巴基斯坦手写处方数据集
收藏资源简介:
巴基斯坦手写处方数据集由巴基斯坦GIFT大学的研究团队创建,旨在解决手写处方中药名识别的难题。该数据集包含来自巴基斯坦不同地区的50名医生的1000张手写处方图像,经过数据增强技术处理后,扩展至9920张图像。数据增强手段包括亮度调整、对比度归一化、平移、剪切、弹性变换、高斯噪声以及裁剪填充等。数据集涵盖了多种手写风格和处方格式,确保了模型的鲁棒性和泛化能力。该数据集主要用于训练和验证基于深度学习的药名提取模型,旨在提高手写处方中药名识别的准确性和效率,解决医疗领域中的实际问题。
The Pakistani Handwritten Prescription Dataset was developed by a research team from GIFT University in Pakistan to address the challenge of drug name recognition in handwritten prescriptions. This dataset initially includes 1000 handwritten prescription images from 50 doctors across different regions of Pakistan, and is expanded to 9920 images via data augmentation techniques. The data augmentation methods involve brightness adjustment, contrast normalization, translation, shearing, elastic transformation, Gaussian noise injection, as well as cropping and padding. The dataset covers diverse handwriting styles and prescription formats, ensuring the robustness and generalization ability of the models. This dataset is primarily utilized for training and validating deep learning-based drug name extraction models, aiming to improve the accuracy and efficiency of drug name recognition in handwritten prescriptions and solve practical problems in the medical field.

- 1Leveraging Deep Learning with Multi-Head Attention for Accurate Extraction of Medicine from Handwritten Prescriptions巴基斯坦GIFT大学计算机科学系 · 2024年



