imread(image_path)
  15. if image is None:
  16. print("Error: 이미지를 로드할 수 없습니다. 파일이 유효한 이미지인지 확인하세요.")
  17. return None, None
  19. # HSV 색상 공간으로 변환
  20. hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
  22. # 마스크 생성
  23. mask1 = cv2.inRange(hsv, lower_red1, upper_red1)
  24. mask2 = cv2.inRange(hsv, lower_red2, upper_red2)
  25. mask = cv2.bitwise_or(mask1, mask2)
  27. # 마스크 저장 (디버깅용)
  28. if save_mask:
  29. if not os.path.exists(output_dir):
  30. os.makedirs(output_dir)
  31. mask_path = os.path.join(output_dir, os.path.basename(image_path).split('.')[0] + '_mask.png')
  32. cv2.imwrite(mask_path, mask)
  33. print(f"마스크가 {mask_path}에 저장되었습니다.")
  35. # 결과 이미지 생성
  36. hsv_flat=hsv.reshape(-1, 3).astype(np.float32)
  37. hsv_flat[:,0] /= 179.0 # Hue 정규화
  38. hsv_flat[:,1] /= 255.0 # Saturation 정규화
  39. hsv_flat[:,2] /= 255.0 # Value 정규화
  40. labels=(mask.reshape(-1) / 255).astype(np.float32)
  42. return hsv_flat, labels
  44. # 다중 이미지 데이터 수집
  45. def collect_training_data_from_directory(directory, lower_red1, upper_red1, lower_red2, upper_red2):
  46. hsv_list = []
  47. labels_list = []
  48. training_dir = r'C://Users//user//Desktop//CVDcoding//training_images'
  49. # 훈련 이미지 디렉토리
  51. if not os.path.exists(training_dir):
  52. print(f"Error: 디렉토리가 {training_dir}에 존재하지 않습니다.")
  53. return None, None
  55. for filename in os.listdir(directory):
  56. if filename.endswith('.jpg') or filename.endswith('.png'):
  57. image_path = os.path.join(directory, filename)
  58. hsv, labels = collect_training_data(image_path, lower_red1, upper_red1, lower_red2, upper_red2, save_mask=True)
  59. if hsv is not None:
  60. hsv_list.append(hsv)
  61. labels_list.append(labels)
  62. if hsv_list:
  63. hsv_data = np.vstack(hsv_list)
  64. labels_data = np.hstack(labels_list)
  65. return hsv_data, labels_data
  66. else:
  67. print("Warning: 처리된 이미지가 없습니다.")
  68. return None, None
  70. # 데이터 증강
  71. def augment_data(hsv_data, labels, num_augmentations=5):
  72. augmented_hsv = []
  73. augmented_labels = []
  74. for i in range(len(hsv_data)):
  75. h, s, v = hsv_data[i]
  76. label = labels[i]
  77. augmented_hsv.append([h, s, v])
  78. augmented_labels.append(label)
  79. for _ in range(num_augmentations):
  80. h_aug = (h + np.random.uniform(-0.028, 0.028)) % 1.0 # Hue ±5/179
  81. s_aug = np.clip(s * np.random.uniform(0.9, 1.1), 0, 1.0)
  82. v_aug = np.clip(v * np.random.uniform(0.8, 1.2), 0, 1.0)
  83. augmented_hsv.append([h_aug, s_aug, v_aug])
  84. augmented_labels.append(label)
  85. return np.array(augmented_hsv, dtype=np.float32), np.array(augmented_labels, dtype=np.float32)
  87. # ANN 훈련
  88. def train_ann(hsv_data, labels, model_path='mlp_model.xml'):
  89. mlp = cv2.ml.ANN_MLP_create()
  90. mlp.setLayerSizes(np.array([3, 20, 10, 1]))
  91. mlp.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
  92. mlp.setTermCriteria((cv2.TERM_CRITERIA_COUNT, 300, 0.001))
  94. mlp.train(hsv_data, cv2.ml.ROW_SAMPLE, labels.reshape(-1, 1))
  95. mlp.save(model_path)
  96. print(f"모델이 {model_path}에 저장되었습니다.")
  98. return mlp
  100. # 실시간 비디오 처리
  101. def process_video_with_ann(model_path='mlp_model.xml'):
  103. # 트랙바 창 생성
  104. cv2.namedWindow("Trackbars")
  106. capture = cv2.VideoCapture(0, cv2.CAP_DSHOW) # 카메라 인덱스에 따라 변경
  107. capture.set(cv2.CAP_PROP_FRAME_WIDTH, 640)
  108. capture.set(cv2.CAP_PROP_FRAME_HEIGHT, 480)
  110. try:
  111. mlp = cv2.ml.ANN_MLP_load(model_path)
  112. except:
  113. print(f"Error: 모델 파일 {model_path}을 로드할 수 없습니다. 먼저 모델을 훈련시키세요.")
  114. return
  116. while True:
  117. start_time = time.time()
  118. ret, frame = capture.read()
  119. if not ret:
  120. print("Error: 프레임을 읽을 수 없습니다.")
  121. break
  123. ori_frame = frame.copy()
  124. hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
  125. hsv_flat = hsv.reshape(-1, 3).astype(np.float32)
  127. hsv_flat[:,0] /= 179.0
  128. hsv_flat[:,1] /= 255.0
  129. hsv_flat[:,2] /= 255.0
  130. _, predictions = mlp.predict(hsv_flat)
  131. mask = (predictions > 0.8).reshape(frame.shape[:2]).astype(np.uint8) * 255
  133. # 적색 픽셀을 흰색으로, 나머지는 검정색으로 표시
  134. masked_frame = np.zeros_like(ori_frame)
  135. masked_frame[mask == 255] = [255, 255, 255]
  137. # 프레임당 처리 시간과 FPS 계산
  138. time_elapsed=time.time() - start_time
  139. fps=1/time_elapsed if time_elapsed > 0 else 0
  141. cv2.putText(masked_frame, f"FPS: {fps:.2f}", (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
  142. cv2.putText(masked_frame, f"Time: {time_elapsed*1000:.2f} ms", (10, 60), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
  144. cv2.imshow("Original Frame", ori_frame)
  145. cv2.imshow("Masked Frame", masked_frame)
  147. if cv2.waitKey(33) == ord('q'):
  148. break
  149. capture.release()
  150. cv2.destroyAllWindows()
  152. # 가중치 분포 시각화
  153. def plot_weights_distribution(model_path='mlp_model.xml'):
  154. try:
  155. tree = ET.parse(model_path)
  156. root = tree.getroot()
  157. weight_elements = root.findall('opencv_ml_ann_mlp/weights/_')
  159. num_layers=len(weight_elements)
  160. if num_layers == 0:
  161. print("Error: 가중치 정보가 없습니다.")
  162. return
  164. fig, axes=plt.subplots(num_layers, 1, figsize=(8 * num_layers, 6))
  165. if num_layers == 1:
  166. axes=[axes] # 단일 레이어의 경우 리스트로 변환
  168. for i, (weight_array, ax) in enumerate(zip(weight_elements, axes)):
  169. print(f"가중치 분포 figure 생성: 레이어 {i+1}")
  170. weight_str = weight_array.text.strip()
  171. weights = np.fromstring(weight_str, sep=' ')
  172. ax.hist(weights, bins=50, edgecolor='black')
  173. ax.set_title(f'Layer {i+1} Weights Distribution')
  174. ax.set_xlabel('Weight Value')
  175. ax.set_ylabel('Frequency')
  176. ax.grid(True)
  178. plt.tight_layout()
  179. plt.show(block=False)
  180. plt.pause(0.001) # 잠시 대기하여 그래프를 업데이트
  182. except Exception as e:
  183. print(f"XML 파싱 오류: {e}")
  185. # HSV 결정 경계 시각화
  186. def plot_hsv_decision_boundary_with_data(mlp, hsv_train, labels_train, V_fixed=128, H_step=1, S_step=5):
  187. V_norm = V_fixed / 255.0
  188. H_values = np.arange(0, 180, H_step)
  189. S_values = np.arange(0, 256, S_step)
  190. H_grid, S_grid = np.meshgrid(H_values, S_values)
  191. H_flat = H_grid.ravel()
  192. S_flat = S_grid.ravel()
  193. V_flat = np.full_like(H_flat, V_norm)
  194. hsv_points = np.vstack([H_flat / 179.0, S_flat / 255.0, V_flat]).T.astype(np.float32)
  195. _, predictions = mlp.predict(hsv_points)
  196. predictions = predictions.reshape(H_grid.shape)
  197. hsv_train_denorm = hsv_train.copy()
  198. hsv_train_denorm[:, 0] *= 179.0
  199. hsv_train_denorm[:, 1] *= 255.0
  200. hsv_train_denorm[:, 2] *= 255.0
  201. mask = np.abs(hsv_train_denorm[:, 2] - V_fixed) < 10
  202. hsv_subset = hsv_train_denorm[mask]
  203. labels_subset = labels_train[mask]
  204. plt.figure(figsize=(10, 8))
  205. plt.contourf(H_values, S_values, predictions, levels=50, cmap='RdBu', alpha=0.8)
  206. plt.colorbar(label='Prediction (0: Non-Red, 1: Red)')
  207. red_points = hsv_subset[labels_subset == 1]
  208. non_red_points = hsv_subset[labels_subset == 0]
  209. if len(red_points) > 0:
  210. plt.scatter(red_points[:, 0], red_points[:, 1], c='red', label='Red Pixels', alpha=0.5, s=10)
  211. if len(non_red_points) > 0:
  212. plt.scatter(non_red_points[:, 0], non_red_points[:, 1], c='blue', label='Non-Red Pixels', alpha=0.5, s=10)
  213. plt.xlabel('Hue (0-179)')
  214. plt.ylabel('Saturation (0-255)')
  215. plt.title(f'ANN Decision Boundary with Training Data at V={V_fixed}')
  216. plt.legend()
  217. plt.grid(True)
  218. plt.show(block=False)
  219. plt.pause(0.001) # 잠시 대기하여 그래프를 업데이트
  221. # 메인 실행
  222. if __name__ == "__main__":
  223. # 적색 범위 설정
  224. lower_red1 = np.array([0, 100, 100])
  225. upper_red1 = np.array([10, 255, 255])
  226. lower_red2 = np.array([170, 100, 100])
  227. upper_red2 = np.array([179, 255, 255])
  229. # 훈련 데이터 수집
  230. training_dir = r'C:/Users//user//Desktop//CVDcoding//training_images' # 훈련 이미지 디렉토리
  231. hsv_data, labels = collect_training_data_from_directory(training_dir, lower_red1, upper_red1, lower_red2, upper_red2)
  233. # 디렉토리 생성 또는 확인
  234. if not os.path.exists(training_dir):
  235. os.makedirs(training_dir)
  236. print(f"디렉토리 {training_dir}를 생성했습니다. 이미지를 추가한 후 다시 실행하세요.")
  237. exit()
  239. if hsv_data is not None:
  240. hsv_train, hsv_test, labels_train, labels_test = train_test_split(hsv_data, labels, test_size=0.2, random_state=42)
  241. # 데이터 균형 조정
  242. red_indices = np.where(labels_train == 1)[0]
  243. non_red_indices = np.where(labels_train == 0)[0]
  244. num_red = len(red_indices)
  246. if num_red == 0:
  247. print("Error: No red pixels in training data.")
  249. else:
  250. # 디버깅: 데이터 크기 확인
  251. print(f"hsv_train shape: {hsv_train.shape}, labels_train shape: {labels_train.shape}")
  252. print(f"red_indices count: {len(red_indices)}, non_red_indices count: {len(non_red_indices)}")
  254. if len(non_red_indices) > num_red:
  255. sampled_non_red_indices = np.random.choice(non_red_indices, size=num_red, replace=False)
  256. else:
  257. sampled_non_red_indices = non_red_indices
  259. selected_indices = np.concatenate([red_indices, sampled_non_red_indices])
  261. # 인덱스 유효성 검증
  262. max_index = hsv_train.shape[0] - 1
  263. if np.any(selected_indices > max_index):
  264. print(f"Error: selected_indices contains invalid indices. Max valid index: {max_index}")
  265. selected_indices = selected_indices[selected_indices <= max_index]
  267. np.random.shuffle(selected_indices)
  269. # 데이터 선택
  270. hsv_selected = hsv_train[selected_indices]
  271. labels_selected = labels_train[selected_indices]
  273. max_samples = 1000 # 시스템 성능에 따라 조정
  274. if len(hsv_selected) > max_samples:
  275. indices = np.random.choice(len(hsv_selected), max_samples, replace=False)
  276. hsv_selected = hsv_selected[indices]
  277. labels_selected = labels_selected[indices]
  279. # 데이터 증강
  280. hsv_aug, labels_aug = augment_data(hsv_selected, labels_selected, num_augmentations=5)
  281. # print("debugging_1")
  283. # ANN 훈련
  284. mlp = train_ann(hsv_aug, labels_aug)
  286. # 테스트 데이터로 평가
  287. _, predictions_test = mlp.predict(hsv_test)
  288. predictions_test = (predictions_test > 0.8).astype(int).flatten()
  289. accuracy = accuracy_score(labels_test, predictions_test)
  290. precision = precision_score(labels_test, predictions_test)
  291. recall = recall_score(labels_test, predictions_test)
  292. print(f"Test Accuracy: {accuracy:.4f}")
  293. print(f"Test Precision: {precision:.4f}")
  294. print(f"Test Recall: {recall:.4f}")
  296. # 실시간 비디오 처리
  297. process_video_with_ann()

AI_HSV_color extraction.py at commit d560e29, no license · at the source

Overview

Authors: Seong-Hyeon Cho1, Do-Hun Baek1, Woo June Choi1,2, Young-Wan Choi1,2
ORCID iDs: Young-Wan Choi
  1. Department of Intelligent Semiconductor Engineering, Chung-Ang University, Seoul 06974, Republic of Korea
  2. School of Electrical and Electronic Engineering, Chung-Ang University, Seoul 06974, Republic of Korea
Institutions: Chung-Ang University (South Korea)
Journal: iScience, volume 29, issue 7, article 116392
Dates: received 3 October 2025; accepted 28 May 2026; published online 12 June 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1016/j.isci.2026.116392 · PMID 42325546 · PMCID PMC13277613 · OpenAlex W7164543655
Open access: gold, a free copy (OpenAlex)
Status: code verified
Methods: Statistics
Keywords: Optics, Optical imaging, Biophysics
Topic: Advanced Optical Imaging Technologies (Media Technology, Engineering), according to OpenAlex
Funding: National Research Foundation; Korea Ministry of Science and ICT (RS-2023-NR076420); Chung-Ang University; Korea Institute for Advancement of Technology (P0017011)
Citations: not cited yet (Europe PMC); 74 references in the paper

Abstract

Color vision deficiency (CVD) impairs the discrimination of specific color pairs, particularly red and green, and curative treatments remain unavailable. Existing assistive approaches, such as tinted glasses, contact lenses, and screen-based recoloring methods, can improve color discrimination, but these approaches either modify the entire visual field globally or are constrained to fixed display environments. Here, we propose an augmented reality (AR) display that captures real-time real-world scenes, identifies regions of perceptual difficulty, and superimposes virtual content to assist users with CVD by integrating artificial neural network-based color tracking. The AR system incorporates a fabricated holographic optical element (HOE) integrated into compact waveguide optics, functioning as a dual-color narrowband coupler within a transparent combiner. Optical measurements confirm selective wavelength coupling, and proof-of-concept demonstrations show spatially selective overlay on the target objects. These results suggest the feasibility of HOE-based AR color vision assistance as a wearable approach for targeted color-discrimination scenarios.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 1 match between paragraphs and lines of code.

qldthe8675/Color-extraction

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: d560e29600f68754ec34abd7411a6979aafd890f, 30 April 2026
Languages: Python (1)
Size: 8 files, 1 script
Software Heritage: not archived
Found in: “Data and code availability”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Matplotlib (1 file), NumPy (1 file), OpenCV (1 file), scikit-learn (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
2 files
At the source: github.com/qldthe8675/Color-extraction

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 1 script, each with its path and the digest of its content;
  • 1 match between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data and code availability

All data reported in this study will be shared by the lead contact upon request.

All original codes used in this study are publicly available on GitHub (https://github.com/qldthe8675/Color-extraction).

Any additional information required to reanalyze the data reported in this study is available from the lead contact upon request.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

  • Authors: added Young-Wan Choi (0000-0002-9608-4549); removed Young-Wan Choi

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 4 authors, 3 keywords, 4 funders, 39 references.

Cite

This paper

Cho, S.-H., Baek, D.-H., Choi, W. J., & Choi, Y.-W. (2026). Dual-color augmented reality waveguide display for color vision assistance using color tracking. iScience, 29(7), 116392. https://doi.org/10.1016/j.isci.2026.116392

BibTeX

@article{cho2026dual,
author = {Cho, Seong-Hyeon and Baek, Do-Hun and Choi, Woo June and Choi, Young-Wan},
title = {{Dual-color augmented reality waveguide display for color vision assistance using color tracking}},
journal = {iScience},
year = {2026},
month = jun,
volume = {29},
number = {7},
pages = {116392},
publisher = {Elsevier},
issn = {2589-0042},
doi = {10.1016/j.isci.2026.116392},
url = {https://doi.org/10.1016/j.isci.2026.116392},
pmid = {42325546},
pmcid = {PMC13277613}
}

RIS

TY - JOUR
AU - Cho, Seong-Hyeon
AU - Baek, Do-Hun
AU - Choi, Woo June
AU - Choi, Young-Wan
TI - Dual-color augmented reality waveguide display for color vision assistance using color tracking
T2 - iScience
J2 - iScience
PY - 2026
DA - 2026/06/12
VL - 29
IS - 7
SP - 116392
SN - 2589-0042
PB - Elsevier
DO - 10.1016/j.isci.2026.116392
UR - https://doi.org/10.1016/j.isci.2026.116392
LA - en
ER -

CSL-JSON

{
"id": "10.1016/j.isci.2026.116392",
"type": "article-journal",
"title": "Dual-color augmented reality waveguide display for color vision assistance using color tracking",
"container-title": "iScience",
"author": [
{
"family": "Cho",
"given": "Seong-Hyeon"
},
{
"family": "Baek",
"given": "Do-Hun"
},
{
"family": "Choi",
"given": "Woo June"
},
{
"family": "Choi",
"given": "Young-Wan"
}
],
"container-title-short": "iScience",
"volume": "29",
"issue": "7",
"page": "116392",
"DOI": "10.1016/j.isci.2026.116392",
"PMID": "42325546",
"PMCID": "PMC13277613",
"ISSN": "2589-0042",
"publisher": "Elsevier",
"URL": "https://doi.org/10.1016/j.isci.2026.116392",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
12
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1007/s00429-026-03166-w [code]
Autoencoders for unsupervised analysis of rat myeloarchitecture.
Journal: Brain structure & function
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[2] doi: [code]
Real-time closed-loop feedback system for mouse mesoscale cortical signal and movement control
Journal: eLife
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[3] doi:10.7554/elife.100880 [code]
An applicable and efficient retrograde monosynaptic circuit mapping tool for larval zebrafish.
Journal: eLife
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[4] doi:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[5] doi:10.3390/diagnostics16172861 [code]
Microstructural Changes in the Corpus Callosum in Different Forms of Sporadic Age-Related Cerebral Small Vessel Disease.
Journal: Diagnostics (Basel, Switzerland)
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[6] doi:10.1002/hipo.70131 [code]
Decoding Medial Entorhinal Cortical Dynamics Produces Planning-Like Alternations in Hippocampal theta Sequences.
Journal: Hippocampus
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[7] doi:10.1038/s43856-026-01817-x [code]
Visual prompt engineering for multimodal and irregularly sampled medical data.
Journal: Communications medicine
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[8] doi:10.1016/j.isci.2026.117375 [code]
Motor priming is associated with widespread recruitment into neural ensembles and more rapid ensemble transitions.
Journal: iScience
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[9] doi:10.1038/s41598-026-61605-4 [code]
Learning precise segmentation of neurofibrillary tangles from rapid manual point annotations.
Journal: Scientific reports
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool
[10] doi:10.1126/sciadv.aed4172 [code]
Locomotion optimizes sensory representations through a computational principle shared by rodents and primates.
Journal: Science advances
In common: OpenCV, scikit-learn, Matplotlib, 1 other tool

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.